Summary

Department of Corrections: Though Improving, the Department Still Does Not Identify and Serve All Parolees Needing Outpatient Clinic Program Services, but Increased Caseloads Might Strain Clinic Resources August 2001 2001-104 rotiduA etatS ainrofilaC S T I D U A E T A T S F O U A E R U B The first five copies of each California State Auditor report are free. Additional copies are $3 each, payable by check or money order. You can obtain reports by contacting the Bureau of State Audits at the following address: California State Auditor Bureau of State Audits 555 Capitol Mall, Suite 300 Sacramento, California 95814 (916) 445-0255 or TDD (916) 445-0255 x 216 OR This report may also be available on the World Wide Web http://www.bsa.ca.gov/bsa/ The California State Auditor is pleased to announce the availability of an online subscription service. For information on how to subscribe, please visit our website at www.bsa.csa.gov/bsa. If you need additional information, please contact David Madrigal at (916) 445-0255, ext. 201. Alternate format reports available upon request. Permission is granted to reproduce reports. C S A ALIFORNIA TATE UDITOR ELAINE M. HOWLE STEVEN M. HENDRICKSON STATE AUDITOR CHIEF DEPUTY STATE AUDITOR August 29, 2001 2001-104 The Governor of California President pro Tempore of the Senate Speaker of the Assembly State Capitol Sacramento, California 95814 Dear Governor and Legislative Leaders: As requested by the Joint Legislative Audit Committee, the Bureau of State Audits presents its audit report concerning the Department of Corrections’ (department) Parole Outpatient Clinic Program (program). This report concludes that the program has failed to serve many of the parolees that the department has determined could most benefit from clinic services. These are primarily mentally ill parolees, but also include other parolees who may pose a threat to public safety, such as sexual offenders. Even though recent changes have resulted in some improvement, the program still failed to serve almost 40 percent of mentally ill inmates who were paroled between October 2000 and March 2001. The program expects additional improvement in its ability to serve mentally ill parolees when it implements a new data management system shortly. However, because the data management system was not yet in use when we concluded our work, we could not assess its effectiveness. Recent changes, both inside and outside the program, will likely increase the demand for program services and highlight the program’s need to increase its effectiveness. Respectfully submitted, ELAINE M. HOWLE State Auditor BUREAU OF STATE AUDITS 555 Capitol Mall, Suite 300, Sacramento, California 95814 Telephone: (916) 445-0255 Fax: (916) 327-0019 www.bsa.ca.gov/bsa CONTENTS Summary 1 Introduction 5 Audit Results The Department Has Failed to Identify All Parolees Requiring Mental Health Treatment 15 The Clinics Do Not Always Perform Needed Prerelease Assessments or Provide Timely Services 19 A New Data Management System May Address Some Program Weaknesses But Could Be More Effective if Linked to Other Department Systems 22 The Clinics Provide Services to Many Parolees With Problems Other Than Mental Illness 26 The Program Needs to Take Additional Actions to Manage Expected Caseload Increases 30 Recommendations 34 Appendix Caseload Information for the Parole Outpatient Clinic Program 37 Response to the Audit Department of Corrections 41 California State Auditor’s Comments on the Response From the Department of Corrections 51 SUMMARY RESULTS IN BRIEF T he Department of Corrections (department) provides treatment and supervision to mentally ill parolees Audit Highlights . . . through its Parole Outpatient Clinic Program (program). However, it has failed to identify and treat a significant number Our review of the Parole of these parolees. Although a new process has effected some Outpatient Clinic Program improvement, between October 2000 and March 2001 the (program) at the Department of Corrections (department) department failed to detect and serve 39 percent of inmates found that: (about 2,400) who had been diagnosed as mentally ill and were (cid:1) being released on parole. For those it did identify, the program The program’s new continuum process, while did not always complete assessments that determine what an improvement over its treatments the parolees will require or provide services within previous process, still does specified periods. By failing to detect and promptly assess not identify and serve mentally ill parolees, the department exposes the public to nearly 40 percent of mentally ill parolees. increased risks. The program believes that contractors it recently hired to assume the responsibility of performing prerelease (cid:1) In 38 of the 83 cases assessments will complete them more efficiently and help clinics we reviewed, social workers did not perform see parolees sooner. prerelease assessments, and 45 parolees were not The program is implementing a new computerized data manage- seen by the clinics within ment system to improve its ability to transfer mentally ill required time frames. inmates to its services once paroled and to schedule appoint- (cid:1) A new data management ments within prescribed time frames. However, because the system, when implemented, program had not yet fully implemented the new system at the may address some of the time of our fieldwork, we could not assess its effectiveness. program’s weaknesses, but it would be more Furthermore, the program plans to continue to use its current effective if linked to other method of identifying mentally ill inmates being released to department computer parole, which relies primarily on a list that the department’s systems. institutional computer systems generate. However, this list does (cid:1) One-third of the parolees not always include all inmates being paroled who require treat- served by the program are ment at the parole outpatient clinics (clinics). Therefore, to more not diagnosed with a effectively recognize all mentally ill parolees, the program mental illness but fit other criteria established by the should link its system directly to other key department systems. department. (cid:1) One of the reasons the program has not been able to treat all The program should mentally ill parolees adequately is that it has used its resources establish caseload standards and use its new to treat many parolees with problems other than mental illness, system to identify its cost such as those with histories of violent crimes or serious sex of serving different types offenses. Between October 2000 and March 2001, more than of parolees so it can manage expected 30 percent (or 3,000) of the parolees served by the program had caseload increases. no diagnosis of mental illness. According to the program, the Youth and Adult Correctional Agency, which provides oversight 1 to the department, requested the program to provide services to paroled sex offenders and violent criminals. However, because the department has failed to identify many of these parolees, they have not received treatment either. If the department had identified all these parolees, the program might lack the capacity to serve them in addition to the mentally ill. Therefore, the department should ensure that the program identifies and has resources available to meet the needs of all the parolees it is required to serve. Further, many recent changes both inside and outside the program will likely increase the demand for clinic services and highlight the program’s need to increase its effec- tiveness. To meet the challenges this expected growth poses, the program needs to establish caseload standards for its clinicians so that it can adequately monitor caseloads and ensure it has sufficient resources. RECOMMENDATIONS To help ensure that the public is adequately protected from parolees who were diagnosed as mentally ill while in prison, the program should fully implement its new data management system. The program should then develop a monitoring process to ensure that its contractors complete prerelease assessments on all mentally ill inmates scheduled for parole and that its clinics see parolees within required time frames. To more effectively identify all the parolees the program will serve, including those with problems other than mental illness, the program should link its new data management system to other department computer systems containing the information it needs to do so. To better identify the costs of treating parolees and justify changing needs for resources, the program should track the amount of time and resources it spends treating the different types of parolees and develop caseload standards for its clinicians. To determine the progress the program has made in identifying and serving mentally ill and other parolees, the department should assess the program one year after implementing the new data management system. The department should submit the completed assessment to the Youth and Adult Correctional Agency. 