The URL can be used to link to this page

Home My WebLink About06-16-20 Public Comments - BudgetFrom:Chris Mehl To:Agenda Subject:FW: Police defunding Date:Tuesday, June 16, 2020 12:16:09 AM Chris Mehl Mayor, City of Bozemancmehl@bozeman.net406.581.4992________________________________________From: webadmin@bozeman.net [webadmin@bozeman.net]Sent: Monday, June 15, 2020 8:16 PMTo: Chris MehlSubject: Police defunding Message submitted from the <City Of Bozeman> website. Site Visitor Name: Amy FabianSite Visitor Email: jafabian6@gmail.com I am writing to encourage you NOT to defund the police! They provide valuable service and protection to ourcommunity. There is no reason to defund them because a few officers acted with bad judgment. As a whole, thepolice are invaluable! From:Karna Ringham To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 4:17:05 AM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Karna Ringham karna.ringham@gmail.com Bozeman, Montana 59715 From:Victoria Wilson To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 7:07:25 AM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Victoria Wilson ladykinga2003@gmail.com Big Sky, Montana 59716 From:Chris Cunningham To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 7:50:20 AM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Chris Cunningham chriscunningham193@yahoo.com Bozeman, Montana 59715 From:Michal Madeline To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 8:27:37 AM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Michal Madeline michal.madeline@gmail.com Belgrade, Montana 59714 From:Web Admin To:Agenda Subject:FW: Feedback for City Of Bozeman Date:Tuesday, June 16, 2020 8:49:01 AM From: webadmin@bozeman.net <webadmin@bozeman.net> Sent: Monday, June 15, 2020 9:27 PM To: Web Admin <webadmin@BOZEMAN.NET> Subject: Feedback for City Of Bozeman You have received this feedback from Linda Hayden < haydenll67@gmail.com > for thefollowing page: https://www.bozeman.net/Home/Components/StaffDirectory/StaffDirectory/775/2052 Thankyou for serving our community. I live on the outskirts of Bozeman but not in the citylimits but since I do most of my shopping in Bozeman I wanted to comment on the budgetconsiderations. I would very much appreciate funding staying the same for police force.I feelvery safe in our city and I'm thankful for all that they do. From:Paige Anderson To:Agenda Subject:Concerns regarding proposed budget Date:Tuesday, June 16, 2020 9:23:28 AM To whom it may concern: I am concerned that the proposed municipal budget for 2021 allots far too much moneyfor policing and far too little for funds that actually strengthen our community. As you areprobably aware, there has been a tidal wave of calls to “Defund the Police” all acrossthis country in response to the murders of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors inthe aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect ourcommunities from crime. This is simply not true. The majority of people arrested everyyear in the United States are arrested on drug charges, and the majority of drug arrestsare for marijuana, a drug that is less dangerous than either alcohol or tobacco(1). Additionally, police solve a small percentage of violent crimes. The most recent datatells us that the Bozeman PD solves less than 30% of violent crimes in ourcommunity(2). What keeps communities safe from crime is not their investments inpolice, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-incomehousing, quality education, stable employment, and adequate mental-health servicesare what keep a community safe. I would very much like to see the Bozeman PD’sbudget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can moreproperly invest in other necessary services(3). Citationshttps://www.nytimes.com/2019/11/05/upshot/is-the-war-on-drugs-over-arrest-statistics-say-no.htmlhttps://www.npr.org/2015/03/30/395799413/how-many-crimes-do-your-police-clear-now-you-can-find-out https://www.bozeman.net/Home/ShowDocument?id=10356 From:Clara Jones To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 9:48:36 AM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Clara Jones claranoel.jones12@gmail.com Livingston, Montana 59047 From:Corinne Moss To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 1:00:39 PM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Corinne Moss cemoss98@gmail.com Bozeman, Montana 59715 From:Cyndy Andrus To:Agenda Subject:Fwd: Please consider Date:Tuesday, June 16, 2020 1:14:49 PM Cynthia Andrus|Deputy Mayor City of BozemanP:406.582.2383|E:candrus@bozeman.net|W: www.bozeman.net Begin forwarded message: From: "Bozeman, MT" <webadmin@bozeman.net>Date: June 16, 2020 at 1:00:19 PM MDTTo: Cyndy Andrus <CAndrus@BOZEMAN.NET>Subject: Please considerReply-To: Kaylon <kaylonmengel@gmail.com> Message submitted from the <City Of Bozeman> website. Site Visitor Name: KaylonSite Visitor Email: kaylonmengel@gmail.com Good morning, Without going deeply into a detailed response... I believe defunding the policedepartment would be a grave mistake. Please do not defund this department. I know many with the same point of view and I hope they reach out to youregarding this important matter. Thank you for giving me a moment of your time. Kaylon Mengel From:agenda@bozeman.net To:Agenda Subject:Thank you for your public comment. Date:Tuesday, June 16, 2020 1:28:57 PM A new entry to a form/survey has been submitted. Form Name:Public Comment Form Date & Time:06/16/2020 1:28 PM Response #:441 Submitter ID:27698 IP address:2600:387:1:805::c4 Time to complete:2 min. , 59 sec. Survey Details Page 1 Public comment may be submitted via the form below, or by any of the following options. Public comment may also be given at any public meeting. Email: agenda@bozeman.net Mail to: Attn: City Commission PO Box 1230 Bozeman, MT 59771 In-person delivery to: Due to City Hall's closure in response to the Covid-19 pandemic, in-person delivery of comments is not available until further notice. First Name Karen Last Name Beach Email Address kay@beachfrontmt.com Phone Number 4065865014 Comments I would just like to express how deeply I support the Bozeman Police Department and all they do to keep us safe! So much so that we FEEL safe living here because they are working so hard behind the scenes. Please do not defund or reallocate their budget to less necessary programs! Thank you! If you would like to submit additional documents (.pdf, .doc, .docx, .xls, .xlsx, .gif, .jpg, .png, .rtf, .txt) along with your comment, you may alternately address agenda@bozeman.net directly to ensure receipt of all information. Thank you, City Of Bozeman This is an automated message generated by the Vision Content Management System™. Please do not reply directly to this email. From:Julie Hitchcock To:Agenda Subject:There’s already an alternative to calling the police — High Country News – Know the West Date:Tuesday, June 16, 2020 3:49:15 PM Although I’m certain you already know about this alternative police effort in Eugene, it certainly is worth noting as you discuss the issues of the day here in Bozeman... worth emulating?? Best regards, Julie Hitchcock https://www.hcn.org/issues/52.7/public-health-theres-already-an-alternative-to-calling-the-police Sent from my iPhone From:Sofia Ferrara To:Agenda Subject:Concern about 2021 Budget Date:Tuesday, June 16, 2020 4:17:21 PM Bozeman City Commission Bozeman City Commission, To whom it may concern, I am concerned that the proposed municipal budget for 2021 allots far too much money for policing and far too little for funds that actually strengthen our community. As you are probably aware, there has been a tidal wave of calls to “Defund the Police” all across this country in response to the murder of George Floyd and Breonna Taylor at the hands of the police, and the acts of police brutality inflicted upon peaceful protestors in the aftermath. These calls are not, and should not be, confined to large urban areas. Many people claim that massive police budgets are necessary to protect our communities from crime. This is simply not true. The majority of people arrested every year in the United States are arrested on drug charges, and the majority of drug arrests are for marijuana, a drug that is less dangerous than either alcohol or tobacco. Additionally, police solve a small percentage of violent crimes. The most recent data tells us that the Bozeman PD solves less than 30% of violent crimes in our community. What keeps communities safe from crime is not their investments in police, but their investments in programs that attack the underlying causes of crime; homelessness, addiction, discrimination, and lack of opportunity. Access to low-income housing, quality education, stable employment, and adequate mental-health services are what keep a community safe. I would very much like to see the Bozeman PD’s budget, which takes up a full ⅓ of the city’s proposal, cut, so that this city can more properly invest in other necessary services. Sofia Ferrara spsofiatlt@gmail.com Bozeman, Montana 59715 From:Chris Mehl To:Agenda Subject:FW: [SENDER UNVERIFIED]End police violence and reimagine public safety Date:Tuesday, June 16, 2020 5:07:31 PM Chris Mehl Mayor, City of Bozemancmehl@bozeman.net406.581.4992________________________________________From: Sherry Schupbach [schupbach_2@msn.com]Sent: Tuesday, June 16, 2020 1:24 PMTo: Chris MehlSubject: [SENDER UNVERIFIED]End police violence and reimagine public safety Dear Mayor Mehl, As a person of faith, I believe police violence and brutality against black men and women must end now,particularly in light of the tragic and senseless recent killing of George Floyd, Breonna Taylor, and so many others. Police violence and systemic racial injustice is tied to America’s original sin of racism and the abhorrent legacy ofwhite supremacy. The time for bold reforms that ensure equal justice under the law extends to everyone is now. As your constituent and a person of faith, I am asking you to do away with policies that lead to the unequaltreatment of and violence against black, Indigenous, and people of color and adopt reforms drawn from evidence-based recommendations in the Task Force on 21st Century PolicingReport<https://cops.usdoj.gov/pdf/taskforce/taskforce_finalreport.pdf>, Campaign Zero<https://www.joincampaignzero.org/#vision>, and more. Specifically, we call on mayors, local officials, and local police precincts to: * Ban chokeholds and neck restraints in all cases * Require officers to de-escalate situations when possible, eliminating or reducing the need to use force * Require officers to give a verbal warning before shooting * Require officers to exhaust all other alternatives, including non-force and less-lethal force options prior toresorting to deadly force * Ban shooting at moving vehicles * Require officers to intervene and stop the use of excessive force by other officers * Establish a use of force continuum that restricts the most severe types of force * Require officers to report each time they use force or threaten to use force against civilians. Every city should adopt these policies immediately. Furthermore, we call on you to begin the process now toreimagine public safety by reinvesting city budgets in programs that expand and enhance restorative justiceprograms, provide alternatives to incarceration, demilitarize the police, provide crisis intervention and de-escalationtraining, increase community engagement, and expand holistic collaborative services. The time is now to finally fix this broken system, locally and nationally. Sincerely, Sherry Schupbach PO BOX 11082 Bozeman, MT 597191082 schupbach_2@msn.com From:Chris Mehl To:Agenda Subject:FW: Email contact from City Of Bozeman Date:Tuesday, June 16, 2020 5:09:59 PM Chris Mehl Mayor, City of Bozemancmehl@bozeman.net406.581.4992________________________________________From: webadmin@bozeman.net [webadmin@bozeman.net]Sent: Tuesday, June 16, 2020 1:02 PMTo: Chris MehlSubject: Email contact from City Of Bozeman Message submitted from the <City Of Bozeman> website. Site Visitor Name: KaylonSite Visitor Email: kaylonmengel@gmail.com Good morning,Without going deeply into a detailed response... I believe defunding the police department would be a gravemistake. Please do not defund this department. I know many with the same point of view and I hope they reach outto you regarding this important matter. Thank you for giving me a moment of your time. Kaylon Mengel From:agenda@bozeman.net To:Agenda Subject:Thank you for your public comment. Date:Tuesday, June 16, 2020 7:03:44 PM A new entry to a form/survey has been submitted. Form Name:Public Comment Form Date & Time:06/16/2020 7:03 PM Response #:442 Submitter ID:27717 IP address:69.163.86.180 Time to complete:3 min. , 18 sec. Survey Details Page 1 Public comment may be submitted via the form below, or by any of the following options. Public comment may also be given at any public meeting. Email: agenda@bozeman.net Mail to: Attn: City Commission PO Box 1230 Bozeman, MT 59771 In-person delivery to: Due to City Hall's closure in response to the Covid-19 pandemic, in-person delivery of comments is not available until further notice. First Name melode Last Name warner Email Address mel.warner.mw@gmail.com Phone Number 4065510457 Comments I am against police defunding. Please reconsider this for the sake of the older generation. Thank you. If you would like to submit additional documents (.pdf, .doc, .docx, .xls, .xlsx, .gif, .jpg, .png, .rtf, .txt) along with your comment, you may alternately address agenda@bozeman.net directly to ensure receipt of all information. Thank you, City Of Bozeman This is an automated message generated by the Vision Content Management System™. Please do not reply directly to this email. From:agenda@bozeman.net To:Agenda Subject:Thank you for your public comment. Date:Tuesday, June 16, 2020 7:22:51 PM A new entry to a form/survey has been submitted. Form Name:Public Comment Form Date & Time:06/16/2020 7:22 PM Response #:443 Submitter ID:27718 IP address:63.153.15.34 Time to complete:5 min. , 40 sec. Survey Details Page 1 Public comment may be submitted via the form below, or by any of the following options. Public comment may also be given at any public meeting. Email: agenda@bozeman.net Mail to: Attn: City Commission PO Box 1230 Bozeman, MT 59771 In-person delivery to: Due to City Hall's closure in response to the Covid-19 pandemic, in-person delivery of comments is not available until further notice. First Name Randy Last Name Wronko Email Address randywronko@gmail.com Phone Number 406-209-7601 Comments Please do not defund our police. I have a family and I want them protected. If anything, they need a pay increase!!! If you would like to submit additional documents (.pdf, .doc, .docx, .xls, .xlsx, .gif, .jpg, .png, .rtf, .txt) along with your comment, you may alternately address agenda@bozeman.net directly to ensure receipt of all information. Thank you, City Of Bozeman This is an automated message generated by the Vision Content Management System™. Please do not reply directly to this email. From:Zoë Unruh To:Agenda Subject:Public comments for general budget funding Date:Tuesday, June 16, 2020 7:58:10 PM Attachments:Overlooked-Undercounted-RELEASED.pdf Public service reductions associated with placement of homeless persons with severe mental illness in supportive housing (1).pdf Law Enforcement Response to the Mentally Ill, An Evaluative Review.pdf The Safer Cities Initiative.pdf Rikers Island Hot Spotters, Defining the Needs of the Most Frequently Incarcerated.pdf Hello Commissioners, Thank you for your words last night. I wanted to supplement my public comments that I madeat the meeting because I felt like the Commissioners did not quite understand what I was asking for. I understand that the City Manager and City Commission believe that the BozemanPolice Department is providing social services for our community, and therefore funding for police means bolstering social services. As a community member I do not want the policeproviding these social services. I've attached several research studies that show that armed response to homelessness, behavioral health crises in public, and mental health crises is moredamaging and costly than it is beneficial. We can't believe that Bozeman is the exception to the rule. As a community I believe we should be moving in the direction of several citiesacross the country who are creating unarmed crisis response teams for this very purpose. Even if they're receiving county or state money for social services, they are additionallycontributing city dollars to protect their community members. I've linked some stories below:Albuquerque San FranciscoEugene (This model has been present for 30 years in Eugene) DenverOakland Portland Thank you again for the opportunity to submit comments. I'm proud of the community that Ilive in, and I'm looking forward to making it even safer for our most marginalized community members. Warm regards, Zoe Unruh521 W. Lamme St -- Zoë UnruhPronouns: She/Her zunruh@gmail.com Law Enforcement Response to the Mentally Ill: An Evaluative Review Abigail S. Tucker, Psy. D. Vincent B. Van Hasselt, PhD Scott A. Russell Rarely does a police ofﬁcer list providing services to the mentally ill as a reason for becoming a law enforcement professional. However, a review of case records illustrates the frustrating, and often tragic, outcome of police service calls for ‘‘mental disturbance.’’ A closer examination of these cases demonstrates the reality that police are usually the initial contact into both the criminal justice and the social service systems for mentally ill persons. Unfortunately, there exists a disconnect in the process from the ﬁrst police contact to the next level of appropriate care due largely to a lack of proper training, resources, and collaborative community support. The purpose of this paper is to provide an overview of the research and public policy on law enforcement response to the mentally ill. An evaluative review of investigative efforts in this area reveals methodological shortcomings in the extant research which (a) prevent deﬁnitive conclusions regarding efﬁcacy of police interventions (e.g., Memphis Crisis Model), (b) have signiﬁcant implications for the development of policy, standard operating procedures, and training of law enforcement personnel, and (c) are potentially relevant to the safety of mentally ill persons who, as subjects or suspects, also become potential victims. Suggestions for directions that future research on policing and the mentally ill might take are offered. [Brief Treatment and Crisis Intervention 8:236–250 (2008)] KEY WORDS: mental illness, police, crisis intervention, community policing, Crisis Intervention Training. Police Response to the Mentally Ill Police ofﬁcers are tasked with responding to the challenges and dangers of society that ordi- nary citizens and social service agencies are not equipped to manage. In addition to their roles as investigators and protectors, police still maintain the responsibility of keeping the peace (Cordner, 2000). Law enforcement has beenrespondingtocallsforservicethatinvolve thementallyillthroughoutthehistoryofthepro- fession. As ﬁrst responders, police are often an From the Center for Psychological Studies, Nova Southeastern University (Tucker, Van Hasselt) and the Crisis Intervention Team, Fort Lauderdale Police Department (Russell) Contact author: Vincent B. Van Hasselt, Ph.D., Professor of Psychology, Nova Southeastern University, Center for Psychological Studies, 3301 College Avenue, Fort Lauderdale, FL 33314. E-mail: vanhasse@nova.edu. doi:10.1093/brief-treatment/mhn014 Advance Access publication on August 21, 2008 ªThe Author 2008. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org. 236 unidentiﬁed, yet signiﬁcant, inﬂuence in the lives of mentally ill persons. Indeed, Teplin’s (2000)reviewofhowofﬁcersmaintainthepeace with reference to the mentally ill underscores law enforcement’s role as a major mental health resource. More recently, however, police de- partments are ﬁnding it necessary to carefully examine, and in many cases modify, their pro- cedures dealing with the mentally ill. The heightened activity in this area is attributable to (a) legal reforms (e.g., deinstitutionalization) that have provided mentally ill persons with increasedindependenceandautonomy,(b)pub- lic policy created to assist law enforcement in maintaining peace and safety, and (c) the relative lack of collaboration between police, social service systems, and the mentally ill individual. Police Contact with the Mentally Ill: Historical Perspectives Thetrendtowarddeinstitutionalizationbetween the 1960s and 1980s is a major reason for the increased contact between the mentally ill and police (Zdanowicz, 2001). Teplin (2000) cited the curtailment of federal mental health funding, combined with legal reforms, which gave mentally ill people the right to live in the community without treatment, as the bases for their greater contact with police. Zdanowicz also cogently points out that legal reforms in the 1970s began criminalizing people with men- tal illness by instituting laws for involuntary treatment and laws that applied to nondanger- ous offenses (e.g., responding verbally to audi- tory hallucinations in public parks, sleeping on park benches). Beginning in the 1950s, police adhered to the ‘‘professional model,’’ which used experts from other ﬁelds (e.g., psychologists, advocacy law- yers) to bolster police reform and response to mentally illness (Cordner, 2000). This model emphasized the goals of (a) training, (b) special- ization, (c) improved communication, and (d) the development of less-than-lethal weaponry (Cordner, 2000). Such goals, while highly com- mendable, were often not realized by police agencies. It seems likely that they were not actualized due to ﬁnancial concerns, a lack of realistic application, and the inability of the consulting professionals to offer heuristic guidelines. When police are currently confronted with a mentally ill person, they have three options: (a) transport them to a receiving psychiatric facility, (b) use informal verbal skills to de- escalate the situation, or (c) arrest (Teplin, 2000). According to Teplin (2000), these possible ac- tions are based on two basic concepts that guide police in all citizen encounters: (a) the duty of the ofﬁcer to protect and serve the community and (b) the governing reforms that stipulate the power of an ofﬁcer to involuntarily protect an irrational person who may be of harm to self or others. More comprehensive and ﬂexible ap- proacheshaverecentlybeendeveloped;however, thesemethodsaretheminorityandhavenotbeen standardized.Examplesincludespecializedpolice trainingandunits,crisisinterventiontraining,and community-collaborative programs. As the widespread media coverage in the past decade has shown, these limited options for po- lice response can lead to cases in which people are killed or injured (Huriash, 2001; Jenne & Eslinger,2003;Vickers,2000).Evenmoretragic is the increase in suicide by cop (sometimes re- ferred to as police-assisted suicide) which is de- ﬁned by Hutson et al. (1998) as ‘‘a term used by law enforcement to describe an incident in which a suicidal individual intentionally en- gages in life-threatening and criminal behavior with a lethal weapon or what appears to be a lethal weapon toward law enforcement ofﬁcers or civilians to speciﬁcally provoke ofﬁcers to shoot the suicidal individual in self-defense or to protect civilians.’’ As a result of such pub- lic attention, the growing population of Law Enforcement Response Brief Treatment and Crisis Intervention 237 / 8:3 August 2008 237 homeless, the increase in mental disturbance service calls, and the evolving goals of law en- forcement agencies, police are now faced with the challenge of how to develop and implement an effective standard operating procedure when dealing with the mentally ill and homeless. Review of Research Thestudiesreviewedbelowwereselectedbased on their (a) empirically oriented focus and (b) attempt to ascertain the utility of specialized training, departmental policies, and/or special- ized programs designed to address the policing of the mentally ill. In a TELEMASP Bulletin (Peck, 2003), 35 po- lice and 6 Sheriff’s departments were surveyed regarding their interaction with mentally ill persons. Results revealed a high frequency of interactionsbetweendepartmentsandmentally ill individuals. Also, there was a consensus that a service call with a mentally ill person was ‘‘... slightly more difﬁcult than the typical police transaction.’’ A wide range of responses was compiled when asked how the departments normally handle the follow-up to a ‘‘mentally disordered’’ service call; most stated that they transport to either a community mental health center (CMHC) or to a hospital. When asked about specialized response policies, less than half of the surveyed departments referenced specialized training or program agreements with local community health centers. Deane et al. (1999) surveyed 174 police de- partments about their methods of dealing with mentally ill persons. This study identiﬁed spe- cializeddepartmentalprogramsaseitherpolice- based specialized police response, police-based specialized mental health response, or mental health-based mental health response. However, results of the survey indicated that over half of the responding departments did not have any typeofspecializedresponseprogramortraining for dealing with mentally ill persons, which is consistent with the ﬁndings of the TELEMAS Bulletin (Peck,2003).Further,themajorityofof- ﬁcerswhowerepartofaspecializedprogram,as well as those from departments with no special- ized response, self-rated their departments as moderately or very effective. Similarly, Hails and Borum (2003) surveyed medium and large law enforcement agencies (n ¼84)andreportedawiderangeofresponses regarding use of training, exposure to mental health support and training, and use of special- ized teams and responses when working with the mentally ill. These investigators found that 32% of the agencies had some type of special- ized response when called to respond to a mentally ill person. Coincidentally, 21% of de- partments surveyed reported a specialized unit orteamandonly8%reportedaccesstoamental health mobile crisis team. Finn and Stalans (2002) used police ofﬁcers from public safety training centers to examine the inﬂuence of ofﬁcer beliefs (using hypothet- ical scripts) in deciding whether to civilly com- mit, verbally resolve the situation on scene, or arrest. This study highlighted the impact of po- lice discretion on decision making when faced withapotentialcaseinvolvingmentalillness.A signiﬁcant number of ofﬁcers from the total sample (n ¼257)was morelikely to utilizecivil commitment if they felt that (a) a community hospital would accept violent persons, (b) the victim was in danger of further harm, (c) the suspect was mentally ill, and (d) the suspect was cooperative. Additionally, it was found that police ofﬁcers felt equipped to recognize the symptoms of mental illness without formal training.Finn andStalans(2002)also concluded that a signiﬁcant factor in decision making, with regard to mental disturbance service calls, is the available options (e.g., hospitals, mental health consultants) police have (or do not have) oncetheyarriveonsceneandassessthatmental illness is a factor. TUCKER ET AL. 238 Brief Treatment and Crisis Intervention / 8:3 August 2008 TABLE 1.Reviewed Research ArticlesAuthors Participants Method Measurements Outcome/ConclusionPeck (2003) 35 Police department and6 sheriff’s departmentParticipant survey Author-created survey:assessing policeinteraction withmentally illCIT-like programsare effectiveHails and Borum (2003) 84 Medium and large lawenforcement agenciesSurvey None Varied responses regardingtraining and use ofspecialized responseFinn and Stalans (2002) 257 Police ofﬁcers fromNorth Georgiain-service trainingSurvey––responses to scripts:analyzed using ANOVANone Ofﬁcers are signiﬁcant morelikely to use civilcommitment (p,.05)based on perceptionsTeplin (2000) and Teplinand Pruett (1992)283 Police ofﬁcers and85 mentally disturbedpersonsObservational, in-ﬁeld study Symptom checklistand ‘‘incidentcoding form’’Mentally ill suspects werearrested more often(46.7%) thannonmentally ill suspects(27.9%)Engel and Silver (2001) 322 Police ofﬁcers and1,849 nontrafﬁc suspectsSystematic observationalin-ﬁeld studyIncident coding form Police are not more likely toarrest mentally ill civilians,7.6% vs. nonmentallydisordered civilians, 18.2%Catalano et al., 2005 State of Florida’s archiveddata on involuntarypsychiatric hospitalizationsinitiated by law enforcementin speciﬁed time periodsbetween 1999 and 2001.Interrupted time-series design None Law enforcement initiatedincreased involuntarypsychiatric hospitalizationsin the weeks following theattacks of September 11,2001. Concluding thatperceived community riskmay increase an ofﬁcer’sjudgment that a personwith mental illness isdangerous.Law Enforcement Response Brief Treatment and Crisis Intervention 239 / 8:3 August 2008 239 TABLE 1.ContinuedAuthors Participants Method Measurements Outcome/ConclusionBrakel and South (1968) Two police counties Observational, in-ﬁeld study Records review of Police are able to discernmental status but lackcommunity collaborationand referral support.1. Hospital records2. Police recordsSheridan and Teplin (1981) 838 Police referred Pre/postuncontrolled Records review of Reduced recidivism anddays as inpatients atCMHC as opposed tostate hospital. (CMHCmean days as inpatient¼33.2 vs. state hospitalmean days asinpatient¼137.5.)1. Demographics2. Pathology of patient3. Type: police contact4. Treatment and recidivismGreen (1997) One Law Enforcement Ofﬁcer(LEO) agency with oneconsulting agency andone in-house programCase study evaluationusing both qualitativeand quantitative analysisQuantitative: author-created incidentformsThe more senior ofan ofﬁcer, the lesslikely they willarrest or refer (–.26,p..05). Police needmore options.Qualitative: interviewsBorum et al. (1998) Three LEO agencies withthree different specializedprograms; totaln¼452Case study evaluationwith cross analysiswithin and betweeneach case.Patrol ofﬁcer survey Memphis CIT model waseffective in maintainingsafety (94.4%) andmeeting needs ofmentally ill (88.8).1. 4-point Likert scale2. Open-ended questionsStrauss et al. (2005) 485 Patients from LouisvilleUniversity Hospitalbrought in by LouisvilleCIT over period of 1 monthChi-squared used to analyzeresults from this time-series studyRecords reviewinterviewData support effectivenessof CIT in identifyingpsychiatric emergenciesDeane et. al (1999) 174 Police departments from194 U.S. cities with aPopulation of 100,000 or moreSurvey analysis comparison Department survey 55% departments had nospecialized and 3% hada Memphis CIT-like model.1. 5-point Likert scaleresponse.2. Open-ended questions TUCKER ET AL. 240 Brief Treatment and Crisis Intervention / 8:3 August 2008 InHonolulu,Green(1997)comparedquantita- tive (148 incident forms assessing the decision apoliceofﬁcermakeswhentheyencounteraper- son who is mentally ill) and qualitative (11 ofﬁ- cers were further assessed with structured and semi-structuredinterviews)responsesofofﬁcers to mentally ill and homeless person dispatch calls. The department had recently joined with acommunityagencyincreatinganin-housepro- gram called Project Outreach. The program was designedwiththeintenttoeducateandaidofﬁ- cers when responding to service calls involving a homeless person. Results indicated that the more years of experience an ofﬁcer had, the less likely he/she would arrest or refer; it was more likelythattheywoulddonothing.Greenoffered that this ﬁnding could be due to of these senior ofﬁcers’ lack of faith in the options available to them regarding mentally ill suspects/civilians. Conclusions of this study highlighted the con- ﬂicting roles of police (i.e., enforcer vs. peace keeper)andthelimitedoptionsavailabletothem withrespecttothesetypeofdisturbancecalls.In addition, this investigation underscored the overlappingnatureofhomelessnessandmental illness with regard to police response in the community. Teplin (2000) examined how police handle mentally ill persons by randomly selecting 283 police ofﬁcers to be observed in a large- scale study. This research employed a symptom checklist and a global rating scale to measure the presence of mental illness in the observed citizen. (Note: ‘‘Observed citizen’’ refers to the civilian to whom the ofﬁcers are dispatched.) In addition, an Incident Coding Form was designedtomeasureofﬁcers’actionswhendeal- ing with the observed citizen. Approximately, 9.3% of the observed citizens were determined to be severely mentally disturbed, with 5.9% being a possible suspect and 3.4% not deemed a suspect in the police investigation. According toTeplin’s(2000)ﬁndings,suspectswithsevere mental illness were more often arrested thanTABLE 1.ContinuedAuthors Participants Method Measurements Outcome/ConclusionSteadman et al. (2000) Same participant sample fromBorum et al. (1998) regardingresponse toRetrospective: to crossanalyze each programregarding response tomental health emergenciesRecords review Memphis CIT responded to95% of mental disturbancecalls (p,.001) andarrested only 6% (p,.001).Watson and Angell (2007) None Analytical review None Support use of proceduraljustice in trainingofﬁcers to respondDupont and Cochran (2000) Outcome data on Memphis CIT Literature review None Supports Memphis CIT model Law Enforcement Response Brief Treatment and Crisis Intervention 241 / 8:3 August 2008 241 hospitalized or otherwise handled. However, nonsuspects were never arrested and were usually dealt with using verbal skills or by pro- viding referral information. Similarly, Engel and Silver (2001) examined how police respond to mentally ill persons andconcludedthattheydonotcriminalizemen- tally ill suspects nor are they more likely to ar- restmentallyillpersons.Thisinvestigationused data from the Police Services Study (PSS) con- ducted in 1977. They compared it to their data collected between 1996 and 1997 and systemat- icallyobservedover300policeofﬁcers,thenre- corded the ofﬁcer’s responses to mentally TABLE 2.Empirical Concerns Concern in Current Research Corrective Options for Future Research 1. Lack of objective and validated assessment tools (a) Use combination of subjective and objective assessments (b) Use validated instruments (even if not validated on police ofﬁcers) (c) Use archival sources, that is, records 2. Lack of comparison sample population (a) Use control groups (b) Diversify sample populations by demographics and rank. 3. Lack of consumer population assessment (a) Measure perceptions within the consumers and consulting agencies. (b) Record review: recidivism, treatment outcome, arrest rates, response rates, etc. Research that focuses on the role of consulting agencies to LEO TABLE 3.Practical Feedback Law Enforcement Agency Feedback Feedback to Be Used in Collaboration with Community Agencies and Professionals 1. Training a. Increased training for all ofﬁcers regarding mental illness, homelessness, dangerousness, and appropriate responses Possible use of procedural justice as training framework b. Intensive and continual training for specialized response teams 1. Referral agencies (hospitals, CMHC) a. Create a no-refusal policy b. Offer 24-h service c. Provide training for receiving personnel regarding type of police referrals, including police perspectives and culture education 2. Specialize community policing a. Separate crisis intervention from umbrella of community policing b. Collaborate with local community agencies, hospitals, and professionals c. Model type suggested: police-based specialized police response 2. Community/social justice system a. Educate about the conﬂicting roles of police ofﬁcers regarding this issue. b. Create policy to support police-based specialized police response teams and community programs that work with the police agencies TUCKER ET AL. 242 Brief Treatment and Crisis Intervention / 8:3 August 2008 disordered suspects. (Note: This study used the term‘‘mentallydisordered’’asopposedto‘‘men- tally ill’’ speciﬁcally to reﬂect an ofﬁcer’s imme- diate discretion in evaluating an individual’s behavior on scene, and not based on a psychiat- ric standard, such as the Diagnostic and Statisti- cal Manual of Mental Disorders.) Usingabivariatecomparisonofarrestratesand characteristics, ofﬁcers were signiﬁcantly less likely to arrest mentally disordered suspects whencomparedtononmentallydisordered sus- pects. Interestingly, mentally disordered sus- pects were more likely to be female, Caucasian, older, homeless, intoxicated, in possession of a weapon, and involved in less serious offences whencomparedtononmentallydisordered sus- pects.SimilarresultswerefoundinOstronetal. (1982).Inbothstudies, itisnoteworthythatthe mentalstatusofthesuspectwasnotrelatedtoar- rest. In other words, police ofﬁcers were able makeadecisionaboutarrestbasedondangerous- ness and necessity, without relying on complex diagnostic assessments. Strauss et al. (2005) further supported the ef- fectiveness of a police-based Critical Incident Team (CIT) in the accurate assessment of psy- chiatric emergencies. These investigators re- viewed data from 485 patients brought to the Louisville University Hospital by the newly de- veloped and trained Louisville Metro Police CIT. The purpose of their study was to compare the psychiatric diagnosis as well as patient de- mographics and dispositions between individ- ualsbroughtinbytheCITandthosebroughtin from non-CIT court order (e.g., psychologists, family members) for psychiatric hospitaliza- tion. Results did not show a discrepancy and praised the CIT training for improving ofﬁcers’ ability to assess a psychiatric emergency (Strauss et al.). Inanotherexaminationofofﬁcers’assessments of imminent dangerousness, Catalano, Kessell, Christy, and Monahan (2005) reviewed the State of Florida’s archival data on involuntary hospitalizations initiated in the weeks follow- ing the attacks of September 11, 2001. Using an interrupted time-series design, they concluded that ofﬁcers presented more men and women for involuntary psychiatric hos- pitalizations in the weeks after the attacks. Their study supports the notion that level of perceived risk to the community is associated with the perception that an individual with mental illness is dangerous (Catalano et al., 2005). In 1963, the American Bar Foundation con- ducted an 18-month study to evaluate the re- sponse of police and other criminal justice agenciesindealingwith thementallyillinrural areas (Brakel & South, 1968). Speciﬁcally, the investigation examined the efﬁcacy of diver- sion programs, which attempted to move the mentally ill person from police custody to a secondary social service (i.e., hospital, shelter). Results conﬁrmed that while police are able to identify serious mental illness, they have min- imal community support to then transport the person to a more appropriate level of care. These ﬁndings suggest a lack of collaborative effort between police agencies and social service systems prior to de-institutionalization. Sheridan and Teplin (1981) examined the ef- forts of one program that served as an identi- ﬁed, ‘‘no-refusal’’ site (a CMHC) where police could transport mentally ill persons for psychi- atric emergences. The study compared police referrals to state hospitals with referrals to the CMHC using several variables, including (a) patientproﬁle,(b)recidivismrates,and(c)length of patient stay. Results revealed reduced recidi- vism rates and shortened length of inpatient staysforCMHCreferrals.Also,itwasnotedthat while the paperwork and length of ofﬁcer time spent remained the same for this program, ‘‘police-friendly’’ advantages included (a) a no- refusal policy, (b) increased rapport between ofﬁcersandCMHCstaff,and(c)24-havailability of the program/CMHC. Law Enforcement Response Brief Treatment and Crisis Intervention 243 / 8:3 August 2008 243 Borum et al. (1998) compared three separate police agencies by their three distinct methods of addressing the mentally ill in crisis. They de- veloped