Center on Juvenile and Criminal Justice   CENTER ON JUVENILE AND CRIMINAL JUSTICE
www.cjcj.org  
Center on Juvenile and Criminal Justice, 54 Dore Street, San Francisco, CA 94103Tel: (415) 621-5661 | Fax: (415) 621-5466

Community-Based Treatment: The Impact of the Homeless Pretrial Release Project

[Press Release]

CONTACT: Gerald Miller
E-mail: [gerald@cjcj.org]
Tel: (415) 621-5661

Introduction

The Homeless Release Project (HRP) was initiated in response to community needs stressed in San Francisco's Continuum of Care: A Five Year Strategic Homeless Plan, which identified the need to monitor individuals discharged from jail in order to help secure housing and access social services. In 1996, HRP received grant monies from the United Way for a four-year pilot program to provide appropriate supervision to homeless persons arrested for misdemeanor offenses. As such, the Homeless Release Project (HRP) serves dual purposes for a socially vulnerable population. As a pretrial release program, HRP plays an important role in reducing the jail population while ensuring compliance with court mandates, and as a model of community corrections, HRP monitors homeless offenders in the community by way of intensive supervision and case management. This is an evaluative study of recidivism rates based on HRP's first year of operation.

In San Francisco, the enormity of homelessness as a salient social problem has long penetrated the county jail and hospital facilities. A Housing Status Assessment of County Bookings, written for the San Francisco Sheriff's Department in 1994, reported that 39% of persons booked into the County Jail were either homeless or temporarily housed. According to the City's Department of Public Health Annual Report (1997-98), San Francisco has disproportionate rates of homelessness, substance abuse and mental illness, including the highest rate of drug emergency room visits in the nation, the highest suicide rate and the second highest rate of homelessness.1 An estimated 30-40% of the homeless in San Francisco suffer from serious mental illness (Tuprin and Tate 1997). In addition, upwards of 70% have substance abuse problems (Tuprin and Tate 1997; Homebase 1997). During fiscal year 1996-7, there were 9,114 involuntary detentions for psychiatric evaluation, giving San Francisco the highest per capita rate of any California county; eighty percent of those detained were estimated to have co-occurring substance abuse disorders and fifty percent were estimated to be homeless. The average length of stay in the hospital was only eighteen hours, and due to a lack of options, homeless individuals are often simply returned to the streets. Homeless populations are also vulnerable to high-risk health practices, such as needle sharing and unprotected sex, and infectious diseases, including hepatitis and tuberculosis (Wojtusik and White 1997). The homeless accounted for 18% of all existing TB cases in San Francisco (Northern California Council for the Community 1998).

The problematic effects of deinstitutionalization of state mental health hospitals in the 1960's and 1970's have been well documented, particularly the burden it has placed on jails due to increased arrests and incarcerations of mentally ill persons (Whitmer 1980; Walsh and Bricout 1996)2. Belcher (1988) concluded that homeless mentally ill offenders are vulnerable to chronic decompensation unless they are supplied with a supportive and structured environment. However, jails are poorly equipped to properly diagnose and treat persons in need. Efforts to integrate mental health services into jails have generated basic services, yet the criminal justice system cannot ensure continued compliance with follow-up care once the offender is released into the community (Kalinich et al. 1988)3.

HRP, like other recent innovations in community corrections, is modeled on enhanced partnerships between judicial administrators and local providers as an effective method for aiding offenders' transition back into their communities (see Leaf et al. 1993). The uniqueness of HRP is that it seeks to remedy the general disconnect offenders face from the community and familial networks (see Irwin 1985) by addressing in concert, by way of intensive case management, individualized mental/medical needs, chronic homelessness and concurring court appearances.

Pretrial Innovation: Alternative Programming in the San Francisco County Jail

Prior to the initiation of HRP services, CJCJ had initiated two population-specific pretrial release programs. In the early 1980's California's fiscal crisis and increasing incarceration rates resulted in serious jail-overcrowding problems throughout the state. At the time, the San Francisco Sheriff's Department was already under a two decades long consent decree to decrease its jail population and improve confinement conditions. In response to this institutional crisis, The Center on Juvenile and Criminal Justice (CJCJ) established the Supervised Misdemeanor Release Program (SMRP) in 1987. SMRP is designed to reduce jail overcrowding by serving as a jail alternative to misdemeanants arrested on bench warrants. Persons arrested for new misdemeanor offenses are regularly released by the Sheriff's Department with a citation, or written promise to appear in court, with the exception of offenses that involve weapons or violence. After a failure to appear, a bench warrant is issued and court approval must be sought for non-financial release. SMRP staff screen the pretrial population, identify and interview eligible misdemeanants and then submit release recommendations to the Court. If the Court accepts the release recommendation, SMRP staff monitor an offender's court compliance until the case is disposed. SMRP staff screened over 2,300 cases in 1999 and made 844 releases with 85% appearing in court.