2 AGENCY COMMENTS The department generally agrees with our conclusions and recommendations. It believes the findings will assist it in completing the implementation of this program. (cid:1) 3 Blank page inserted for reproduction purposes only. 4 INTRODUCTION BACKGROUND T he Department of Corrections (department), the largest state agency in terms of staffing, is divided into two functional areas—Field Operations and Support Services— as depicted in Figure 1. Field Operations is responsible for the supervision and care of inmates during incarceration as well as oversight of parolees. The functions of Support Services include legal affairs, facilities management, and information systems management. Parole and Community Services Division The Parole and Community Services Division within the department’s Field Operations area is responsible for supervising and providing services to felons who have been paroled. The level of supervision a parolee receives is based on factors related to the offender’s history of violence and current service needs. Case assessments dictate the placement of selected parolees in a category of supervision intended to prevent, detect, or interrupt behavior likely to endanger the community or themselves. The Parole and Community Services Division oversees 136 parole offices and four regional parole outpatient clinics (clinic) that provide services statewide. The parole agent is responsible for the following tasks: supervising parolees; arranging services such as housing, medical care, employment, counseling, education, and social activities; investigating alleged parole violations; and helping assess a parolee’s risk to the community and the type of services he or she requires. The department employs more than 1,700 parole agents who are responsible for the supervision of approximately 121,000 parolees throughout the State. 5 6 FIGURE 1 Department of Corrections Organizational Chart as of December 2000 Director Field Support Operations Services Parole and Health Care Office of Legal Office of Institutions Community Office of Services Financial Affairs Correctional Division Services Compliance Division Management Division Planning Division m h a e e In n d a s l d i t t c h i a t m u l s , t e e i d n o r e v n t n i a c a l t e l a s l, 3 1 3 c 6 o 3 s c f r t 8 a r o a e c t m c c i e a l t i m t m i p o ie r u n p i s s n a s o i l t n y s out 1 p 4 4 3 a o 6 t p r i e f e a f p g i n r c a i o e o t ro l s n c e l s l e in ics R H D es u iv o m i u si a r o c n n e s in (i f n o I c r n m lu S f D o d a y i t r e v s i m o t s is e n o i a m o f t s f n y i e s o s n t n d e e m r s M Fa a c B D n i u l i a i v s t g i i i e n s e i s e o m s a n s e n n d t unit) Note: Functions appearing in bold type are those we discuss in this report. Source: Department of Corrections. Parole Outpatient Clinic Program History According to the department, it established the Parole Outpa- tient Clinic Program (program) in 1954 as a result of an academic study that found that parolees who participated in intensive psychotherapy with specially trained parole agents and clinicians were less prone to recidivism, committing other crimes and returning to prison. The department originally operated two clinics: a northern clinic primarily responsible for providing individual therapy to parolees and a southern clinic that provided group therapy. The mission of the program is to reduce the symptoms of men- tal illness among parolees, lower the rate of recidivism, and improve public safety. To accomplish these goals, the program supplies clinical services such as evaluations of mental health status, medication management, and individual or group therapy. In 1991 the department divided the program into four regions, with their respective headquarters in Sacramento, Oakland, Los Angeles, and Diamond Bar, as shown in Figure 2. Services are provided either at a central outpatient clinic, as in Los Angeles, or at the various parole unit offices within each region. The clinics target parolees of three groups: (1) those who are diagnosed in prison as mentally ill, violent offenders with identified mental disorders, or serious sex offenders; (2) those who must attend the clinics as a condition of parole; and (3) those who exhibit symptoms of mental illness after they are released from prison. Current department policy requires the clinics to see all parolees who leave prison with diagnosed mental illnesses. However, parole agents can also refer parolees who they believe have mental health needs or those who have in the past committed offenses that were sexual or violent in nature and might pose a threat to the community. No statute defines the role or responsibility of the program. Consequently, the department has established regulations and policies to specify the population the program should serve and the services it should provide. Current regulations and policies mandate that the clinics serve all parolees who were diagnosed as mentally ill while in prison and allows the program discretion in determining whether the clinics treat parolees who commit- ted serious sex offenses or violent crimes when mental illness was a contributing factor. Additionally, the program has received verbal direction from the Youth and Adult Correctional Agency, which oversees the department, to treat all parolees with records of serious sex offenses. 7 FIGURE 2 Department of Corrections’ Parole Regions and Offices 1 DEL Parole Regions and Offices NORTE SISKIYOU Region I Region II Region III Region IV 2 Region Headquarters 3 4 5 6 7 9 10 8 12 11 13 14 15 16 18 17 19 20 21 22 23 24 25 Correctional Institutions 26 1 Pelican Bay State Prison 18 California Substance Abuse 27 29 2 High Desert State Prison Treatment Facility 28 30 3 California Correctional Center 19 California State Prison, Corcoran 4 Folsom State Prison 20 Avenal State Prison 5 California State Prison, Sacramento 21 North Kern State Prison 6 California Medical Facility 22 Wasco State Prison 31 7 California State Prison, Solano 23 California Men's Colony 8 San Quentin State Prison 24 California Correctional Institution 33 32 9 Mule Creek State Prison 25 California State Prison, Los Angeles County 10 Northern California Women's Facility 26 California Institution for Men 11 Deuel Vocational Institution 27 California Institution for Women 12 Sierra Conservation Center 28 California Rehabilitation Center 13 Valley State Prison for Women 29 Chuckawalla Valley State Prison 14 Central California Women's Facility 30 Ironwood State Prison 15 Correctional Training Facility 31 California State Prison, Calipatria 16 Salinas Valley State Prison 32 Richard J. Donovan Correctional Facility at Rock Mountain 17 Pleasant Valley State Prison 33 California State Prison, Centinela Source: Department of Corrections. 