an ofﬁcer survey in order to measure police perceptions of (a) how incidents are han- dled when dealing with the mentally ill, (b) helpfulnessofthelocalmentalhealthresources, and (c) the department’s overall effectiveness. The three models of police programs included (a) police-based specialized police response, which uses sworn ofﬁcers with exclusive men- tal health training to work with the mental health system as a specialized team (i.e., Mem- phis Crisis Intervention Team; CIT), (b) police- based specialized mental health response, whichusesnon-swornemployeesofthedepart- ment who are mental health professionals to provide consultation on an as-needed basis for the ofﬁcers on duty (i.e., Birmingham Com- munity Service Ofﬁcers), and (c) mental health- based specialized mental health response, which uses established, independent, and mobile mental health crises teams to work col- laboratively with police when needed (i.e., Knoxville program). Results indicated that the police-based spe- cialized response (i.e., Memphis CIT) as the model perceived as most effective in prepared- ness, meeting the needs of the mentally ill, reducing the numbers of mentally ill persons being jailed, minimizing ofﬁcer time spent on these incidents, and ensuring community safety. Training appears to be the key in the Memphis CIT model (they originally brought trainers from the Galveston, Texas police department to initiate their program), and approximately 10% of their department is now CIT trained (Peck, 2003). Steadman et al. (2000) conducted a more exclusive study to address the outcome quality of police responses to the mentally ill. They utilized the same three police agencies from their original 1998 effort, and used arrestrate,treatmentmodality(psychiatrichos- pitalization), or on-scene resolution as their de- pendent variables (Borum et al., 1998). Using a cross-site comparison, the sample included approximately 100 dispatch calls from each site all dispatched as a ‘‘mental disturbance’’ (Borum et al., 1998). Results showed that the Memphis CIT (police-based specialized police response) was superior to the other two special- ized responses with a 95% rate of responding, compared to Knoxville (mental health-based response)witha40%rateofresponse,andonly a 28% on-scene response for Birmingham (police-based mental health response). Overall, these specialized programs diverted persons from jail, as reﬂected by a 7% overall rate of arrest, with Memphis CIT leading that low arrest outcome rate at 2%. Dupont and Cochran (2000) reviewed the lit- erature not only in support of the police-based specialized police response model (Memphis CIT) but also to reﬂect challenges and changes that appear to hinder its growth and efﬁcacy. It should be noted that they cited research that used their own specialized response model (Memphis CIT; Borum et al., 1998; Deane et al., 1999, Steadman et al., 2000) as well as their own research (Dupont & Cochran, 2000). More recently, Watson and Angell (2007) dis- cussedtheimpactof‘‘proceduraljustice’’onthe outcome of police interactions with mentally ill populations. As deﬁned in the article, proce- dural justice is an approach to handling situa- tions (including police response incidents) with the mentally ill populations that focuses on ‘‘how ofﬁcer behaviors may shape cooperation or resistance’’ (p. 787). The three components of procedural justice include (a) participation (e.g., two-way communication and active lis- tening), (b) dignity, and (c) trust in thegoalthat all parties are working together in the best interest of the individual and public safety. These investigators cited the Broward County Mental Health Court Evaluation (Broward County, FL) and studies that support the use TUCKER ET AL. 244 Brief Treatment and Crisis Intervention / 8:3 August 2008 of cooperation, destigmatization, and other as- pects of procedural justice. They conclude by highlighting the importance of fairness in pro- cedural justice and the potential positive impli- cations of incorporating procedural justice in police training. Police Response: The Reality of Mental Disturbance Service Calls Matthews (1970) reviewed the 1968 American BarFoundationstudyasoneoftheﬁrstcompre- hensive investigations on this topic (Brakel & South, 1968). The review indicated that police are often blamed for the misfortunes of this growing problem. Speciﬁcally, Matthews describes how police are often ‘‘pigeonholed’’ into making a medical decision with little train- ing and few, if any, response options. In addi- tion, this exploratory investigation provided recommendations for increased ofﬁcer training, specialized response squads, improved commu- nity mental health facilities, and adequate legal authority for emergency detention and admission. Expanding on the work of Matthews (1970), Sheridan and Teplin (1981) examined the util- ity of a police referral program that designated an intake unit at a CMHC. They found that streamlining the process of how police refer mental disturbance service calls to hospitals bolstered the program’s effectiveness. Addi- tionally, they showed that a collaborative response between law enforcement and the CMHC reduced recidivism rates in referred psychiatric patients. Teplin’s research in the early 1980s mirrored the abovementioned study with in-ﬁeld inves- tigators who rotated between each of the ofﬁ- cer’s 24-h shifts and between all districts of thecitybeingexamined(Teplin,2000,Teplin& Pruett, 1992). This work veriﬁed that police are not arresting mentally disabled persons at an increased rate when all variables (e.g., dangerousness of scene, available resources, potential victims) are taken into account. It ap- pears that ofﬁcers are arresting when they feel there is no alternative (e.g., involuntary hospi- talization) to control the situation and maintain community safety (Teplin, 2000). This study also underscored the lack of options for ofﬁcers respondingtoamentallyillpersonandthehigh frequency with which ofﬁcers attempt to resolve situations by informal means (i.e., with- out arrest or hospitalization). Specialized Police Response Models and the Role of Crisis Intervention Training Deane et al. (1999) provided a preliminary ep- idemiology for the subject of police response to the mentally ill, including the existence of spe- cialized programs, perceptions of effectiveness in managing the mentally ill, and the availabil- ity within each department for ofﬁcers to re- ceive specialized training. They reported that although more than 50% of departments na- tionwide do not have a specialized program/ response, the majority of departments self-rate themselves as effective in managing these types of service calls. This stands in contrast to research that points to the effectiveness of specialized response programs (Borum et al., 1998; Green, 1997; Sheridan & Teplin, 1981; Steadman et al., 2000). However, more recent ef- fortssuggestthatthenumberoflawenforcement agencies reporting specialized training and units for dealing with the mentally ill popula- tions is increasing (Hails & Borum, 2003). Nev- ertheless, the abovementioned surveys have their limitations and, generally, concede that a standardized procedure or ‘‘model’’ for police response to the mentally ill is problematic. Borum et al. (1998) provided a framework that supported the police-based specialized po- lice response (i.e., Memphis CIT) using the variables of perceived preparedness, quality of response to the mentally ill, diversion from Law Enforcement Response Brief Treatment and Crisis Intervention 245 / 8:3 August 2008 245 jail, ofﬁcer time spent on these calls, and com- munity safety. In addition to providing evi- dence supporting the need to develop specialized response programs, their study offered a structure for further evaluation of such specialized approaches, including (a) the variable of community safety, (b) the needs of the civilian in crisis, (c) realistic perceptions, and (d) repercussions of ofﬁcer response. Steadman et al. (2000) expanded on their ear- lier work (Borum et al., 1998) by reﬁning out- come variables and using a more homogonous sample. Their study continued to support the Memphis CIT model (i.e., police-based special- ized police response) with ﬁndings of higher response rates and fewer arrests. The outcome variables of this investigation shifted from subjective perceptions of police ofﬁcers and used arrest rates, treatment modalities, and referral sources to bolster their ﬁndings in sup- port of police-based specialized police response models. Although the work of Strauss et al. (2005) does not address the generalized effectiveness of CIT training, it supports an important aspect of CIT. They provided support for the conten- tion that CIT training provides ofﬁcers with the ability to accurately evaluate the need for psy- chiatric emergency services. It is important to note that additional research conﬁrmed that of- ﬁcers (CIT trained or not) are generally able to identify when mental illness is the primary fac- tor in a disturbance call (Brakel & South, 1968; Finn&Stalans,2002).Anintegralcomponentof CIT training is the utilization of crisis interven- tion and active listening skills (i.e., paraphras- ing, reﬂecting emotions, open-ended questions) that assist an ofﬁcer in de-escalation and appro- priate community resources (Cochran, 2008). It would appear that to comprehensively address police response to mental illness, it is important to include both psychoeducation concerning mental health issues as well as crisis interven- tion skills training. Additional Barriers and Concerns Deﬁning Training in the Field of Law Enforcement In a continuation of their work, Dupont and Cochran (2000) reviewed their ﬁndings and made suggestions for future work. First, they proffered that basic ofﬁcer training will prove to be inadequate in addressing this growing and volatile, problem without continual review and experiential learning. Expanding on their concern for the generalization of training ef- fects, Dupont and Cochran (2000) point out that the survival model of training regarding self- defense and ﬁrearms not only exceeds the real-life frequency of such an occurrence (particularly when compared to the rate of in- cidents involving mentally ill individuals in crisis) but also seems to inappropriately mold an ofﬁcers’ perception of dangerousness. Refer- ring to the 21-foot rule of weapon drawing, DupontandCochran(2000)questionthevalidity of this rule when dealing with a subject sus- pected of mental disturbance. Dupont and Cochran (2000) refer to the common mispercep- tion that all police ofﬁcers have the same man- dated training and available resources. Steadmanetal.(2001)concurthatforspecialized response programs to work effectively, training is a crucial element. However, Steadman et al. (2001) emphasized the importance of consulting mental health professionals and other adminis- trativeandsocialservicesystemsinlawenforce- ment training. The Mental Health Care System as a Barrier The mental health care system itself appears tobeabarrierforprogressregardingthisissueof policing the mentally ill (Dupont & Cochran, 2000; Matthews, 1970). Social service agencies often refuse to admit intoxicated or psychotic TUCKER ET AL. 246 Brief Treatment and Crisis Intervention / 8:3 August 2008 persons referred by police. In addition, the ‘‘revolving door’’ phenomenon of recidivism supports the reality of overworked and under- paid staff in receiving facilities, such as hospitals (Dupont & Cochran, 2000). More spe- ciﬁcally,DupontandCochran(2000)addressthe challenges of responsibility and organized de- liveryofcarewhenpolicerespondtomentaldis- turbance service calls. The ﬁrst dilemma concerns an ofﬁcers’ attempt to avoid making an arrest by transporting to a receiving agency and continuing their patrol duties. Further, manytreatmentfacilitiesrequirepolicecustody in the waiting area for individuals transported for a mental disturbance. There is also a lack of a systematic and hier- archical structure that links ﬁrst responders (e.g., police or emergency medical services) with the appropriate level of care to the mental health system (e.g., medical vs. psychiatric hos- pitals, social service shelters vs. drug rehabili- tation centers). Steadman et al. (2001) address this problem and recommend the following: (a) use of designated drop-off sites and (b) po- lice-friendly procedures that include (a) a no- refusal policy, (b) an intake process with streamlined paperwork, and (c) consistent pro- cedural steps. If not addressed, these overlap- ping problems reduce the options an ofﬁcer has available. Implications of the Research: Empirical and Practical The purpose of this paper was to review the extant research in order to evaluate and poten- tially enhance the current development of specialized programs, as well as to provide direction for further study in this area. Our review of work in this ﬁeld revealed three overlapping concerns that should be addressed infutureresearchregardingthepoliceresponse to the mentally ill. Clinical and Actuarial Assessment Several studies used self-rating surveys to mea- sure ofﬁcer perceptions (Borum et al., 1998; Deane et al., 1999; Steadman et al., 2000). Indeed, there is a lack of validated assessment methodsforlawenforcementofﬁcers,andobser- vational studies have been negatively critiqued. Mastrofski and Parks (1990) provide in-depth analysis of observational research, speciﬁcally, in examining the activity of police ofﬁcers. The work provides evidence that observational studies can be scientiﬁcally bolstered, in the absence of validated assessment tools. Because psychosocial research mandates parametrically sound evaluation instruments, studies in this area should not adhere to a lesser standard. In addition, records review was used in only two studies; further, this strategy was merely employedtoeitheridentifytheirsampleortode- scribethepatientpopulation(Sheridan&Teplin, 1981;Steadmanetal.,2000).Recordsreviewisan easilyaccessible,veriﬁable,andthoroughassess- mentstrategythatshouldbeincludedinresearch regarding police department response. Theroleofclinicalassessmentinthecontextof thispaperrefers totheuse,proﬁciency, and efﬁ- cacyofcrisisassessmentwhenworkingwithmen- tally ill persons. A review of speciﬁc crisis assessment and intervention techniques useful forworkingwiththementallyillisabroadtopic thatextendsbeyondthescopeofthispaper.How- ever,itshouldbenotedthattheabilityofofﬁcers to both identify the need for mental health serv- ices, and appropriately utilize basic crisis inter- vention techniques on-scene, is a strength of the CIT model. This again highlights the impor- tance of interagency collaboration and the coop- erativeuseofavailablementalhealthcrisisteams. Expanding Samples Another concern characterizing research in this area were the samples employed. Police participants often reﬂected a diverse police Law Enforcement Response Brief Treatment and Crisis Intervention 247 / 8:3 August 2008 247 population, with the exception of ofﬁcer ranks (Borum et al., 1998; Teplin, 2000; Teplin & Pruett, 1992). As noted in Bellah (2002), the ac- ceptanceofspecializedresponsetothementally ill needs to be addressed ﬁrst, and foremost, with supervisory and command staff. In addi- tion to developing a representative police sam- ple, research in this area can be improved by employing appropriate control groups to further validate results. Collaboration Commondependentvariablesincludedcommu- nity safety and the subjective perception of the ofﬁcers (Borum et al., 1998; Sheridan & Teplin, 1981; Steadman et al., 2000). Much of the re- search indicated that cross-training, ongoing and frequent communication, and collaborative program development would best serve police agencies if they included consulting mental health professionals (Borum, 2000; Dupont & Cochran, 2000; Steadman et al., 2001). Yet, only onestudy(Sheridan&Teplin)surveyedaconsul- tingmentalhealthagencyandthementallyillsub- jects/consumers themselves. Further, results of thoseinvestigationswerenotcomparedtopolice perceptions of the program (Sheridan & Teplin). Indeed, ofﬁcer perception is critical and directly impacts program efﬁcacy. However, the goal should be a collaborative examination of ofﬁcer perception, community safety, the safety of the mentally ill and homeless, and the availability and perception of the social service system. Other Need Areas Thecurrentliteraturerevealsalackoffollow-up studieswithdifferentagenciesandparticipants, despite the replicable methodologies many investigationsoffered.Strongerassessmentcom- ponents,morediversiﬁedsamples,andinclusion of control groups would greatly enhance the study of this topic. However, at a basic level of empirical interest, follow-up efforts would provide important evidence regarding the long- termutilityofspecializedresponseprograms.Un- fortunately, only one investigation (Sheridan & Teplin, 1981) developed a pre/postprogram de- sign;andthiseffortonlyassessedoutcomevaria- bles for the mentally ill and the psychiatric provider. Beyondlongitudinalandreplicableinvestiga- tions in this area, additional work is needed to focus on such variables as (a) the amount and type of specialized training, (b) the extent of community acceptance of specialized teams, (c) the effects of a no-refusal dispatch policy/no- refusal acceptance policy combined with a spe- cialized team/training, and (d) the impact of continuous, positive contact between a special- izedresponseteamandthechronicallymentally ill and homeless. Practical Applications Overall, this review provides support for deployingspecializedlawenforcementresponse programstoaddresstheneedsofmentallyillper- sons. Most research, to date, has focused on the Memphis CIT model. In 1988, the Memphis Po- liceDepartmentjoinedwiththeirlocalchapterof theNationalAlliancefortheMentallyIllandin- stituted a new community-policing approach thatcreatedspecializedteamsofofﬁcerstrained tomanagepeopleincrisis:theCrisisIntervention Team(Vickers,2000).TheMemphisCITconsists ofuniformpatrolofﬁcerswhovolunteerandre- ceive a minimum of 40 h of specialized training, preparingthemtobecalleduponwhenaservice call involves a mentally disordered person. Research shows that the Memphis CIT model ofpoliceresponsetothementallyillisfunctional, generally accepted by police departments, and mostimportantly,effective(Dupont&Cochran, 2000). While this model has demonstrated efﬁcacy, perhaps, even greater utilization of this model would be achieved by (a) deter- mining, speciﬁc factors most responsible for TUCKER ET AL. 248 Brief Treatment and Crisis Intervention / 8:3 August 2008 effectivenessofthemodel,and(b)garneringfur- therdatasupportingthereplicabilityandgener- alizability of the approach. In addition, the current literature does little to address speciﬁc training variables (e.g., length and format) that might have a bearing on eventual skills applied in the ﬁeld. An important element in this area is collabo- rationamongmentalhealthprofessionals,social service agencies, and law enforcement. Coggins and Reddy-Pynchon (1998) examined the de- velopment of the United States Secret Service Mental Health Liaison Program designed to as- sist the Secret Service in ascertaining threat po- tential and in following up with appropriate intervention. These investigators stated that appraisal of theirprogram supportedits success and was based on (a) careful review of the pro- visions offered by the consulting entity (e.g., local mental health correspondent, specialized responseprogram),(b)evaluationofviewpoints of both the mentally ill and police, and (3) ex- amination of the consultant or specialized pro- gram’s records, reports, data, and/or any other evaluation products used in their program. Police agencies are evolving to create opera- tionalandeffectiveprogramsthatmeettheneeds of the community and the intrinsic attributes of law enforcement culture. The scope of research and program development in the future should include empirical research and interagency collaboration, with an overarching goal of real- istic and efﬁcacious ofﬁcer response options and improved outcome for the mentally ill. Summary Police ofﬁcers are responsible for maintaining and enforcing public order. Their role as both ﬁrst responders and peacekeepers remains a challenge in many ways, as has been illustrated by this review. The law enforcement response to mental disturbance calls with ethical, practi- cal, and effective strategies requires interagency collaboration. A review of relevant research highlights several concerns with regard to research methodology, public policy, police interventions, as well as a frequent lack of community-wide collaboration. However, the research also illustrates numerous examples of effective police-based interventions and collab- orative policies and procedures. In particular, current research supports the utilization of spe- cialized law enforcement response to meet the needs and demands of the mentally ill popula- tion with safety and dignity. Acknowledgments Conﬂict of Interest:None declared. References Bellah, J. (2002). Recognizing mental illness.Law & Order, 50(4), 48–51. Borum, R. (2000). Improving high risk encounters betweenpeoplewithmentalillnessandthepolice. Journal of the American Academy of Psychiatry and the Law, 28,332–337. Borum, R., Deane, M. W., Steadman, H. J., & Morrissey, J. (1998). Police perspectives on responding to mentally ill people in crisis: Perceptions of program effectiveness.Behavioral Sciences and the Law, 16,393–405. Brakel, S. J., & South, G. R. (1968). Diversion from the Criminal Process in the Rural Community: Final Report of the American Bar Foundation Project on Rural Criminal Justice.American Bar Foundation.Reprinted (1969) from the American Criminal Law Quarterly, 7(3), 122-173. Broward Coalition for the Homeless (2002).Outreach 2002, Year End Report. Broward Coalition for the Homeless, Inc. Catalano,R.A.,Kessell,E.,Christy,A.,&Monahan,J. (2005). Involuntary psychiatric examinations for danger to others in Florida after the attacks of September 11, 2001.Psychiatric Services, 56,858. Cochran, S. (2008).Fighting the Stigma in Law Enforcement.Retrieved April 21, 2008, from Law Enforcement Response Brief Treatment and Crisis Intervention 249 / 8:3 August 2008 249 http://www.memphispolice.org/ Crisis%20Intervention%20Team%202.htm. Coggins, M. H., & Reddy Pynchon, M. (1998). Mental health consultation to law enforcement: Secret service development of a mental health liaison program.Behavioral Sciences and the Law, 16,407–422. Cordner, G. W. (2000). A community policing approach to persons with mental illness.Journal of the American Academy of Psychiatry and the Law, 28,326–331. Deane, M. W., Steadman, H. J., Borum, R., Veysey, B. M., & Morrissey, J. P. (1999). Emerging partnerships between mental health and law enforcement.Psychiatric Services, 50,99–101. Dupont, R., & Cochran, S. (2000). Police response to mental health emergencies––Barriers to change. Journal of American Academy of Psychiatry and the Law, 28,338–344. Engel, R. S., & Silver, E. (2001). Policing mentally disordered suspects: A reexamination of the criminalization hypothesis.Criminology, 39, 225–253. Finn, M. A., & Stalans, L. J. (2002). Police handling of the mentally ill in domestic violence situations. Criminal Justice and Behavior, 29,278–307. Green,T.M.(1997).Policeasfrontlinementalhealth workers: The decision to arrest or refer to mental health agencies.International Journal of Law and Psychiatry, 20,469–486. Hails, J., & Borum, R. (2003). Police training and specialized approaches to respond to people with mentalillness.CrimeandDelinquency,49(1),52–62. Huriash, L. J. (2001, October 15). Police team to focus on the mentally ill.Sun-Sentinel, pp.5B. Hutson, H. R., Anglin, D., Yarbrough, J., Hardway, K., Russell, M., Strote, J., Canter, M., & Blum, B. (1998). Suicide by cop.Annals of Emergency Medicine, 32(6), 665–669. Jenne, K., & Eslinger, D. F. (2003). Baker act: Without reform, problems mount.Sun-Sentinel, pp. 25A. Mastrofski, S., & Parks, R. B. (1990). Improving observational studies of police.Criminology, 28, 475–496. Matthews, A. R., Jr. (1970). Observations on police policy and procedures for emergency detention of the mentally ill.Journal of Criminal Law, Criminology and Police Science, 61,283–295. Ostrom, E., Parks, R. B., Whittaker, G. (1982).Police Services Study, Phase II, 1977: Rochester, St. Louis, and St. Petersburg [Computer ﬁle]. 3 rd ICPSR version. Bloomington, IL: Ostrum, E., Parks, R. B., and Whittaker, G., Indiana University [producers], (1982). Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], (2001). Peck, L. W., Jr. (2003). Law enforcement interactions with persons with mental illness. TELEMASP Bulletin, 10(1), 1–10. Sheridan, E. P., & Teplin, L. A. (1981). Police- referred psychiatric emergencies: Advantages of community treatment.Journal of Community Psychology, 9,140–147. Steadman, H. J., Deane, M. W., Borum, R., & Morrissey, J. P. (2000). Comparing outcomes of major models of police responses to mental health emergencies.Psychiatric Services, 51,645–649. Steadman, H. J., Stainbrook, K. A., Grifﬁn, P., Draine, J., Dupont, R., & Horey, C. (2001). A specialized crisis response site as a core element of police-based diversion programs.Psychiatric Services, 52,219–222. Strauss, G., Glenn, M., Reddi, P., Afaq, I., Podolskava, A., Rybakova, T., et al. (2005). Psychiatric disposition of patients brought in by crisis intervention team police ofﬁcers. Community Mental Health Journal, 41,223–230. Teplin, L. A. (2000, July). Keeping the peace: Police discretion and mentally ill persons.National Institute of Justice Journal, 244,8–15. Teplin,L.A.,&Pruett,N.S.(1992).Policeasstreetcorner psychiatrist: Managing the mentally ill.International Journal of Law and Psychiatry, 51,139–156. Vickers, B. (2000). Memphis, Tennessee, police department’s crisis intervention team.U.S. Department of Justice: Bulletin From the Field, Practitioner Perspectives, July,1–4. Watson, A. C., & Angell, B. (2007). Applying procedural justice theory to law enforcement’s response to persons with mental illness. Psychiatric Services, 58,787. Zdanowicz, M. (2001). A sheriff’s role in arresting the mental illness crisis.Sheriff, 53(3), 2–4. TUCKER ET AL. 250 Brief Treatment and Crisis Intervention / 8:3 August 2008 Overlooked in the Undercounted The Role of Mental Illness in Fatal Law Enforcement Encounters Online at TACReports.org/overlooked-undercounted Doris A. Fuller Chief of Research and Public Affairs Treatment Advocacy Center H. Richard Lamb, M.D. Emeritus Professor of Psychiatry and Behavioral Sciences Keck School of Medicine of the University of Southern California Michael Biasotti Chairman, Committee on Untreated Severe Mental Illness and Past President New York State Association of Chiefs of Police, John Snook Executive Director Treatment Advocacy Center © 2015 Treatment Advocacy Center Overlooked in the Undercounted THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS December 2015 EXECUTIVE SUMMARY Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 1 An estimated 7.9 million adults in the United States live with a severe mental illness that disorders their thinking.1 Treatment in most cases can control psychiatric symptoms common to these diseases, but the system that once delivered psychiatric care for them has been systematically dismantled over the last half-century. Today, half the population with these diseases is not taking medication or receiving other care on any given day.2 Hundreds of thousands of these men and women live desperate lives.3 They sleep on the streets, over- flow emergency rooms and, increasingly, overwhelm the criminal justice system. Numbering somewhat fewer than 4 in every 100 adults in America,4 individuals with severe mental illness generate no less than 1 in 10 calls for police service5 and occupy at least 1 in 5 of America’s prison and jail beds.6 An estimated 1 in 3 individuals transported to hospital emergency rooms in psychiatric crisis are taken there by police.7 Individuals with mental illness also make up a disproportionate number of those killed at the very first step of the criminal justice process: while being approached or stopped by law enforcement in the com- munity. Enormous official and public attention has become focused on the official undercounting of fatal police shootings; barely noted in the uproar has been the role of severe mental illness – a medical condition that, when treated, demonstrably reduces the likelihood of interacting with police or being arrested, much less dying in the process.8,9,10,11 By all accounts – official and unofficial – a minimum of 1 in 4 fatal police encounters ends the life of an individual with severe mental illness. At this rate, the risk of being killed during a police incident is 16 times greater for individuals with untreated mental illness than for other civilians approached or stopped by officers (see Methodology). Where official government data regarding police shootings and mental illness have been analyzed – in one U.S. city and several other Western countries – the findings indicate that mental health disorders are a factor in as many as 1 in 2 fatal law enforce- ment encounters.12,13,14,15,16 Given the prevalence of mental illness in police shootings, reducing encounters between on-duty law en- forcement and individuals with the most severe psychiatric diseases may represent the single most immedi- ate, practical strategy for reducing fatal police shootings in the United States. Evidence-based treatment for severe mental illness exists. The disproportionate risk for criminal justice involvement associated with mental illness occurs chiefly among the less than 2% of the adult population with untreated severe mental illness. Treating the untreated is a proven practice for reducing the role of mental illness in all criminal justice involvement, including in deadly law enforcement encounters. But – in a data- and cost-driven world – making the case to invest in any solution requires reliable data about the scope and nature of the problem to be addressed. Reliable data about fatal law enforcement encounters in general do not exist, much less data about the role of mental illness in them. Here’s why: • More than a half-dozen federal databases tasked with tracking and/or reporting the number of fatal law enforcement encounters in the U.S. have been developed in recent decades, but not one ex- ists that produces complete and reliable statistics (see Appendix A: Federal Government Homicide Databases). Underreporting is so endemic that one audit of the government’s efforts concluded “the current data collection process results in a significant underestimation and potentially a biased picture of arrest-related deaths in the United States.”17 We can learn the average prenatal litter size of a feral cat in America18 but not the number of civilians killed during encounters with law enforcement. 2 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS • Common themes run through the databases’ individual shortcomings: reliance on methods guar- anteed to produce incomplete or inconsistent data, lack of centralized oversight to create account- ability, insufficient funding. One national database was incomplete by design (Police Use of Deadly Force, 1970-1979); another mandated by Congress but abandoned after a short life (National Data Collection on Police Use of Force Database). None is supplied with universal data (Supplementary Homicide Report, National Violent Death Reporting System and others). • In the absence of reliable official data, more than a dozen news organizations, nonprofits and individu- al bloggers are operating independent databases (see Appendix B: Independent Databases Tracking Fatal Law Enforcement Encounters). These coalesce around the probability that approximately 1,000 people die annually in officer-involved shootings – more than double the number any federal agency has ever reported.19,20 Because these databases rely on published anecdotes or crowdsourcing, they, too, inevitably understate the actual numbers. And because, like government databases, each uses different methods, they also inevitably arrive at different numbers. • Despite their limitations, independent databases have proven to be so much more complete and accurate than any government source that a bizarre feedback loop of incomplete information has emerged: Because government agencies themselves lack complete and accurate data on fatal law enforcement encounters, they collect their data from media and other online datasets. But because media and online sources don’t have definitive data from government, they generate incomplete data. Before expanding its methodology to include Google Alerts and other online resources, the Bureau of Justice Statistics (BJS) estimates that its Arrest-Related Deaths (ARD) program captured only half of the fatalities it was created to track. By expanding its methods, the agency reports that completeness has improved but estimates that 31%-41% of likely fatal law enforcement encounters are still not captured.21 These gaps pale by comparison with the information void surrounding the role of mental illness in fatal police encounters. The ARD program operated by the BJS is the only federal database that has ever set out to systematically collect and publish mental health information about police homicide victims. It is currently suspended because an audit determined that the available data did not meet the agency’s quality standards.22 Among the independent databases, only three – including the Treatment Advocacy Center’s Preventable Tragedies Database – report directly on the role of mental illness in fatal police shootings. In their analy- ses, both The Washington Post and The Guardian newspapers reported that at least 25% of the fatali- ties involved individuals with severe mental illness.23,24,25 Official studies in Las Vegas, Nevada, Australia, Canada and the United Kingdom report the prevalence to range from 33% to more than 50%.12,13,14,15,16 Reducing fatal law enforcement shootings of people whose encounters with police are the result of psy- chiatric disease is in the best interest of the individuals involved and society. The Treatment Advocacy Center makes the following recommendations to foster solutions that will reduce this loss of life and the many social costs associated with it: • Treat the untreated. Shifting the responsibility for responding to acutely ill individuals from mental health professionals to police has criminalized mental illness at enormous cost to individuals with the most severe psychi- atric diseases, the criminal justice system and society. The mental illness treatment system must be restored sufficiently so those with mental illnesses receive treatment before their actions provoke a police response. Lawmakers need to enact and implement five public policies to achieve these goals: o Increase the number of treatment beds for individuals suffering from acute or chronic psychiat- ric conditions. o Reform treatment laws that erect barriers to treatment for at-risk individuals, including laws that require courts to wait until individuals become violent, suicidal or gravely ill before intervention becomes possible. Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 3 o Make full use of laws that provide access to treatment for individuals too ill to seek treatment for themselves. o Expand the use of court-ordered outpatient treatment (assisted outpatient treatment or AOT), assertive community treatment (ACT) and other evidence-based practices that demonstrably reduce the likelihood of individuals with severe mental illness becoming police calls. o Make treatment funding decisions that consider both the cost of treatment and taxpayer savings that result from providing treatment that reduces criminal justice involvement, home- lessness and emergency medical services and other public costs. • Accurately count and report the number of fatal police encounters. The U.S. government does not possess a comprehensive, accurate database of fatal police encoun- ters. Factors that contribute to the absence include the lack of reporting requirements that are realistically funded, the absence of standardized definitions and methods to produce consistent data, and the lack of centralized oversight. Congress must enact legislation to direct resolution of these issues and to fund the operation of a reliable federal database of fatal police interactions nationwide. • Accurately count and report all incidents involving use of deadly force by law enforcement. Fewer than half of police shootings result in a death.26 Counting only fatal shootings produces an in- complete picture of the use of deadly force and its attendant impacts, including injury and disability. A system for tracking and incentivizing the reporting of all use of deadly force by law enforcement must be established and maintained. • Systematically identify the role of mental illness in fatal police shootings. Severe mental illness is an identifiable factor in at least 25% and as many as 50% of all fatal law enforcement encounters, but its role has been rendered virtually invisible by the failure of the government to track or report its presence. Questions to identify psychiatric factors must be in- cluded in the official surveys used to capture data about both fatal and nonfatal police shootings. TERMINOLOGY Many different terms and definitions are used officially and unofficially to characterize the death of a civilian that results from action by a law enforcement officer. Even though firearms are not the only weapons involved in fatal encounters with law enforcement, “police (or law enforcement) shooting” and/or “officer-involved shooting” have emerged in discussions by the public, media and law enforcement itself as the most common terms for describing civilian deaths that result from law enforcement interactions. For that reason, these terms are used throughout Overlooked in the Undercounted. Government terminology such as “justifiable homicide,” “arrest-related death” and “law enforcement homicide” is used when discussing the databases or publications within which they are used. A glossary of commonly used terms and definitions, including terms to describe categories in which the federal government may collect data, may be found in Appendix C. INTRODUCTION The Treatment Advocacy Center and the National Sheriffs’ Association in September 2013 released a joint report entitled Justifiable Homicides by Law Enforcement Officers: What is the Role of Mental illness? After combining the best official information available at the time with an extensive survey of academic journals, media stories and other unofficial sources, the authors concurred with unofficial estimates that “at least half of the people shot and killed by police each year in this country have mental health prob- lems.”27 Concluded the report, “As a consequence of the failed mental illness treatment system, an in- creasing number of individuals with untreated serious mental illness are encountering law enforcement officers, sometimes with tragic results.”28 Three recommendations were made: • The collection of data on justifiable homicides is a “legitimate federal responsibility (that) should be financed by the Department of Justice,” including the role of psychiatric disease in such deaths. • The responsibility for individuals with serious mental illness must be returned to the mental health treatment system from law enforcement, which has been co-opted as a substitute. • Wider use of court-ordered outpatient treatment (widely known as assisted outpatient treatment or AOT) is needed to ensure adherence by individuals with demonstrated dangerousness who are living in the community. Those recommendations are as timely today as they were then. However, since the 2013 report was published, dramatic developments have created an urgent need to update the analysis and expand our recommendations. The developments include the following: • Virtually universal criticism from law enforcement, lawmakers, media and the public about the federal government’s failure to accurately track the number of civilians killed during encounters with law en- forcement. FBI Director James B. Comey in October told a gathering of politicians and law enforce- ment executives, “It’s ridiculous – it’s embarrassing” that the United States cannot produce accurate police shooting numbers.29 • Proliferation of media and other independent databases typically coalescing around estimates that roughly 1,000 civilians die annually from the use of lethal force by law enforcement – more than double the number that federal databases have ever reported.19,20,30 • Introduction of new bills in Congress to produce complete and accurate fatal police shooting data (see Appendix D: History of Homicide Data Collection in the United States). • The U.S. Bureau of Justice Statistics’ (BJS’s) suspension of the Arrest-Related Deaths Program – the only federal database that systematically sought to identify the variable of mental health in what BJS calls “law enforcement homicide.” The action was taken after an audit of the source found that the number of incidents was being undercounted by half because of incomplete or inconsistent source data.31 • Publication of research based on government data in the U.S., Australia, Canada and the United Kingdom reporting that severe mental illness is implicated in up to 50% of deadly law enforcement encounters.12,13,14,15,16 4 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS Overlooked in the Undercounted examines the role of mental illness in fatal police shootings in three parts: • The Overlooked – The demonstrable role of mental illness in the use of deadly force by law enforcement • The Undercounted – What we do and don’t know about fatal police encounters – and why • Discussion and Recommendations – Practical approaches to reducing fatal police shootings and the many social costs associated with them THE OVERLOOKED The overrepresentation of individuals with mental illness in fatal police encounters is entirely consis- tent with their overrepresentation in virtually every corner of the criminal justice system – from the calls made to local police departments to the isolation cells of prisons. After a half-century of closing public psychiatric hospitals without replacing them with community-based facilities – a trend called “deinstitu- tionalization” – several hundred thousand people with mental illnesses such as schizophrenia and severe bipolar disorder cycle chronically through the streets, hospital emergency rooms and, most frequently, the criminal justice system.3 Deinstitutionalization transformed America’s mental health system. Medications made it possible for many individuals who once would have been hospitalized to live safely in their communities. But the community centers that were supposed to replace hos- pitals were not built in sufficient numbers to meet the need – or, when they were built at all, they were operated for patients with milder psychiatric symptoms than those of the people who were once hospitalized. The consequences are on public display in the many settings where people who are untreated and unable to seek treatment are found – in bus stations and libraries, on the streets, in emergency rooms. When individuals with severe mental illness become disruptive or dangerous, law enforce- ment is forced onto the mental health front line because there is no alternative.32,33,34 The spread of “community policing,” which put more officers on the streets at roughly the same time hospitals were emptying, likely also contributed to increased contact between police and individuals with severe mental illness.35 In this environment, studies consistently find that 10%-20% of law enforcement calls involve a mental health issue.36 Most of these calls result from behavior that falls under the all-purpose umbrella of “public nuisance” – vagrancy, loitering or urinat- ing in public, trespass – or from individuals endangering themselves. When officers respond, the symptoms of psychosis, paranoia and/or suicidal thinking make these subjects less predictable and the threat they pose more difficult to assess. If alcohol intoxication and/or other substance use is involved – co-occurring conditions that are common among people with severe psychiatric disease – the situation becomes even less stable. The resulting mix can be deadly. The three independent databases in which fatal police shootings have been analyzed for mental health factors have all found mental illness reported in at least 25% of the deadly encounters. • The Washington Post: A real-time database of fatal police encounters reported that 25% of the deadly shootings by police nationwide from January 1 to September 30, 2015, involved “signs of mental illness.”23 • The Guardian: The Counted, also a real-time database, reported in an analysis of several subpopula- tions that 26% of the 464 fatal law enforcement encounters and deaths in custody from January 1 to May 31, 2015, included a mental illness factor.24 • The Jim Fisher True Crime blog, where former FBI investigator and blogger Jim Fisher analyzed le- thal police shootings in 2011: Mental illness was reported to be present in 25% of the officer-involved fatalities.25 Given that all three of these databases operate without benefit of complete data, their findings inevitably understate the magnitude of the problem. The Treatment Advocacy Center’s 2013 report on the topic surveyed academic journals, media reports and other unofficial sources and concurred with published speculation that “at least half of the people shot and killed by police each year in this country have mental health problems.”27 Where it exists, official research has reached similar conclusions. Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 5 “The transfer of responsibility for persons with mental illness from mental health professionals to law enforcement officers is both illogical and unfair and harms both the patients and the officers.” Treatment Advocacy Center • From Nevada: An analysis of officer-involved shootings in Las Vegas for the Office of Community Oriented Police Services of the U.S. Department of Justice found that 54% of fatal shootings involved “mentally impaired” individuals.37 • From Australia: Victoria state police launched five independent reviews to investigate police shoot- ing deaths from 1990 to 2004. The Office of Police Integrity found that more than half the people fatally shot by Victoria police “were experiencing a mental disorder.”12 Nationwide in Australia, from 1990 to 1997, at least one-third of the 41 people shot and killed by Australian police were found to be suffering from “a diagnosed mental illness (requiring psychiatric treatment) or depression.”14 • From the United Kingdom: The Police Complaints Authority in 2003 reported that just under half the people involved in police fatalities in England and Wales had mental health conditions.12 • From Canada: One-third of the individuals killed by law enforcement in British Columbia from 1980 to 1994 were found to have a recorded mental health history, most frequently schizophrenia.38 When officers’ perceptions of the decedents’ condition were taken into account, behaviors indicative of mental illness were reported present in half the fatalities.39 In short, whatever the actual number of deaths from officer-involved homicides, every credible source – official, academic or private – consistently finds that the sliver of the adult population with untreated severe mental illness (half the 3.3% of the total adult population with schizophrenia or severe bipolar disorder) is victim in not less than 25% of fatal police shootings – and more likely closer to half of them. Yet the question, “What is the role of mental illness in fatal police encounters?” is not being systemati- cally asked, much less answered. The FBI’s Supplementary Homicide Report to the Uniform Crime Report seeks information about where, how and why homicides are committed, including “justifiable homicides.” Victim and perpetrator age, sex, race and relationship of victim to offender and homicide circumstances such as “lovers’ triangle,” “child killed by babysitter” and “brawl due to influence of alcohol.” Not a single question solicits mental health information.40 6 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS THE UNDERCOUNTED Government Databases To some degree, the failure to track the role of mental illness in fatal police encounters is symptomatic of the failure to systematically track fatal police encounters, period. It is now common knowledge that the government doesn’t know how many people die every year in fatal encounters with law enforcement in the United States. As The Guardian reported in March 2015, “The top crime-data experts in Washington had determined that they could not properly count how many Americans die each year at the hands of police. So they stopped.”41 This should not surprise us. Federal laws mandating the collection of crime statistics – including homicide data – date from the 19th century but ran afoul of funding, staffing, political and other obstacles from the beginning (see Appendix D: History of Homicide Data Collection in the United States). It was not until 1930 that the first Uniform Crime Report (UCR) was published, initially by the International Association of Chiefs of Police but soon by the FBI, which continues to publish it annually today. “Justifiable homicides” (the term used by the FBI for fatalities that result from cause) were not reported in the UCR because, being justified, they were not crimes. It was only in 1991 – after the FBI had included “justifiable homicides” in its Supplemental Homicide Re- port – that the first report on fatal police encounters was incorporated in the UCR.42 Since then, five additional federal statistical systems have been authorized to count the number of fatalities resulting from police use of force in the U.S., and every one of them has fallen victim to one or more recurring circumstances. “To stop violent deaths, we must first understand all the facts.” Centers for Disease Control and Prevention • Absence of standard definitions No two government databases define fatal law enforcement encounters in the same way. Some in- clude all law enforcement officers, among them school police and security guards. Others exclude all law enforcement agencies except local and state police. A category like “suicide by cop” – a widely used term to describe fatalities that result when an individual appears to have intentionally provoked a lethal response from officers – has no useful, common definition at all. • Lack of universal participation Even when the databases themselves have been mandated by Congress, the flow of data to the data- base typically relies on voluntary participation, which is not universal. The BJS reported participation in its detailed 2015 audit of data collected by the ARD program. The “data quality profile” reported, for example, that 36 states had submitted ARD data every year since collection began in 2003, Geor- gia had never reported, Wyoming had reported only once, and Maryland had reported only twice.43 • Lack of standardized collection and reporting methods It was not participation alone that was uneven. Those states that had reported to the ARD developed their data in a variety of ways. In 2011, 26 of the 45 states reporting relied on media reports and online resources such as Google Alerts as their primary source of information about arrest-related deaths, but no two reporting centers used the exact same procedures even for searching media reports. The other 19 participating states relied instead on law enforcement reports, coroner reports and/or other sources, again with no uniformity. Even within reporting centers, different methods of collecting data were used from one case to the next.44 • Lack of central oversight More than 18,000 law enforcement agencies and 3,000 counties track fatal law enforcement en- counters in some way. In addition to collecting the data differently, they are required to report the information to a wide assortment of different data centers. During 2011, data were reported to a state criminal justice agency in 16 states, to a state law enforcement or public safety agency in 14 states, and to any one of six other offices in the remaining states that reported, including universities in four. No federal clearinghouse exists to establish standard definitions or methods to enforce consistency or participation, and no “standardized mode for data collection, definitions, scope, agency participa- tion” exists among the participating states.45 • Non-funding Congress has from the 1870s been more eager to mandate data collection than to fund it. The Na- tional Data Collection on Police Use of Force database was started in 1996 and terminated in 2000 for lack of funding. Funding for the Deaths in Custody Reporting Program (DCRP), under which the suspended ARD effort operated, expired after six years. Funding rarely trickles down to the individual law enforcement agencies that are the sources of the raw data. The UCR is one of two federal databases that collect general homicide data. The second is operated at the Centers for Disease Control and Prevention (CDC). The CDC’s Fatal Injury Reports – a component of the National Vital Statistics System (NVSS) – is the only U.S. government source of homicide statistics to which reporting is mandatory. The NVSS is based entirely on local death certificates that coroners or medical examiners are required by law to file for all “sudden unexpected deaths,” including homicides. Fatal law enforcement encounters are termed “homicide by legal intervention” in the NVSS, a category that also includes deaths from lethal force used in jails and prisons. The NVSS routinely reports more fatal law enforcement encounters than the UCR does – a disparity gen- erally attributed to differences in how the two agencies define the category and how they collect data – but there is no related checkbox in the reporting forms for “homicide by legal intervention.” Medical examiners/coroners may note that a homicide resulted from a law enforcement encounter, but this is not systematic, and the CDC has officially excluded police-involved shootings from its NVSS because “lack of an organized effort to collect detailed data on police use of lethal force results in little knowledge about these incidents.”46 Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 7 Some individual police departments and states have stepped into the breach left by the federal govern- ment. The New York City Police Department has been producing a detailed annual firearms discharge report that includes “adversarial conflict” with civilians since 1971 (2014: 35 shootings, 8 deaths).47 In New Jersey, the Newark Police Department publishes detailed lethal force data every month (year-to-date: 8 firearms discharges, no deaths).48 California in September 2015 launched Open Justice, a “transpar- ency initiative” to collect and report data on any death that occurs while an individual is being arrested, in custody of law enforcement or incarcerated.49 Meanwhile, the Public Safety Data Portal operated by the nonprofit Police Foundation includes selected data tables from national and local datasets. By mid- November 2015, the portal included data on officer-involved shootings in five U.S. cities and Los Angeles County.50 Not one of these databases collects or reports mental health factors. Emblematic of the obstacles to generating complete, reliable information about fatal law enforcement encounters, much less the role of mental illness in them, is the ARD program, a component of the DCRP. The DCRP was mandated by the Death in Custody Reporting Act (DCRA) of 2000, which required the U.S. Attorney General to make quarterly collections of individual death records for all inmates who die in state prisons or local jails and for “any person who is in the process of arrest” at the time of death, including those attempting to elude officers.17 Deaths due to suicide, accidental injury, illness or natural causes that occur during interaction with state or local law enforcement personnel were to be included; federal law enforcement agencies and prisons were exempt. Data collection began in 2000 with data on jail deaths. It was subsequently expanded to report on deaths in state prisons and juvenile correctional facilities and, in 2003, to arrest-related deaths. The first ARD program report was issued in 2007 for data collected from 2003 to 2005; a second was is- sued in 2011, for data collected from 2003 to 2009. Unique among all the databases, the ARD sought to identify mental health factors (see graphic). But the ARD has fared no better than its predecessors. Funding incentives included in the DCRA were misdirected to agencies that did not generate the data the BJS needed to collect52 and then, in 2006, the funding authorization expired altogether. After that time, the BJS has continued collecting and publish- ing data without funding, “as they represent a unique na- tional resource for understanding mortality in the criminal justice system,”53 but the data quality issues encountered by BJS eventually proved insurmountable. When an audit was conducted, it found the data available for the report “did not meet BJS data quality standards.” In March 2014, the bureau formally “suspended data collection and publi- cation of the ARD data until further notice.”17 Following renewed attention to fatal police shootings after the death of Michael Brown in Ferguson, Missouri, Congress voted in December 2014 to restore funding for the DCRA of 2000. The BJS is now developing a pilot ARD program in which Google Alerts, media and open-source websites are used to identify potentially eligible cases, and individual police departments and medical examin- ers are contacted directly for details.54 In the pilot, mental health data continue to be solicited and retained – creat- ing the possibility that statistical information about the role of mental illness in arrest-related deaths someday will be available from the U.S. government. BJS will resume continuous data collection in 2016 after the pilot is complete.55 In the meantime, two bills introduced in Congress in 2015 represent new attempts to legislate 8 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS Sample questions from the Arrest-Related Death Report Bureau of Justice Statistics51 statistical accuracy: the National Statistics on Deadly Force Transparency Act of 2015 introduced by Rep. Steve Cohen (D-TN), and the Police Reporting of Information, Data and Evidence Act, co-introduced by Senators Barbara Boxer (D-CA) and Cory Booker (D-NJ). Until one or more of these initiatives comes to fruition, the U.S. government will not be a source of cred- ible statistics surrounding deadly law enforcement encounters, and the role of mental illness in them will not merit so much as a footnote. Independent Databases Having amply reported on the failure of the government to accurately or completely document officer- involved homicides, a number of news organizations, private organizations and individual bloggers have dedicated themselves to doing the job themselves. Like government agencies, they employ differing methods that result in death tallies that can vary significantly. Most begin with news reports. Many sup- plement the news with legwork – reporters assigned to track down and confirm fatal shooting informa- tion, going door to door to interview neighbors if need be. Some supplement with crowdsourcing, asking the public to submit online forms to report incidents from their own communities. The Washington Post unveiled its database of “those incidents in which a police officer, while on duty, shot and killed a civilian” on May 30, 2015.56 The database is built from news reports, police records, open sources on the Internet, a decade of FBI and CDC records and the newspaper’s own reporting. The Guardian launched what it calls The Counted on June 1, 2015, “because the federal government does not currently publish a comprehensive database.”57 The Counted combines its own reporting with veri- fied crowdsourcing in an interactive platform the news site claims “is recording every killing by police in the U.S.” Smaller, regional media outlets also have stepped in to fill the gaps in data nationally or locally. The Las Vegas Review Journal in Nevada hosts “Deadly Force: When Las Vegas Police Shoot, and Kill,” searchable for dozens of factors in police shootings dating from 1990.58 The Portland Press Herald in Maine operated Police Use of Deadly Force in Maine 1990-2012, drawing from attorney general records to report “all incidents in which a person was shot at, injured or killed by Maine police agencies, or in which a person committed suicide in the course of an armed confrontation with Maine police,” dating back to 1990.59 A National Public Radio station in Southern California has launched a searchable database of officer- involved shootings in Los Angeles County.60 Nonprofits and some individuals also operate databases that endeavor to quantify and identify dead civilians from lethal encounters with law enforcement. Among those publishing current data are the following: • Fatal Encounters: “people who are killed through interactions with police,” from 2000 to present61 • Gun Violence Archive: “officer-involved shootings,” from 2013 to present62 • Killed by Police: fatal use of deadly force by U.S. law enforcement officers, whether “in the line of duty or not, regardless of reason or method,” found to be “justified” or not, from May 2013 to present63 Among all the official and unofficial homicide and police homicide databases, only the Preventable Trag- edies Database published by the Treatment Advocacy Center focuses exclusively on violent acts in which the victim or the perpetrator is reported to have a severe mental illness, typically untreated.64 Starting with published stories from the early 1980s through the present, the database can be searched for inju- ries and deaths in which severe mental illness was reported, including police shootings. Because many incidents in the database predate the Internet and all rely upon media sources in which a searchable term for psychiatric factors is included, the Preventable Tragedies Database – like all the other official and unofficial databases – is doomed to be incomplete. Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 9 10 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS DISCUSSION AND RECOMMENDATIONS The national conversation around fatal law enforcement encounters has been largely focused on the number of fatalities. Until the conversation expands to include the dynamic role of severe mental illness in these tragedies, promising avenues for reducing fatal officer-involved deaths and the many associated social costs remain largely unexplored. Exploration begins with defining the problem. Accurate, reliable data about the frequency, circumstanc- es and results of the use of lethal force by law enforcement are essential to taking that step. The recurring issues that have undermined every federal effort to operate a reliable federal database of fatal police shootings must be addressed by Congress, a system for tracking and incentivizing reporting of all use of deadly force by law enforcement must be established and maintained, and questions to identify psychiat- ric factors must be incorporated into official surveys used to capture data about both fatal and non-fatal police shootings. It warrants noting that many proven or promising police practices for reducing fatal law enforcement encounters exist, including strategies specifically addressing the unique challenges and risks associated with responding to individuals in psychiatric crisis. Examining such practices in detail is beyond the scope of this report, but the following brief descriptions illustrate that proven and practical options exist. • Restrictive deadly force policies In 1982, an 18-month study of the police use of deadly force entitled A Balance of Forces was pub- lished based on the Police Use of Deadly Force joint project of the FBI, National Institutes of Justice and International Association of Chiefs of Police. The study concluded “there is sufficient evidence to indicate that lives can possibly be spared if certain administrative guidelines are clearly developed, presented, understood, and enforced by law enforcement management.”65 A set of “model” deadly force guidelines was proposed. More than 30 years later, President Obama’s Task Force on 21st Centu- ry Policing released proposals for some of the same measures. Wider implementation of clear deadly force policies and the de-escalation tactics they typically incorporate is a proven practice that would reduce civilian fatalities with or without the presence of severe mental illness. • “Chronic Consumer Stabilization” The city of Houston in 2007 set out to “closely examine recent deadly encounters and identify pos- sible solutions” for reducing encounters between residents with severe and persistent mental illness and a history of frequent encounters with the Houston Police Department – the at-risk population de- scribed in this report. The Chronic Consumer Stabilization Initiative (CCSI) began as a six-month pilot in February 2009 by engaging 30 of the city’s “most problematic consumers” for the police depart- ment in “needed mental health services in order to reduce contacts with law enforcement.” The pro- gram, which was expanded to 60 participants following the pilot, significantly decreased monthly calls to police for service, overall interactions between the clients of the initiative and law enforcement, the number of deadly encounters between police and mentally ill subjects, and hospital admissions. One of the principle tactics was diverting 911 calls by providing clients in the initiative with cell phone numbers for their case managers. After the trial, assisted outpatient treatment (AOT) was added to the program to increase treatment compliance.66 CCSI achieved its results while dramatically reducing taxpayer costs for participating clients – a win-win that any jurisdiction should covet. • De-escalation training and implementation The Memphis Police Department in 1988 pioneered a program for providing intensive training to active-duty officers on mental illness conditions, symptoms, treatments and de-escalation tactics, in- cluding proven techniques to communicate with and calm agitated individuals in acute psychiatric crisis. Trademarked “Crisis Intervention Training (CIT),” the model by 2013 was in use where approxi- mately half the U.S. population lives.67 De-escalation techniques such as those used in CIT have been documented to produce “positive outcomes for police, offenders, and the community,” including a decreased likelihood of arrest, an increased likelihood the individual will receive treatment and “sig- nificantly” greater likelihood that verbal engagement or negotiation will be the highest level of force used in encounters between law enforcement and individuals in psychiatric crisis.68,69,70 • Co-responder teams for psychiatric emergency response Los Angeles in 1991 took the concept of providing specialized mental health training to law en- forcement a step further and pioneered a “co-responder team” concept called SMART (Systemwide Mental Assessment Response Team). Initially a pilot, SMART paired specially trained officers with mental health professionals to respond to psychiatric emergency calls. More than 60% of the 101 subjects engaged by SMART during a sample period were characterized by “acute and chronic severe mental illness, a high potential for violence, a high incidence of serious substance abuse, and long histories with both the criminal justice and the mental health system.” Yet only 2% of the subjects were arrested, and 19% of the encounters ended without transportation to either a hospital or jail because of the field evaluation and intervention. The authors concluded that the approach resulted in more treatment and less incarceration for the population.71 More recently, San Diego County in 2015 funded co-responder PERT (Psychiatric Emergency Response Team) units “to provide the most clinically appropriate resolution to the crisis by linking people to the least restrictive level of care that is appropriate and to help prevent the unnecessary incarceration or hospitalization of those seen.”72 Outcome research on co-responder teams is relatively scant, but it defies logic that bringing a mental health professional to the scene of a mental health crisis could worsen the situation. As effective as these and other policing strategies may be, however, they all require that individual with mental illness deteriorate sufficiently to become a police incident before they are activated. The most proven and predicable practice of all for reducing fatal police shootings and the role of mental illness in fatal police shootings – and throughout the overwhelmed criminal justice system – is far more straight- forward: treat the symptoms and avoid the encounter altogether. The Treatment Advocacy Center recommends the following proven practices to improve access to men- tal illness treatment and reduce encounters between law enforcement and individuals in psychiatric crisis. • Increase the number of hospital beds for acute and chronic psychiatric treatment. The number of public psychiatric beds in America has plunged more than 90% since the 1950s while the U.S. population has nearly doubled. At 14.1 beds per 100,000 people, the per capita bed popu- lation at last report stood at the same level it did in 1850, when the concept that imprisoning those with mental illness was inhumane first took hold.73 In 1955, when there were 350 beds for 100,000 people, about 4% of the inmate population in U.S. prisons and jails was mentally ill. Today, it is well established that roughly 20% of all inmates have a serious psychiatric disease, but individual facilities report that up to 50% of the prisoners in their facilities have a mental illness. In 44 states, at least one jail or prison holds more inmates with mental illness than the state’s largest remaining psychiatric hospital.74 More than 75 years ago, Lionel Penrose wrote a seminal study theorizing that if psychiatric hospital populations are reduced, prison populations will grow, and vice versa.75 This theory remained a matter of controversy for decades, even as that very phenomenon unfolded. In recent years, a new generation of researchers has examined the relationship between psychiatric hospital beds and incarceration over extended periods of time and found statistically significant inverse relationships between them.76,77 Incarceration is the ultimate result of an encounter with law enforcement that leads to arrest. If adequate hospital treatment options reduce incarceration, it is because fewer people with mental illness are being arrested – and being at risk for a fatal encounter in the process. • Expand the use of court-ordered outpatient treatment for at-risk individuals in the community. Assisted outpatient treatment or AOT authorizes court-ordered mental health treatment – including medication – for individuals with severe mental illness who, because of their inability to stay in treatment voluntarily, have a history of poor outcomes (e.g., repeated hospitalization, incarceration, suicide at- tempts). AOT has been deemed an evidence-based practice for reducing crime and violence by the U.S. Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 11 Department of Justice, U.S. Health and Human Services and the Substance Abuse and Mental Health Services Administration. The House Appropriations Committee has approved $15 million to fund up to 50 new AOT programs in the current fiscal year, and Congress has authorized the same amount for ad- ditional programs for the next three years. With funding, this will expand local use of AOT nationwide and reduce the interaction of law enforcement with the high-risk participants in the programs. • Reform treatment laws that erect barriers to treatment for high-risk individuals. Deinstitutionalization resulted in part from the development of more effective medications and also from an array of legal, social, economic and political forces that converged to produce wholesale changes in state civil commitment laws. The purpose of these changes was to make it more difficult for authorities to use involuntary treatment options, including hospitalization, for individuals who were in psychiatric crisis but unable or unwilling to seek treatment. Thirty states have eased these restric- tions to some degree in the last 20 years as the consequences and costs of leaving this population to deteriorate in the community without treatment have become evident, but barriers remain. Even in states where reform has taken place, many legal barriers exist to timely and effective treatment of the diseases most likely to result in law enforcement attention. Among these barriers are the following: o Five states (Connecticut, Maryland, Massachusetts, New Mexico and Tennessee) remain with- out laws specifically authorizing the use of court-ordered outpatient treatment (AOT) for at-risk individuals with a history of criminal justice involvement and/or other negative consequences of non-treatment. All five since 2012 have actively considered legislation to add AOT to their treatment options. It is time for them to enact authorizing laws. o Thirty-four states fail to recognize the inability to seek needed psychiatric care and to make an informed medical decision as a basis for civil commitment. Reforms must be passed to make timely treatment possible in those states. 12 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS METHODOLOGY The risk of a fatal encounter with law enforcement for persons with untreated severe mental illness rela- tive was calculated as a ratio of two proportions: the proportion of total contacts in which a civilian was approached or stopped by police that resulted in a fatality for persons with severe untreated mental ill- ness, divided by the same proportion for persons without severe untreated mental illness. In 2011, the most recent year for which police contact data is available, 14.7% of U.S. residents age 16 or older were stopped or approached by police.78 Applying this percentage to the U.S. Census estimate for the total U.S. population age 16 and over in 2011 results in a total of 36,199,814 police-initiated contacts. The estimated number of fatal law enforcement encounters for that year is 999, which amounts to 2.76 civilian deaths per 100,000 police-initiated contacts.79 Of these fatalities, by the most conservative estimates, 25% of the deceased will be identified publicly as suffering from a severe mental illness, typically untreated.23,24,25 With 2% of U.S. adults estimated to have untreated severe mental illness, the death rate for individuals with serious mental illness killed during law enforcement interactions is 0.0345%, or 16 times greater than the death rate of 0.00211% for those without such a condition. The findings should be considered in the context of their data limitations. The number of total fatali- ties, mentally ill decedents and untreated decedents among them are estimates developed and cross- referenced from a survey of government, academic, open-source and media resources in the U.S. and other countries. Additionally, risk ratios may be distorted by the small number of fatalities relative to the millions of police contacts. Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 13 The Treatment Advocacy Center gratefully acknowledges the research and editorial contributions of Dr. E. Fuller Torrey, principal investigator for our 2013 examination of the role of mental illness in fatal law enforcement encounters; the technical feedback provided by the Bureau of Justice Statistics and the statistical support of Jennifer R. Schroeder Ph.D., Schroeder Statistical Consulting LLC. ACKNOWLEDGMENTS 14 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS Federal Government Homicide Databases (listed chronologically from inception) Uniform Crime Report (UCR) – based on crime reports submitted by local police agencies – FBI (from 1929) • Voluntary at the federal level but mandated by 35 individual states • Incomplete: the UCR is submitted by an estimated 97% of local and state law enforcement agencies reported in 2014; level of detail reported varies among agencies; federal agencies not required to report • Defines “justifiable homocide” as a killing committed, either by police or private citizens, while the victim is in the process of committing a felony • Includes “justifiable homicides” among the incidents subject to a Supplementary Homicide Report (SHR) to the UCR (since 1991). Submission of an SHR is voluntary, and data are incomplete (e.g., Florida reported 98 “law enforcement homicides” to the Arrest-Related Deaths program from 2003 to 2005 but none to the SHR)78 • Neither the UCR nor the SHR includes queries about mental illness Fatal Injury Reports within the National Vital Statistics System (NVSS) – data on all sudden, unexpected deaths, supplied by coroners and medical examiners on locally filed death certificates – Centers for Disease Control and Prevention (from 1933, when uniform collection of death certificates began) • Mandatory: state laws require that a medical examiner or coroner investigate all homicides and other sudden, unexpected deaths • Universal: all U.S. jurisdictions report • Describes “homicide by legal intervention”; does not request information about such deaths but accepts volunteered information • Does not include mental health queries Police Use of Deadly Force, 1970-1979 – data on lethal use of force by law enforcement officers during a nine-year period – joint project of the FBI, National Institute of Justice and International Association of Chiefs of Police; stored in the National Archive of Criminal Justice Data (1982) • Voluntary • Incomplete by design: data from 57 cities of 250,000 people or more • Included fatal law enforcement encounters involving on-duty and off-duty officers • Did not include mental health queries National Incident-Based Reporting System (NIBRS) – a component of the UCR that collects incident-based details about crimes known to the police – FBI (from 1984) • Voluntary • Implemented but incomplete: collects and reports crime data from approximately half the states • Excludes fatal law enforcement encounters because “justifiable homicide is not an actual ‘offense’” • Does not include mental health queries National Data Collection on Police Use of Force Database – database and annual summary publication on the use of excessive force by law enforcement officers – Bureau of Justice Statistics and National Institute of Justice with International Association of Chiefs of Police (from 1996-2000) • Mandated by the Violent Crime Control and Law Enforcement Act of 1994 • Focused on “police use of force,” including fatal encounters • Published once: Police Use of Force in America: 2001 • Ended in 2000 due to lack of funding • Did not include mental health queries, but some agencies provided such information in the comments section, and data about “emotionally disturbed subjects” were included in the published report APPENDIX A Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 15 Deaths in Custody Reporting Program (DCRP) – an annual national census of civilians who die during the process of ar- rest or while in the custody of state or local law enforcement personnel; includes accidental, suicidal, homicidal and natural causes of death – Bureau of Justice Statistics (from 2000) • Mandated but not funded or enforced • Implemented for jail and prison deaths; suspended for arrest-related deaths (see below) • Includes mental health queries Arrest-Related Deaths (ARD) Program – a component of the DCRP that quantifies and describes the circumstances sur- rounding civilian deaths that take place during an arrest or while in the custody of law enforcement – Bureau of Justice Statistics (from 2003) • Mandated but not funded • Produced two reports based on incomplete FBI data (2009 and 2011) • Suspended in 2014 because the source data were too inconsistent and incomplete to meet Bureau of Justice Statistics data quality standards; however, data collection continues utilizing expanded sources • Included mental health queries National Violent Death Reporting System (NVDRS) – state-based surveillance system that collects detailed information regarding violent deaths from multiple sources (law enforcement, coroners and medical examiners, vital statistics, crime laboratories) about all violent deaths – Centers for Disease Control and Prevention (CDC), building on the privately funded National Violent Injury Statistics System pilot program (from 2003) • Voluntary • Incompletely implemented: currently operates in 32 states • Includes “death from legal intervention” by on-duty law enforcement officers • Linked to the NVDRS Restricted Access Database, a restricted online database for public health research • Does not include mental health queries but does include queries about more than 20 other personal factors Web-based Injury Statistics Query and Reporting System (WISQARS) – searchable online tool for accessing fatal and non- fatal injury, violent death and cost-of-injury data from the NVDRS and the NVSS Fatal Injury Reports – CDC National Center for Injury Prevention and Control (from ca. 2000) • Based on mandatory reports to the CDC • Universal: data from 1981 or later • Includes “death from legal intervention” data from the NVDRS and NVSS • Does not include mental health queries Wide-ranging Online Data for Epidemiological Research (WONDER) Database – searchable online tool of mortality and population data from the Underlying Cause of Death database – CDC Center for Surveillance, Epidemiology and Laboratory Services (from mid-1990s) • Based on mandatory reports to the CDC • Universal • Includes “death from legal intervention” • Does not include mental health queries Police Data Initiative – a “public-safety open data portal” serving as a “central clearinghouse option,” containing data on a variety of police-community interactions, including use of force – White House (from 2015) • Voluntary • Incomplete: 26 police departments participating (as of October 27, 2015) • Contents under development APPENDIX A (continued) 16 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS Independent Databases Tracking Fatal Law Enforcement Encounters Deadly Force – data for 1990-2011 – media (Las Vegas Review Journal) • Used information from shooting reports obtained from each law enforcement agency to report dozens of details about fatal law enforcement encounters, including the agency involved, incident type, outcome, type of force used and others • Does include searchable fields for incident calls involving “mentally ill person,” “suicide or suicide attempt” and “shooting subject” who was “mentally ill” or “suicidal” Deadspin U.S. Police Shootings Data – data for 2011 to 2014 – media (Gawker Media) • Used a public submission form in an effort to “catalogue every police-involved shooting in America” for 2011-2013; collected date, name, age, gender, race/ethnicity, injured/killed, armed/unarmed, city, county, state, agency and number of shots • Provides a brief summary and a link to a story about the incident. • Did not report mental health information Fatal Encounters – data for 2000 to present – media (Reno News & Review) • Uses public information and crowdsourcing to compile a national database of “people who are killed through interactions with police” • Does report mental health information but inconsistently Gun Violence Archive – data for 2013 to present – nonprofit (Gun Violence Archive) • Uses automated queries, manual research, police blotters, police media outlets and others to identify and validate incidents of gun-related violence in America • Does not report mental health information Jim Fisher True Crime – data for 2011 – blog (Jim Fisher, former FBI agent, author and professor of criminal justice at Edin- boro University of Pennsylvania: “I collected this data myself because the U.S. Government doesn’t”) • Used public information on the Internet to identify any police-involved shooting in 2011 • Did not report mental health information Killed by Police – data for May 1, 2013 to present – Facebook page • Uses public information from “corporate news” to report the occurrence of deadly use of force U.S. law enforcement officers, whether “in the line of duty or not, regardless of reason or method,” found to be “justified” or not • Does not specifically report mental health information but links to news coverage that may contain such information Mapping Police Violence – data for 2014 to present on fatal police shootings of black civilians – private (research collaborative) • Aggregates data from KilledByPolice.net, FatalEncounters.org and the Deadspin U.S. Police Shootings Database; validates these data with social media, obituaries, police reports and other sources • Does not report mental health information Preventable Tragedies Database – data from before 2000 to present – nonprofit (Treatment Advocacy Center) • Data from news reports on violent acts associated with severe mental illness, including law enforcement shootings in which either the officer or a civilian is injured or killed by any method • Includes only incidents in which a mental health condition is publicly reported The Counted – data for 2015 only – media (The Guardian) • Data from news reports and “verified crowdsourced information” • Periodically does report mental health factors; reported that, through May 31, 2015, mental illness was present in 26% of deadly law enforcement encounters in 2015 The Homicide Report – data for 2007 to present – media (Los Angeles Times) • Data from Los Angeles County Coroner’s Office, supplemented by reporting and crowdsourcing • Reports on all homicide data in Los Angeles County but also includes a search function for officer-involved homicides in Los Angeles County • Does not report mental health information APPENDIX B Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 17 The Washington Post – data for 2015 only – media (The Washington Post) • Uses news reports, police records, open sources on the Internet and original reporting • Reports only on officer-involved shootings while on duty; also tracks officer fatalities • Does report mental health information by systematically analyzing and reporting the presence of “signs of mental illness” in “deadly shootings” by on-duty police; reported mental illness signs in 26% of fatal law enforcement encounters through September 30, 2015 List of killings by law enforcement officers in the United States – data for 2009 to present – (Wikipedia) • Lists people killed by “nonmilitary law enforcement officers ... whether in the line of duty or not, and regardless of reason or method ...; implies neither wrongdoing nor justification on the part of the person killed or the officer involved” • Does not report mental health information APPENDIX B (continued) 18 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS Glossary Many terms are used to describe the killing of a civilian by a law enforcement officer, and no two describe the same circum- stances, even within the same federal agency. The following terms are often used interchangeably to describe deadly encoun- ters between law enforcement agents and civilians. However, because each term applies to a different category of victims killed under differing circumstances, statistics reported for them are not directly comparable. Arrest-related deaths • Established by: Bureau of Justice Statistics pursuant to the Death in Custody Reporting Act of 2000; is not synonymous with “in the process of arrest” as used by law enforcement, which involves Mirandizing and handcuffing • Definition: “civilian deaths that occurred prior to, during or following an arrest event or noncriminal incident and that were attributed to: o “any use-of-force by state or local law enforcement” o “injuries sustained while attempting to elude law enforcement or injuries incurred while in custody” o “self-imposed events, such as suicides, accidents caused by the decedent, and intoxication” o “medical conditions or illness”2 Deaths in the process of arrest • Established by: Congress in the Death in Custody Reporting Act of 2000 • Definition: “all deaths of persons in the physical custody or under the physical restraint of law enforcement officers”; refined and expanded as “arrest-related deaths” Homicide by legal intervention (legal intervention deaths) • Established by: Centers for Disease Control and Prevention for use in Fatal Injury Reports and the National Violent Death Reporting System • Definition: decedents killed by a police officer or other peace officer (a person with specified legal authority to use deadly force), including military police, acting in the line of duty; excludes homicides by off-duty officers and legal executions; includes homicides by federal officers Justifiable homicide • Established by: FBI in the Uniform Crime Report • Definition: “the killing of a felon by a peace officer in the line of duty (or) the killing (during the commission of a felony) of a felon by a private citizen”; excludes fatalities involving off-duty officers or non-felon suspects and deaths that otherwise would qualify when they occur on federal or Indian lands Police shootings, or officer-involved shootings • Established by: common usage • Definition: popularly used to describe shootings by all law enforcement officers, whether police or not; technically includes all weapons discharge, such as accidental weapon discharge, the shooting of animals, shootings that injure but do not kill and shootings that miss their target Use of lethal/deadly force • Established by: common usage • Definition: a use of force that a reasonable person would consider likely to cause death or serious bodily harm; in practice, typically refers to use of a firearm APPENDIX C Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 19 History of Homicide Data Collection in the United States 1870 Congress authorizes the collection of crime statistics in the United States, but “plans never got off the ground”. 1871 The National Police Association (forerunner of the Interna- tional Association of Chiefs of Police) calls for the collection of crime statistics for police use. 1900 Ten states, the District of Columbia and several cities begin reporting deaths to the Census Bureau, including “prevent- able causes of death,” homicide among them. 1927 The International Association of Chiefs of Police (IACP) forms a Committee on Uniform Crime Records. An advisory group that includes FBI Director J. Edgar Hoover and a technical staff is assembled. 1929 The IACP develops and adopts the voluntary Uniform Crime Report (UCR) program to classify and collect crime data and report reliable uniform crime statistics for the nation. Fatal law enforcement encounters are not included as a category because, as “justifiable homicides,” they are not crimes. 1930 In January, the IACP publishes the first UCR, with police de- partments from 400 cities in 43 states participating. In June, the FBI is appointed to collect, publish and archive the UCR. 1933 Uniform reporting of births and deaths, including homicides, to the Census Bureau, expands to all states. 1940 The UCR begins accepting voluntary reports of fatal law en- forcement encounters from police departments. 1946 Responsibility for reporting births and deaths is transferred from the Census Bureau to the Public Health Service and the National Office of Vital Statistics (subsequently to merge as the National Center for Health Statistics and, in 1987, become part of the Centers for Disease Control and Prevention [CDC], where it remains to this day). 1949 CDC reporting expands to include “injury by intervention of police.” 1962 The first Supplementary Homicide Report with incident and victim/perpetrator information is published. “Justifiable homi- cides” are not included. 1984 Development of the National Incident-Based Reporting System begins in an effort to gather more information about crimes. Fatal law enforcement interactions are not included. 1991 Deadly law enforcement interactions are reported in the first “justifiable homicide” data report by the UCR; includes data available since 1976. 1994 The Violent Crime Control and Law Enforcement Act of 1994 mandates that the Department of Justice and the attorney general produce an annual summary of uses of “excessive” force by law enforcement. 1996 The first National Data Collection on Police Use of Force Report is published in response to the Violent Crime Control and Law Enforcement Act of 1994. 1999 A short-lived, privately funded pilot program, known as the National Violent Injury Statistics System, begins collecting data on “homicides and other types of violent death (sui- cide).” 2000 The Deaths in Custody Reporting Act of 2000 is enacted with funding in reaction to prison confinement deaths. The Bureau of Justice Statistics (BJS) begins quarterly data collection to cover all inmate deaths in local jails to comply with require- ments of the law. 2002 Congress allocates funds to the CDC to initiate the National Violent Death Reporting System. 2003 Lawmakers insert a provision to the Deaths in Custody Reporting Act of 2000 to create the Arrest-Related Deaths (ARD) program on deaths caused by accident, suicide, homi- cide or natural causes (including illness) that occur in the pro- cess of arrest, during transfer or while individuals are detained in jail or prison. Data collection is initiated. 2006 Federal funding for the Deaths in Custody Reporting program expires. The BJS continues collecting data without funding. APPENDIX D 20 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS 2014 March Publication of the Arrest-Related Deaths Report is suspended after the BJS “conducted an assessment of the validity and reliability of the ARD data” and found it “did not meet BJS data quality standards.” December The Deaths in Custody Reporting Program is reauthorized as the Deaths in Custody Reporting Act, which requires states to report to the Department of Justice any time a civilian is killed by a police officer while in custody or during the course of an arrest. 2015 January Rep. Steve Cohen (D-TN) calls for the National Statistics on Deadly Force Transparency Act of 2015, which requires the attorney general to collect data on deadly force. The act includes what data should be captured and a Byrne Justice Assistance Grant penalty for failure to report. May President Obama announces the White House Police Data Initiative, in which 21 police departments nationwide agree to release data on use of force, pedestrian and vehicle stops, and officer-involved shootings. June U.S. Senators Cory Booker (D-NJ) and Barbara Boxer (D-CA) introduce the Police Reporting of Information, Data and Evidence Act. APPENDIX D (continued) Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 21 ENDNOTES 1. Treatment Advocacy Center. (2014). Prevalence of untreated serious mental illness by state. Retrieved from http://www.treatmentadvocacycenter.org/storage/documents/Research/sz%20and%20bp%20numbers%20by%20state.pdf 2. Wang, P. S., Lane, M., Olfson, M. l., Pincus, H. A., Wells, K. B., & Kellser, R. C. Twelve-month use of mental health services in the United States: Results from the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 629–640. Available from http://www.ncbi.nlm.nih.gov/pubmed/15939840 3. Treatment Advocacy Center. (n.d.). Consequences of non-treatment. Retrieved November 11, 2015, from http://www.treatmentadvocacycenter.org/problem/consequences-of-non-treatment 4. National Institute of Mental Health. (n.d.). Bipolar disorder among adults. Retrieved November 11, 2015, from http://www.nimh.nih.gov/health/statistics/prevalence/bipolar-disorder-among-adults.shtml 5. Chappell, D. (Ed.). (2013). Policing and the mentally ill: International perspectives. Boca Raton, FL: CRC Press. 6. Torrey, E. F., Zdanowicz, M. T., Kennard, A. D., Lamb, H. R., Eslinger, D. F., Biasotti, M. C., & Fuller, D. A. (2014). The treatment of persons with mental illness in prisons and jails: A state survey. Arlington, VA: Treatment Advocacy Center. 7. Lamb, H. R., Shaner, R., Elliott, D. M., DeCuir, W. J., & Foltz, J. T. Outcome for psychiatric emergency patients seen by an outreach police-mental health team. Psych Services, 46(12), 1267–1271. 8. New York State Office of Mental Health. (2005). Kendra’s Law: Final report on the status of assisted outpatient treatment. Retrieved from http://bi.omh.ny.gov/aot/files/AOTFinal2005.pdf 9. Swartz, M. S., Swanson, J. W., Steadman, H. J., Robbins, P. C., & Monahan, J. (2009). New York State assisted outpatient treatment program evaluation. Durham, NC: Duke University School of Medicine. Retrieved from http://www.macarthur. virginia.edu/aot_finalreport.pdf 10. MacLeod A., Pate, L., & Dempsey, C. (Presenters). (2012.) Innovative law enforcement strategies for interacting with people with mental illnesses that frequently require emergency and crisis services. (webinar). Retrieved from https:// csgjusticecenter.org/law-enforcement/webinars/webinar-archive-innovative-law-enforcement-strategies-for-interacting-with- people-that-frequently-require-emergency-and-crisis-services/ 11. Thomas, S. (2013). Core requirements of a best practice model for police encounters involving people experiencing mental illness in Australia: A Victorian perspective. In D. Chappell (Ed.), Policing and the mentally ill: International perspectives (pp. 121–136). Boca Raton, FL: CRC Press. 12. Kesic, D. (2013). The role of mental disorders in use of force incidents between the police and the public. In D. Chappell (Ed.), Policing and the mentally ill: International perspectives (pp. 153–170). Boca Raton, FL: CRC Press. 13. Ibid. 14. Clifford, K. (2013). Mental health crisis and interventions and the politics of police use of deadly force. In D. Chappell (Ed.), Policing and the mentally ill: International perspectives (pp. 171–195). Boca Raton, FL: CRC Press. 15. Kesic, The role of mental disorders. 16. Kesic, The role of mental disorders. 1 7. Planty, M., Burch, A. M., Banks, D., Couzens, L., Blanton, C., & Cribb, D. (2015). Arrest-Related Deaths program: Data quality profile. Washington, DC: Bureau of Justice Statistics. Retrieved from http://www.bjs.gov/content/pub/pdf/ardpdqp. pdf 18. Wallace, J. L., & Levy J. K. (2006). Population characteristics of feral cats admitted to seven trap-neuter-return programs in the United States. Journal of Feline Medicine and Surgery 8(4): 279–284. Available from http://www.ncbi.nlm.nih.gov/ pubmed/16603400 19. The Washington Post. (n.d.). [Data file]. Retrieved November 11, 2015, from https://www.washingtonpost.com/graphics/ national/police-shootings/ 20. Swaine, J., & Laughland, O. (2015, November 16). Number of people killed by US police in 2015 at 1,000 after Oakland shooting. The Guardian. http://www.theguardian.com/us-news/2015/nov/16/the-counted-killed-by-police-1000 22 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS 21. Planty et al., Arrest-Related Deaths program. 22. Ibid. 23. Lowery, W., Kindy, K., Alexander, K. L., & Rich, S. (2015, May 30). Distraught people, deadly results. The Washington Post. http://www.washingtonpost.com/sf/investigative/2015/06/30/distraught-people-deadly-results/ 24. Swaine, J., Laughland, O., & Lartey, J. (2015, June 1). Black Americans killed by police twice as likely to be unarmed as white people. The Guardian. http://www.theguardian.com/us-news/2015/jun/01/black-americans-killed-by-police-analysis 25. Fisher, J. (2013, December 25). Police involved shooting statistics: A national one-year summary. [Blog post]. http:// jimfishertruecrime.blogspot.com/2012/01/police-involved-shootings-2011-annual.html 26. White, M. D. (2006). Hitting the target (or not): Comparing characteristics of fatal, injurious, and noninjurious police shootings. Police Quarterly, 9(3), 303–330. doi:10.1177/1098611105277199 27. Torrey, E. F., Kennard, A. D., Eslinger, D. F., Biasotti, M. C., & Fuller, D. A. (2013). Justifiable homicides by law enforcement officers: What is the role of mental illness? Arlington, VA: Treatment Advocacy Center and National Sheriffs’ Association. 28. Ibid. 29. Davis, A. C., & Lowery, W. (2015, October 7). FBI director calls lack of data on police shootings ‘ridiculous,’ ‘embarrassing.’ The Washington Post. https://www.washingtonpost.com/national/fbi-director-calls-lack-of-data-on-police-shootings- ridiculous-embarrassing/2015/10/07/c0ebaf7a-6d16-11e5-b31c-d80d62b53e28_story.html 30. Killed by Police 2015. (2015). [Data file]. Retrieved November 16, 2015, from http://killedbypolice.net/ 31. Planty, et al., Arrest-Related Deaths program. 32. Torrey, E. F. (2014). American psychosis: How the federal government destroyed the mental illness treatment system. New York: Oxford University Press. 33. Thomas, Core requirements of a best practice model. 34. Lamb et al., Outcome for psychiatric emergency patients. 35. Thomas, Core requirements of a best practice model. 36. Chappell, D (Ed.). (2013). Policing and the mentally ill. 37. Stewart, J. K., Fachner, G., King, D. R., & Rickman, S. (2012). Collaborative reform process: A review of officer-involved shootings in the Las Vegas Metropolitan Police Department. Washington, DC: Department of Justice Community Oriented Policing Services and CNA. http://www.cops.usdoj.gov/pdf/e10129513-Collaborative-Reform-Process_FINAL.pdf 38. Parent, R. B. (1996). Aspects of police use of deadly force in British Columbia: The phenomenon of victim-precipitated homicide (Master’s thesis). Retrieved from research repository: http://summit.sfu.ca/item/7114 39. Kesic, The role of mental disorders. 40. Federal Bureau of Investigation. (n.d.) Supplementary homicide report (OMB Form No. 1110-0002). Retrieved from https:// www.fbi.gov/about-us/cjis/ucr/nibrs/addendum-for-submitting-cargo-theft-data/shr 41. McCarthy, T. (2015, March 18). The uncounted: Why the U.S. can’t keep track of people killed by police. The Guardian. http://www.theguardian.com/us-news/2015/mar/18/police-killings-government-data-count 42. Publication began in 1991, but data collection began in 1976. 43. Planty et al., Arrest-Related Deaths program. 44. Ibid. 45. Ibid. ENDNOTES (continued) Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS n 23 46. Shields, R. T., & Ward, B. W. (2008). Comparison of the National Violent Death Reporting System and Supplementary Homicide Report: Potential benefits of integration. Justice Research and Policy, 10(2). http://phpa.dhmh.maryland.gov/ ohpetup/docs/Comparison_NVDRS_SHR-2008.pdf 47. New York City Police Department. (2014). Annual Firearms Discharge Report. Retrieved from http://www.nyc.gov/html/ nypd/downloads/pdf/analysis_and_planning/nypd_annual_firearms_discharge_report_2014V3.pdf 48. Newark Police Department. (n.d.). [Police activity statistics data file]. Retrieved November 23, 2015, from http:// newarkpdonline.org/comstat/transparency.php 49. Open Justice. (n.d.). [Data file]. Retrieved from http://openjustice.doj.ca.gov/index.html 50. Police Foundation. (n.d.). Public safety open data portal [Data file]. Retrieved November 20, 2015, from http:// publicsafetydataportal.org/officer-involved-shooting-data/ 51. Bureau of Justice Statistics. (2012.). Arrest-Related Death report CJ-11A addendum. Retrieved from http://www.bjs.gov/ content/pub/pdf/cj11a_12.pdf 52. Bureau of Justice Statistics. (2015). Deaths in Custody Reporting Program and annual survey of jails, 2016–2020. http:// www.bjs.gov/content/pub/pdf/dcrpasj1620sol.pdf 53. Bureau of Justice Statistics, Deaths in Custody Reporting Program. 54. M. Planty, personal communication, November 13, 2015. 55. Ibid. 56. Kindy, K. (2015, May 30). Fatal police shootings in 2015 approaching 400 nationwide. The Washington Post. https://www. washingtonpost.com/national/fatal-police-shootings-in-2015-approaching-400-nationwide/2015/05/30/d322256a-058e- 11e5-a428-c984eb077d4e_story.html 57. Robertson, I. (2015, August 1). The most police-related killings of the year were in July 2015. The Guardian. http://www. vibe.com/2015/08/police-killings-july-2015/ 58. Deadly force: When Las Vegas police shoot, and kill. (n.d.). Las Vegas Review-Journal. Retrieved November 18, 2015, from http://www.reviewjournal.com/news/deadly-force 59. Police use of deadly force in Maine 1990–2012 [Data file]. (n.d.). Retrieved from http://www.pressherald.com/interactive/ maine-police-deadly-force-lethal-database/ 60. Explore the Data. (n.d.). [Data file]. Available from http://projects.scpr.org/officer-involved/explore/ 61. Fatal Encounters. (n.d.). [Data file]. Available from http://www.fatalencounters.org 62. Gun Violence Archive. (n.d.). [Data file]. Retrieved from http://www.gunviolencearchive.org 63. Killed by Police 2015. (2015). [Data file]. Retrieved from http://killedbypolice.net/ 64. Preventable Tragedies Database. (n.d.). [Data file]. Available from http://www.treatmentadvocacycenter.org/problem/ preventable-tragedies-database 65. Matulia, K. J. (1982). A balance of forces: Executive summary. Washington, DC: National Institute of Justice. 66. MacLeod et al., Innovative law enforcement strategies. 67. Stettin, B., Frese, F. J., & Lamb, H. R. (2013). Mental health diversion practices: A survey of the states. Arlington, VA: Treatment Advocacy Center. Available from http://tacreports.org/diversion-study 68. Thomas, Core requirements of a best practice model. 69. Lamb et al., Outcome for psychiatric emergency patients. 70. Compton, M. T., Bakeman, R., Broussard, B., Hankerson-Dyson, D., Krishan, S., Stewart-Hutto, T., D’Orio, B. M., Oliva, J. R., & Watson, A. C. (2014). The police-based crisis intervention team (CIT) model: I. Effects on officers’ knowledge, attitudes, and skills. Psychiatric Services, 65(4), 517–522. Available from http://www.ncbi.nlm.nih.gov/pubmed/24382628 ENDNOTES (continued) 24 n Overlooked in the Undercounted: THE ROLE OF MENTAL ILLNESS IN FATAL LAW ENFORCEMENT ENCOUNTERS 71. Lamb et al., Outcome for psychiatric emergency patients. 72. Community Research Foundation. (n.d.). Psychiatric Emergency Response Team. Retrieved November 16, 2015, from http://www.comresearch.org/pert.php 73. Torrey, E. F., Fuller, D. A., Geller, J., Jacobs, C., & Ragosta, K. (2013). No room at the inn: Trends and consequences of closing public psychiatric hospitals. Arlington, VA: Treatment Advocacy Center. Available from http://tacreports.org/bed- study 74. Torrey et al., The treatment of persons with mental illness. 75. Planty et al., Arrest-Related Deaths program. 76. Lamb, H. R. (2015). Does deinstitutionalization cause criminalization? The Penrose hypothesis. JAMA Psychiatry, 72(2):105- 106. doi:10.1001/jamapsychiatry.2014.2444 77. Hartvig, P., & Kjelsberg, E. (2009). Penrose’s law revisited: The relationship between mental institution beds, prison population and crime rate. Nordic Journal of Psychiatry, 63:51-56. 78. Langton, L., & Durose, M. (2013). Police behavior during traffic and street stops, Washington, DC: 2011. Bureau of Justice Statistics. Retrieved from http://www.bjs.gov/content/pub/pdf/pbtss11.pdf 79. Banks, D., Couzens, L., & Planty, M. (2015). Assessment of coverage in the Arrest-Related Deaths program. Bureau of Justice Statistics. Retrieved from http://www.bjs.gov/content/pub/pdf/acardp.pdf ENDNOTES (continued) © 2015 Treatment Advocacy Center The Treatment Advocacy Center is a national nonprofit organization dedicated exclusively to eliminating barriers to the timely and effective treatment of severe mental illness. The organization promotes laws, policies and practices for the delivery of psychiatric care and supports the development of innovative treatments for and research into the causes of severe and persistent psychiatric illnesses, such as schizophrenia and bipolar disorder. View publication statsView publication stats Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=rhpd20 Housing Policy Debate ISSN: 1051-1482 (Print) 2152-050X (Online) Journal homepage: https://www.tandfonline.com/loi/rhpd20 Public service reductions associated with placement of homeless persons with severe mental illness in supportive housing Dennis P. Culhane , Stephen Metraux & Trevor Hadley To cite this article: Dennis P. Culhane , Stephen Metraux & Trevor Hadley (2002) Public service reductions associated with placement of homeless persons with severe mental illness in supportivehousing, Housing Policy Debate, 13:1, 107-163, DOI: 10.1080/10511482.2002.9521437 To link to this article: https://doi.org/10.1080/10511482.2002.9521437 Published online: 31 Mar 2010. Submit your article to this journal Article views: 1636 View related articles Citing articles: 186 View citing articles Public Service Reductions Associated withPlacement of Homeless Persons with SevereMental Illness in Supportive HousingDennis P. Culhane, Stephen Metraux, and Trevor HadleyUniversity of PennsylvaniaAbstract This article assesses the impact of public investment in supportive housing for home- less persons with severe mental disabilities. Data on 4,679 people placed in such hous- ing in New York City between 1989 and 1997 were merged with data on the utilization of public shelters, public and private hospitals, and correctional facilities. A series of matched controls who were homeless but not placed in housing were similarly tracked. Regression results reveal that persons placed in supportive housing experience marked reductions in shelter use, hospitalizations, length of stay per hospitalization, and time incarcerated. Before placement, homeless people with severe mental illness used about $40,451 per person per year in services (1999 dollars). Placement was associated with a reduction in services use of $16,281 per housing unit per year. Annual unit costs are estimated at $17,277, for a net cost of $995 per unit per year over the first two years. Keywords:Homelessness; Housing Introduction Placing homeless persons with severe mental illness (SMI) into subsi- dized permanent housing with social service support promises to sub- stantially reduce the demand for shelter among those placed. This housing provides a more humane alternative to living on the streets and in shelters, and providers report retention rates in such housing to be upwards of 70 percent in the first year after placement. However, little empirical evidence has been gathered to quantify the degree to which supportive housing supplants shelter use among the formerly homeless with SMI. Furthermore, it can similarly be assumed that homeless persons with SMI, once placed in supportive housing, reduce their use of acute psychiatric and medical services, and are arrested and incarcerated less often. However, such assumptions are somewhat more tenuous, and a similar dearth of empirical evidence exists both on the demand for these services among homeless persons with SMI and on the impact of supportive housing on this level of demand. The study reported here examines service use by formerly homeless persons with SMI before and after being placed into New York/New Housing Policy Debate ·Volume 13, Issue 1 107© Fannie Mae Foundation 2002. All Rights Reserved.107Culhane.qxd 6/18/02 11:49 AM Page 107 108 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyYork (NY/NY) housing, a large housing program in New York City(NYC). Administrative data from public health, psychiatric, criminaljustice, and shelter service providers are used to assess the aggregatelevel of service demand, pre- and postintervention, for the study groupand for a matched set of controls. The extent to which reductions inservices are present and attributable to NY/NY housing placement isassessed, and the degree to which service reductions offset supportivehousing costs is measured.Background In 1990, New York State (NYS) and NYC agreed to jointly fund and develop 3,600 community-based permanent housing units for homeless persons with SMI under what became known as the New York/New York Agreement to House the Homeless Mentally Ill (Hevesi 1999; Kennedy 1995, 1997).1 This initiative was in response to problems with homelessness and community mental health services that were per- ceived to have reached crisis proportions in NYC. The NY/NY agree- ment was designed to target those who were among the most chronic and difficult to serve among the homeless population and to ease demands on public shelter and psychiatric treatment services. The agreement provided housing and psychosocial services in a variety of configurations collectively known as NY/NY housing. There are two general models: The first, supportive housing, includes scattered-site housing with community-based service support and single-room occu- pancy (SRO) housing (independent housing linked to either community- based or site-based service support). The second, community residence facilities, includes community residences, long-term treatment facilities, and adult homes (Center for Urban Community Services 1995; Lipton et al. 2000). In general, supportive housing emphasizes “normality” in housing in terms of separating services from housing arrangements and giving tenants a choice in their housing arrangements and mental health service regimens. By contrast, community residences take a more clinical approach that integrates housing and services delivery by hav- ing services available on site and participation mandated by the resi- dence agreement. Supportive housing maintains that such housing is appropriate for persons with mental illness regardless of the severity of impairment, while the community residence model places people in increasingly less restrictive living arrangements as they progress Fannie Mae Foundation 1 This initiative is also now referred to as NY/NY I, since a second initiative to provide additional units under a similar state/city–financed structure (NY/NY II) was estab- lished in 1999. Culhane.qxd 6/18/02 11:49 AM Page 108 Public Service Reductions Associated with Supportive Housing 109through their treatment regimens (Bebout and Harris 1992; Carling1993).To be eligible for this housing, tenants must have a diagnosis of SMIand have been recently homeless in shelters or on the streets. Aftergoing through an application and assessment with the NYC HumanResources Administration (HRA) to determine NY/NY housing eligibil-ity, the prospective tenant then applies through one of the nonprofitagencies that administer the actual units funded under the agreement.Thus, NY/NY eligibility, housing availability, agency eligibility guide-lines, and tenant preference all factor into the placements providedunder the agreement. Literature review Studies that focus on supportive housing interventions for homeless persons with mental illness consistently find high rates of retention in these programs. Lipton, Nutt, and Sabatini (1988) followed 49 home- less persons with mental illness, half of whom were provided program housing. After one year, they found that 69 percent of the experimental group were living in permanent housing, as opposed to 30 percent of the control group. Drake et al. (1997) similarly report improved hous- ing outcomes for a group of dually diagnosed2 homeless persons who were provided residential treatment, compared with a control group given standard treatment. Caton et al. (1993) and Murray et al. (1997) report high rates of housing retention for participants in transitional and/or continuum-model programs, although neither of the studies included comparable control groups. In the most comprehensive review of supportive housing studies to date, Ridgway and Rapp (1998) report that supportive housing for homeless persons with mental illness reduced homelessness and im- proved housing stability among program participants. Research from a McKinney Demonstration Program (Center for Mental Health Services 1994), in which the National Institute of Mental Health sponsored five supportive housing projects in four cities, found increased rates of sta- ble housing among the experimental groups—formerly homeless per- sons with mental illness who received supportive housing and case management services—compared with similar groups of controls who were provided with standard treatment services (Shern et al. 1997). Housing Policy Debate 2 In this article, “dually diagnosed” refers to comorbid diagnoses of serious mental ill- ness and substance abuse disorder. Culhane.qxd 6/18/02 11:49 AM Page 109 110 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyIn the San Diego McKinney project, Hurlburt, Wood, and Hough (1996)report on 362 persons who were homeless and severely mentally ill andwho were randomly assigned to four groups that varied Section 8 rentalsubsidies and case management services. They found that Section 8manipulation had a dramatic impact on subsequent housing stability,but that enhanced case management manipulation had no significantimpact. Only 30 percent of the study participants who did not receive arent subsidy achieved stable independent living, compared with 57 per-cent of those who did.Both Goldfinger et al. (1999) and Dickey et al. (1996) report on an initiative that provided two types of housing in Boston for homelesspersons with mental illness: independent apartments with community- based services and so-called evolving consumer households, where the tenants lived communally and with gradually diminishing levels of staff assistance. Three-quarters of the total subjects were stably housed at the end of the 18-month follow-up period. The subjects in the group homes had fewer days homeless than the supported housing group, but otherwise no significant differences in housing outcomes or services use were found (Dickey et al. 1996; Goldfinger et al. 1999). Three studies have looked at the housing provided by the NY/NY agree- ment. Lipton et al. (2000) found that after one, two, and five years, 75 percent, 64 percent, and 50 percent of the almost 3,000 persons placed had remained in the program across all types of NY/NY housing configurations. Tsemberis (1999) and Tsemberis and Eichenberg (2000) have also found high rates of housing retention by NY/NY recipients, but in addition, found that tenants of one supported scattered-site housing program affiliated with NY/NY had a substantially higher retention rate after five years (88 percent) than other NY/NY support- ive housing programs (55 percent). High tenant-retention rates among housing interventions regardless of the particular configuration of services and housing have been a common finding in the housing programs examined so far. However, these similar outcomes belie the disparate costs involved in the two approaches. Community residence models, with their incorporation of site-based staff and services that work exclusively with the tenants, have substantially higher associated service costs than supportive hous- ing models; these decouple residency and services to a greater extent, and their tenants make greater use of existing community services. Persons with SMI could also be expected to reduce their use of hospital services following a housing placement, because persons who are receiv- ing services would be in a better position to engage in regular outpa- tient regimens that could prevent the need for hospitalization. Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 110 Public Service Reductions Associated with Supportive Housing 111Furthermore, if they are hospitalized, access to housing and supportcould reduce the length of stay in a hospital. In a study of public hospi-tal records in NYC, Salit et al. (1998) found homelessness to be associ-ated with substantial excess stays and costs per hospital stay. Lewis andLurigio (1994), in a study of state hospital patients living in Chicago,found that when poor persons with mental illness seek psychiatric hos-pitalization, they often do so more as a short-term housing arrange-ment than for psychiatric reasons. In a review of the literature,Rosenheck (2000) found that enhancing services, either housing or casemanagement, can generate reductions in the use of inpatient mentalhealth services, especially among heavy hospital users. These reduc-tions, however, may be at least partially offset by increased use of out-patient services (Averyt and Kamis-Gould 2000) and support services such as case management, which are needed to effect such inpatient reductions (Rosenheck 2000). This leads to the question of whether providing a service such as sup- portive housing is cost-effective in reducing homelessness among per- sons with mental illness. Rosenheck (2000) found that for all but the heaviest service users, enhanced interventions cost more than the sav- ings they generate. However, studies of reductions in service utilization and associated cost savings have typically focused on one type of service and on a single service system and have not integrated multiple pro- viders and multiple systems. Integrating costs accrued by homeless persons across multiple providers in such systems as shelters, mental health services, medical care, and criminal justice would allow for a more comprehensive assessment of the cost of homelessness from which to estimate savings. By tracking people across multiple systems, the estimated public expenses associated with homelessness would likely increase, as would the estimated reductions in service use follow- ing the receipt of targeted housing. Thus, greater cost-effectiveness may be demonstrated. Data and methods Data sources The data used in this study come from administrative databases main- tained by eight different agencies. These databases are collected in computerized management information systems and track service uti- lization over time. As such, they represent comprehensive banks of data on users, both their characteristics and their patterns of use. Because these databases contain client identifiers, they can be linked across systems to identify how services received through one system may affect services in others. Administrative databases are the only Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 111 112 Dennis P. Culhane, Stephen Metraux, and Trevor Hadleypractical means of obtaining information on a large number of home-less persons over an extended period of time and with accurate data onservice consumption across multiple systems (Culhane and Metraux1997). Databases used for this analysis come from the following sources: 1. NYC HRA, with records for 4,679 persons either placed in housingdeveloped under the NY/NY agreement or deemed eligible forNY/NY housing and placed in community-based housing. The data-base includes demographic and identifying information, as well asthe date and type of housing for placements through 1997. Noinformation on the duration of discrete NY/NY placements is available. 2. NYC Department of Homeless Services (DHS), with records for all shelter users and shelter use since 1986 for its single-adult shelter network. 3. NYS Office of Mental Health (OMH), with a database of lifetime records of inpatient stays in the state psychiatric hospital system for anyone who was an inpatient in a state hospital from 1990 through 1996. 4. NYS Department of Health, Office of Medicaid Management (here- after referred to as Medicaid), with records of Medicaid-reimbursed inpatient and outpatient health care claims for persons with records of shelter use and/or NY/NY housing placements for the years 1993 through 1997. 5. NYC Health and Hospitals Corporation (HHC), with records of inpatient stays in municipal hospitals between 1989 and 1996 for all persons with a DHS shelter record. 6. U.S. Department of Veterans Affairs (VA), with records of inpatient stays in the VA hospital system between 1992 and 1999 for all per- sons with records of DHS shelter utilization or NY/NY placement. 7. NYS Department of Correctional Services (NYSDOCS), with a data- base on state prison utilization for persons with a NY/NY housing placement, and a set of control observations selected from the DHS shelter system. Data used in this study were from 1988 through April 15, 1997. 8. NYC Department of Corrections (NYCDOC), with a database on NYC jail utilization for persons with a NY/NY housing placement, Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 112 Public Service Reductions Associated with Supportive Housing 113and a set of control observations selected from the DHS shelter sys-tem. Data used in this study were from 1988 to 1999.These databases were merged by matching five common identifiers:first and last names, sex, date of birth, and Social Security number.Segments of the first four identifiers were combined to create a uniqueidentifier that was used to match cases across databases. Also, SocialSecurity numbers (when available) were used to provide additionalmatches when the other identifiers were missing or contained erro-neous data.3 Matched control groups By comparing each individual’s history of service use in the two-year periods immediately before and after his or her NY/NY placement, it is possible to estimate the changes in service use for persons with NY/NY placements across these seven service systems. In addition, each person placed in NY/NY housing was matched to an individual control observa- tion with similar characteristics to assess service use in the absence of a supportive housing placement. Because of the difficulty in consis- tently pairing case (NY/NY) and control observations with similar pre- intervention service use patterns across the seven service systems, different control groups were used to analyze different service systems. Appendix A contains a more detailed overview of the sampling frames for the respective control groups. To construct the matched-pair control group for each analysis, the fol- lowing criteria were used to select observations for the control groups based on similarities with specific control observations: 1. Demographics. Gender, race (black/nonblack), and age. Ages of those in the control pool are within five years of the case. 2. Indicators of mental illness and substance abuse. For the DHS control group, these indicators are based on data from DHS records and reflect the assessments of DHS social service staff and self- disclosure by shelter users at the intake interview (no standardized criteria for determining mental illness or substance abuse problems are used). For the OMH, HHC, Medicaid, and VA control groups, these indicators are based on Diagnostic and Statistical Manual of Mental Disorders, 4th Edition,diagnoses that accompany hospitalizations. Housing Policy Debate 3 Further details on this process are available from the authors. Culhane.qxd 6/18/02 11:49 AM Page 113 114 Dennis P. Culhane, Stephen Metraux, and Trevor Hadley3. Similar service use for the two-year period up to the NY/NY place-ment date, based on the number of stays and days spent in servicefacilities (shelter, hospital, or prison) during the two-year interven-tion period, and the length of time between the last service use andNY/NY placement.This case-control matching process has two parts. First, matching onelements of the first two groups of criteria (demographics and diagnosisindicators) limits the numbers of potential matches between each caseobservation and the pool of controls. Then, for each match, using thecase observation’s NY/NY placement date (which varies for each obser-vation) as a surrogate intervention date for each potential controlobservation means that the control observation with the most similar pattern of preintervention service use (i.e., days and episodes for the two-year preintervention period) is selected to pair up with each case. For each case-control pair, the case observation’s NY/NY placement date represents the intervention point that separates pre- and post- intervention periods for the case and control observations. Analysis methods Each analysis follows a parallel set of procedures. First, descriptive sta- tistics that facilitate comparisons of raw pre- and postintervention serv- ice use among the aggregate group with NY/NY housing placements are provided. Second, descriptive statistics on pre- and postintervention service use are provided for the case and control groups, with paired comparison t tests used to assess whether the differences between and within groups across intervention periods are statistically significant. Following these two analyses, the effect of a NY/NY housing placement on the reduction in postintervention service use, measured in days, is estimated with multivariate least squares regression models, using a generalized estimating equations (GEE) methodology. With its use of maximum likelihood estimation and an iterative generalized least squares algorithm, GEE can accommodate nonindependent observa- tions such as matched pairs. Such a data structure normally violates the assumptions associated with ordinary least squares (OLS) regres- sion, but this approach corrects the attenuated standard error values that would otherwise result from using OLS regression (Allison 1999).4 In each regression model, the dependent variable is the difference, for each observation, in the number of days accrued in each of the service systems across the pre- to postintervention periods. The covariate of Fannie Mae Foundation 4 Regression analyses were computed using GENMOD with the REPEATED option in SAS statistical software, version 8.1. Culhane.qxd 6/18/02 11:49 AM Page 114 Public Service Reductions Associated with Supportive Housing 115primary interest estimates the effect of getting a NY/NY placement,with all other factors held constant. Along with this NY/NY covariate,the control variables in the models are as follows:1. The variables used to match the control groups2. When applicable, a set of dichotomous variables to control for theyear of NY/NY placement3. A set of measures for preperiod service use, including serviceepisodes, service days consumed, and (when available) cost ofservices54. Measures of preperiod DHS shelter use Results System-specific effects Use of DHS shelter services.Almost three-quarters of those placed in NY/NY housing have some record of having stayed in a DHS shelter at some point between 1987 and 1999.6 These 3,365 persons with DHS and NY/NY records were matched with control observations from the NYC DHS single-adult shelter system, and the resulting dataset, with 3,338 matched pairs, provides the basis for a case-control comparison of shelter usage for two-year periods before and after the NY/NY inter- vention point. A total of 27 case observations (0.8 percent) could not be matched with a control observation. The descriptive results (table 1) show a dramatic 85.6 percent pre/post placement decline in the mean number of shelter days used by persons with NY/NY placements, from 137.0 days per placement to 19.8 days Housing Policy Debate 5 Measures of preperiod use of services control for two phenomena: first, the prerequi- site that higher degrees of preintervention services must have higher differences in levels of pre/post use of services and second, the effects of prior use of services on the likelihood of engaging in subsequent use of services. These anticipated effects run counter to each other, since the former would associate higher preperiod use of serv- ices with greater reductions in postintervention use, and the latter would associate preperiod use of services with lower or negative differences in pre/post use of services. While this could lead to difficulty in interpreting the services use coefficients, it should control for these effects when considering the effects of NY/NY placement. 6 Of the 4,679 persons with NY/NY housing placements, 3,365 (71.9 percent) also had DHS shelter records, even though all of them, by NY/NY eligibility criteria, must have been homeless before housing placement. Of those without shelter records, some may have used shelters not covered by the DHS database (approximately 20 percent of all NYC shelter beds), some may have used shelters outside NYC, and some may have stayed exclusively in nonshelter (street) arrangements during their periods of homelessness. Culhane.qxd 6/18/02 11:49 AM Page 115 116 Dennis P. Culhane, Stephen Metraux, and Trevor Hadleyper placement. In the case-control comparison, the unadjusted casegroup decline (85.6 percent) is consistent with that of the entire NY/NYgroup and far outpaces the 6.4 percent unadjusted decline experiencedby the controls. The pre/post declines for both groups are statisticallysignificant (at p < 0.01) using paired-comparison t tests. In the prein-tervention period, the mean per placement number of shelter days usedamong the cases is heavier (p < 0.0001) among the case group, but thisrelationship inverts in the postintervention period (p < 0.0001) as thecontrols become, on average, the heavier shelter users. In the regression model shown on table 2, the NY/NY placement is stillassociated with a 115.3-day reduction in shelter days used from the pre-to the postintervention period (95 percent confidence interval [CI], 107.7 to 123.0 days) after controlling for other factors, especially heavy shelter use.7 Taking this 115.3-day reduction and averaging it out over all 4,679 persons with NY/NY placements yields an estimated reduction per NY/NY placement of 82.9 shelter days (95 percent CI, 77.4 to 88.5 days) over the two-year period. 8 This represents an adjusted reduc- tion of 60.5 percent, compared with the average preintervention shelter usage by the NY/NY placements (137.0 from table 1). Use of OMH inpatient state psychiatric hospital services. Of the 2,396 persons receiving a NY/NY placement from 1992 to 1994, 897 (37.4 per- cent) had some record of an inpatient OMH state hospital stay. Of this subgroup, 630 observations also had a record of DHS shelter use and were matched with DHS controls. These 630 case observations pro- vided the basis for the OMH case-control analysis, with 570 (90.5 per- cent) of these observations matched with control observations selected from DHS shelter users. Fannie Mae Foundation 7 Heavy shelter use is interpreted as a combination of two covariates in the model, “shelter days accrued” and “any shelter use.” These coefficients have opposite signs, meaning that the value of preintervention shelter days is associated with decreased reductions in the number of postintervention shelter days with few preintervention shelter days accrued, and increased reductions with many preintervention shelter days accrued (see footnote 4). Thus, if an observation had five preintervention shelter days, the combined coefficient of the coefficients shelter days accrued and any shelter use (0.75 5 – 34.48) would be a postperiod increase of 30.73 shelter days used (all else held equal), while a preintervention stay of 100 days (0.75 100 – 34.48) would lead to a combined coefficient associated with a decrease of 40.52 shelter days used. Equilib- rium here would be at 46 days. 8 This average is computed by dividing estimated aggregate reduction in shelter days attributed to NY/NY (115.3 3,365) by the total number of NY/NY placements (4,679). Culhane.qxd 6/18/02 11:49 AM Page 116 Public Service Reductions Associated with Supportive Housing 117Table 1.Shelter Days Consumed by Persons in NY/NY Housing and Controlsin the Two-Year Periods before and after the NY/NY InterventionNY/NY NY/NY Controls(Total 1989–97) (Matched Pairs) (Matched Pairs)N 4,679 3,338 3,338Total service users 3,365 3,338 3,338Pre-NY/NY intervention (two years)Total persons with shelter records 2,786 (59.5%) 2,750 (82.4%) 2,265 (67.9%)Total days sheltered 641,171 636,319 544,700Mean days (all persons) 137.0 190.6 130.9Mean days (shelter users) 230.2 231.4 240.5 Post-NY/NY intervention (two years) Total persons with shelter records 782 (16.7%) 776 (23.2%) 1,754 (51.4%) Total days sheltered 92,421 91,751 408,883 Mean days (all persons) 19.8 27.5 122.5 Mean days (shelter users) 118.2 118.2 233.0 Note:Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention differences in shelter days, t = 27.3 (3,337 df and p< 0.0001), and for postin- tervention differences, t = –26.2 (3,337 dfand p< 0.0001). Using paired-comparison t tests, pre/post differences yield, within the NY/NY group, t= 46.04 (3,337 df and p< 0.0001), and within the control group, t = 2.6 (3,337 dfand p < 0.01). Table 2.Regression Model Estimating Effects on Changes in Shelter Days Used in the Two-Year Periods before and after the NY/NY Intervention (N = 3,338 Matched Pairs) Parameter Estimate Lower Upper Covariate (Days Saved) (95%) CI (95%) CI Intercept –44.13*** –61.35 –26.92 Received NY/NY placement 115.33*** 107.66 123.01 Shelter days accrued in two-year 0.75*** 0.72 0.78 preintervention period Any shelter use in two-year preinter- –34.48*** –41.77 –27.19 vention period NY/NY placement in 1996–97 –24.83*** –36.32 –13.34 NY/NY placement in 1994–95 –9.54 –19.97 0.89 NY/NY placement in 1992–93 3.70 –6.35 13.76 NY/NY placement before 1992 Reference Category Age at NY/NY placement –0.12 –0.42 0.19 Male –13.35*** –20.26 –6.44 Black race 1.97 –4.72 8.67 DHS mental illness indicator –8.14* –15.06 –1.21 DHS drug use indicator 0.02 –7.04 7.09 *p < 0.05. **p < 0.01. ***p< 0.001. Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 117 118 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyDescriptive results of state hospital use, shown in table 3, show largereductions in pre/post state hospital use, measured in days, among thetotal NY/NY group (59.9 percent reduction), as well as among the morerestricted case group (57.0 percent reduction). Looking at the case-con-trol comparison, the preintervention state hospital use, measured indays, is (by design) very similar (i.e., with statistically nonsignificantdifferences). By contrast, the difference in postintervention state hos-pital use is both substantial and statistically significant (p < 0.0001).Comparing within groups, the NY/NY group shows significant pre/postreductions in state hospital use (p < 0.0001), while the reductions instate hospital days used are nonsignificant for the control group. Fur-ther, for the NY/NY group, far fewer persons experienced postinterven-tion hospital episodes than in the control group, and the mean number of hospital days per hospitalized person also declined after the interven- tion. While persons hospitalized also declined in the latter time period for the control group, the average number of days hospitalized increased substantially for the control group. Table 3.OMH State Hospital Days Consumed by Persons in NY/NY Housing and Controls in the Two-Year Periods before and after the NY/NY Intervention NY/NY NY/NY Controls (Total 1992–94) (Matched Pairs) (Matched Pairs) N 2,396 570 570 Total service users 897 570 570 Pre NY/NY intervention (two years) Total persons hospitalized 634 (26.4%) 406 406 Total days hospitalized 137,215 78,250 78,940 Mean days (all persons) 57.3 137.3 138.5 Mean days (hospital users) 216.4 192.7 194.4 Post NY/NY intervention (two years) Total persons hospitalized 353 240 335 Total days hospitalized 55,070 33,623 74,869 Mean days (all persons) 23.0 59.0 131.4 Mean days (hospital users) 156.0 140.1 223.5 Note:Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, among preintervention state hospital days, t= –1.8 (569 dfand p = 0.07), and for postinterven- tion differences, t = –7.7 (569 dfand p < 0.0001). Using paired-comparison ttests, pre/post differ- ences yield, within the NY/NY group, t= 9.3 (569dfand p < 0.0001), and within the control group, t= –1.8 (569 df and p = 0.37). Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 118 Public Service Reductions Associated with Supportive Housing 119A multivariate regression model (table 4) shows that, holding other factors constant, a NY/NY placement is associated with a statisticallysignificant estimated reduction of 75.3 days (95 percent CI, 55.7 to95.0 days). Averaging this adjusted reduction for the case group acrossall the 2,396 NY/NY placements from 1992 to 1994 yields an estimatedreduction of 28.2 days per NY/NY placement (95 percent CI, 20.8 to35.6 days).9 Compared with the 57.3 days of mean preintervention statehospital use by the NY/NY group (table 3), this reflects a 49.2 percentadjusted reduction. Table 4.Regression Model Estimating Effects on Changes in State HospitalDays Used in the Two-Year Periods before and after the NY/NY Intervention (N = 570 Matched Pairs) Parameter Estimate Lower Upper Covariate (Days Saved) (95%) CI (95%) CI Intercept –101.40** –165.69 –37.12 Received NY/NY placement 75.33*** 55.66 95.00 Days between last preintervention OMH 0.16*** 0.11 0.21 exit and NY/NY placement (gap)a No preintervention period OMH inpatient –70.49*** –99.90 –41.09 record Hospital days in preintervention period 0.75*** 0.66 0.83 Hospital stays in preintervention period 1.76 –15.19 18.70 Shelter days in preintervention period –0.03 –0.09 0.03 NY/NY placement in 1992 Reference Category NY/NY placement in 1993 2.51 –17.51 22.53 NY/NY placement in 1994 12.08 –9.98 34.15 Age at NY/NY placement 0.42 –0.60 1.45 Male –8.20 –26.98 10.59 Black race 3.90 –15.15 22.94 295 diagnosis (schizophrenia) –46.44*** –63.61 –29.26 296 diagnosis (affective disorders) –35.17*** –55.93 –14.41 Drug/Alcohol dependency diagnosis –7.39 –25.84 11.07 aFor those with no preintervention OMH inpatient record, the gap is set at 731 days. *p < 0.05. **p < 0.01. ***p < 0.001. Housing Policy Debate 9 This is estimated by multiplying 75.3 by the 897 persons with both NY/NY place- ments and OMH records and then dividing by the 2,396 NY/NY placements in 1992–94. This assumes that the pre/post reduction in state hospital use is the same for NY/NY placements with and without shelter records. Comparisons of these two subgroups (Metraux, Culhane, and Hadley 2000) indicate that use of OMH inpatient services is in fact somewhat higher among nonshelter users in the preintervention period and that this group has higher pre/post intervention period reductions compared with counter- parts with DHS shelter records. Thus, this 75.3-day reduction extrapolation for non- shelter users is likely to be a conservative estimate. Culhane.qxd 6/18/02 11:49 AM Page 119 120 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyUse of NYC public hospitals (HHC).HHC granted access to inpatienthospital records from 1989 to 1996 for all persons with a history ofDHS shelter use. These parameters limit the analysis to the 1,984 per-sons who had a NY/NY placement between 1991 and 1994 and who alsohad a DHS shelter record. Of these, 855 (43.1 percent) had at least onerecord of inpatient hospitalization through HHC that was not reim-bursed through Medicaid,10 and these observations were matched withcontrols selected from persons who had both a DHS shelter record andat least one HHC hospitalization record. The resulting case-controlgroup, consisting of 791 matched pairs (92.5 percent of those with HHCrecords), are used for further analysis on NY/NY housing placementsand their impact on hospitalizations.11 HHC hospital utilization, summarized in table 5, shows another sub- stantial unadjusted pre/post placement decline for the NY/NY group. Among all NY/NY placements (first column), total users decline 68.6 percent from the pre- to the postintervention period, while hospi- tal days consumed declines even more sharply at 79.9 percent. Looking at the case-control groups (second and third columns), for the cases there is a similar pre/post decline in persons hospitalized, 68.9 percent, and again a larger decline in days consumed, 78.0 percent. Compara- tively for the controls, both the pre/post declines in persons hospital- ized and days consumed, 49.5 percent and 53.4 percent, respectively, are substantial but considerably lower than the declines for case obser- vations. While the cases have a significantly higher preintervention number of hospital days used (p < 0.01), their number of postinterven- tion hospital days used is significantly lower than that of the control group (p < 0.0001). Regression model results (table 6) show that, after controlling for differ- ences in the included covariates, NY/NY placement is associated with a greater pre/post differential of 8.1 days (95 percent CI, 4.6 to 11.6 days). Averaging this 8.1-day reduction over the 1,984 observations results in Fannie Mae Foundation 10 Hospitalizations that are included in both the HHC and Medicaid datasets (i.e., a Medicaid-reimbursed inpatient stay occurring in an HHC hospital) are omitted from the HHC analysis and included in the subsequent Medicaid analysis. 11 A separate analysis of hospital stays finds that over three-quarters of the hospitaliza- tions fall into nine Diagnosis Related Groups (DRGs), all of which correspond to treat- ment for either mental health or substance abuse issues. The Psychosis DRG (430) alone accounts for over half of all hospitalizations by persons receiving NY/NY place- ments, during both the pre- and postintervention periods (Metraux, Culhane, and Hadley 2000). DRG is a categorization system for hospital stays that are medically related with respect to diagnosis and treatment and that are statistically similar in length of stay. Culhane.qxd 6/18/02 11:49 AM Page 120 Public Service Reductions Associated with Supportive Housing 121a reduction of 3.5 days per placement (95 percent CI, 2.0 to 5.0 days).12Compared with the 16.5 days of mean preintervention HHC use by theNY/NY group (table 5), this reflects a 21.2 percent adjusted reduction.Table 5.HHC Hospital Days (non-Medicaid) Consumed by Persons in NY/NY Housing and Controls in the Two-Year Periods before and after the NY/NY InterventionNY/NY NY/NY Controls(Total 1991–94) (Matched Pairs) (Matched Pairs)N (NY/NY placements with shelter 1,984 791 791 record) Inpatient (non-Medicaid) HHC 855 791 791 record: 1989–96 Pre-NY/NY intervention (two years) Total persons hospitalized 549 (27.7%) 515 (65.1%) 515 (65.1%) Total days hospitalized 32,823 27,014 26,456 Mean days (all persons) 16.5 34.2 33.4 Mean days (hospital users) 59.8 52.5 51.4 Post-NY/NY intervention (two years) Total persons hospitalized 175 (8.8%) 160 (20.2%) 260 (32.9%) Total days hospitalized 6,610 5,937 12,330 Mean days (all persons) 3.3 7.5 15.6 Mean days (hospital users) 37.8 37.1 47.4 Note:Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention HHC hospital days, t = –2.6 (790 dfand p< 0.01), and for postintervention differences, t = 5.0 (790 df and p < 0.0001). Using paired-comparison ttests, pre/post differences yield, within the NY/NY group, t= 15.2 (790 dfand p < 0.0001), and within the control group, t = 9.6 (790 df and p< 0.0001). Housing Policy Debate 12 This is estimated by multiplying 8.1 by the 855 persons with NY/NY placements, DHS records, and HHC records (from which the control group was selected) and then dividing by the 1,984 persons with NY/NY placements and DHS records. For this analy- sis, it is assumed that NY/NY placements without DHS shelter records have HHC hos- pital use patterns that are the same as those of the persons with DHS records used in this case-control analysis. Culhane.qxd 6/18/02 11:49 AM Page 121 122 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable 6.Regression Model Estimating Effects on Changes in HHC HospitalDays (non-Medicaid) Used in the Two-Year Periods before and after the NY/NY Intervention (N = 791 Matched Pairs)Parameter Estimate Lower UpperCovariate (Days Saved) (95%) CI (95%) CIIntercept –12.67* –23.57 –1.77Received NY/NY placement 8.05*** 4.55 11.55Days between last preintervention HHC 0.03*** 0.01 0.04exit and NY/NY placement (gap)aNo preintervention period HHC inpatient –15.31*** –20.39 –10.23recordHospital days in preintervention period 0.94*** 0.87 1.01 Hospital stays in preintervention period –1.34 –3.59 0.90 Shelter days in preintervention period 0.00 –0.01 0.01 NY/NY placement in 1991 Reference Category NY/NY placement in 1992 7.48* 1.78 13.17 NY/NY placement in 1993 3.17 –1.87 8.21 NY/NY placement in 1994 5.73* 0.58 10.87 Age at NY/NY placement –0.11 –0.28 0.07 Male 2.47 –1.06 6.00 Black race 2.02 –1.25 5.29 295 diagnosis (schizophrenia) –7.28*** –10.44 –4.11 296 diagnosis (affective disorders) –8.27*** –12.57 –3.96 Drug/Alcohol dependency diagnosis –6.19*** –9.39 –2.99 aFor those with no preintervention HHC inpatient record, the gap is set at 731 days. *p< 0.05. **p < 0.01. ***p < 0.001. Use of Medicaid-reimbursed inpatient and outpatient services.This analysis looks at claims records, both inpatient and outpatient, for medical and psychiatric health services that were eligible for reim- bursement under the NYS Medicaid program.13 Medicaid data were available for 1993 through 1997. To provide full two-year pre- and postintervention periods of claims records, only the cohort placed in NY/NY housing in 1995 and a set of matched controls were included in this analysis. Fannie Mae Foundation 13 The Medicaid inpatient claims data include hospital stays that are duplicated in the HHC database but are not used for the HHC analysis. Over three-quarters of the health care services provided in both outpatient and inpatient settings involved a pri- mary diagnosis of either mental illness or substance abuse. The inpatient claims, which allow up to seven diagnoses, showed at least one diagnosis involving mental illness or substance abuse 92 percent of the time. More details can be found in Metraux, Cul- hane, and Hadley (2001a). Culhane.qxd 6/18/02 11:49 AM Page 122 Public Service Reductions Associated with Supportive Housing 123Inpatient servicesAs has been the pattern, unadjusted inpatient service use reimbursedby Medicaid (table 7) drops substantially in the post-NY/NY placementperiod. Of the 733 persons who were in the 1995 NY/NY cohort, 502(68.5 percent) had a Medicaid claims record from 1993 to 1997. Thepercentage of this cohort using inpatient services dropped 22.4 percentbetween the pre- and postperiods, while the number of inpatient daysconsumed dropped a more drastic 39.9 percent. The cost of services,also included in the Medicaid data, drops proportionately as well.Table 7. Inpatient Hospital Days Reimbursed by Medicaid for Persons in NY/NY Housing and Controls in the Two-Year Periods before and after the NY/NY Intervention NY/NY NY/NY Controls (Total 1995) (Matched Pairs) (Matched Pairs) N 733 457 457 Medicaid service users 502 457 457 Pre-NY/NY intervention (two years) Total persons hospitalized 406 (55.4%) 372 (81.4%) 372 (81.4%) Total days hospitalized 25,892 21,157 19,210 Mean days (all persons) 35.3 46.3 42.0 Mean days (hospital users) 63.8 56.9 51.6 Total amount billed to Medicaid $12,538,656 $10,525,629 $10,025,685 Mean amount billed (all persons) $17,106 $23,032 $21,938 Mean amount billed (hospital $30,883 $28,295 $26,951 users) Post-NY/NY intervention (two years) Total persons hospitalized 316 (43.1%) 280 (61.3%) 313 (68.5%) Total days hospitalized 15,558 13,542 19,137 Mean days (all persons) 21.2 29.6 41.9 Mean days (hospital users) 49.2 36.4 51.4 Total amount billed to Medicaid $8,070,885 $7,109,844 $10,738,287 Mean amount billed (all persons) $11,011 $15,558 $23,497 Mean amount billed (hospital $25,541 $19,112 $28,866 users) Notes:For the number of inpatient days (non-HHC) reimbursed by Medicaid: Between the NY/NY and control groups, paired-comparison t tests assessing difference yield, for the preintervention period, t = –4.8 (456 df and p< 0.0001), and for postintervention differences, t = 3.7 (456 df and p< 0.001). Using paired-comparison ttests, pre/post differences yield t = 6.0 (456 df and p< 0.0001) within the NY/NY group and t= 0.05 (456 dfand p= 0.96) within the control group. For the billing of inpatient days (non-HHC) reimbursed by Medicaid: Between the NY/NY and control groups, paired-comparison t tests assessing difference yield, for the preintervention period, t = –1.3 (456 df and p= 0.20), and for the postintervention period, t = 4.5 (456 df and p< 0.0001). Using paired-comparison ttests, pre/post differences yield, within the NY/NY group, t = 5.1 (456 df and p< 0.0001), and within the control group, t = –0.84 (456 df and p < 0.40). Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 123 124 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyAlso in table 7, the case group’s pre/post drop in service use is in con-trast to virtually no change in the number of days used and costsaccrued by the controls over this time. Compared with the controls, thecases have a significantly higher number of days consumed in the pre-intervention period (p < 0.0001), but nonsignificant cost differencesand a significantly lower number of postintervention days consumedand costs accrued in the postintervention period (p < 0.001).Separate regression models are presented in table 8 for pre/postchanges in days and costs. Controlling for all other factors in themodel, a NY/NY placement is still significantly associated with pre/postreductions of 12.6 days (95 percent CI, 6.2 to 18.9 days) and $7,983(95 percent CI, $4,608 to $11,358). Averaged over the total number of NY/NY placements in 1995, this leads to an estimated reduction of 8.6 days (95 percent CI, 4.2 to 13.0 days) and $5,467 (95 percent CI, $3,156 to $7,779).14 These adjusted reductions reflect 24.4 percent and 31.9 percent declines from the mean preintervention levels of inpatient days used and costs accrued, respectively, by the overall group of NY/NY placements studied here. Outpatient services In contrast to the reductions in inpatient services that have been docu- mented so far, table 9 shows that the number of outpatient visits and the costs increase, by 95.2 percent and 114.1 percent, respectively, for the 1995 NY/NY cohort. Looking at the case-control group, the same group as was used for the inpatient analysis, the significant and substantial increase among the cases is matched by a modest, nonsignificant pre/post increase in the number of outpatient visits by the control group. These pre/post changes in visits and costs, when adjusted through a multi- variate model (table 10), yield an increase of 68.9 visits (95 percent CI, 47.1 to 90.6) and $5,612 (95 percent CI, $3,871 to $7,352) associated with NY/NY placement. Averaging this over all 733 NY/NY placements in 1995 results in increases of 47.2 visits (95 percent CI, 32.3 to 62.1 visits) and $3,843 (95 percent CI, $2,651 to $5,035). These adjusted amounts reflect proportional increases of 75.9 percent and 81.5 percent over the mean number preintervention outpatient visits consumed and costs accrued, respectively. Fannie Mae Foundation 14 This is estimated by multiplying the reductions in days and costs associated with NY/NY in the case group, 12.6 and $7,983, respectively, by the 502 persons with NY/NY placements and Medicaid inpatient records (from which the control group was selected) and then dividing by the 733 persons with NY/NY placements in 1995. Culhane.qxd 6/18/02 11:49 AM Page 124 Public Service Reductions Associated with Supportive Housing 125Table 8.Regression Model Estimating Effects on Changes in Medicaid-Reimbursed Inpatient Hospital Days Used and Related Costs in the Two-YearPeriods before and after the NY/NY Intervention (N = 457 Matched Pairs)Reduction in Cost Stays (Days) Reduction ($)Parameter 95% CI Parameter 95% CICovariate Estimate Lower Upper Estimate Lower UpperIntercept 5.67 – 14.99 26.32 –807 –11,986 10,373Received NY/NY 12.56 *** 6.19 18.93 7,983 *** 4,608 11,358placementGap—Hospital to NY/NY interventiona 0.01 –0.01 0.04 11.6 * 0.2 22.9 No preintervention –27.33 *** –39.95 –14.71 –15,063 *** –21,916 –8,210 Medicaid record Medicaid days 0.84 *** 0.69 1.00 –10.4 –95.0 74.2 (preintervention) Medicaid stays –1.95 –5.74 1.85 –999.6 –3,213.5 1,214.3 (preintervention) Amount billed to Medicaid 0.00 0.00 0.00 0.8 *** 0.7 1.0 (preintervention) Shelter days 0.01 –0.01 0.02 3.0 –6.3 12.2 (preintervention) Age –0.01 –0.43 0.41 27.6 –188.9 244.1 Male –4.50 –11.44 2.44 –1,543 –5,400 2,314 Black race –8.28 * –15.30 –1.26 –4,853 * –8,680 –1,026 295 diagnosis –21.60 *** –29.24 –13.96 –7,947 *** –12,177 –3,717 (schizophrenia) 296 diagnosis –21.07 *** –29.24 –12.91 –11,524 *** –16,152 –6,895 (affective disorder) Chemical dependency –13.74 *** –20.15 –7.33 –9,979 *** –13,402 –6,556 diagnosis aIn records where there is no preintervention period Medicaid inpatient stay, a value of 731 days is inserted. *p < 0.05. **p< 0.01. ***p< 0.001. Use of VA hospitals.This analysis examines inpatient VA hospital data from 1992 through 1999 across pre- and postintervention periods for the 2,496 persons with NY/NY placements in the years 1994–97 and, when applicable, the controls matched to individual observations.15 Housing Policy Debate 15 Judging from the information available, approximately 20 percent of the persons receiving NY/NY placement claim veteran status. These proportions are somewhat higher for men (27 percent) and very small for women (2 percent). This suggests that approximately one fifth of the persons with NY/NY placements are eligible for VA serv- ices and that a smaller number will likely actually use these services. The DRGs for hospitalizations for persons with NY/NY placements show that over 75 percent of the stays involved treatment for mental illness or substance abuse or both (Metraux et al. 2000). Culhane.qxd 6/18/02 11:49 AM Page 125 126 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable 9.Outpatient Visits Reimbursed by Medicaid for Persons in NY/NY Housing and Controls in the Two-Year Periods before and after the NY/NY InterventionNY/NY NY/NY Controls(Total 1995) (Matched Pairs) (Matched Pairs)N 733 457 457Medicaid service users 502 457 457Pre-NY/NY intervention (two years)Total persons with outpatient 461 (62.9%) 419 (91.7%) 410 (89.7%)visits Total outpatient visits 45,615 42,623 37,323 Mean visits (all persons) 62.2 93.3 81.7 Mean visits (hospital users) 98.9 101.7 91.0 Total amount billed to Medicaid $3,448,239 $3,246,487 $2,796,755 Mean amount billed (all persons) $4,704 $7,104 $6,120 Mean amount billed $7,480 $7,748 $6,821 (hospital users) Post-NY/NY Intervention (two years) Total persons with outpatient 483 (65.9%) 440 (96.3%) 374 (81.8%) visits Total outpatient visits 89,042 80,913 40,109 Mean visits (all persons) 121.5 177.1 87.8 Mean visits (hospital users) 184.4 183.9 107.2 Total amount billed to Medicaid $7,382,207 $6,587,614 $3,218,494 Mean amount billed (all persons) $10,071 $14,415 $7,043 Mean amount billed $16,013 $14,972 $8,606 (hospital users) Notes:For the number of outpatient visits (non-HHC) reimbursed by Medicaid: Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention differences, t= 1.5 (456 dfand p = 0.15), and for postintervention differences, t= 7.7 (456 dfand p < 0.0001). Using paired-comparison ttests, pre/post differences yield, within the NY/NY group, t= –8.2 (456 dfand p< 0.0001), and within the control group, t = –0.9 (456 df and p= 0.35). For the billing of outpatient visits (non-HHC) reimbursed by Medicaid: Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention differences, t= 1.4 (456 dfand p = 0.15), and for postintervention differences, t= 7.6 (456 dfand p < 0.0001). Using paired-comparison ttests, pre/post differences yield, within the NY/NY group, t= –9.2 (456 dfand p< 0.0001), and within the control group, t = –1.7 (456 df and p= 0.10). Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 126 Public Service Reductions Associated with Supportive Housing 127Table 10.Regression Model Estimating Effects on Changes in Visits andCosts related to Medicaid Outpatient Use in the Two-Year Periods before andafter the NY/NY Placement (N = 457 Matched Pairs)Reduction in Cost Visits Reduction ($)Parameter 95% CI Parameter 95% CICovariate Estimate Lower Upper Estimate Lower UpperIntercept –0.97 –66.20 64.25 –4,130 –9,394 1,134Received NY/NY –68.88 *** –90.62 –47.14 –5,612 *** –7,352 –3,871placementGap—Visit to NY/NY 0.11 ** 0.04 0.18 10.02 *** 5.00 15.04interventiona No preintervention –64.36 ** –110.25 –18.47 –5,427 ** –8,808 –2,047 Medicaid record Medicaid visits 0.53 *** 0.26 0.81 –3.69 –26.99 19.61 (preintervention) Amount billed to Medicaid 0.00 0.00 0.00 0.58 *** 0.30 0.87 (preintervention) Shelter days –0.05 –0.14 0.05 –2.90 –9.33 3.53 (preintervention) Age –1.24 –2.67 0.19 –10.99 –118.89 96.92 Male 32.64 * 6.86 58.42 2,637 * 525 4,749 Black race 0.79 –22.25 23.82 545 –1,296 2,385 295 diagnosis –25.96 –52.75 0.82 –2,820 ** –4,867 –774 (schizophrenia) 296 diagnosis –38.09 ** –66.53 –9.64 –2,641 * –4,725 –557 (affective disorders) Chemical dependency 6.83 –22.38 36.04 1,009 –1,306 3,324 diagnosis aIn records where there is no preintervention period Medicaid inpatient stay, a value of 730 days is inserted. *p < 0.05. **p < 0.01. ***p< 0.001. Among the NY/NY placements, 323 (12.9 percent) had some record of VA inpatient hospitalization between 1992 and 1999. Of these, 255 (10.2 percent) had records of hospitalization, and these observa- tions, whether or not they had shelter records, were matched with per- sons who had DHS shelter records (but not necessarily in the two-year preintervention period). Of the 323 observations with VA records, 294 (91 percent) were matched with control observations. Table 11 shows that both NY/NY and control groups had (by design) virtually identical numbers of days of preintervention VA inpatient hospital use and that this use declined significantly for both cases and controls in the post- intervention period. However, the decline was substantially greater among the cases than among the controls, leading to statistically sig- nificant postintervention case-control differences (p < 0.001). Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 127 128 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable 11.VA Inpatient Days Consumed by Persons in NY/NY Housing andControls in the Two-Year Periods before and after the NY/NY InterventionNY/NY–DHS NY/NY Controls(Total 1994–97) (Matched Pairs) (Matched Pairs)N 2,496 294 294Total service users 323 294 294Persons with preintervention 255 (10.2%) 229 (77.9%) 229 (77.9%)VA hospitalization Total preintervention hospital 19,578 15,332 15,130days Mean preintervention hospital 7.8 52.1 51.5 days (all persons) Mean preintervention hospital 76.8 67.0 66.1 days (hospital users) Persons with postintervention 169 (6.8%) 153 (52.0%) 180 (61.2%) VA hospitalization Total postintervention hospital 8,053 7,651 12,289 days Mean postintervention hospital 3.2 26.0 41.8 days (total) Mean postintervention hospital 47.7 50.0 68.3 days (hospital users) Note:Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention state hospital days, t = –0.8 (293 dfand p= 0.41), and for postintervention differences, t = 3.7 (293 df and p < 0.001). Using paired-comparison t tests, pre/post differences yield, within the NY/NY group, t = 6.9 (293 df and p < 0.0001), and within the control group, t = 2.3 (293 dfand p < 0.05). The regression model results on table 12 show a significant 14.4-day reduction in VA hospital use (95 percent CI, 5.6 to 23.1 days), all other factors held equal. When averaged over all 2,496 NY/NY placements made during the years 1994–97, this effect becomes considerably more diluted, resulting in an estimate of 1.9 days saved (95 percent CI, 0.7 to 3.0 days) per NY/NY placement. This represents an adjusted 24.4 per- cent decrease in mean preintervention VA hospital days used attributa- ble to the effect of a NY/NY placement.16 Incarceration in NYSDOCS prisons.The last type of institution included in this study consists of incarceration facilities: state prisons and city jails. For state prisons, data include incarcerations up to April 15, 1997, so all NY/NY placements made before April 15, 1995, are included in this analysis. Because of the availability of records, the Fannie Mae Foundation 16 This was estimated by multiplying 14.4 by the 323 persons with NY/NY placements and Medicaid inpatient (non HHC) records (from which the control group was selected) and then dividing by the 2,496 persons with NY/NY placements and Medicaid inpatient records. Culhane.qxd 6/18/02 11:49 AM Page 128 Public Service Reductions Associated with Supportive Housing 129Table 12.Regression Model Estimating Effects on Changes in VA HospitalDays Used in the Two-Year Periods before and after the NY/NY Intervention(N = 294 Matched Pairs)Parameter Estimate Lower UpperCovariate (Days Saved) (95%) CI (95%) CIIntercept 34.63* 0.54 68.72Received NY/NY placement 14.37** 5.60 23.14Gap—VA to interventiona 0.03 0.00 0.06Pre-NY/NY shelter record –18.31 –37.40 0.78VA days (preintervention) 0.77*** 0.62 0.91VA stays (preintervention) 0.02 –4.89 4.93 Pre-NY/NY shelter days 0.01 –0.01 0.04 Placement in 1994 Reference Category Placement in 1995 –1.45 –12.47 9.57 Placement in 1996 9.73 –0.67 20.12 Placement in 1997 9.26 –3.15 21.68 Age –0.62* –1.11 –0.13 Male –19.15* –34.97 –3.34 Black race –4.14 –12.60 4.31 295 diagnosis (schizophrenia) –15.40** –25.96 –4.84 296 diagnosis (affective disorders) –26.78*** –37.45 –16.12 Chemical dependency diagnosis –10.93* –19.43 –2.44 aFor those with no preintervention VA inpatient record, the gap is set at 731 days. *p < 0.05. **p < 0.01. ***p< 0.001. case-control group used in the DHS shelter analysis is again used here to assess differences in pre- and postintervention days spent incarcer- ated. Thus, the case group includes only those persons with DHS shel- ter records, and 44 pairs were omitted from the analysis because the control observation was incarcerated at the intervention point and thus would have biased the pre/post analyses.17 Table 13 shows that low proportions of observations in either group have records of incarceration. Despite this, the NY/NY placements show substantial reductions in the use of state prisons. In the case- control comparison, the state prison utilization rate for the two groups is very similar (and statistically nonsignificant) in the preintervention period, suggesting that, for the purposes of this analysis, the groups are comparable. In the postintervention period, the NY/NY group shows a substantial, statistically significant reduction in the number of days incarcerated (p < 0.0001), while the control group fails to show any statistically significant change in the number of persons incarcerated or in the total number of days the group was incarcerated. Housing Policy Debate 17 In these 44 pairs, the case observation did not necessarily have an incarceration record. Culhane.qxd 6/18/02 11:49 AM Page 129 130 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable 13.NYSDOCS Prison Days Used by Persons in NY/NY Housing and Controls in the Two-Year Periods before and after the NY/NY InterventionNY/NY–DHS NY/NY Controls(Total up to 4/1/95) (Matched Pairs) (Matched Pairs)N 3,196 2,285 2,285Total service users 109 94 136Pre-NY/NY intervention periodPersons incarcerated 87 (2.7%) 75 (3.3%) 74 (3.2%)Time incarcerated (days) 29,569 25,490 25,241Time incarcerated (days per 9.3 11.2 11.0total persons)Time incarcerated (days per 339.9 339.9 341.1 person incarcerated) Post-NY/NY intervention period Persons incarcerated 36 (1.1%) 32 (1.4%) 78 (3.4%) Time incarcerated (days) 7,818 6,938 26,236 Time incarcerated (days per 2.4 3.0 11.5 total persons) Time incarcerated (days per 217.1 216.8 336.4 person incarcerated) Note:Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention state incarceration days, t = –0.05 (2,294 dfand p = 0.96), and for postinter- vention differences, t = 5.2 (2,294 df and p < 0.0001). Using paired-comparison ttests, pre/post differences yield, within the NY/NY group, t= 5.2 (2,294 df and p < 0.0001), and within the control group, t= –0.2 (2,294 df and p= 0.83). After multiple regression was used to control for various factors in table 14, it was determined that having a NY/NY placement is associ- ated with a reduction of 7.9 days (95 percent CI, 4.8 to 11.0 days). This estimate differs from those in the previous models in that placements in the case and control groups were included regardless of whether they had a state prison record. Since case-control data are unavailable for the incarceration of persons without a DHS shelter record, this adjusted reduction is used, without further adjustment, as the per placement reduction in prison use attributable to NY/NY. Taking this reduction estimate as a proportion of average preintervention prison use, 7.9 days represents an 84.8 percent reduction in the mean prein- tervention days spent incarcerated by the NY/NY group. Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 130 Public Service Reductions Associated with Supportive Housing 131Table 14.Regression Model Estimating Effects on Changes in NYSDOCSIncarceration Days in the Two-Year Periods before and after the NY/NY Intervention (N = 2,285)Parameter Estimate Lower UpperCovariate (Days Saved) (95%) CI (95%) CIIntercept –24.56*** –33.75 –15.36Received NY/NY placement 7.89*** 4.81 10.97Any incarceration (preintervention) –48.98** –84.54 –13.41Days incarcerated (preintervention) 1.04*** 0.95 1.12DHS days (preintervention) 0.01** 0.00 0.01Age at NY/NY intervention 0.35*** 0.21 0.50 Male –3.68** –6.46 –0.89 Black 1.88 –1.39 5.15 Mental illness indicator 2.43 –0.80 5.66 Drug/Alcohol dependency indicator 0.47 –2.81 3.75 NY/NY placement in 1990 Reference Category NY/NY placement in 1991 –3.35 –9.96 3.25 NY/NY placement in 1992 –0.80 –6.24 4.64 NY/NY placement in 1993 –0.71 –5.83 4.41 NY/NY placement in/after 1994 –0.31 –5.01 4.40 *p < 0.05. **p < 0.01. ***p< 0.001. Incarceration in NYCDOC jails.Analysis of incarceration data from NYCDOC augments the NYSDOCS analysis, which does not cover any incarceration episodes in county or municipal corrections facilities. This analysis of NYCDOC incarceration records for Riker’s Island and other local jail facilities follows the same case-control group that was exam- ined in the NYSDOCS and DHS analyses.18 As shown in table 15, the number of persons incarcerated, as well as the time spent in jail, declined significantly for the NY/NY group between the pre- and postintervention periods. Persons spending time in jail represented 12.0 percent of the total NY/NY group in the prein- tervention period, but only 8.2 percent of this group in the postinter- vention period. The total number of days incarcerated fell 39.8 percent after the housing placements. The average number of persons and days Housing Policy Debate 18 The data set used here is larger than the one in the NYSDOCS analysis because information was available through 1999 (see appendix A). There are no duplicate incar- ceration records between this and the NYSDOCS analyses, although some of the records examined here immediately precede state prison records used in the other analysis. Of the 1,590 offenses related to the study group over this time, by far the most frequently occurring types involved possession or sale of drugs (27.2 percent), offenses related to assault (12.8 percent), theft (11.6 percent), and larceny (8.9 percent). Some 39 percent of these offenses were charged as felonies (Metraux, Culhane, and Hadley 2001b). Culhane.qxd 6/18/02 11:49 AM Page 131 132 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable 15.NYCDOC Jail Days Used by Persons in NY/NY Housing and Controlsin the Two-Year Periods before and after NY/NY InterventionNY/NY–DHS NY/NY Controls(1989–97) (Matched Pairs) (Matched Pairs)N 4,679 3,284 3,284Total service users 766 607 716Pre-NY/NY intervention periodPersons jailed 563 (12.0%) 441 (13.4%) 480 (14.7%)Time in jail (days) 46,574 36,165 41,481Time in jail (days per total 10.0 11.0 12.6persons)Time in jail (days per person 82.7 82.0 86.4 jailed) Post-NY/NY intervention period Persons jailed 383 (8.2%) 308 (9.4%) 457 (13.9%) Time in jail (days) 28,027 21,711 37,828 Time in jail (days per total 6.0 6.6 11.5 persons) Time in jail (days per person 73.2 70.4 82.8 jailed) Note:Between the NY/NY and control groups, paired-comparison ttests assessing difference yield, for preintervention city jail incarceration days, t= 1.4 (3,283 df and p < 0.17), and for postinter- vention differences, t = 4.8 (3,283 df and p < 0.0001). Using paired-comparison ttests, pre/post differences yield, within the NY/NY group, t= 4.8 (3,283 df and p < 0.0001), and within the control group, t= 1.1 (3,283 df and p < 0.29). incarcerated fell such that not only did fewer persons get jailed after their housing placements, but, for those incarcerated, the average time spent behind bars also fell.19 These pre/post dynamics are not replicated in the control group. While the NY/NY and control groups are compara- ble in their use of jails in the preintervention period, the magnitude of the reduction per placement realized by the NY/NY group (4.4 days) is statistically significant (p < 0.0001), as opposed to the smaller and nonstatistically significant pre/post reduction for the control group (1.1 days). When multiple regression is used to control for other factors in table 16 and everything else is held constant, NY/NY placement is associated with a 3.8-day reduction per placement (95 percent CI, 1.8 to 5.8 days). As in the NYSDOCS analysis, the estimated regression model includes all of the DHS case-control observations, regardless of whether they have a record of jail use, and 3.8 days represents the number of days Fannie Mae Foundation 19 A total of 54 matched pairs were omitted because the control observation was incar- cerated during the intervention date. Culhane.qxd 6/18/02 11:49 AM Page 132 Public Service Reductions Associated with Supportive Housing 133Table 16.Regression Model Estimating Effects on Changes in NYC Jail Daysin the Two-Year Periods before and after the NY/NY Intervention (N = 3,284 Matched Pairs)Parameter Estimate Lower UpperCovariate (Days Saved) (95%) CI (95%) CIIntercept –18.21*** –24.17 –12.26Received NY/NY placement 3.81*** 1.79 5.84Any jail (preintervention) –22.32*** –28.61 –16.03Days jailed (preintervention) 0.93*** 0.88 0.98Any shelter use (preintervention) 1.80 –1.11 4.71DHS days (preintervention) 0.01*** 0.00 0.01 Age at NY/NY intervention 0.25*** 0.17 0.33 Male –4.41*** –6.48 –2.35 Black –0.03 –2.09 2.04 Mental illness indicator –1.64 –3.73 0.45 Drug/Alcohol dependency indicator –0.33 –2.61 1.94 NY/NY placement in 1990 Reference Category NY/NY placement in 1991 3.05 –1.46 7.56 NY/NY placement in 1992 4.61* 0.28 8.94 NY/NY placement in 1993 2.93 –1.60 7.46 NY/NY placement in 1994 0.39 –4.12 4.91 NY/NY placement in 1995 –0.12 –4.84 4.59 NY/NY placement in 1996 or 1997 2.49 –1.89 6.87 *p < 0 .05. ** p< 0 .01. ***p < 0 .001. per placement (without further adjustment) attributed to a NY/NY placement. This represents a 38.0 percent decrease in the mean prein- tervention period number of incarceration days used by the case group. Cumulative system effects The results of the system-specific analyses have thus far been pre- sented separately by agency over the two-year postplacement period. For purposes of facilitating an interpretation of the cumulative effects of the intervention within and across systems, summary results are provided in tables 17 and 18. Table 17 estimates the costs of service utilization by intervention group members in the two years before their housing placement by multiply- ing service days used by the average per diem service cost (in 1999 dol- lars). These costs are then annualized by dividing by two. The results show that, per placement per year, the total mean cost of service uti- lization for the two-year pre-NY/NY placement period was $40,451. The bulk of those expenditures occurred in health services (86 percent, Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 133 134 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable 17.Summary of Mean Two-Year Pre-NY/NY Intervention Period Service Use across Seven Service Providers Mean Days Used (Two Years Cost AnnualizedService Provider Pre-NY/NY) Per Diem* (Two Years)* Cost*DHS 137.0 $68 $9,316 $4,658OMH 57.3 $437 $25,040 $12,520HHC 16.5 $755 $12,458 $6,229Medicaid (inpatient) 35.3 $657 $23,192 $11,596Medicaid (outpatient stays) 62.2 $84 $5,225 $2,613VA 7.8 $467 $3,643 $1,822Department of Corrections 9.3 $79 $735 $368 (state) Department of Corrections 10.0 $129 $1,290 $645 (city) Total $80,899 $40,451 *In 1999 dollars. Totals reflect rounding. or $34,778) and in emergency shelter services (11 percent, or $4,658). Criminal justice services (incarceration costs only) accounted for only 3 percent, or $1,012 per year. Table 18 summarizes estimates of the cost reductions in service use, based on pre/post placement comparisons and as adjusted by the case- control regression analyses. Cost savings are again imputed based on estimated per diem costs by service system in 1999 dollars. Results indi- cate that placement in NY/NY housing is associated with a $12,146 net reduction in health, corrections, and shelter service use annually per person over each of the first two years of the intervention. Half of those cost reductions are associated with reduced use of state psychiatric inpa- tient services, and another quarter (23 percent) are associated with reduced use of emergency shelter services. Half of the remaining quar- ter in net savings is associated with reduced use of NYC public hospitals (10.9 percent of the total) and VA hospitals (3.7 percent of the total). Reductions in costs associated with Medicaid inpatient services out- weigh, by $843 (6.9 percent of the total cost reductions), the increases in Medicaid outpatient services. Taken together, about 95 percent of the cost reductions are associated with reductions in health and shelter services. The criminal justice system costs account for the remaining 4.5 percent of the total cost reductions associated with a supportive housing placement. Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 134 Public Service Reductions Associated with Supportive Housing 135Table 18.Summary of Estimated Cost Reductions Associated with Reductionsin Service Use Attributable to NY/NY Housing, by TypeDays Saved Cost Cost Annualized(Two Years Reductions Per Reductions* CostService Provider Pre/Post) 95% CI Diem* (Two Years) Reductions*DHS 82.9 77.4–88.5 $68 $5,637 $2,819OMH 28.2 20.8–35.6 $437 $12,323 $6,162HHC 3.5 2.0–5.0 $755 $2,643 $1,322Medicaid (inpatient) 8.6 4.2–13.0 $657 $5,650 $2,825Medicaid (outpatient visits) –47.2 –62.1 to –32.3 $84 –$3,965 –$1,983VA 1.9 0.7–3.0 $467 $887 $444Department of Corrections (state) 7.9 4.8–11.0 $79 $624 $312 Department of Corrections (city) 3.8 1.8–5.8 $129 $490 $245 Total $24,289 $12,146 *In 1999 dollars. Totals reflect rounding. Do reductions in service use offset the costs of supportive housing? One of the primary purposes of the previous analyses was to determine whether reductions in service use attributable to a housing placement offset the costs of the intervention. To compare the costs of the inter- vention with the reduced service system costs, both sets of costs must be computed in comparable terms. In the previous cross-system analy- sis, the reductions in service use were calculated in terms of annualized average cost reductions per placement in the two-year period after housing placement. Alternatively, housing cost figures, given the an- nual budgeting process by which they are calculated by city and state officials (the methodology for deriving the housing costs is provided in appendix B), are measured in annual costs per housing unit. Each measure has its usefulness, the former for service system planners who need to take account of client turnover and project costs for a pool of placements and the former for housing planners, who need to project costs based on fully occupied units of housing, irrespective of turnover.20 Housing Policy Debate 20 Because these measures are not directly comparable, they must be converted, taking into account client turnover, to produce annualized cost and cost-reduction estimates. Because tenant-level data on length of housing tenure for each placement were not available for this analysis, aggregated data on longevity of placement in NY/NY hous- ing, presented in Lipton (1996), are used. These data indicate that NY/NY tenants stay in housing, on average, for 17.9 months of the two-year postintervention period. The annualized length of tenure is therefore 8.95 months, or 0.746 a year. The inverse of Culhane.qxd 6/18/02 11:49 AM Page 135 136 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyPer placement per year.Table 19 shows the conversion of annual costsper housing unit into annualized costs per placement, both overall andas broken down by the two housing models featured in NY/NY.21 Theannualized cost per placement, averaged over the two-year postplace-ment period and derived by multiplying the annual costs per housingunit by the average annual length of tenure (0.746 years ), is $13,570.This can then be compared with the adjusted service cost reductionsreported in the last column of table 18, which show an annualizedreduction of $12,146 per placement. The result is a net annual cost of$1,425 associated with a NY/NY placement. It is noteworthy that thesupportive housing units, which comprise two-thirds of the units devel-oped, have an annualized cost per placement of $12,889 and thereforeoperated at a lower average annualized net cost of $744 per placement. Table 19.Estimated Annual Costs per Unit and Annualized Costs per Placement of NY/NY Housing, by Housing Type Number Annualized Annualized per Housing Type of Units per Unit Cost* Placement* Community residence (mean) 1,384 $19,662 $14,668 Supportive housing (mean) 2,231 $17,277 $12,889 Weighted mean 3,615 $18,190 $13,570 *In 1999 dollars. Per housing unit per year.Alternatively (and inversely), one can con- vert the annualized service utilization reductions, reported in terms of placements (table 18) into annualized reductions per housing unit (table 20). These reductions are then expressed in terms of average annualized service cost reductions per housing unit by multiplying the annualized per placement service reductions by the annualized number of tenants per housing unit (1.34 per year). This procedure yields turnover-adjusted cost reductions per housing unit per year of $16,281, imputing an assumption of year-round housing occupancy. This figure can be compared with the estimated cost per housing unit per year, Fannie Mae Foundation this number, 1.34, produces the annualized average number of tenants per housing unit. These numbers are used to compute both annualized per placement costs from the annual housing unit costs in appendix B and annualized per housing unit service cost reductions from the per placement service use reductions in table 18. These two computations reflect inverse procedures and are equally valid approaches for comparing the service cost savings and housing costs associated with the intervention (all figures are in 1999 dollars). 21 See appendix B for the specific housing programs and also table B.8. Culhane.qxd 6/18/02 11:49 AM Page 136 Public Service Reductions Associated with Supportive Housing 137as shown in table 19 (and as computed in appendix B), which alsoassumes year-round occupancy. Comparing the average annual cost of a housing unit ($18,190) with the comparable measure for the serv-ice utilization reductions yields a net cost of $1,908 per unit per year.Again, the supportive housing units, which comprise two-thirds of theunits developed, operate at a lower net cost of $995 per housing unitper year. The net cost attributable to year-round occupancy of NY/NYhousing can be calculated by multiplying the annualized per unit netcost by 3,615 (the total number of housing units developed), yielding anet annual cost of $6,897,420 per year.Finally, multiplying the turnover-adjusted cost reductions by 3,615 foreach category of service produces an estimate of the annual cost reduc- tions (or increases) accruing to each service type attributable to a year- round housing placement, as shown in the last column of table 20. These figures provide useful information about the impact of the inter- vention on aggregate service use annually by service type and demon- strate that annual service use was reduced by $58.9 million. This compares to the annual cost of the NY/NY intervention (including oper- ating, service, and debt service costs) of approximately $65.8 million. Table 20.Annualized Cost Reductions per Placement and per Housing Unit, and Total NY/NY Housing Units (N = 3,615), by Service Type Annualized Annualized Cost Total Cost Cost Reductions Reductions per Reduction by Service Provider per Placement* Housing Unit* NY/NY Units* DHS $2,819 $3,779 $13,660,436 OMH $6,162 $8,260 $29,860,094 HHC $1,321 $1,771 $6,401,361 Medicaid (inpatient) $2,825 $3,787 $13,689,511 Medicaid (outpatient stays) –$1,982 –$2,657 –$9,604,464 VA $444 $595 $2,151,555 Department of $312 $418 $1,511,903 Corrections (state) Department of $245 $328 $1,187,232 Corrections (city) Total $12,146 $16,281 $58,857,628 *In 1999 dollars. Totals reflect rounding. Discussion The placement of homeless people with SMI in supportive housing is, as expected, associated with substantial reductions in homelessness. Not only do homeless people with severe mental disabilities who are placed in housing have marked reductions in shelter use, they also Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 137 138 Dennis P. Culhane, Stephen Metraux, and Trevor Hadleyexperience marked reductions in their use of hospital and correctionalfacilities. Although other studies tracking the placement of homelesspeople in housing have found comparable housing retention rates, onlya few, more limited analyses have assessed reductions in collateral serv-ices (Averyt and Kamis-Gould 2000; Proscio 2000). By contrast, thisstudy provides a uniquely broad and more comprehensive test, employ-ing a case-control study design and examining the impact of a compara-tively large number of housing placements on seven major publiclyfinanced service systems.It is important to note that this study was able to quantify for the firsttime in the published literature the extent of service use by homelesspeople with SMI before a housing placement. Results show that such persons are extensive users of publicly funded services, particularly inpatient health services, and that they accumulate an average of $40,451 per year in health, corrections, and shelter system costs. While the costs of services before housing placement are comparably high for the cases and controls, due to the matching criteria of this study, it is not clear whether these costs can be generalized to all homeless people with SMI, and they certainly cannot be generalized to homeless people irrespective of their mental health. Nevertheless, in light of this high cost for such a significant number of persons, the importance of the effect found here—that the supportive housing intervention signifi- cantly reduces these costs—is further reinforced. Strictly on the basis of the direct cost reductions measured here and compared with the annual cost of the housing, the NY/NY initiative was a sound investment of public resources. The $6.9 million net annual cost, or $1,908 per housing unit per year, represents approxi- mately 10 percent of the annual overall cost of providing this housing. However, supportive housing units, which were the more common type developed under this initiative and which better represent the trend in housing development for people with mental illness, operated at a more modest per unit cost of $995 per year, or 5 percent of the overall hous- ing unit cost. In other words, 95 percent of the costs of the supportive housing (operating, service, and debt service costs) are compensated by reductions in collateral service attributable to the housing place- ment. This modest cost is particularly striking, given the magnitude of the initiative, which required an original capital investment of $200 million and which costs $65.8 million annually (including service, operating, and debt service costs). It should be noted that the service reductions measured here represent a conservative assessment of the impact of the initiative on service use and costs. First, by limiting the analysis to the impact on service use in Fannie Mae Foundation Culhane.qxd 6/18/02 11:49 AM Page 138 Public Service Reductions Associated with Supportive Housing 139the first two years postintervention, the study has included the stabi-lization period associated with entry into housing. As in other serviceinterventions for people with SMI, service use often increases tem-porarily following placement, because tenants’ unmet health and psy-chiatric needs are more likely to be identified and treated once theyreceive regular, periodic case management services (Pollio et al. 2000).If this were the case here, one would expect service use to decline andstabilize over time, producing net cost savings in successive years.However, this possibility must be balanced against the possibility thatpeople may be engaged in services more intensively before a housingplacement, in part to prepare them for it. This area deserves furtherstudy. This study did not include all direct or indirect costs associated with service use by the homeless persons eventually placed in housing. Street outreach services, soup kitchens, and services provided by drop- in centers were not included. Health services funded by the federal Health Care for the Homeless program were not included. Other clini- cal and social services provided at shelters that are funded by grants from the Department of Housing and Urban Development’s (HUD’s) McKinney Act programs were also not included, and neither were the costs of uncompensated care provided by private hospitals. Not included as well are the social costs of homelessness, which are far more difficult to enumerate or to associate with individual persons. They include the costs of crime to crime victims, to the courts, and to the police, and the private and public costs of accommodating homeless- ness (or not) in public spaces. Finally, many of the potential benefits of the housing initiative were not measured here. Residents of supported housing are more likely to secure voluntary or paid employment (HUD 1994) and to experience an improved quality of life. Investments in supported housing have also been shown to be associated with improved neighborhood quality and property values (Arthur Andersen LLP et al. 2000). Lastly, the social value of reduced homelessness and of providing greater social protec- tion for those who are disabled, while not possible to translate into eco- nomic terms, constitutes an important if less tangible benefit to society. Were all such costs and benefits included, these unmeasured costs of homelessness and benefits of the housing intervention would have increased its already significant net benefit (and potential cost savings). Although this study was limited to one locality and cannot be general- ized to all urban areas, the results have important public policy impli- cations. Research suggests that on any given day, as many as 112,000 single adults with SMI are homeless in the United States and that as Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 139 140 Dennis P. Culhane, Stephen Metraux, and Trevor Hadleymany as 280,000 single adults are chronically homeless.22 If such per-sons, or even significant proportions of them, are extensive users ofacute care health services, public shelters, and criminal justice systems,then the results of this study suggest that an aggressive investment in supportive housing is warranted. While such housing may not beappropriate or effective for every person who is homeless and mentallyill, enough would likely benefit that their placement in housing couldsignificantly offset the costs of a targeted initiative, such as was demon-strated here. In effect, the results presented here indicate that policymakers could substantially reduce homelessness for a large, visible segment of the homeless population—often thought to be stubbornlybeyond the reach of the social welfare safety net—at a very modest costto the public. However, while reductions in service use may nearly cover the costs of supportive housing intervention in the aggregate (assuming that the results given here can be generalized beyond NYC), it remains a major public policy challenge to shift funds from one set of purposes (health, jails, prisons) to another (housing or housing support services). Different levels of government pay for different activities, and some will have to do so regardless of whether homeless people are using them (jails and prisons, for example). Moreover, legislative committees with responsibil- ity for housing cannot appropriate funds from health committees for housing or housing support purposes, regardless of the savings in health costs that might justify the expenditure. So, the challenge facing propo- nents of a national strategy to increase the supply of supportive housing will be to determine how costs can be paid in one area (for housing or housing support services), when the bulk of the savings from the inter- vention will accrue elsewhere (state mental health services, Medicaid, etc.). In New York, a complex package of federal, state, and city re- sources was required to pay for the operating, service, and debt service costs of the NY/NY initiative (see appendix B). Similarly, a national strategy will require the participation of various levels of government and multiple agencies within each level of government. Fannie Mae Foundation 22 The Burt, Aron, and Lee (2001) analysis of a 1996 federal survey of homeless persons suggests that as many as 840,000 people were homeless at one point in time in the United States that year. One-third of those were people in families, leaving approxi- mately 560,000 single adults. A meta-analysis of epidemiological research estimates that approximately 20 percent of homeless adults without children have SMI (Lehman and Cordray 1993), yielding an estimated 112,000 persons with SMI as homeless for that study period. Longitudinal research in two large U.S. cities (Philadelphia and New York) finds that people who are chronic shelter users, with or without a mental disabil- ity, represent approximately 50 percent of the daily shelter-using population (Kuhn and Culhane 1998) or an estimated 280,000 of the adults during the Burt, Aron, and Lee one-day study period. Given the differing sampling frames underlying their derivation, these figures must be understood as gross estimates only. Culhane.qxd 6/18/02 11:49 AM Page 140 Public Service Reductions Associated with Supportive Housing 141Operating costsA substantial hurdle that must be overcome in developing and sustain-ing permanent supportive housing is bridging the gap between thecosts of operating the housing and the extremely low incomes of pro-spective tenants. Supportive housing providers typically address thisgap through a direct housing subsidy to the tenant or housing unitand/or income supports to the tenants. The NY/NY initiative drew onboth strategies to cover the operating costs, relying on a combination offederal Section 8 subsidies, supplemental security income payments bythe state, and some direct state support, resulting in an average perunit subsidy of $4,600 (see appendix B, derived from table B.3). Historically, HUD has been the primary source of housing subsidies. An especially potent source of operating subsidies for supportive hous- ing serving homeless persons, including those with SMI, has been the McKinney-Vento Homeless Assistance Act, which authorized operating subsidies in various forms under its three major programs: Section 8 Moderate SRO Rehabilitation, the Supportive Housing Program, and Shelter Plus Care. Federal primacy and initiative in the provision of operating subsidies are likely necessary if supportive housing for home- less persons with SMI is to be taken to scale. Even in relatively wealthy states, there is little evidence of an inclination to displace or even sig- nificantly add to the federal role in this regard (Twombley et al. 2001). Although the federal investment in incremental housing subsidies slowed to a trickle in the mid-1990s (DeParle 1996), significant oppor- tunities in this area may be on the horizon. In enacting HUD’s fiscal year (FY) 2001 budget, Congress explicitly stated its goal that “HUD and local providers increase the supply of permanent supportive hous- ing for chronically homeless, chronically ill people over time until the need is met (estimated 150,000 units)” (U.S. Senate 2000, 52–53). To that end, Congress maintained its recent requirement that 30 percent of McKinney-Vento funds (about $300 million per year based on recent annual appropriations) be targeted to permanent housing for homeless persons with disabilities. If this funding level is maintained, then this investment alone could result in subsidizing 96,000 new units of sup- portive housing over the next 10 years.23 Housing Policy Debate 23 This figure was derived from an estimated cost of $6,100 per unit (based on HUD’s FY 1999 estimates for the Shelter Plus Care program), assuming five-year terms and an annual inflation adjustment of 2 percent. Some 9,643 subsidies per year would result. Over a decade, this would translate into 96,433 incremental subsidies, assuming that any subsidies expiring during this period were renewed from another source. Culhane.qxd 6/18/02 11:49 AM Page 141 142 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyThe FY 2001 VA–HUD appropriations bill also authorized substantialchanges in the statute that allow local housing authorities to converttenant-based Section 8 vouchers into project-based subsidies linked tospecific units in order to spur new development in tight housing mar-kets or for special populations (Public Law 106–377, Sec. 232. 114Stat. 1441A at 31). The changes streamlined a previously underusedtool for housing development24 and also increased the ceiling on such“project-basing” of vouchers to 20 percent of the total tenant-basedportfolio (the previous limit was 15 percent). Nationally, this couldtranslate into over 300,000 potential project-based operating subsi-dies.25 Even in the absence of incremental vouchers, the new project-basing statutory authority can be of significant use in adding afford-able units to serve special populations, like homeless persons with SMI, who often cannot access housing in a tight rental market even with a tenant-based subsidy (Sard 2001).26 Thus, opportunities already exist to finance operating subsidies for permanent supportive housing. Of course, other issues remain to be resolved, including local resistance to the siting of such housing, the capacity of states and localities to develop it, and the ongoing financial burden of renewing operating subsidies (though this final factor is an issue in all federally subsidized affordable housing). Capital and debt service With respect to capital and debt service costs, the NY/NY initiative used a combination of city and state bonds, valued at nearly $200 mil- lion, and, secondarily, federal tax credits, valued at approximately $5 million, 27 to fund the capital costs for acquisition, development, and Fannie Mae Foundation 24 The NYC Housing Authority has been a notable exception among local housing agen- cies in its willingness in recent years to stimulate the development of new housing by project-basing a portion of its tenant-based Section 8 portfolio. The new statutory pro- visions could encourage other housing authorities to follow its example. 25 While Congress has appropriated between 50,000 and 100,000 incremental vouchers each year for the past half decade, the Bush budget proposes only 34,000 this year. 26 Evidence suggests that homeless households, even without SMI, struggle to use hous- ing subsidies in the private market. See “Judge Orders City” (2001) (noting that the city-funded rent subsidy program intended to house 460 homeless families had to date housed only 11 because private landlords were reluctant to accept them as tenants). 