During the early stages of the program, SMRP staff recognized a growing number of homeless defendants who were not eligible for citation release because they lacked an address. In response, CJCJ worked with the Sheriff's Department to establish the "No Local" Citation Project in 1991. This project targeted homeless offenders charged with misdemeanor offenses or infraction warrants. Because the "No Local" project did not release persons charged with bench warrants, court approval for the release was not required. Over the next six years, more than 1700 persons were released on their "promise to appear" in court with a compliance rate of 76%. Due to the project's success, the San Francisco Sheriff's Department changed its citation policies in 1997 to no longer exclude homeless persons.

HRP: A Community Based Treatment Model

A homeless offender presents a number of unique challenges beyond the lack of a stable address. Figure 1 (Appendix) illustrates how HRP employs the community-based treatment (CBT) model which serves as the blueprint for providing individualized care to homeless offenders. The HRP caseworker develops a care plan with the client and submits oral or written progress reports at all subsequent court dates. During the primary screening stage, a SMRP staff member identifies homeless offenders arrested on misdemeanors in the jail. The secondary screening stage entails conducting a preliminary needs assessment, which deter-mines the offender's existing relationships with community providers and collects information on where the offender can be found in the community. This preliminary data is then submitted to the court for a recommendation for release from jail and HRP program participation. Once the release is approved, jail staff determine a temporary housing plan which could include supplying the client with a hotel voucher, and provides the HRP case manager with information on the client and his or her court date.

On the initial court date following release, the HRP case manager meets with the client and the assigned Public Defender to determine the status of the pending criminal case. HRP staff accompany all clients to their court dates and strive to gain their active participation in what can be an alienating and quick-paced process. Immediately following their first court date together, the case manager conducts a more thorough needs assessment, collaborating with the client on designing a care plan which includes short and long-term goals, such as obtaining temporary/permanent housing, entering a substance abuse program or accessing medical treatment. Often short-term goals such as accessing mental health or substance abuse treatment will become incorporated into a court-mandated diversion program. However, the plan is created in collaboration with the client, and usually contains components, such as seeking medical care, which are irrelevant to the criminal proceedings.

Once the care plan begins to be implemented, the HRP case manager ensures that the client appears at all subsequent court dates. The case manager often spends the majority of his time outside of court working with clients in shelters, encampments, hotels and the street.

Clients are also invited to drop in at the office. Staff strive to make the office as inviting as possible by not requiring appointments, providing food, clothing, temporary storage, the use of the phone and the office safe for holding cash.

Beyond facing homelessness and pending criminal matters, the majority of HRP's clients are also suffering from medical fragility, mental illness, and/or substance abuse. Approximately 85% of HRP clients are dealing with substance abuse issues and 50% have been diagnosed with a co-occurring mental illness. Often HRP clients can display disruptive behavior, inhibiting their ability to access services. So while assistance can sometimes entail a referral and a bus token, it often means accompanying a client to an appointment. Over the last two years, CJCJ has received supplemental foundation funding to allow us to hire peer advocates (former offenders who are in recovery) to assist the HRP case manager. By accompanying our clients to appointments, the peer advocate helps them control their frustrations with the often difficult intake processes of social service agencies. The use of peer advocates brings a special understanding of client issues to service delivery. The shared experiences of the client and case worker often help to establish the client/case worker relationship, thus facilitating the most positive and successful program outcomes.

Data Summary Points

While the experimental and comparison groups had similar prior San Francisco arrest histories, the experimental group showed a decrease in recidivism rates after participating in the Homeless Release Project:


Methodology

This study analyzes the recidivism rates of two populations of offenders: HRP's successful graduates from the project's first year of operation (experimental group) and offenders who did not receive HRP services (comparison group). Rates of recidivism were determined by reviewing individual San Francisco criminal histories from the point of staff contact (7/1/96 to 6/30/97) through March 1999, representing a study period range from 20 to 32 months. This analysis differentiates recidivism between arrest rates (which indicate police contact with no charges filed by District Attorney) and re-offense rates (which indicate that charges were filed by the District Attorney). We further broke down re-offenses into felony and misdemeanor categories.