8 For mentally ill parolees, the program has established criteria for admission to the clinics that are consistent with the criteria for admission into the prison system’s mental health program. Specifically, the inmate or parolee must be diagnosed with a psychiatric disorder as described in the Diagnostic and Statistical Manual of Mental Disorders (manual). The manual, published by the American Psychiatric Associa- Psychiatric Disorders Found in the tion, provides standard criteria for psychiatrists Diagnostic and Statistical Manual of and psychologists in diagnosing mental disorders. Mental Disorders ••••• Schizophrenia On entering the prison system, an inmate must ••••• Delusional disorder undergo an initial health screening at a prison ••••• Schizophreniform disorder reception center. In addition to a basic physical health screening, the inmate must go through a ••••• Schizoaffective disorder mental health screening. Based on this screening, ••••• Substance-induced psychotic disorder some inmates are further evaluated to determine ••••• Psychotic disorder due to a general whether they are experiencing one of the disor- medical condition ders included in the manual. If an inmate is ••••• Psychotic disorder not otherwise specified determined to have a psychiatric disorder, the ••••• Major depressive disorders clinician must then determine the inmate’s level ••••• Bipolar disorders I and II of functioning for placement in an appropriate mental health program in the prison. In its inmate tracking system, the department records the mental health program to which the inmate is assigned and later uses this information to identify parolees who require clinic services. The department has four programs for its mentally ill inmates: • The Correctional Clinical Case Management System (CCCMS) serves inmates who are diagnosed with at least one of the clinical psychiatric disorders listed in the manual, who have stable functioning in the community, and who score above 50 on the Global Assessment of Functioning.1 Generally, these are the inmates with the least severe mental illnesses. • The Enhanced Outpatient Program (EOP) serves inmates who are diagnosed with at least one of the clinical psychiat- ric disorders listed in the manual; who demonstrate acute onset or significant deterioration of a serious mental disorder, dysfunctional or disruptive social interaction, or impairment of activities of daily living; and who score 50 or less on the Global Assessment of Functioning. These are inmates with more severe mental illnesses. 1The Global Assessment of Functioning score is a standard 100-point scale used by the mental health profession to measure a patient’s ability to function in society. 9 • Crisis Beds serves inmates who may not have been diag- nosed with a disorder from the manual but need services to assist them through periods of crisis in their lives. Inmates in this program are considered to be more severely mentally ill. • In an inpatient setting, the Department of Mental Health serves inmates who are severely mentally ill. Before October 2000 parole agents referred parolees to the clinics for services. When the department released an inmate on parole, the parole agent reviewed the inmate’s records. If the parole agent identified a history of mental illness or violations of a sexual nature in the parolee’s past, the parole agent was to refer the parolee to the clinic for evaluation. A parole agent also could refer a parolee who did not meet the mental health criteria but who, in the parole agent’s view, exhibited signs of mental instability or other problems that the program could treat. In July 2000 this process was enhanced by the Mental Health Services Continuum Program (continuum process), described in the next section. Figure 3 depicts the established and new processes. The Continuum Process In July 2000 the program implemented the continuum process, which applies to inmates released on parole on or after October 1, 2000. The continuum process is intended to ensure that the mental health treatment inmates receive while incarcer- ated continues when they are paroled to the community. The goals of the continuum process are the following: (cid:127) Reduce the symptoms of mental illness among parolees by providing timely, cost-effective mental health services. (cid:127) Optimize the level of individual functioning of mentally ill parolees in the community. (cid:127) Reduce recidivism. (cid:127) Improve public safety. Parolees who receive mental health treatment in prison and those referred by parole agents and found through clinic evalua- tions to meet the criteria required for admission to the prison mental health system are the focus of the continuum process. The process is an expansion of established services to help the department ensure continuity of care for parolees after their release into the community. 10 11 FIGURE 3 Comparison of the Program’s Established and New Processes Parole clinic Parole agent assesses needs of identifies need parolee. The clinic Process for All Inmates for services and Department then develops a Paroled Before refers to parole releases inmate service plan within October 2000* clinic on parole 30 days of referral Approximate 120 days before 90 days before 30 days before 3-10 days after Within 7 days Parole date Timeline parole parole parole parole date after parole date Offender Social worker Social worker Department Parole clinic Process for Information assesses needs of updates releases inmate establishes Mentally Ill Inmates Systems Unit mentally ill inmate information and on parole a service plan Paroled After provides list of establishes an for the parolee October 2000 mentally ill appointment for and begins to inmates scheduled the mentally ill provide services to be paroled inmate at a within 120 days parole clinic once paroled * This continues to be the process used to refer parolees to parole clinics, except for those parolees diagnosed as mentally ill in prison. The program originally designed the continuum process to identify all inmates who participated in the CCCMS and EOP while in prison so that the Parole and Community Services Division could alert parole agents and the clinics of these inmates’ needs. According to the program, it will soon include parolees who received Crisis Beds services or were Department of Mental Health inpatients while incarcerated. The program receives a printout once a month from the department’s Offender Information Systems unit, which lists all inmates who have been diagnosed with mental illness and are eligible for parole within the next 120 days. A social worker assesses each inmate before the release from prison. The inmate must meet with the social worker at 90 days and 30 days before being released. At the 90-day visit, the social worker verifies the inmate’s information to determine to which region the inmate will be paroled and reviews the inmate’s health and other records. At the 30-day visit, the social worker updates any information before the inmate’s release, gives the inmate infor- mation about the clinic the inmate will attend, and follows up with the prison to ensure that the inmate will receive any necessary medication when released. After this, the social worker forwards the assessment to the appropriate region, where the clinic starts a file on the parolee. After receiving the assessment and opening a file on the parolee, the clinic starts to develop a treatment plan geared toward the needs of the parolee. The guidelines for the continuum process indicate that the parolee must have an initial meeting at the clinic within 3 to 10 business days after being paroled, depend- ing on the severity of the parolee’s mental illness. During the initial evaluation, a clinician evaluates the parolee’s mental status and develops a treatment plan and goals for the parolee. The treatment plan must address the specific needs of the parolee and