27 Approximately $25 million in federal tax credit expenditures were involved in the NY/NY initiative, but only 20 percent of these funds supported capital costs; 80 percent was used to fund operating reserves (see appendix B). Culhane.qxd 6/18/02 11:49 AM Page 142 Public Service Reductions Associated with Supportive Housing 143rehabilitation. The average debt service cost per unit per year for theNY/NY initiative is approximately $4,900 (see table B.7).Several factors will affect whether sufficient capital investment/debtservice can be obtained to develop supportive housing at the scale tomeet the need among homeless person with SMI. Nationally, competi-tion with other low-income housing programs for federal tax credits andcompetition with other state and local purposes for bond funds will posea challenge to state and local leaders who must balance demands forhousing for the homeless with other public interests. It remains to beseen whether the federal or state governments are willing to establish a priority for supportive housing for the homeless in this competitiveprocess or will allocate new dollars, given the potential for offsetting cost savings. The availability of capital/debt service funding is also likely to vary significantly across different regions of the country. To the extent that policy makers perceive existing affordable housing programs, including the Low-Income Housing Tax Credit and HOME programs, as insufficient to produce capital for housing that extremely low income households (below 30 percent of the area median income [AMI]) can afford, supportive housing providers would clearly benefit from a new production program targeted to that population.28 In the 106th Congress, senators from both major parties introduced bills directed to this purpose,29 and such a program was nearly enacted as part of the FY 2001 HUD appropriations bill. It remains to be seen whether the momentum will carry over into this congressional session or whether the new administration will identify such an initiative as a priority for HUD. Housing Policy Debate 28 Data from the 1999 American Housing Survey indicate that there are more than 5 million fewer housing units affordable and available (i.e., not occupied by a household of higher income) to households earning below 30 percent of the AMI than there are such households. Put another way, there are 7.7 million households with incomes below 30 percent of the AMI and 4.9 million units affordable to such households, but 2.6 mil- lion of those units are already occupied by households with incomes above 30 percent of the AMI—so only 2.3 million of the affordable units are actually available. (Dolbeare, C., unpublished data. Available from the author.) 29 S.2997, National Affordable Housing Trust Fund Act of 2000 (introduced by Senator John Kerry, D-MA), available at <http://www.thomas.loc.gov>; S.3033, The Housing Needs Act of 2000 (introduced by Senator Christopher Bond, R-MO), also available at <http://www.thomas.loc.gov>. Culhane.qxd 6/18/02 11:49 AM Page 143 144 Dennis P. Culhane, Stephen Metraux, and Trevor HadleySupportive servicesThe challenge of identifying funding for the services that must accom-pany this housing may well prove greater than finding the resources forthe housing. In the case of the NY/NY initiative, the NYS OMH paidfor nearly all of the services associated with this housing, at an averagecost of approximately $9,100 per unit per year.30 Results from thisstudy indicate that the expenditures by the state OMH also providedample returns, given that the plurality of cost reductions attributableto the intervention were reductions in OMH inpatient hospital costs($8,260 per housing unit, see table 20). Other inpatient health servicespaid by Medicaid, public hospitals, and the VA experienced combinedcost reductions as well ($6,153 per housing unit per year, see table 20). However, only Medicaid paid for some of the housing support services (at a net savings of $1,130 per housing unit per year). From the perspective of developing a national strategy, the question for proponents will be how to motivate other states or health service payers (and potential savers) to make the same commitment as the NYS OMH and, secondarily, Medicaid did under this initiative. One possible mechanism is to make housing support services (or more of them) reimbursable by Medicaid. Unfortunately, some of these services, such as intensive case management or community treatment teams, are already reimbursable by Medicaid at the states’ option, and many states do not avail themselves of it. Another option is to increase funds or create a set-aside for housing support service funds in the federal Mental Health Block Grant. However, states have successfully opposed federal mandates on block grant funds and may oppose such a mandate for this program. A third alternative would be to create a new program at the federal level that would provide matching funds from the Department of Health and Human Services (DHHS) for funds commit- ted by HUD through its set-aside in the McKinney-Vento Act for per- manent, supportive housing. In this case, services could be specifically targeted to housing for homeless people with mental disabilities.31 Whatever the specific mechanism chosen, the provision of support services will be a necessary component to any national strategy to address the housing problems of homeless people with SMI. Fannie Mae Foundation 30 For this calculation, tenant contributions, where applicable, were deducted from average state service contract amounts to produce a net service cost (see appendix B). 31 “The 106th Congress Wrap-up” contains language of agreement reached in the appropriations process but dropped at the later stages. Culhane.qxd 6/18/02 11:49 AM Page 144 Public Service Reductions Associated with Supportive Housing 145ConclusionIn sum, acquiring the resources for supportive housing will requirelocal, state, and federal leadership in all three areas that comprise theessential elements of this intervention: operating subsidies, capital/debtservice, and supportive services. The federal government, through newprograms, matching funds, and set-asides within existing programs,can provide the incentives that engage states’ interest. However, onlyexecutive leadership at the state level can compel various agencies towork together for the common, multijurisdictional purpose of develop-ing permanent supportive housing for homeless persons with SMI. Ourresearch has demonstrated the compensatory cost reductions of a sup- portive housing initiative, but only political will and leadership can act on such findings to guide the next initiative through the intergovern- mental and interagency maze. Of course, there are some caveats to this study—it is post hoc and quasi- experimental. Therefore, the extent to which cases and controls are truly comparable could not be addressed fully by random assignment. Comparability problems were reduced by matching cases and controls according to a variety of available demographic, service, and diagnostic criteria, and by statistically correcting for differences that may have remained. However, the extent to which unmeasured differences between the study groups may persist cannot be fully ascertained, nor can the possibility of a selection effect in the study sample be elimi- nated. Whether housing providers select for heavier service users or for less severe cases could not be determined, although every effort was made to produce results generalizable to the population of homeless per- sons with SMI from which the intervention group was drawn. Despite this limitation, it is also clear that there exists a relatively large pool of homeless persons with SMI for whom this housing is effective in achiev- ing housing stability and providing offsetting reductions in the use of collateral services. There are also caveats on the use of administrative data. Given the large volume of data entered into the databases of the service systems studied here and a level of quality control on data entry that is not as stringent as is usual for scientific studies, these administrative sources can be prone to missing data, keystroke errors, and erroneous informa- tion. Although missing data did not present a problem in these analy- ses, it is more difficult to ascertain the quality of the data along the other dimensions. Despite these potential problems, the only source that can inform a study of service use covering a large study group over an extended period of time, as is done here, is administrative data. Housing Policy Debate Culhane.qxd 6/18/02 11:49 AM Page 145 146 Dennis P. Culhane, Stephen Metraux, and Trevor Hadley Fannie Mae Foundation Future research should specify the effects of the various housing typeson patterns of service use. Causes of attrition from supportive housingand the housing transitions of those who exit supportive housing alsodeserve careful attention. For although most people remain stablyhoused two years after placement, a third of the clients exit this hous-ing and represent a substantial group that should be further studied.Future research could also benefit from replicating this study method,in that integrated administrative records provide a wealth of informa-tion on the utilization patterns and costs of a population that has otherwise proven costly and difficult to track and study. In particular,applying this method to studying patterns of homelessness and serviceuse among the majority of homeless persons who do not have an SMIcould likewise prove informative as to the potential efficacy of the vari- ous policies and intervention strategies that would target them. Fur- ther fruitful areas of study could examine the same group studied here but follow their service use over longer pre- and postintervention peri- ods (as additional data become available) and/or through service sys- tems not covered in this article. Culhane.qxd 6/18/02 11:50 AM Page 146 Public Service Reductions Associated with Supportive Housing 147 Table A.1.Selection Factors for Constructing the Matched-Pair Case-Control Groups Used in the AnalysesIntervention TotalService Time Years NY/NY MatchedProvider Frame Selected Placements Pairs RestrictionsDHS 1987–99 1989–97 4,679 3,338 NY/NY placements (cases) without a DHS shelter record are omitted (N = 1,341)OMH 1990–96 1992–94 2,396 570 NY/NY placements (cases) omitted include thosewithout a state hospital inpatient record (N = 1,499)with a state hospital record but without a DHSshelter record (N = 267)without an appropriate control match (N = 60) Medicaid 1993–97 1995 733 457 NY/NY placements (cases) omitted include thosewithout a Medicaid claim record (N = 231)without an appropriate control match (N = 45)HHC 1989–96 1991–94 2,396 791 NY/NY placements (cases) omitted include those(non-without a DHS shelter record (N = 412)Medicaid) with a DHS record but without an HHC inpatientrecord (N = 1,920)without an appropriate control match (N = 64)VA 1992–99 1994–97 2,496 294 NY/NY placements (cases) omitted include thosewithout a VA hospital inpatient record (N = 2,173)without an appropriate control match (N = 29)NYSDOCS 1987– 1989– 3,196 2,285 NY/NY placements (cases) omitted include those4/15/97 4/15/95 without a DHS shelter record (N = 911)whose matched control observation was incarcerated on the placement date (N = 44)NYCDOC 1987–99 1989–97 4,679 3,284 NY/NY placements (cases) omitted include thosewithout a DHS shelter record (N = 1,341)whose matched control observation wasincarcerated on the placement date (N = 54)Housing Policy Debate Appendix ASummary of Control Group Selection across Seven PublicServices SystemsCulhane_app.qxd 6/18/02 11:58 AM Page 147 148 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyAppendix B Estimating direct federal, state, and city expenditures on theNY/NY supportive housing initiativeIntroductionIn 1990, NYS and NYC agreed to collaborate on what became known asthe NY/NY supportive housing initiative. The agreement committedthe state and the city to jointly fund construction, operating, and socialservice costs for 3,600 community-based housing units in NYC for per- sons who were severely mentally ill and homeless. While much of the funding for this program comes from the state and the City of New York, as well as from the federal government, an array of private non- profit organizations administer these housing units. This appendix pro- vides estimates of the federal, state, and city outlays for the NY/NY initiative, in the aggregate, per year, per housing unit, and per housing placement. This compilation and disaggregation of costs is intended to provide a quantitative benchmark for evaluating returns from the pro- gram, as measured by the reductions in the use of collateral services that were detailed in the article. Data and methods Data on the distribution of housing units and both the total and the per unit operating, debt, and service costs were constructed from budget documents and in consultation with state and city administra- tors involved in financing and administering the programs.32 Inflation adjustments and unit-cost calculations were also verified by personal communication with program administrators. All figures are reported in 1999 dollars. Unless otherwise noted, all cost estimates assume full, year-round occupancy of housing units. Actual costs will differ for spe- cific sites and specific service contracts. Fannie Mae Foundation 32 Final confirmation of data on construction costs and debt service costs was obtained by personal communication with Michael Newman of the NYS OMH (March 5, 2001). Figures for the tenant contributions deducted from OMH-developed sites were con- firmed in consultation with Christopher Roblin of the NYS OMH (April 4, 2001, per- sonal communication). The value of federal Section 8 subsidies and the service cost estimates for the city-developed sites were confirmed by personal communication with Peter Bittle of the NYC Department of Mental Health (April 4, 2001). Information on the financing of city Department of Housing Preservation and Development (HPD)– developed units was obtained by personal communication with Timothy O’Hanlon of the NYC HPD (March 5, 2001). Information on the city HRA-developed units was obtained from David Mittelman of the NYC DHS (March 25, 2001). Culhane_app.qxd 6/18/02 11:58 AM Page 148 Public Service Reductions Associated with Supportive Housing 149Housing and service configurationsThe housing developed under the NY/NY initiative encompasses a variety of housing and service configurations and consists of severaldistinct models that combine housing with rehabilitative or supportservices. The residential continuum includes, on the one end, support-ive housing models that provide private, individual apartment-style living accommodations with varied levels of service support. In the sup-portive housing model, the intensity of services provided depends onneed as expressed by the tenant and can change as the tenant’s needschange. Tenants’ housing tenure is based on a lease arrangement andis not contingent on a prescriptive service plan. On the other end of the continuum are community residence living arrangements that provide more intense, structured regimens of supportive services. Community residential programs are supervised, and housing is part of a structured treatment plan. Both supportive housing and community residence approaches are tied into bodies of research literature that provide theo- retical rationales and evaluations for each approach. More relevant to the purposes of this inquiry is the fact that the two models are associ- ated with different sources and amounts of funding. Table B.1 shows the number of housing units, grouped by different housing and service configurations, that are funded under NY/NY. Under the community residences subheading, community residence/ SROs provide extended-stay housing in SRO living units with on-site services for those whose self-maintenance and socialization skills are minimal. Community residences are single-site facilities with either pri- vate or shared bedrooms, meals provided, access to on-site rehabilitative services, and 24-hour staff coverage. Community residences often target special populations, such as people with co-occurring mental health and substance abuse problems, and they seek to eventually place residents in less service-intensive, more permanent housing arrangements. The supportive housing grouping includes various state- and city- administered programs that range from scattered-site, individual apart- ments to clustered apartments or SRO units in a single development. In both cases, services are available on a periodic or as-needed basis. For capital (state) supportive housing, construction was initiated by NY/NY funding, while rental (state) supportive housing used existing units in the private rental market. For city-funded supportive housing, the Department of Housing Preservation and Development (HPD) administers NYC–HPD housing units, the HRA administers NYC–HRA units, the NYC–NYS units are administered by the city with state capi- tal funding, and the three NYC rental units use existing, private apart- ments that receive state funding for services. Housing Policy Debate Culhane_app.qxd 6/18/02 11:58 AM Page 149 150 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable B.1.NY/NY–Funded Housing: Number of Units under Different Housing/Service ConfigurationsHousing UnitsCommunity residencesCommunity residences/SROs 713Community residences 671Supportive housingCapital (state) 285Rental (state) 520Capital (NYC–HPD) 1,087Capital (NYC–HRA) 258 Capital (NYC–NYS) 78 Rental (NYC) 3 Total units 3,615 Operating and service costs Because of variations in housing layout and services offered, different housing configurations have varying levels of operating and service costs. Table B.2 shows the levels of service cost for the different types of housing. Service costs represent funding for social and related serv- ices provided to NY/NY tenants—services that are provided on-site, brokered by case management staff, or arranged with outside providers. Depending on the unit type, federal, state, and city funding can pay for these services. The community residence units have higher average service costs because of the greater intensity of the services provided.33 For the supportive housing subheading, however, while tenants receive less intense services than their counterparts in the community resi- dence housing, the differences in per unit funding reflect different lev- els and sources of money but not necessarily different intensities of service provision. Operating costs, shown in table B.3, reflect costs needed for building upkeep, apartment maintenance, utilities, and so on and are most often provided in the form of rental subsidies to supplement the rent tenants Fannie Mae Foundation 33 Average service costs for community residence units are $19,200 per unit. However, because Medicaid pays for the services in 128 of these units, at an average annual cost of $17,478 per unit, and because those costs are accounted for in the services utilization analysis as client outpatient costs paid by Medicaid, they have been removed here. This reduces the average service cost per community residence unit to $15,865, from non- Medicaid sources. Culhane_app.qxd 6/18/02 11:58 AM Page 150 Public Service Reductions Associated with Supportive Housing 151pay. In federally subsidized units (Section 8), all of which are operatedby the city, tenants pay 30 percent of their income, here assumed to bethe income for a person receiving Supplemental Security Income (SSI)and living alone ($617/month), toward rent. Thirty percent of thisamount ($185) has been deducted from the maximum fair market rentfor an SRO allowed in NYC ($500), and the difference is calculated asthe federal Section 8 contribution ($3,800).Table B.2.Service Costs for NY/NY Housing Type, by Funding SourceState City Per Unit Housing Units Funding* Funding* Subtotal* Community residences Community residences/SROs 713 $10,500 $0 $10,500 Community residences 671 $15,865 $0 $15,865 Supportive housing Capital (state) 285 $8,400 $0 $8,400 Rental (state) 520 $4,800 $0 $4,800 Capital (NYC–HPD) 1,087 $9,400 $900 $10,300 Capital (NYC–HRA) 258 $9,400 $900 $10,300 Capital (NYC–NYS) 78 $9,400 $900 $10,300 Rental (NYC) 3 $9,400 $900 $10,300 *In 1999 dollars. Table B.3.Operating Costs for NY/NY Housing Type, by Funding Source Federal (Section 8) State City Per Unit Housing Units Funding* Funding* Funding* Subtotal* Community residences Community residences/SROs 713 $0 $5,700 $0 $5,700 Community residences 671 $0 $4,200 $0 $4,200 Supportive housing Capital (state) 285 $0 $5,700 $0 $5,700 Rental (state) 520 $0 $5,000 $0 $5,000 Capital (NYC–HPD) 1,087 $3,800 $0 $0 $3,800 Capital (NYC–HRA) 258 $3,800 $0 $0 $3,800 Capital (NYC–NYS) 78 $3,800 $0 $0 $3,800 Rental (NYC) 3 $3,800 $0 $0 $3,800 *In 1999 dollars. Housing Policy Debate Culhane_app.qxd 6/18/02 11:58 AM Page 151 152 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyTable B.4 combines the net unit operating and service costs and, inapplicable categories, deducts tenant contributions from this cost. In the case of many NYS OMH-administered units—all communityresidence/SROs and Capital (state) units—the Section 8 standard fortenant rent contribution (one-third of income) is applied. However, ten-ant rent contributions are deducted from state operating and servicecontracts after providers report tenant rent collections. Thus, a sepa-rate column in table B.4, Less Tenant Contribution, shows the impactof deducting tenant contributions from the operating and service costcontracts of these state-administered units (again, assuming full, year-round occupancy, by persons receiving SSI income for an adult livingalone). The deduction for community residences is higher ($6,000annually) because additional funds are deducted from tenant income to cover the costs of board. The Total Cost column reflects the per unit costs multiplied by the number of units for the particular type of hous- ing, less tenant contributions, where applicable. Capital costs Some 11 percent of the NY/NY units consist of existing, private rental housing, and the rest are located in buildings that have been specifi- cally constructed or rehabilitated under the auspices of the NY/NY program. In the latter case, the city or the state provided the capital. Table B.5 provides the capital costs, both per unit and total, broken down by different subcategories of NY/NY housing. All of the capital costs per unit are budgeted at $70,000, except for HPD-administered units, which are set at $50,000 per unit. HPD unit costs were lower for several reasons: HPD began its property acquisition process earlier than the state did, and real estate values were relatively more depressed then; HPD also acquired some properties at essentially no cost (properties it owned through tax foreclosure—in rem buildings); and HPD’s developments involved much larger buildings than the state’s projects, so development costs per unit were lower (Timothy O’Hanlon, personal communication).34 Fannie Mae Foundation 34 Approximately half (N = 508) of the HPD units received revenue from the sale of federal tax credits. For purposes of calculating the per unit per year cost, the tax credits are assumed to pay out over a 15-year period, amounting to a $3,333 per unit per year cost. Although the cost of the tax credits is not included in the tabulation of debt serv- ice costs here, given that only 20 percent of the tax credit revenue was used for capital support, it is figured into the final costs per unit by source, in table B.7, combined with the debt service amounts. Culhane_app.qxd 6/18/02 11:58 AM Page 152 Public Service Reductions Associated with Supportive Housing 153 Table B.4.Combined Service and Operating Costs, Less Tenant Contributions, for NY/NY Housing Operating Service Lessper Unit per Unit Tenant Per Unit TotalHousing Units Subtotal* Subtotal* Contribution* Cost* Cost*Community residencesCommunity residences /SROs 713 $5,700 $10,500 $2,200 $14,000 $9,982,000Community residences 671 $4,200 $15,865 $6,000 $14,065 $9,437,615Supportive housingCapital (state) 285 $5,700 $8,400 $2,200 $11,900 $3,391,500Rental (state) 520 $5,000 $4,800 $0 $9,800 $5,096,000Capital (NYC–HPD) 1,087 $3,800 $10,300 $0 $14,100 $15,326,700Capital (NYC–HRA) 258 $3,800 $10,300 $0 $14,100 $3,637,800Capital (NYC–NYS) 78 $3,800 $10,300 $0 $14,100 $1,099,800Rental (NYC) 3 $3,800 $10,300 $0 $14,100 $42,300Total 3,615$48,013,715*In 1999 dollars.Housing Policy Debate Culhane_app.qxd 6/18/02 11:58 AM Page 153 154 Dennis P. Culhane, Stephen Metraux, and Trevor Hadley Table B.5.Capital Costs Allocated to Various Types of NY/NY–funded HousingPer Unit Costs* Combined Costs*Housing Units State City Total State City TotalCommunity residenceCommunity residences /SROs 713 $70,000 $0 $70,000 $49,910,000 $0 $49,910,000Community residences 671 $70,000 $0 $70,000 $46,970,000 $0 $46,970,000Supportive housingCapital (state) 285 $70,000 $0 $70,000 $19,950,000 $0 $19,950,000Rental (state) 520 $0 $0 $0 $0 $0 $0Capital (NYC–HPD) 1,087 $0 $50,000 $50,000 $0 $54,350,000 $54,350,000Capital (NYC–HRA) 258 $0 $70,000 $70,000 $0 $18,060,000 $18,060,000Capital (NYC–NYS) 78 $70,000 $0 $70,000 $5,460,000 $0 $5,460,000Rental (NYC) 3 $0 $0 $0 $0 $0 $0Total 3,615 $122,290,000 $72,410,000 $194,700,000*In 1999 dollars.Fannie Mae Foundation Culhane_app.qxd 6/18/02 11:58 AM Page 154 Public Service Reductions Associated with Supportive Housing 155NYC and NYS both financed capital costs through a separate series ofbond issues. On the basis of available information, on $81 million of the$130.5 million in state capital costs, the aggregated interest rate onstate bond issues was 6.339 percent. Similarly, interest rates on bondsissued by the city were between 6 percent and 7 percent; the higherrate was the basis for these calculations.35 A 25-year amortizationschedule and these interest rates were used to estimate annual debtservice costs and allocate them across the different types of housingbased on the number of units and the capital costs (see table B.6). Boththe city and state incur these debt service costs on behalf of the hous-ing provider as part of their assistance to NY/NY. Calculating average total costs per NY/NY housing unit and per housing placement The debt service, social service, and operating costs are combined and averaged across housing types in table B.7 to come up with more com- plete cost estimates. Each of these estimates, which aggregate two or more subtypes of housing, represents a weighted mean as determined by the number of housing units and specific costs associated with each subtype. Federal tax credit costs have been added, along with the debt service costs in this table, although only 20 percent of federal tax credit revenue was applied to capital expenses. The remaining 80 percent was used to fund operating reserves. Tax credit costs are assumed to pay out over the first 15 years of the project but are assumed as a constant annual cost here (see also footnote 34). The total combined cost per unit per year, for all NY/NY units, is esti- mated at $18,190.36 Breaking down this estimate, community residence housing costs more per unit than supportive housing ($19,662 versus $17,277, respectively). Social services and operating costs account for 73 percent of the total estimated per unit cost, and NYS provides, on average, 78 percent of the estimated per unit cost. Housing Policy Debate 35 Information on the state’s NY/NY bond issues was provided through personal com- munication with Michael Newman at the NYS OMH; information on the interest rates for bond issues was provided by the city. 36 This number excludes Medicaid-paid services for the 128 community residence units mentioned earlier. Culhane_app.qxd 6/18/02 11:58 AM Page 155 156 Dennis P. Culhane, Stephen Metraux, and Trevor Hadley Table B.6.Estimated Debt Service on State and City Bond Issues to Fund Capital Costs for NY/NY HousingPer Unit Costs* Total Costs*Housing Units State City Total State City TotalCommunity residencesCommunityresidences/SROs 713 $5,630 $0 $5,630 $4,013,976 $0 $4,013,976Community residences 671 $5,630 $0 $5,630 $3,777,730 $0 $3,777,730Supportive HousingCapital (state) 285 $5,630 $0 $5,630 $1,604,550 $0 $1,604,550Rental (state) 520 $0 $0 $0 $0 $0 $0Capital (NYC–HPD) 1,087 $0 $4,293 $4,293 $0 $4,666,369 $4,666,369Capital (NYC–HRA) 258 $0 $5,997 $5,997 $0 $1,547,171 $1,547,171Capital (NYC–NYS) 78 $5,630 $0 $5,630 $439,117 $0 $439,117Rental (NYC) 3 $0 $0 $0 $0 $0 $0Total 3,615 $9,835,373 $6,213,540 $16,048,913*In 1999 dollars.Fannie Mae Foundation Culhane_app.qxd 6/18/02 11:58 AM Page 156 Public Service Reductions Associated with Supportive Housing 157Table B.7. Estimated Unit Costs to Federal, State and City Sources for NY/NYHousing Averaged over Housing TypesNet Service and Operating CostsaHousing Type Units Federal State City SubtotalCommunity residence 1,384 $0 $14,032 $0 $14,032Supportive housing 2,231 $2,429 $9,813 $575 $12,817Total 3,615 $1,499 $11,428 $355 $13,282Debt Service CostsaUnits Federal State City Subtotal Community residence 1,384 $0 $5,630 $0 $5,630 Supportive housing 2,231 $759 b $916 $2,785 $4,460 Total 3,615 $468 $2,721 $1,719 $4,908 Total Costsa Units Federal State City Subtotal Community residence 1,384 $0 $19,662 $0 $19,662 Supportive housing 2,231 $3,188 $10,729 $3,360 $17,277 Total 3,615 $1,967 $14,149 $2,074 $18,190 aIn 1999 dollars. bThis per unit cost reflects the cost associated with the federal tax credit, paid over the first 15 years of each project. Overall, 20 percent went for debt service, and the remainder went for operating reserves, to be used in the event that Section 8 support stopped on units after the initial five-year commitment. To facilitate comparisons of the housing costs reported here and reduc- tions in the use of collateral services reported in the article, housing costs have been converted into annualized costs per placement in table B.8. The annualized per unit costs (from table B.7) were con- verted into per diem costs by multiplying them by 0.746, the average annualized length of tenure over the first two years following place- ment (Lipton 1996). This yields an annualized per placement cost of $13,570. (This approach is consistent with the one used in the service utilization analysis.) Table B.8.Estimated per Annum, per Diem, and per Placement per Year Costs Annual per Annualized per Housing Units Unit Cost Placement Cost Community residence (mean) 1,384 $19,662 $14,668 Supportive housing (mean) 2,231 $17,277 $12,889 Weighted mean 3,615 $18,190 $13,570 Housing Policy Debate Culhane_app.qxd 6/18/02 11:58 AM Page 157 158 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyAlternatively, one could also compare housing costs and service systemcost reductions by converting the per placement per year service utiliza-tion reductions reported in the article into per housing unit per yearnumbers, by applying the same set of assumptions. As shown intable B.9, the utilization reductions by service type, adjusted for case-control differences, can be expressed in terms of the annualized costreductions per placement (also shown in the last column of table 18).These annualized per placement reductions can be expressed in terms of annualized cost reductions per unit by adjusting for the housingretention rate (multiplying the annualized per placement reductions by1.34—the annualized number of tenants per average housing unit). Thisprocedure yields turnover adjusted and annualized cost reductionsattributable to the full-year housing placement of $16,281 per unit per year. This figure can then be compared with the estimated costs of the housing units presented in this appendix, which have already been cal- culated in terms of the per unit per year costs and (with the exception of table B.8) have also assumed year-round occupancy. Table B.9.Annualized Cost Reductions Adjusted for Housing Turnover: Per Unit and for Total NY/NY Housing Units (N = 3,615), by Type Annualized Annualized Turnover/ Service Provider Cost Reductions* Adjusted Cost Reductions* DHS $2,819 $3,779 OMH $6,162 $8,260 HHC $1,321 $1,771 Medicaid (inpatient) $2,825 $3,787 Medicaid (outpatient visits) –$1,982 –$2,657 VA $444 $595 Department of Corrections (state) $312 $418 Department of Corrections (city) $245 $328 Total $12,146 $16,281 *In 1999 dollars. Conclusion This appendix article has presented estimates of costs borne by NYC, NYS, and the federal government for the construction, operation, and service provision associated with NY/NY housing. City and state pro- gram administrators served as sources for the cost information, and costs are assessed in 1999 dollars. It must also be reiterated that these cost estimates are not comprehensive, but rather reflect the assistance provided by the federal, state, and city governments to nonprofit hous- ing and service providers to administer this housing. Additional funding Fannie Mae Foundation Culhane_app.qxd 6/18/02 11:58 AM Page 158 Public Service Reductions Associated with Supportive Housing 159may come from the nonprofit agencies themselves and from tenant rentcontributions that are otherwise not deducted here.37NY/NY housing reflects diverse housing and service configurations,which correspond to a wide range of expenses. In the process of com-bining the different types of housing, average costs also become lessrepresentative of individual types of housing. While the average figuresare useful for comparing NY/NY housing costs with potentially offset-ting cost reductions brought on by reductions in the use of collateralservices such as psychiatric hospitals, other public and private hospi-tals, homeless shelters, and corrections programs, future researchshould be refined by breaking down cost calculations by specific pro-grams and by specific tenancy histories. Authors Dennis P. Culhane is Associate Professor of Social Welfare Policy at the Center for Men- tal Health Policy and Services Research at the University of Pennsylvania. Stephen Metraux is Senior Research Coordinator at the Center for Mental Health Policy and Services Research at the University of Pennsylvania. Trevor Hadley is Professor of Psy- chology in Psychiatry at the University of Pennsylvania. The authors gratefully acknowledge the generous assistance of the following individu- als and their agencies or organizations (or former agencies/organizations) in obtaining data access, providing data, reviewing preliminary results, and/or securing funding for this effort: Gail Clott, Jill Berry, and Susan Wiviott (NYC DHS); Leon Cosler, Peter Gallagher, Dellie Glaser, and Thomas Fanning (NYS Department of Health); Bruce Fredrick, Susan Jacobsen, and Steven Greenstein (NYS Department of Correctional Services); Eric Sorenson (NYC Department of Corrections); Sharon Salit, Laray Brown, and Ava Quint (NYC HHC); Robert Rosenheck (U.S. Department of Veterans Affairs); Barry Brauth, Michael Newman, Christopher Roblin (NYS OMH); Timothy O’Hanlon (NYC Department of Housing Preservation and Development); Peter Bittle (NYC Department of Mental Health); Frank Lipton (NYC HRA); Brad Race and Robert King (NYS Executive Offices); Sharon Salit, David Gould, and Lenore Glickhouse (United Hospital Fund of New York); Tracey Rutnik, Stephanie Jennings, James Carr, Steven Hornburg, and Lawrence Small (the Fannie Mae Foundation); Sandra Newman (Johns Hopkins University); and Julie Sandorf, Constance Tempel, Ted Weerts, Roger Clay, Cynthia Stuart, Heidie Joo, James Krauskopf, Jonathan Harwitz, Ted Houghton, and Richard Ravitch (the Corporation for Supportive Housing). This research was sponsored by the Fannie Mae Foundation, United Hospital Fund of New York, Conrad Hilton Foundation, Rhodebeck Charitable Trust, and the Corpora- tion for Supportive Housing. Housing Policy Debate 37 Medicaid-paid services to tenants, delivered in 128 of the community residence units, are excluded. Culhane_app.qxd 6/18/02 11:58 AM Page 159 160 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyReferencesAllison, Paul D. 1999. Logistic Regression Using the SAS System: Theory and Applica-tion.Cary, NC: SAS Institute.Arthur Andersen LLP, University of Pennsylvania Center for Mental Health Policy andServices Research, Kay E. Sherwood, and TWR Consulting. 2000. The Connecticut Sup-portive Housing Demonstration Program Evaluation Report. New Haven, CT: Corpora-tion for Supportive Housing. Averyt, June M., and Edna Kamis-Gould. 2000. The Effect of Supportive Housing onTenants, Including Utilization and Costs of Services. Unpublished paper. University ofPennsylvania, Center for Mental Health Policy and Services Research.Bebout, Richard R., and Maxine Harris. 1992. In Search of Pumpkin Shells: Residential Programming for the Homeless Mentally Ill. In Treating the Homeless Mentally Ill: A Task Force Report of the American Psychiatric Association,ed. H. Richard Lamb, Leona L. Bachrach, and Frederick I. Kass, 159–82. Washington, DC: American Psychiatric Association. Burt, Martha R., Laudan Y. Aron, and Edgar Lee. 2001. Helping America’s Homeless: Emergency Shelter or Affordable Housing?Washington, DC: Urban Institute Press. Carling, Paul J. 1993. Housing and Supports for Persons with Mental Illness: Emerging Approaches to Research and Practice. Hospital and Community Psychiatry44(5): 439–49. Caton, Carol L., Richard J. Wyatt, Alan Felix, Jeffrey Grunberg, and Boanerges Dominguez. 1993. Follow-up of Chronically Homeless Mentally Ill Men. American Jour- nal of Psychiatry150:1639–42. Center for Mental Health Services. 1994. Making a Difference: Interim Status Report of the McKinney Demonstration Program for Homeless Adults with Serious Mental Illness. Washington, DC. Center for Urban Community Services. 1995. A Guide to Supportive Housing Models for People with Mental Illness and Other Special Needs. New York. Culhane, Dennis P., and Stephen Metraux. 1997. Where to from Here: A Policy Research Agenda Based on the Analysis of Administrative Data. In Understanding Homelessness: New Policy and Research Perspectives,ed. Dennis P. Culhane and Steven P. Hornburg, 341–60. Washington, DC: Fannie Mae Foundation. DeParle, Jason. 1996. The Year That Housing Died. New York Times Magazine, Octo- ber 20, p. 52. Dickey, Barbara, Olinda Gonzalez, Eric Latimer, Karen Powers, Russell K. Schutt, and Stephen M. Goldfinger. 1996. Use of Mental Health Services by Formerly Homeless Adults Residing in Group and Independent Housing. Psychiatric Services 47(2):152–58. Drake, Robert E., Nancy A. Yovetich, Richard R. Bebout, Maxine Harris, and Gregory J. McHugo. 1997. Integrated Treatment for Dually Diagnosed Homeless Adults. Journal of Nervous and Mental Disease 185(5):298–305. Fannie Mae Foundation Culhane_app.qxd 6/18/02 11:58 AM Page 160 Public Service Reductions Associated with Supportive Housing 161Goldfinger, Stephen M., Russell K. Schutt, George S. Tolomiczenko, Larry J. Seidman,Walter E. Penk, Winston Turner, and Brina Caplan. 1999. Housing Placement and Sub-sequent Days Homeless among Formerly Homeless Adults with Mental Illness. Psychi-atric Services 50:674–79.Hevesi, Dennis. 1999. Building Homes for the Single Homeless. New York Times, April28, Sec. 11, p. 1.Hurlburt, Michael S., Patricia A. Wood, and Richard L. Hough. 1996. Providing Inde-pendent Housing for the Mentally Ill Homeless: A Novel Approach to Evaluating Long-Term Housing Patterns. Journal of Community Psychology 24:291–310.Judge Orders City to Shelter Two Suffering Homeless Families. 2001. New York Times,April 12, p. B2. Kennedy, Randall. 1997. Doors That Offer Hope May Shut: A Program That Has Housed the Mentally Ill Lapses. New York Times, October 4, p. B1. Kennedy, Shawn G. 1995. New Look for S.R.O.s: Decent Housing. New York Times, March 28, p. B1. Kuhn, Randall S., and Dennis P. Culhane. 1998. Applying Cluster Analysis to Test a Typology of Homelessness by Pattern of Shelter Utilization: Results from the Analysis of Administrative Data. American Journal of Community Psychology 26(2):207–32. Lehman, Anthony F., and David S. Cordray. 1993. Prevalence of Alcohol, Drug and Mental Disorders among the Homeless: One More Time. Contemporary Drug Problems 20:355–83. Lewis, Dan A., and Arthur J. Lurigio. 1994. The State Mental Patient and Urban Life. Springfield, IL: Charles C. Thomas. Lipton, Frank R. 1996. Summary Client Placement Report of the New York–New York Agreement to House Homeless Mentally Ill Individuals.New York: New York City Human Resources Administration, Office of Health and Mental Health Services. Lipton, Frank R., Suzanne Nutt, and Albert Sabatini. 1988. Housing the Homeless Mentally Ill: A Longitudinal Study of a Treatment Approach. Hospital and Community Psychiatry39(1):40–45. Lipton, Frank R., Carol Siegel, Anthony Hannigan, Judith Samuels, and Sherryl Baker. 2000. Tenure in Supportive Housing for Homeless Persons with Severe Mental Illness. Psychiatric Services 51(4):479–86. Metraux, Stephen, Dennis P. Culhane, and Trevor R. Hadley. 