The experimental group consists of forty-one individuals who were released from the county jail through HRP from July 1, 1996 through June 30 1997, and completed the program. Program completion is defined as maintaining contact with project staff and remaining on supervised release until the pending criminal matter is disposed after judicial review. The average length of stay on the caseload for these individuals was approximately five months, and their average duration of homelessness prior to their participation was two and a half years.

The comparison group was derived from two separate pools of homeless persons charged with misdemeanors. The first group consists of eighteen individuals who were released through HRP between July 1, 1996 and June 30, 1997 but failed to appear in court. Similar to the experimental group, these individuals were interviewed in the jail, had misdemeanor bench warrants and self-identified as homeless. However, after they did not appear in court, staff was unable to locate them, so they never participated in the program. To provide a valid comparison with the 41 persons in the experimental group, we decided to draw additional persons from the No Local Citation Program. We drew twenty-three non-random persons from the No Local database for the same time period of July 1, 1996 to June 30, 1997. Clients were selected to match the experimental group based on race, gender and offense incident (see table 1).

Table 1: Prior San Francisco Criminal Histories4
  Comparison Experimental

No Arrests

 1 1
Felonies 22 21
Misdemeanors 13 13

Table 2: Demographic Breakdown of Offenders
  Comparison Experimental
Gender    
Male 36 36
Female 5 5
Race/Ethnicity    
White 15 13
African American 17 21
Latino/a 5 4
Asian/Pacific Islander 2 2
Native American 2 1
Offense Incident (at point of staff contact)    
Drug 14 5
Theft 13 10
Prostitution 3 10
Batteries 3 9
Vehicle Codes 3 3
Trespass 3 0
Weapons 0 1
Vandalism 2 2
Obstructing 0 1

Prior Arrest Histories

We ran prior arrest histories for both groups after they were selected and found that offenders in both the experimental and comparison groups had highly comparable prior criminal histories. While the group selection process itself was non-random, the similarity in prior arrest histories supports our contention that these groups are comparable in likelihood of re-offending without the mediation of HRP.

Finding 1: Decrease in Overall Arrest Rates for HRP Participants
One of the most significant results was the number of HRP graduates who had no arrests during or following their participation in HRP. Fifteen (or 37%) of the experimental group did not recidivate as compared to only eight individuals (or 20%) in the comparison group. In other words, persons who did not receive services were almost twice as likely to be rearrested. The general deduction we make from this finding is that when the individualized needs of homeless offenders are met such as housing, benefits assistance and mental health and substance abuse treatment participants are better equipped to avoid future criminal behavior.

Finding 2: Decrease in Re-offense Rates for HRP Participants
As mentioned previously, the reoffense rate is defined as a new felony or misdemeanor charge filed by the District Attorney. Within the experimental group, eighteen participants were arraigned on new offenses, resulting in a recidivism rate of 44%, whereas within the comparison group twenty-nine members were arraigned on a new offense, resulting in a recidivism rate of 71%.

Finding 3: Decrease in the Number of Serious Re-offenses for HRP Participants
To further examine the re-offense rates of both the experimental and comparison groups, the charts below illustrate the differences in the seriousness of the re-offenses. The eighteen HRP participants who were arraigned on a new offense committed fewer felonies than those in the comparison group. Within the comparison group, twenty-nine persons had re-offenses: twenty-four (or 83%) were arraigned on felony counts and five (or 17%) were arraigned on misdemeanors. Within the experimental group, eighteen persons had re-offenses: eight (or 44%) were arraigned on felonies and ten (56%) were arraigned on misdemeanors. Although the experimental group had more misdemeanor re-offenses (ten as compared to 5), of importance here is that the HRP participants were arraigned on fewer re-offenses, and these tended to be of a far less serious nature than those of the comparison group.