can include services such as medication manage- ment and individual or group therapies. Each clinic has a staff of psychiatrists, psychologists, and social workers who provide the treatment services for each parolee. Depending on the level of severity of the parolee’s disorder, goals can be as simple as the parolee taking all medications prescribed or coming to the next clinic meeting. To help ensure that it operates effectively and is meeting its goals, the program has provided for ongoing program monitoring as part of its continuum process. These efforts will 12 include quarterly compliance reviews that an external contractor will complete as well as evaluations that program management will complete. Services for Sex Offenders and Some Violent Criminals In addition to serving the mentally ill, the clinics also provide services to parolees with histories of serious sex offenses. In order to keep track of specific high-risk parolees, the California Penal Code, Section 290, requires parolees who have committed certain sex crimes, such as rape and incest, to register as sex offenders. According to the program, in 1994 the department began requiring all parolees with histories of sex offenses cov- ered under the provisions of Section 290 to attend clinics for more treatment. Additionally, the clinics treat parolees with histories of violent offenses when parole agents determine such services might be helpful. Services provided to these parolees normally consist of group or individual therapy to teach them to control their socially unacceptable tendencies. The depart- ment believes that it is prudent to have additional supervision of these high-risk parolees while they are in the community to ensure that they do not commit similar crimes during parole. Beginning in January 2001 the department received additional funding to implement a new Specialized Sex Offender Contain- ment Program (containment program). Using contractors, the containment program will focus on administering greater levels of supervision and recidivism prevention programs for parolees who are high-risk sex offenders. Participating parolees will not be required to attend the clinics unless they also have a mental health need. However, as of June 2001 the department advised us that it is still in the process of retaining contractors to provide the services under the containment program and, therefore, no parolees have yet been treated in this program. SCOPE AND METHODOLOGY The Joint Legislative Audit Committee (audit committee) requested that the Bureau of State Audits review and evaluate the goals of the program and determine whether the department has adopted reasonable strategies to achieve these goals. The audit committee also asked us to examine a sample from the clinics’ case files to determine whether referrals were adequately supported and based on established guidelines. Finally, the audit committee requested that we profile the status of parolees referred to the clinics during fiscal years 1998-99 to 2000-01. 13 To evaluate the goals of the program and determine whether the department had adopted reasonable strategies to achieve the goals, we reviewed laws, regulations, and policies related to the clinics, and we interviewed key clinic staff. We also compared data from clinic information systems to information from the department’s offender-based information system to identify characteristics of the parolees that the clinics serve and determine whether all mentally ill parolees were receiving clinic services. To determine whether referrals were adequately supported and based on established guidelines, we sampled 100 case files. Our sample was drawn only from those inmates paroled between October 2000 and March 2001, which corresponds to the imple- mentation of the program’s continuum process. We selected our sample of 100 case files based on the number of parolees served by each region. We reviewed each file to determine five things: (1) whether it included a prerelease assessment, (2) whether the initial appointment was within the required time frame, (3) whether individual goals were established for each parolee, (4) whether each case was an appropriate referral, and (5) whether information in the case file matched information in both the clinic and prison databases. In determining whether the clinic scheduled initial appointments promptly, we did not fault the clinics in those cases in which parolees did not show up for scheduled appointments, as long as the appointments were within the required time frames. To develop a profile of the status of parolees referred to the clinics within the past three years, we analyzed data contained in the program’s four regional database management systems. We present this information in the Appendix to this report. (cid:1) 14 AUDIT RESULTS THE DEPARTMENT HAS FAILED TO IDENTIFY ALL PAROLEES REQUIRING MENTAL HEALTH TREATMENT T he Department of Corrections (department) has failed to identify and treat through its Parole Outpatient Clinic Program (program) a large number of parolees who had been diagnosed as mentally ill when in prison. Before October 2000 the department relied on parole agents to refer parolees for evaluation and treatment. This process was not effective, and many mentally ill parolees received no treatment at the parole outpatient clinics (clinic). Although the program implemented a new process for inmates scheduled for parole on or after October 1, 2000, which has improved identification of parolees who need mental health services, it is still falling far short of its goal of serving all mentally ill parolees. The Former Process Failed to Identify and Serve Many Mentally Ill Parolees Between July 1998 and September 2000, nearly 24,000 inmates with diagnosed mental illnesses were paroled. However, based on our analysis of department records, it did not identify and thus did not provide services to nearly half of these parolees. During this period, the clinics relied on parole agents to identify Between July 1998 and refer parolees who they felt needed clinic services, including and September 2000, those who had been diagnosed as mentally ill. However, this 24,000 inmates with method was not always effective because most parole agents are diagnosed mental not mental health professionals and do not always have the illnesses were paroled— information in the parolees’ files needed to properly identify almost 12,000 of these mental illnesses. As a result, many mentally ill parolees received parolees did not receive no treatment services from the clinics once they were paroled. clinic services. As Table 1 shows, almost 12,000 of the nearly 24,000 parolees who were diagnosed as mentally ill while in prison did not receive clinic services when they were paroled. Especially troubling is the number of more severely mentally ill parolees— 1,130, or 35 percent—who were not served by the program, because they can pose a greater risk to public safety. 