2000. Supportive Housing and Inpatient Hospital Use in the New York City Health and Hospitals Corporation among Homeless Persons with Severe Mental Illness: Assessing the Impact of “New York/New York” on the Use of Public Hospitals. Unpublished paper. University of Penn- sylvania, Center for Mental Health Policy and Services Research. Metraux, Stephen, Dennis P. Culhane, and Trevor R. Hadley. 2001a. The Impact of NY/NY Supportive Housing Placements on Medicaid-Funded Health Care. Unpublished paper. University of Pennsylvania, Center for Mental Health Policy and Services Research. Housing Policy Debate Culhane_app.qxd 6/18/02 11:58 AM Page 161 162 Dennis P. Culhane, Stephen Metraux, and Trevor HadleyMetraux, Stephen, Dennis P. Culhane, and Trevor R. Hadley. 2001b. NY/NY Housingand the Use of New York City Jail Services: An Analysis Merging Administrative Data.Unpublished paper. University of Pennsylvania, Center for Mental Health Policy andServices Research.Metraux, Stephen, Robert Rosenheck, Dennis P. Culhane, and Trevor R. Hadley (2000).The Impact of NY/NY Supportive Housing Placements on Inpatient Use of VeteransAdministration Hospitals. Unpublished paper. University of Pennsylvania, Center forMental Health Policy and Services Research.Murray, Ruth, Marjorie Baier, Carol North, Mary Lato, and Cary Eskew. 1997. One-YearStatus of Homeless Mentally Ill Clients Who Completed a Transitional Residential Pro-gram. Community Mental Health Journal 33(1):43–50.Pollio, David E., Edward L. Spitznagel, Carol S. North, Sanna Thompson, and Douglas A. Foster. 2000. Service Use over Time and Achievement of Stable Housing in a Men- tally Ill Homeless Population. Psychiatric Services 51(12):1536–43. Proscio, Tony. 2000. Supportive Housing and Its Impact on the Public Health Crisis of Homelessness. New York: Corporation for Supportive Housing. Ridgway, Priscilla, and Charles A. Rapp. 1998. The Active Ingredients of Effective Sup- ported Housing: A Research Synthesis. Lawrence, KS: University of Kansas, School of Social Welfare. Rosenheck, Robert. 2000. Cost-Effectiveness of Services for Mentally Ill Homeless Peo- ple: The Application of Research to Policy and Practice. American Journal of Psychiatry 157(10):1563–70. Salit, Sharon A., Evelyn M. Kuhn, Arthur J. Hartz, Jade M. Vu, and Andrew L. Mosso. 1998. Hospitalization Costs Associated with Homelessness in New York City. New Eng- land Journal of Medicine 338(24):1734–40. Sard, Barbara. 2001. Revision of the Project-Based Voucher Statute.Washington, DC: Center on Budget and Policy Priorities. Shern, David L., Chip J. Felton, Richard L. Hough, Anthony F. Lehman, Stephen M. Goldfinger, Elie Valencia, Deborah Dennis, Roger Straw, and Patricia A. Wood. 1997. Housing Outcomes for Homeless Adults with Mental Illness: Results from the Second- Round McKinney Program. Psychiatric Services 48(2):239–41. Tsemberis, Sam. 1999. From Streets to Homes: An Innovative Approach to Supported Housing for Homeless Adults with Psychiatric Disabilities. Journal of Community Psy- chology 27(2):225–41. Tsemberis, Sam, and Rhonda F. Eichenberg. 2000. Pathways to Housing: Supported Housing for Street-Dwelling Homeless Individuals with Psychiatric Disabilities. Psychi- atric Services 51(4):487–93. Twombley, Jennifer G., Harry Lawson, Carla H. Smith, and Sheila Crowley. 2001. A Report on State-Funded Rental Assistance Programs: A Patchwork of Small Measures. Washington, DC: National Low-Income Housing Coalition. Fannie Mae Foundation Culhane_app.qxd 6/18/02 11:58 AM Page 162 Public Service Reductions Associated with Supportive Housing 163U.S. Department of Housing and Urban Development, Office of Policy Developmentand Research. 1994. National Evaluation of the Supportive Housing DemonstrationProgram.Washington, DC.U.S. Senate. 2000. Departments of Veterans Affairs and Housing and Urban Develop-ment, and Independent Agencies Appropriations Bill, 2001. Senate Report 106 410.World Wide Web page <http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=106_cong_reports&dcid=fr:sr410.106pdf> (accessed February 23, 2001). Housing Policy Debate Culhane_app.qxd 6/18/02 11:58 AM Page 163 TheRikersIslandHotSpotters:DeﬁningtheNeeds oftheMostFrequentlyIncarcerated Ross MacDonald, MD, Fatos Kaba, MA, Zachary Rosner, MD, Allison Vise, BA, David Weiss, MD, Mindy Brittner, MD, Molly Skerker, BA, Nathaniel Dickey, MPH, MA, and Homer Venters, MD, MA The United States has the highest rate of incarceration in the world, with a nearly 5-fold increase in the prison population since 1978 1 and approximately 6.9 million people under the supervision of adult correctional systems at the end of 2013. 2 Though the causes of this growth are complex, the “war on drugs”and “deinstitutionalization”of inpatient psychiatric hospitals have been proposed as key drivers of growth in the incarcerated population over this time. The war on drugs refers to a law enforcement approach to the problem of sub- stance abuse that historically focused on harsh penalties for drug possession and all aspects of distribution and sale. 3 Deinstitutionalization refers to the process by which long-term psychiatric hospitals were closed with the in- tention of diverting mental health resources to community-based care, though the corollary expansion of these services did not occur. 4 Both the war on drugs and deinstitutionaliza- tion contribute to the de facto criminalization of 2 societal problems that also lie squarely within the purview of public health: substance use disorders and mental illness. 5---7 Hot spotting, a practice in part inspired by targeted policing, refers to identifying and focusing on the highest users of health care services in a population and offering tailored, intensive case management in an effort to reduce costs and improve care. 8 Interventions tailored to this group have, on balance, dem- onstrated improved cost and health out- comes.9---12 The high-user population identified in this manner has been shown to have high rates of mental illness, substance use, and homelessness.13---15 The Bureau of Correctional Health Services of the New York City Department of Health and Mental Hygiene is responsible for the provision of medical, mental health, and dis- charge planning services in the New York City jail system. With an average daily population of approximately 11000 persons and more than 60000 admissions per year, theNew York City jail system is among the largest in the world. Most admissions are relatively short, with a median length of stay of approximately 13 days, though a subset of patients stays for longer periods as they go through trial. Those sentenced to less than a year serve their time in 1 “sentenced”facility on Rikers Island and those with longer sentences are transferred to New York State prisons. All patients admitted to the New York City jail system undergo a full history and physical examination before being housed. The intake examination screens all patients for chronic medical and mental illness including substance abuse and sets the trajectory for follow-up care while incarcerated. We sought to characterize the medical and mental health conditions of the most frequently returning persons to New York City jails. The primary goal of this investigation was to inform the care we provide for this population, but the analysis also raises questions that cut across disciplines, including the nature of our patients’ relationship with the criminal justice system. METHODS We designed this study to better understand the patterns of incarceration and profile of the most frequently jailed persons in the New York City jail system. We were specifically interested in characterizing the demographics, medical and mental health status, and criminal justice involvement of this group as they compared with a randomly selected cross- section of the New York City jail population. Starting with all patients admitted to the jail system in 2013, we defined the frequently incarcerated group as a rank-ordered sample of the 800 most-frequently returning patients since November 2008, when the electronic health record (EHR) was implemented. This group will hereafter be referred to as the frequently incarcerated group or the hot spot- ters. We randomly selected a control group of Objectives.We used “hot spotting” to characterize the persons most fre- quently admitted to the New York City jail system in 2013. Methods.We used our Correctional Health Services electronic health record to identify 800 patients admitted in 2013 who returned most since November 2008. We compared them to a randomly selected control group of 800 others admitted in 2013, by using descriptive statistics and cross-tabulations, including data through December 2014. Results.The frequently incarcerated individuals had a median of 21 incarcer- ations (median duration 11 days), representing 18713 admissions and $129 million in custody and health costs versus $38 million for the controls. The frequently incarcerated were signiﬁcantly older (42 vs 35 years), and more likely to have serious mental illness (19% vs 8.5%) and homelessness (51.5% vs 14.7%) in their record. Signiﬁcant substance use was highly prevalent (96.9% vs 55.6%). Most top criminal charges (88.7%) for the frequently incarcerated were mis- demeanors; assault charges were less common (2.8% vs 10.4%). Conclusions.Frequently incarcerated persons have chronic mental health and substance use problems, their charges are generally minor, and incarceration is costly. Tailored supportive housing is likely to be less costly and improve outcomes. (Am J Public Health.2015;105:2262–2268. doi:10.2105/AJPH.2015. 302785) RESEARCH AND PRACTICE 2262 |Incarceration |Peer Reviewed |MacDonald et al.American Journal of Public Health |November 2015, Vol 105, No. 11 800 among the remainder of the same pool of individuals who had a jail admission in 2013, after we excluded the 800 most-frequently incarcerated. Once these 2 groups had been defined, analysis for each included data through 2014 for a total review period of approximately 6 years (74 months). We extracted data on demographics as well as jail admission and discharge dates from the EHR. Though not readily available to end- users, charges are available alongside demo- graphic data in the EHR and we extracted top charges for each individual. We categorized charges comprising less than 1% of the total into groups of related charges. We defined serious mental illness according to criteria established by the New York State Office of Mental Health. 16 We also obtained data for medical and mental health status and resource use from the EHR during the study period. We defined use as the total number of clinical staff encounters for each group divided by the total number of person-months incarcerated. We defined clinical encounters as face-to-face encounters with physicians, physician assis- tants, nurse practitioners, licensed clinical so- cial workers, or nurses, excluding recurring tasks (e.g., wound care, medicine administration, blood glucose monitoring, cell-side checks). Because of changes in EHR workflow, we calculated use only for the period from April 2011 through the end of 2014. We defined Medicaid status as the most recent result of a Medicaid status query done at admission, a process implemented in August 2013. We excluded persons in the 2 groups who did not have any admissions after this date from analysis of the Medicaid status variable. We derived categorical variables in this analysis from the EHR and these included mental health diagnosis, serious mental illness designation, gender, race/ethnicity, homeless- ness, drug and alcohol use, HIV status, anti- psychotic medication prescription, alcohol withdrawal treatment, Medicaid status, and top charges. We defined significant drug or alcohol use as evidence of drug use excluding users of only marijuana or only alcohol (unless requiring treatment of withdrawal), whereas any drug or alcohol use includes those patients who reported use of marijuana alone or alcohol alone (regardless of need for with- drawal treatment). We derived continuous variables from the EHR and these included length of stay, number of days between incarcerations, and number of clinical staff encounters. We calculated length of stay from jail admission and discharge dates, creating an artificial discharge date (December 31, 2014) for those persons who were still in jail at that time. In a similar way, we created time between incarcerations by calcu- lating the days between previous discharge and subsequent admission dates. We calculated cost estimates (reported in 2011 dollars) based on the methodology employed by the New York/New York III housing evaluation, which derived daily custody and health care costs for the New York City jail system as reported in the annual New York City Mayor’s Manage- ment Reports from 2005 to 2010. 17 We used descriptive statistics, the indepen- dent t test, and cross-tabulation to explore differences between the frequently incarcer- ated group and the control group. We de- termined statistical significance of differences by using the v2 test with a threshold of signif- icance defined as a 2-sided P value of less than .05. We used SPSS version 19 (IBM, Somers, NY) for statistical analysis. RESULTS In 2013, there were 79618 incarcerations among 57194 individuals in New York City jails. Among these individuals, slightly more than one third (37.3%) had only 1 incarcera- tion since November 15, 2008, whereas 53.5% had between 2 and 7 incarcerations. Over the 74 months of the study period, the 800 patients comprising the frequently incar- cerated group experienced 18713 incarcera- tions, whereas the control group had 3108, corresponding to a median of 21 incarcerations in the frequently incarcerated group and 3 in the control group (Table 1). Though the me- dian jail stays were similar in the 2 groups (11 vs 13 days, not tested for significance), the frequently incarcerated group had shorter stays in jail with a mean of 28 days versus 49 for the control group (P<.001) and shorter mean intervals between stays (60.9 days vs 246.2 days;P<. 001; Table 1). The median time between all incarcerations in the fre- quently incarcerated group was only 32 days. In addition, the frequently incarcerated group represented only 0.3% of all persons incarcerated during the study period, but accounted for 3.5% of all incarcerations during TABLE 1—Patterns of Incarceration: New York City Correctional Health Services Electronic Health Records, 2008–2014 Variable Frequently Jailed Group (n=800) Control Group (n=800) No. of incarcerations 18713 3108 Mean 23.4* 3.9 Median 21 3 Range 16–66 1–17 Length of stay, days Mean 28* 49 Median 11 13 Range 0–656 0–962 Sum of years incarcerated 1422.5 415.2 Cost of incarceration, a $ 129105794 37679178 No. of days between incarcerations Mean 60.9* 246.2 Median 32 131 Range 0–996 0–1851 aCalculation based on per day jail cost in 2011 dollars of $248.65 as used in the New York/New York III housing evaluation (John Volpe, New York City Department of Health and Mental Hygiene Bureau of Mental Health, e-mail communication, January 21, 2015). *P<.001. RESEARCH AND PRACTICE November 2015, Vol 105, No. 11 |American Journal of Public Health MacDonald et al.|Peer Reviewed |Incarceration |2263 this time. Over the study period, the 800 frequently incarcerated persons spent 1423 person-years incarcerated at an estimated cost of $129 million. Frequently incarcerated individuals were significantly older (42 vs 35 years), and more likely to be non-Hispanic Black (72.6% vs 57.9%), to be diagnosed as seriously mentally ill (19% vs 8.5%), to receive antipsychotic prescriptions in jail (37.0% vs 15.6%), and to have mention of homelessness in their charts (51.5% vs 14.7%;P <.001 for all; Table 2). In addition, the vast majority (96.6%) of the frequently incarcerated group had evidence of significant drug or alcohol use compared with 55.6% of the control group (P<.001). Report of any drug or alcohol use was also higher in the frequently incarcerated group (99.4% vs 78.8%) as well as crack or cocaine use (83.5% vs 30.4%), heroin or opiate use (36.6% vs 22.3%), and alcohol use requiring alcohol with- drawal treatment (22.1% vs 4.4%;P<.001 for all; Table 2). A higher proportion of the frequently incarcerated persons reported HIV/AIDS (10.9% vs 4.3%) and overall service use in jail was higher for medical (mean5.6vs4.0visitspermonth)and mental health (mean 2.0 vs 1.8 visits per month;P<.001 for both; Table 2). A high percentage of the frequently incarcerated persons had some relationship with Medicaid in the past (95.9%), and they were more likely to have such a relationship than the control group (78%;P<.001). The majority of persons in the frequently incarcerated group had a favorable Medicaid status (i.e., active, suspended, or applied; 68.7%). When we compared criminal charges, the top (most serious) charges faced by frequently incarcerated persons were qualitatively differ- ent than those of the control group. Two charges, petit larceny (29.9%) and criminal possession of controlled substances in the seventh degree (residue or small quantity of drug; 23.8%), constituted more than half of the top charges among the frequently incarcerated group, whereas top charges in the control group were more varied. The remainder of the top charges that constituted 1% or more of all charges for the frequently incarcerated group were as follows: criminal trespass in the second degree (5.7%), theft of services (e.g., public transportation fare evasion; 5.5.%), assault in the third degree (2.1%), criminal sale of a con- trolled substance in the third degree (1.9%), criminal possession of stolen property in the fifth degree (1.7%), criminal trespass in the third degree (1.7%), criminal possession of marijuana in the fifth degree (1.3%), criminal mischief in the fourth degree (1.1%), and resisting arrest (1.0%; Table 3). A total of 3.1% of charges were missing and the remaining 20% of charges were made up of 143 varied charges that each accounted for less than 1% of all charges. These varied charges were categorized as follows: theft or robbery (4.7%), administrative (3.6%), fraud (3.1%), disorder (2.7%), drugs (2.6%), prostitution (2.3%), vio- lent (1.2%), weapons (0.6%) and vehicular (0.3%; Table 4). Any assault charge constituted 10.4% of the control group’s charges com- pared with only 2.8% of the charges among the frequently incarcerated (Figure 1). A TABLE 2—Demographics, Clinical Characteristics and Health Care Use: New York City Correctional Health Services Electronic Health Records, 2008–2014 Frequently Jailed Group Control Group Variable No. % (95% CI) No. % (95% CI) Gender Male 709 88.6 (86.40, 90.80) 721 90.1 (88.03, 92.17) Female 91 9.9 (7.83, 11.97) 79 11.4 (9.20, 13.60) Race/ethnicity Hispanic 152 19.0 (16.28, 21.72) 261 32.6 (29.35, 35.85) Non-Hispanic Black 581 72.6* (69.51, 75.69) 463 57.9 (54.48, 61.32) Non-Hispanic White 48 6.0 (4.35, 7.65) 59 7.4 (5.59, 9.21) Other or unknown 17 2.4 (1.34, 3.46) 19 2.1 (1.11, 3.09) Mental illness Serious mental illness 152 19.0* (16.28, 21.72) 68 8.5 (6.57, 10.43) Antipsychotic prescriptions 296 37.0* (33.65, 40.35) 125 15.6 (13.09, 18.11) Homeless (missing) 409 51.5* (48.03, 54.97) 111 14.7 (12.19, 17.21) Medicaid status a 724 477 Not queried or not found 30 4.1* (2.66, 5.54) 105 22.0 (18.28, 25.72) Ever a relationship 694 95.9* (94.46, 97.34) 372 78 (74.28, 81.72) Active, suspended, or applied 477 68.7 (65.25, 72.15) 230 61.8 (56.80, 66.74) Closed 198 28.5 (25.14, 31.86) 123 33.1 (28.32, 37.88) Rejected or other insurance 19 2.7 (1.49, 3.91) 19 5.1 (2.86, 7.34) Alcohol or drug use Any drug or alcohol use 795 99.4* (98.86, 99.94) 630 78.8 (75.97, 81.63) Signiﬁcant drug or alcohol use 775 96.9* (95.70, 98.10) 445 55.6 (52.16, 59.04) Cocaine use 668 83.5* (80.93, 86.07) 243 30.4 (27.21, 33.59) Heroin or opiate use 293 36.6* (33.26, 39.94) 178 22.3 (19.42, 25.18) Alcohol withdrawal in jail 177 22.1* (19.22, 24.98) 35 4.4 (2.98, 5.82) Ever in methadone maintenance 146 18.3 (15.62, 20.98) 132 16.5 (13.93, 19.07) Medical conditions HIV+ 85 10.9* (8.74, 13.06) 34 4.3 (2.89, 5.71) Hepatitis C 146 18.3* (15.62, 20.98) 59 7.4 (5.59, 9.21) Diabetes 71 8.9* (6.93, 10.87) 33 4.1 (2.73, 5.47) Epilepsy 70 8.8* (6.84, 10.76) 43 5.4 (3.83, 6.97) Note.CI|=|conﬁdence interval. Mean age=42 years* in frequently jailed group; 35 years in control group. No. of visits for medical services per 30 person-days=5.6* in frequently jailed group; 4.0 in control group. No. of mental health visits per 30 person-days=2.0* in frequently jailed group; 1.8 in control group.aP<.001 only for those patients (n=1201) who were admitted after August 2013. *P<.001. RESEARCH AND PRACTICE 2264 |Incarceration |Peer Reviewed |MacDonald et al.American Journal of Public Health |November 2015, Vol 105, No. 11 preponderance of top charges (88.7%) in the frequently incarcerated group were misde- meanors compared with only slightly more than half (54.9%) in the control group (P<.001; Figure 2). DISCUSSION We described the basic characteristics of a population of individuals caught in the re- volving door of frequent incarceration in the New York City jail system. Consistent with the literature on high users of health care services, this group had a higher prevalence of mental illness (e.g., meeting criteria for serious mental illness, receiving antipsychotic medication) compared with the overall jail population. However, less than 40% of the frequently incarcerated group fit this description. More strikingly, substance use was almost universally prevalent in this group and by many measures was more severe than in the control group. We also found evidence of homelessness in more than half the charts of these patients despite not formally screening for housing status, which makes this likely to be a signifi- cant underestimate. Finally, we noted their charges to be suggestive of persons who pose little public safety threat, with low-level theft, possession of small quantities of drugs, trespassing, and fare evasion accounting for approximately two thirds of the top charges against them. We also found a preponderance of misdemeanors and fewer assault-related charges than the control group. Together these data present a picture of a population whose significant substance use, in conjunction with homelessness and often mental illness, promotes frequent incarceration for relatively minor transgressions. It is impor- tant that our study demonstrated a higher proportion of non-Hispanic Blacks in the fre- quently incarcerated group, given that mass incarceration has disproportionately affected communities of color and has been postulated to exacerbate health disparities. 18 With the persistent lack of housing and prevalence of concomitant health issues of this population, it is unlikely that their repeated incarceration is an effective strategy from a criminal justice or public health perspective. The 4 basic principles generally used to justify incarceration are retribution, rehabilitation, deterrence, and incapacitation (separation from the public). 19 Repeated incarceration has failed to modify the behavior that is leading to recurrent arrest, suggesting that this is not an effective strategy for rehabilitation. Our clinical experience with this group leads us to believe that they have acclimated to jailing over the years such that jailing no longer serves as retribution (as it is not perceived as punishment) and does not have a significant deterrent effect on their future behavior. Their minor charges suggest that separation from the public is not necessary for this group, nor is it achieved as they spend most of their time in the community. Incarceration does not address broader social issues inthe community, such as poverty, homelessness, and lack of effective access to medical and mental health care and thus may serve to propagate rather than interrupt a cycle of maladaptive behavior. There is little public health value to repeated incarceration of this group, as jail carries significant risks to health and has not been shown to improve behavioral health outcomes. Detoxification alone does not represent ade- quate treatment of substance use disorders, TABLE 3—Hot Spotters’ Individual Charges Constituting ‡1% of All Top Charges: New York City Correctional Health Services Electronic Health Records, 2008–2014 Top Charge Percentage of Total Top Charges Petit larceny 29.9 Criminal possession of a controlled substance in the seventh degree a 23.8 Criminal trespass in the second degree 5.7 Theft of services b 5.5 Assault in the third degree 2.1 Criminal sale of controlled substance in the third degree 1.9 Criminal possession of stolen property in the ﬁfth degree 1.7 Criminal trespass in the third degree 1.7 Criminal possession of marijuana in the ﬁfth degree 1.3 Criminal mischief in the fourth degree 1.1 Resisting arrest 1.0 Other or missing 24.3 Total 100 aSmall quantity of drug or drug residue. bPublic transportation fare evasion. TABLE 4—Hot Spotters’ Top Charges Constituting <1%, by Category: New York City Correctional Health Services Electronic Health Records, 2008– 2014 Top Charge Percentage of Total Top Charges Theft or robbery a 4.7 Administrativeb 3.6 Fraudc 3.1 Uncategorized or missing 3.1 Disorderd 2.7 Drugse 2.6 Prostitutionf 2.3 Violentg 1.2 Weaponsh 0.6 Vehiculari 0.3 Total 24.2 aGrand larceny, robbery, burglary, etc. bAdministrativecode,criminalcontempt,bailjumping, etc. cCriminal possession of a forged instrument, fraudu- lent accosting, tampering with physical evidence, etc. dDisorderly conduct, menacing, obstructing govern- mental administration, etc. eCriminal sale of marijuana, criminal possession of a controlled substance, criminal sale of a controlled substance, etc. fProstitution, loitering for the purpose of engaging in a prostitution offense, patronizing a prostitute, etc.gMurder, rape, assault, etc.hCriminal possession of a weapon, various degrees.iUnauthorized use of a vehicle, operating a motor vehicle while intoxicated, illegal possession of a vehi- cle identiﬁcation number, etc. RESEARCH AND PRACTICE November 2015, Vol 105, No. 11 |American Journal of Public Health MacDonald et al.|Peer Reviewed |Incarceration |2265 and the risk of death in the immediate post- release period is known to be increased, driven largely by overdose risk. 20,21 Periods of absti- nence following incarceration have been shown to be shorter than those following treatment.22 The stressful jail environment can lead to mental health decompensation, and suicide is a leading cause of death in jails and prisons.23---25 Violence is prevalent in jail, in- cluding traumatic brain injury, which may potentiate behavioral problems and substance use.26,27 Moreover, self-harm is common in jail and is promoted by features of the jail envi- ronment such as solitary confinement.28 Jails have been shown to drive the community-level epidemiology of some communicable diseases, with the highest incidence among the fre- quently incarcerated. 29---32 Still, patients in the New York City jail have free and robust access to medical and mental health care. Whether the intervention as a whole is health-promoting depends on whether the value of the access to care outweighs the health risks of jail. This remains an active area of inquiry. 33,34 Re- gardless, the health-promoting aspects of the jail intervention could be replicated more efficiently in settings with fewer attendant health risks. The huge costs associated with the security apparatus ensure that jail represents the most expensive setting to provide these interventions, demonstrated by the estimated $129-million cost of incarceration for this group over the study period, which equates to more than $161000 per person over 6 years. Supportive housing interventions tailored to serve similar populations have been shown to reduce incarceration, reduce homeless shelter 29.9 23.8 5.7 5.5 2.80 5 10 15 20 25 30 35 40 45 50 Petit larceny Criminal possession of substance to the seventh degree Criminal trespass in the second degree Theft of services Assault of any typePercentage Charges Hot spotters Control FIGURE 1—Most common top charges and pooled assault charges: New York City Correctional Health Services electronic health records, 2008–2014. 88.7 54.9 0 10 20 30 40 50 60 70 80 90 100 Misdemeanors Felonies PercentageGroup Hot spotters Control FIGURE 2—Top charges, misdemeanors versus felonies: New York City Correctional Health Services electronic health records, 2008–2014. RESEARCH AND PRACTICE 2266 |Incarceration |Peer Reviewed |MacDonald et al.American Journal of Public Health |November 2015, Vol 105, No. 11 use, improve substance use indicators, reduce medical and psychiatric hospitalization, and cost less than usual care. 17,35 We have dem- onstrated that frequent incarceration is associ- ated with homelessness, minor charges, and the key comorbidities that define eligibility criteria typically used in supportive housing interventions, namely substance use and men- tal illness. The frequently incarcerated group defined here shared clinical characteristics with populations targeted for supportive housing interventions, but this group had many more incarcerations than those in the supportive housing study populations to date, which sug- gests that even more pronounced cost savings may be possible. 17,35 Thus, we would argue that the most fre- quently returning jail cohort should be specif- ically targeted for supportive housing and that the criminal justice system should have tools to divert this group to housing rather than send them to jail for minor charges. This would represent a novel approach to support- ive housing, which generally relies on broad categories of eligibility rather than targeting a cohort of previously identified individuals deemed to be at highest need. A targeted approach would ensure that barriers to entry for this group (disabling substance use, mental illness, frequent incarceration itself) would not preclude them from access in favor of other, better compensated individuals who also meet eligibility criteria, but may be better equipped to successfully apply for supportive housing. Though they pose little public safety risk, they likely have other barriers to retention in hous- ing that will require interventions with inten- sive services. This group may be entrenched in the so-called “institutional circuit”with custo- dial institutions purportedly meant to address their underlying problems (jails, hospitals, shelters, drug treatment centers, etc.) instead promoting continued homelessness by provid- ing a rotating host of temporary housing solutions that functionally become perma- nent.36 Interventions targeting this group should have a specific goal of interrupting this cycle by promoting permanent housing and minimizing interventions (especially jailing) that may reinitiate the cycle. Programs and policymakers must take into account that these patients will continue to struggle with their substance use, 22 such that zero-tolerance policies will not be successful in achieving cost savings or health benefits for this population. A shift in expectations will also be required of the criminal justice system, which has traditionally employed urine drug screening as part of probation or parole strat- egies. Strategies that focus on functional status will be more effective at measuring progress forthis group than those that focus on complete abstinence. Next steps in the investigation of potential interventions for the frequently incarcerated include assessing the level of contact this group has had with supportive housing to date as well as conducting qualitative interviews to explore their life circumstances and trajectory. We also plan to explore patterns of incarcera- tion among this group to potentially identify patients earlier on the trajectory toward fre- quent incarceration and perform a more so- phisticated analysis of their charges. At the same time, the Bureau of Correctional Health Services will participate in efforts for diversion of these patients from jail to treatment by leveraging several new initiatives under way in New York City. 37 j About the Authors All of the authors are with the New York City Department of Health and Mental Hygiene, Bureau of Correctional Health Services, Queens, NY. Correspondence should be sent to Ross MacDonald, 42-09 28th St, Office 10-79, Queens, NY 11101-4132 (e-mail: rmacdonald@nychhc.org). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints”link. This article was accepted May 16, 2015. Contributors R. MacDonald developed the concept and led article writing and revision. F. Kaba managed the data analysis with the assistance of Z. Rosner, A. Vise, D. Weiss, M. Brittner, and M. Skerker. H. Venters and N. Dickey contributed to article revisions and writing. Acknowledgments The authors would like to acknowledge the contribution of John Volpe and the members of the New York City RxStat Collaboration, including Denise Paone, Hillary Kunins, Daliah Heller, Blythe McCoy, Pamela Phillips Lum, Kaitrin Roberts, and Chauncey Parker. Human Participant Protection This study was deemed to be exempt from institutional review board review as research involving the collection or study of existing data, documents, records, patholog- ical specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that participants cannot be identified, directly or through identifiers linked to the participants. References 1. Carson AE. Prisoners in 2013. Bureau of Justice Statistics. 2014. Available at: http://www.bjs.gov/index. cfm?ty=pbdetail&iid=5109. Accessed February 2, 2015. 2. Glaze LE, Kaeble D. Correctional populations in the United States, 2013. 2014. Bureau of Justice Statistics. Available at: http://www.bjs.gov/index.cfm? ty=pbdetail&iid=5177. Accessed February 2, 2015. 3. Drucker E.A Plague of Prisons: The Epidemiology of Mass Incarceration in America. New York, NY, and London, England: The New Press; 2011: xiv, 189. 4. Sisti DA, Segal AG, Emanuel EJ. Improving long- term psychiatric care: bring back the asylum.JAMA. 2015;313(3):243---244. 5. Dumont DM, Brockmann B, Dickman S, Alexander N, Rich JD. Public health and the epidemic of incarcer- ation.Annu Rev Public Health. 2012;33:325---339. 6. Cloud DH, Parsons J, Delany-Brumsey A. Address- ing mass incarceration: a clarion call for public health.Am J Public Health. 2014;104(3):389---391. 7. Rich JD, Wakeman SE, Dickman SL. Medicine and the epidemic of incarceration in the United States.N Engl JMed. 2011;364(22):2081---2083. 8. Gawande A. The hot spotters: can we lower medical costs by giving the neediest patients better care?New Yorker. January 24, 2011: 40---51. 9. Tricco AC, Antony J, Ivers NM, et al. Effectiveness of quality improvement strategies for coordination of care to reduce use of health care services: a systematic review and meta-analysis.CMAJ. 2014;186(15):E568---E578. 10. Shumway M, Boccellari A, O’Brien K, Okin RL. Cost- effectiveness of clinical case management for ED fre- quent users: results of a randomized trial.Am J Emerg Med. 2008;26(2):155---164. 11. Green SR, Singh V, O’Byrne W. Hope for New Jersey’s city hospitals: the Camden initiative.Perspect Health Inf Manag. 2010;7:1d. 12. Englander H, Michaels L, Chan B, Kansagara D. The care transitions innovation (C-TraIn) for socioeconomi- cally disadvantaged adults: results of a cluster random- ized controlled trial.J Gen Intern Med. 2014;29(11): 1460---1467. 13. Mandelberg JH, Kuhn RE, Kohn MA. Epidemiologic analysis of an urban, public emergency department’s frequent users.Acad Emerg Med. 2000;7(6):637---646. 14. Hunt KA, Weber EJ, Showstack JA, Colby DC, Callaham ML. Characteristics of frequent users of emer- gency departments.Ann Emerg Med. 2006;48(1):1---8. 15. Williams BC, Paik JL, Haley LL, Grammatico GM. Centralized care management support for“high utilizers” in primary care practices at an academic medical center. Care Manag J. 2014;15(1):26---33. 16. Criteria for serious and persistent mental illness. New York Office of Mental Health. Available at: http:// www.omh.ny.gov/omhweb/guidance/serious_ persistent_mental_illness.html. Accessed February 6, 2015. 17. Levanon Seligson A, Lim S, Singh T, et al. New York/New York III Supportive Housing Evaluation: in- terim utilization and cost analysis. A report for the New RESEARCH AND PRACTICE November 2015, Vol 105, No. 11 |American Journal of Public Health MacDonald et al.|Peer Reviewed |Incarceration |2267 York City Department of Health and Mental Hygiene in collaboration with the New York City Human Resources Administration and the New York State Office of Mental Health, 2013. Available at: http://www.nyc.gov/html/ doh/downloads/pdf/mental/housing-interim-report.pdf. Accessed February 2, 2015. 18. Dumont DM, Allen SA, Brockmann BW, Alexander NE, Rich JD. Incarceration, community health, and racial disparities.J Health Care Poor Underserved. 2013;24(1): 78---88. 19. Williams BA, Sudore RL, Greifinger R, Morrison RS. Balancing punishment and compassion for seri- ously ill prisoners.AnnInternMed. 2011;155(2): 122---126. 20. Lim S, Seligson AL, Parvez FM, et al. Risks of drug-related death, suicide, and homicide during the immediate post-release period among people released from New York City jails, 2001---2005.Am J Epidemiol. 2012;175(6):519---526. 21. Binswanger IA, Stern MF, Deyo RA, et al. Release from prison—a high risk of death for former inmates [erratum in N Engl J Med. 2007;356(5):536].N Engl J Med. 2007;356(2):157---165. 22. Nosyk B, Anglin MD, Brecht ML, Lima VD, Hser YI. Characterizing durations of heroin abstinence in the California Civil Addict Program: results from a 33-year observational cohort study.Am J Epidemiol. 2013;177 (7):675---682. 23. Selling D, Solimo A, Lee D, Horne K, Panove E, Venters H. Surveillance of suicidal and nonsuicidal self- injury in the New York City jail system.J Correct Health Care. 2014;20(2):163---167. 24. Hayes LM. National study of jail suicide: 20 years later.J Correct Health Care. 2012;18(3):233---245. 25. Brittain J, Axelrod G, Venters H. Deaths in New York City jails, 2001---2009.Am J Public Health. 2013;103(4):638---640. 26. Ludwig A, Cohen L, Parsons A, Venters H. Injury surveillance in New York City jails.Am J Public Health. 2012;102(6):1108---1111. 27. Kaba F, Diamond P, Haque A, MacDonald R, Venters H. Traumaticbrain injury among newly admitted adolescents in the New York City jail system.J Adolesc Health. 2014;54(5):615---617. 28. Kaba F, Lewis A, Glowa-Kollisch S, et al. Solitary confinement and risk of self-harm among jail inmates.Am J Public Health. 2014;104(3):442---447. 29. Stuckler D, Basu S, McKee M, King L. Mass in- carceration can explain population increases in TB and multidrug-resistant TB in European and central Asian countries.Proc Natl Acad Sci U S A. 2008;105 (36):13280---13285. 30. Awofeso N. Prisons as social determinants of hepatitis C virus and tuberculosis infections.Public Health Rep. 2010;125(suppl 4):25---33. 31. Baussano I, Williams BG, Nunn P, Beggiato M, Fedeli U, Scano F. Tuberculosis incidence in prisons: a systematic review.PLoS Med. 2010;7(12):e1000381. 32. Gough E, Kempf MC, Graham L, et al. HIV and hepatitis B and C incidence rates in US correctional populations and high risk groups: a systematic review and meta-analysis.BMC Public Health. 2010;10:777. 33. Patterson EJ. The dose---response of time served in prison on mortality: New York State, 1989---2003.Am J Public Health. 2013;103(3):523---528. 34. Spaulding AC, Seals RM, McCallum VA, Perez SD, Brzozowski AK, Steenland NK. Prisoner survival inside and outside of the institution: implications for health-care planning.Am J Epidemiol. 2011;173(5):479---487. 35. Aidala AA, McAllister W, Yomogida M, Shubert V. Frequent Users Service Enhancement “FUSE”initiative: New York City FUSE II evaluation report. Available at: http://www.csh.org. Accessed February 6, 2015. 36. Hopper K, Jost J, Hay T, Welber S, Haugland G. Homelessness, severe mental illness, and the institutional circuit.Psychiatr Serv. 1997;48(5):659---665. 37. City of New York. Mayor’s Task Force on Behavioral Health and the Criminal Justice System. Action plan. 2014. Available at: http://www1.nyc.gov/assets/ criminaljustice/downloads/pdf/annual-report-complete. pdf. Accessed February 6, 2015. RESEARCH AND PRACTICE 2268 |Incarceration |Peer Reviewed |MacDonald et al.American Journal of Public Health |November 2015, Vol 105, No. 11