Finding 4: Greater Number of Discharges for HRP Participants
HRP graduates who were arrested were more likely to have their cases discharged by the District Attorney's Office than persons in the comparison group. Of the 26 individuals in the experimental group whose criminal histories indicated police arrests, seven (or 27%) were never arraigned while only three (or 9%) of the thirty-three comparison group arrestees were not discharged by the District Attorney. While an indepth study of this discharge process was beyond the scope of this analysis, we can reasonably deduct that the rearrests for the HRP experimental group were more diminimus than those in the comparison group.5

Discussion

Homelessness is a multifaceted social problem that is further complicated when intertwined with the judicial system. A source of amelioration is to remedy the serious disconnect between external community providers and the criminal justice system. Homeless offenders typically lack ties to community resources, which undermines their ability to comply with court demands. As such, homeless offenders naturally pose a special challenge to pretrial release standards of court compliance, especially for urban jails that process large numbers of misdemeanants.

The Homeless Release Project serves as organizational linkage between a homeless person's detainment, subsequent court appearances and social services. The data illustrate that the differences in arrest rates and seriousness of offense between offenders who participated in HRP and the comparison group are attributable to this unique approach to community corrections. In addition, individualized treatment, which is the hallmark of the community-based treatment model, yields a positive long term impact on the institutional level: reduced rate of re-offenses and reduced costs of over detainment.

Researchers have hypothesized that the homeless are monitored more closely and arrested more frequently due to their social status and physical appearance (Dunford 1987; Snow et al 1989). Persons who live their lives in the public domain are the most vulnerable to police scrutiny. The high number of police contacts for both the experimental and comparison groups must be understood within this context. While more indepth evaluative studies need to be conducted, our preliminary data on the effectiveness of the Community Based Treatment model points to the need for structural as well as community alliances between judicial procedures and social services providers to reduce recidivism rates of homeless individuals.

References

Belcher, John R. 1988. "Are Jails Replacing the Mental Health System for the Homeless Mentally Ill?" Community Mental Health Journal 24: 185-195.

City and County of San Francisco. 1995. Continuum of Care: A Five Year Strategic Homeless Plan 1995-2000.

Dunford, Franklyn W. 1987. Crime Among Homeless Persons. Boulder, CO: Institute of Behavioral Sciences, University of Colorado.

Irwin, John. 1985. The Jail: Managing the Underclass. Berkeley: University of California Press.

HomeBase. 1997 (May). Homelessness in San Francisco at a Glance.

Kalinich, Dave, Paul Embert and Jeffrey D. Senese. 1988. "Integrating Community Mental Health Services Into Local Jails: A Policy Perceptive." Policy Studies Review 7: 660-670.

Northern California Council for the Community. 1998. "Building a Healthier San Francisco: A Citywide Collaborative Community Assessment." Health, Social and Economic Indicators Report: Volume 1.

Snow, David A., Susan G. Baker and Leon Anderson. 1989. "Criminality and Homeless Men: An Empirical Assessment." Social Problems 36: 532-549.

Tuprin, Allen M.D. and James Tate. 1997. "Mental Health and Homelessness." San Francisco Medicine (January).

Walsh, Joseph and John Bricout. 1996. "Improving Jail Linkages of Detainees with Mental Health Agencies: The Role of Family Contact." Psychiatric Rehabilitation Journal 20: 73-76.

Whitmer, Gary E. 1980. "From Hospitals to Jails: The Fate of California's Deinstitutionalized Mentally Ill." American Journal of Orthopsychiatry 50: 65-75.

Wojtusik L. and M.C. White. 1997. "Health Status, Needs and Health Care Barriers Among the Homeless." Department of Community Health Systems, University of California School of Nursing.

Endnotes

  1. There are an estimated 11,000 to 14,000 homeless persons living in San Francisco (Homebase 1997).
  2. Walsh and Bricout (1996) studied how family contacts act as linkages to mental health agencies once the offender is released from jail; this work acknowledges the effectiveness of community ties in ensuring an offender's "continuity of care" (p. 73).
  3. Also see Steadman, H.J. et al. 1989. The Mentally Ill in Jail: Planning for Essential Services. New York: The Guilford Press for further discussion on the homeless mentally ill in jails.
  4. Prior criminal history figures do not include charges discharged by the District Attorney.
  5. The number of District Attorney discharges includes one individual from each group where the DA pursued motions to revoke probation in lieu of prosecuting the new offense.
Suggested citation format for this report: Castellano, Ursula A. and Riker, Alissa. (2000) Community-Based Treatment: Impact of the Homeless Pretrial Release Project.

This site and its contents © 2002 Center on Juvenile and Criminal Justice