15 TABLE 1 The Parole Outpatient Clinics Served Only About Half of the Designated Parolees Between July 1998 and September 2000 Mental Health Parolees Released Parolees Seen at Parolees Not Seen at Severity While Between July 1998 the Clinics the Clinics in Prison and September 2000 Number Percent Number Percent Less severe 20,587 10,106 49.1 10,481 50.9 More severe 3,227 2,097 65.0 1,130 35.0 Totals 23,814 12,203 51.2 11,611 48.8 For our report, we categorized the mental health of parolees who were part of the Correctional Clinical Case Management System as less severe and those in either the Enhanced Outpatient Program, Crisis Beds, or the Department of Mental Health inpatient program as more severe. We describe the different mental health classifications in the Introduction of this report. Although an Improvement, the New Process Still Does Not Identify All Mentally Ill Parolees In July 2000 the program implemented the Mental Health Services Continuum Program (continuum process) to ensure that mental health treatments provided to inmates while they were incarcerated continued when they were paroled to the community. Although the continuum process has increased the proportion of mentally ill parolees that the clinics see, a signifi- cant number are still not being served. The clinics are supposed to see all mentally ill parolees, but they failed to schedule appointments for 39 percent of such inmates paroled during the period October 2000 through March 2001, including 2,166 less severe mentally ill parolees and 251 with more severe mental illnesses. As part of the continuum process, the program is supposed to identify mentally ill inmates scheduled for parole and assess the needs of those inmates 90 days before the inmates are paroled. Thus, when the process began in July 2000, social workers should have assessed inmates to be paroled on or after October 1, 2000. However, as evidenced by the large percentage of parolees who met the criteria but were not scheduled for appointments, the program did not supply all inmates with the mental health services they needed to successfully make the transition from the prisons to parole despite the new process. 16 The continuum process begins by identifying mentally ill inmates who will be paroled. To do this, the program receives a monthly listing from the department’s Offender Information Systems unit identifying mentally ill inmates with parole dates up to 120 days in the future. The program manager reviews the list, sorts it by region, and then forwards the list to each region. Social workers periodically visit the prisons to assess the needs of each inmate scheduled for parole. However, as presented in Table 2, we discovered that almost 39 percent of parolees diag- nosed as mentally ill who were paroled between October 2000 and March 2001 did not receive services at the parole clinics. TABLE 2 The Parole Outpatient Clinics Program Still Fails to Serve Many Designated Parolees Parolees Seen at Parolees Not Seen at Mental Health Severity Parolees Released Between the Clinics the Clinics While in Prison October 2000 and March 2001 Number Percent Number Percent Less severe 5,514 3,348 60.7 2,166 39.3 More severe 756 505 66.8 251 33.2 Totals 6,270 3,853 61.5 2,417 38.5 The program had opened case files in one of its four clinics for approximately 1,100 of the more than 2,400 mentally ill parolees not served by the clinics during this time. The program usually opens a case file when it receives either a social worker’s prerelease assessment or a parole agent’s referral. However, although the program had opened cases for these parolees, it had not scheduled appointments. Most of the 1,100 parolees (88 percent) were less severely mentally ill, although the 12 percent more severely ill parolees whom the program did not serve are significant when one considers the threat they may pose to the community. The program had no record in its four regional data management systems of the remaining 1,300 parol- ees, almost 10 percent of whom had more severe mental illnesses. To determine why the program had not scheduled appointments for the 1,100 cases with open files, we selected 10 cases at random and reviewed them. For 2 cases, the clinics had valid explanations: One parolee was returned to custody the day after being paroled, and in the other case, the clinic had scheduled an appointment four months after the inmate was paroled. However, the clinics failed to record these circumstances in the program’s current data management system. For the other 17 8 cases, the clinics did not have valid reasons for not scheduling appointments. Three of these 8 cases had been transferred between regions or clinicians, but the clinics had not completed the transfers. The clinics had not scheduled appointments for another 3 cases. Clinicians did not see the remaining two parolees because the clinics did not receive either prerelease assessments or parole agent referrals. A contributing factor to nearly 1,300 parolees having no records Originally, the continuum at the clinics is the failure of the continuum process to identify process did not include all mentally ill inmates. Originally, the specified population of some of the more severely the continuum process included only inmates in the Correc- mentally ill parolees tional Clinical Case Management System and the Enhanced who may pose more Outpatient Program while in prison. It did not include the significant risks to the inmates receiving inpatient Department of Mental Health public. However, the treatment or participating in the Crisis Beds program. Although program intends to the number of parolees in the latter two programs is relatively amend its process to small compared to all mentally ill parolees, their exclusion is include these parolees. troubling because they are more severely mentally ill and there- fore may pose a more significant risk to the public. This exclusion is in direct contrast to the program’s goals of reducing the symptoms of mental illness, lowering the rate of recidivism, and improving public safety. The program has advised us that, to fulfill its mission, it will amend its process to ensure that inmates in these categories are included in the continuum process. The fact that in a six-month period more than one-third of the targeted population was not served indicates that the program is still not adequately managing the transition of cases from the prisons to parole. When we asked what it believed were the causes of these parolees not being identified, the program stated that staffing shortages during this time often prevented social workers from completing prerelease assessments. Because the clinics use these assessments to establish cases in their systems, if they are not completed, the cases may not be identified. The program also manually distributes the list it receives each month from the department’s Offender Information Systems unit to notify its clinics of inmates who might soon be paroling to their regions. However, the program also stated that it has discovered that these listings are not always complete. The program believes that some of the identification problems that are occurring will be eliminated once the recently hired contractors assume the task of completing prerelease assessments. However, 18 as we discuss later, the use of contractors does not resolve the problem related to the incomplete lists the program receives from the department’s Offender Information Systems unit. THE CLINICS DO NOT ALWAYS PERFORM NEEDED PRERELEASE ASSESSMENTS OR PROVIDE TIMELY SERVICES As part of the continuum process, the department established guidelines requiring all inmates diagnosed with mental illness to be assessed before leaving prison on parole and requiring the clinics to see the newly released parolees within specified time frames. The department established these guidelines to ensure that parolees needing mental health services continued to receive timely services when paroled to the community. Of the 100 cases we reviewed that were opened after implementation of the continuum process, all met the criteria established by the program for eligibility for services. Nonetheless, as indicated in Table 3, social workers did not perform prerelease assessments for 38 of the 83 mentally ill parolees whose cases we reviewed; 5 of these 38 were severely mentally ill parolees. Further, clini- Social workers did not cians saw 45 of these 83 parolees outside prescribed time frames; prepare prerelease 6 of these 45 were severely mentally ill. A study conducted by assessments for 38 of the the department in 1999 concluded that parolees who participate 83 mentally ill parolees in services such as the clinics’ are more likely to succeed in their whose cases we reviewed; transition to the community than those who do not participate. 5 of these 38 were Therefore, when the program fails to complete a prerelease severely mentally ill. assessment or see a parolee within the required time frame, it exposes the public to increased risks. The remaining 17 cases we reviewed involved parolees with problems other than mental illness, for which the program has not established prerelease assessment or timeliness requirements. Failure to Conduct Prerelease Assessments or to Follow Program Guidelines Can Delay Treatment for Mentally Ill Parolees One of the key elements of the continuum process is an assess- ment of mentally ill inmates conducted before their parole to better identify and continue their mental health treatment. However, we discovered that out of the 83 case files of mentally ill parolees that we sampled, 38 (more than 45 percent) did not receive such an assessment. According to the department’s guidelines, an assessment is required for mentally ill inmates paroled on or after October 1, 2000. Prerelease assessment is 19 TABLE 3 The Department Has Failed to Properly Assess and Promptly Serve Mentally Ill Parolees Parolees Not Scheduled for Parolees Not Assessed Appointments Within Required Before Parole Time Frames Parolee Type Sample Number Percent of Sample Number Percent of Sample Less severely mentally ill 72 33 45.8 39 54.2 More severely mentally ill 11 5 45.5 6 54.5 Total mentally ill 83 38 45.8 45 54.2 Other (mostly sex offenders)* 17 Total parolee cases reviewed 100 38 45 *The program has not established requirements for prerelease assessments or time frames for parolees with problems other than mental illness. necessary to determine the mental health needs and level of services the inmate will require once paroled. The assessment also alerts the clinic that an inmate will soon be paroled to that region, giving the clinic an opportunity to obtain any additional information it requires from the parole agent before the initial meeting with the parolee. Between July 2000 and May 2001, the clinics relied on their own social workers to perform assessments. According to the department’s mental health administrator, the program was unable to adequately fill its social worker positions during this period. Consequently, assessments were not always completed as program guidelines required. The program has contracted with the health departments of San Diego and Kern counties to perform these prerelease assessments beginning in May 2001. According to the program, using contractors should result in more assessments being completed. When they did complete prerelease assessments, social workers did not always follow program guidelines. In some assessments, a social worker added a box indicating that the parolee should be seen within 30 days, although program guidelines allow for a maximum of 10 business days. In another case, a social worker filled out an assessment after the inmate was already paroled. Failure to perform prerelease assessments or follow guidelines hinders the clinic’s ability to see the parolees within required time frames because, by the time the clinic becomes aware of them, the parolees may have already been on parole for several days. 20 The department is confident that many of the issues of missing or inaccurate assessments that we discovered should be addressed by the contractors now conducting assessments, because the contractors will have the flexibility to increase the number of social workers to complete the assessments as the designated population increases. Additionally, the program said that all social workers hired by the contractors have laptop computers with a standardized assessment form to ensure that the information collected is consistent with program standards. Prompt Initial Appointments With Parolees Are Critical to Successful Transition to the Community To ensure that parolees receive a continuum of care between imprisonment and release to the community, the program established a policy requiring clinics to see parolees with more severe mental illnesses within 3 business days after their release and those with less severe mental illnesses within 10 business days. However, of the 83 cases of mentally ill parolees we reviewed, 45 were not seen within the required Of the 83 cases we time frames. In 28 cases, parolees were seen within 30 days after reviewed, 45 parolees parole, but for the other 17, initial appointments did not occur were not seen within the until between 32 to 119 business days after parole. In at least required time frames—17 6 of the 45 cases, the clinic did not set up initial appointments; of whom were seen from rather, parolees met with clinicians on a drop-in basis for 32 to 119 business days reasons such as the parolee being out of medicine. The program, after parole. of course, cannot ensure that parolees show up for their appointments. Therefore, for our testing of this requirement, we did not fault the clinics when parolees did not show up for initial appointments, as happened in 8 of the 83 cases we reviewed, as long as the clinics had scheduled appointments within the required time frame. The purpose of these time frames is for clinicians to see parolees as soon as possible to ease their transition to the community and increase their opportunity for success. A study conducted by the department in 1999 on the effects of the program’s services on the success of parolees concluded that mentally ill parolees have a greater chance of succeeding on parole when they receive mental health treatment soon after being paroled. Therefore, by beginning treatment for mentally ill parolees promptly, the clinics can help the parolees succeed during their parole terms. 21 A NEW DATA MANAGEMENT SYSTEM MAY ADDRESS SOME PROGRAM WEAKNESSES BUT COULD BE MORE EFFECTIVE IF LINKED TO OTHER DEPARTMENT SYSTEMS Another piece of the program’s continuum process is a new computerized data management system that, when imple- mented, will improve communication among the four regional clinics as well as with the contractors performing prerelease assessments in the State’s 33 prisons. Additionally, the system will standardize some practices within the program and allow it to better identify mentally ill parolees. If the system works as described by the program, it will address many of the weak- nesses we identified in this report. However, we believe the program could make its data management system even more effective if it linked the system to key department systems, giving the program quick access to vital information. Because the program had not yet implemented the new data manage- ment system when we completed our fieldwork, we were not able to assess its effectiveness in addressing the program’s problems. The program should complete further assessment of the effectiveness of the system at least one year after implementing it. The Clinics Will Use the New System to Create a Comprehensive Mental Health Record for Each Parolee The program’s new data management system is designed to provide a complete mental health record for each parolee who receives services from the clinics. Adapted from a proven system The program’s new data currently used by another program within the department, this management system is new data management system is based on commonly used based on a proven system database management software. Because the program is using currently used by another funds from its continuum process to finance the development of program within the this system, it is planning to use the system only to track its department. mentally ill parolees. The clinics will continue to record the treatments they provide to all other parolees they serve in their existing system. Under the new system, information on mentally ill inmates will be entered into the system by contracted social workers perform- ing mental health assessments 90 days before the inmates’ parole dates. The system will collect a wide range of information on each inmate, such as employment and drug use histories, and mental health assessments. The social worker will also establish, before the inmate is paroled, an initial appointment at 22 the appropriate clinic. All information will be input into a standardized evaluation form, so consistent information should be obtained on all inmates. When the social worker enters this information into the computer, it will be available to all four clinics as well as the headquarters office in Sacramento. Based on the region to which the inmate is being paroled, the system will identify the appropriate clinic and notify it that a parolee is coming to that region. The system will provide the region with the parolee’s identification number and parole date as well as the date of the scheduled initial appointment. Following the initial appointment, the clinic will establish a The program believes the treatment plan, including goals for the parolee to work toward new system will improve in treatment, and enter it in the system. As treatment progresses, the identification of clinicians will record information, such as notes on patient mentally ill parolees and history and medications prescribed, in the various screens that provide a standard the system includes. This becomes the record of the mental method of collecting health services the parolee received while in the program. pertinent information Clinicians will update the system throughout the time the about them. parolee receives services at the clinic. When the case is closed, the clinician will enter a code into the system to indicate the reason for closure. The Program Believes the New System Will Address Some of Its Weaknesses According to the program, its new data management system, along with full implementation of its continuum process, will address many of the weaknesses we have identified in this report. The program believes that the system will improve identification of mentally ill inmates coming out of prison on parole by providing a systematic method of assigning the regions to which the inmates will be paroled and then handing off the cases to those regions. The system will also provide a standardized method for collecting similar information on inmates scheduled for parole. Additionally, because it will include the prerelease assessments being completed by its contractors, the system will allow the program to better monitor its contractors. Further, because the social worker completing prerelease assessments will also use the system to set up initial clinic appointments, the clinics should be able to see patients more promptly. On a larger scale, the system should enable the program to better assess its entire operations because it will be able to consolidate the activities of all four regions, something the program’s current system cannot do. 23 Linking the New System to Other Department Systems Could Improve Program Efficiency Although the data management system that the program is implementing should greatly improve its effectiveness, by linking this system to the department’s systems, which contain key information, its effectiveness could be enhanced. The department’s Offender Information Systems unit identifies mentally ill inmates due for parole in a monthly report. This unit queries two different department computer systems: one to obtain the mental health services the parolee received while in prison and the other to obtain the parole date and other infor- mation. Using this information, the unit develops a list of parolees who will be paroled in the next 120 days and sends it to the program, which provides the list to the social workers who perform the prerelease assessments. As discussed earlier, these social workers enter the information from their assessments into the system. This transfer of information to the program is the same method in use since July 2000, when the program implemented its continuum process. However, this information-gathering method has not consistently provided complete information to the program. As we discussed earlier, under this process, the program failed to identify and serve almost 39 percent of men- tally ill inmates beginning parole terms between October 2000 and March 2001. At least part of this was due to problems identifying all mentally ill inmates about to be paroled. Accord- ing to the program, the computer program developed to extract the information from the department’s systems did not include all specified mentally ill inmates, so the lists the unit produced for the clinics were incomplete. Although the program advised us that it sometimes is able to obtain more complete information from individual institutions, it cannot rely on that information always being available. Therefore, if the program continues to use this method, it may continue to fail to identify parolees in need of its services or at least delay the delivery of services to the parolee. If the program automated this exchange of information, it could reduce the chances of its failing to identify inmates and, therefore, not providing them with needed services. As part of our fieldwork, we requested and received a download of data from the systems that contain the information the program would need. Although these are older technology systems, the department was able to convert the data into a file format 24 similar to what the program would need. Because this informa- tion is available, the program should be able to create a bridge between its data management system and the department’s systems using its existing computer software. Even if it is unable to establish an automated bridge between these systems, the program could receive the information on compact disks, as we did, and load this information into its new system on a daily or weekly basis. Also, as the department investigates options to replace its aging computer systems, it should consider the needs of the program. This information would also help the program by providing more timely and up-to-date information to the clinics as well as If linked to other key to the contractors performing the prerelease assessments. department systems, the Currently, the program updates information on inmates being new system would paroled only once a month. If it establishes a direct link to these provide more timely and department systems, the program could update information up-to-date information much more often, even daily. Because the new system shares on parole dates and information with all four regions as well as with the contractors, mental health status of the latest information would then be available to each of them. inmates nearing parole. The program could use this information to evaluate future workload or to ensure that all inmates who will be paroled have been assessed before their parole dates. Also, because the program will be able to see in the system when and whether assessments are completed, the new system could also assist the program in monitoring the contractors who perform prerelease assessments. The New System Has Not Yet Been Implemented Although the program has developed the new data management system and done some user testing to ensure that it will function as planned, it has not yet been fully implemented. On June 29, 2001, the program received final approval from the Department of Information Technology to complete the procurement of the system. The program advised us that it has already negotiated the purchase of the needed hardware, and the system should be operating by the end of August 2001. Additionally, the program has purchased the personal computers to establish the computer network. Because the program had not yet fully implemented the complete system before the end of our fieldwork, we were not able to assess its effectiveness in addressing the weaknesses we have identified in our report. Once the program has implemented the system and has operated it for a year, the department should assess the system’s effectiveness. 25 THE CLINICS PROVIDE SERVICES TO MANY PAROLEES WITH PROBLEMS OTHER THAN MENTAL ILLNESS Although the program primarily serves parolees who are mentally ill, it also provides treatment to parolees with other problems, particularly sex offenders and violent criminals. Further, the clinics also serve mentally ill parolees who were not diagnosed as such during their incarceration. Whom the program serves and its exact responsibilities are not defined in statute; therefore, the department has issued regulations and policies to guide the program in fulfilling its mission success- fully. However, the department has not ensured that the program has the funding and resources needed to do this. One-Third of the Parolees the Clinics Served Had Not Been Diagnosed With Mental Illness The clinics began as a means to increase public safety by treating and monitoring parolees with mental health needs. The depart- ment recognizes that many parolees have unique treatment needs that community mental health programs might not be The population served by able to meet and that those needs could affect public safety. This the clinics reflects the is reflected in the population that the clinics serve. Figure 4 department’s recognition indicates that 66 percent of the parolees for whom the clinic of parolees’ unique opened cases between October 2000 and March 2001 were treatment needs. diagnosed as mentally ill at the time of their parole. (See the Appendix for additional information about these parolees.) The remaining 34 percent, a significant proportion of parolees the program serves, were not diagnosed with mental illness. In addition to providing services to mentally ill parolees, the clinics have the responsibility of treating parolees who have prior serious sex offenses or have committed certain violent crimes. Indeed, nearly 3,100 parolees who the clinics served did not have a mental health classification when they were paroled. Some of these parolees may be sex offenders or violent offenders with mental disorders who, as a condition of parole, must attend the clinics. From the clinics’ four regional databases, we were able to determine that almost 1,500 of these 3,100 were sex offenders. We were not able to determine why the remaining 1,600 cases were referred to the clinics because the clinics did not collect such data. 26 FIGURE 4 Cases Opened by the Program Between October 2000 and March 2001 Other 1,602 (18%) Sex Offenders 1,472 (16%) Mentally Ill 6,080 (66%) Source: Regional Parole Clinic databases. Although we cannot determine why these cases were referred to the clinics, a look at the commitment offense, the offense that caused the person to be incarcerated, provides some insight. Table 4 lists the commitment offenses for the 1,600 parolees served by the program who were neither mentally ill nor serious sex offenders. TABLE 4 Commitment Offenses for Parolees Without a Diagnosed Mental Illness or Serious Sex Offense Record Who Were Served by the Program Commitment Offense Number of Parolees Percent of Total Violent offenses 433 27 Property offenses 406 25 Drug charges 380 24 Sex offenses for which the parolee is not required by law to register as a sex offender 160 10 Threats and stalking 53 3 Weapons 48 3 Unknown or not indicated 74 5 Other 48 3 Totals 1,602 100 27 This analysis shows that 27 percent of these parolees were violent offenders. Many of the remaining commitment offenses do not appear to fit into the categories of parolees that the program serves. In these cases, the parolees likely had records of sex offenses or violent crimes, or parole agents had concerns about the parolees’ mental health even though no illness was diagnosed in prison. Table 4 shows only the violations that resulted in the parolees’ latest incarcerations; previous violations do not appear. If the Department Refers to the Clinics More Parolees Who Are Not Mentally Ill, the Clinics May Lack the Resources Required to Serve Them Adequately For the clinics to adequately serve parolees who are not mentally ill, the department needs to ensure that the program has the appropriate resources and funding. As shown in Table 5, the clinics did not treat 38 percent of parolees with sex offenses who were paroled between July 1998 and September 2000. This statistic increased to 66 percent between October 2000 and March 2001, after the continuum process was implemented, even though the responsibility of identifying and referring these parolees remained with parole agents. Although a new, sepa- rately funded program instituted by the department will treat certain high-risk sex offenders, the clinics still must serve parolees not covered under the new program. TABLE 5 The Parole Outpatient Clinic Program Fails to Serve Many Paroled Sex Offenders Parolees Seen at Parolees Not Seen at the Clinics the Clinics Sex Offender Percent Percent Period Parolees* Number of Total Number of Total July 1998 to September 2000 10,376 6,436 62.0 3,940 38.0 October 2000 to March 2001 3,128 1,051 33.6 2,077 66.4 * We did not include in these figures sex offenders who have also been diagnosed as mentally ill because they would already be included in the program. 28 Because the department has determined that sex offenders as well as certain violent criminals should receive clinic services, the program needs a better identification process so that it can ensure it is treating all parolees who need its services. One option would be for the clinics to implement a continuum process for other populations as it has for the mentally ill popu- lation. This would provide a systematic method for identifying inmates with these characteristics and including them in the clinics once paroled. However, this added process could significantly increase the clinics’ workload and could require additional resources. As shown on Table 5, 66 percent of the sex offender population did not receive clinic services between October 2000 and March 2001. Based on this analysis, if the program had imple- mented an effective identification process and treated all these parolees, it would have increased the number of sex offenders coming into its program by nearly 200 percent. Therefore, if the department continues to require the clinics to serve these parolees, it should ensure that they have the processes and resources they need to fulfill that responsibility. In fiscal year 1999-2000 the program received funding beyond its budget of $8.7 million: A $6 million appropriation to fund We could not determine the continuum process designated to be used to serve only how much the program mentally ill parolees. However, according to the program, even spends on parolees who though the State has agreed to increase this funding as the are not mentally ill number of mentally ill parolees increase, the $6 million aug- because it does not track mentation does not fully fund its services to the mentally ill. the amount of resources Because the clinics must also treat other populations, the it uses to serve different program’s original $8.7 million budget, which is not restricted to types of parolees. funding services to only the mentally ill, may be strained. The $6 million was meant to enhance the services that the program already provided to mentally ill parolees. Therefore, the funding available to the program to serve parolees who are not mentally ill is only a portion of the $8.7 million. We could not determine exactly how much it spends on these parolees because, as we discuss later in this report, the program does not track the amount of resources it uses to serve different types of parolees. 29 THE PROGRAM NEEDS TO TAKE ADDITIONAL ACTIONS TO MANAGE EXPECTED CASELOAD INCREASES Changes both within and outside the program are likely to increase the demand on its services. If its continuum process achieves its goals, the program could experience a significant increase in the number of mentally ill parolees it serves. Addi- tionally, state voters recently passed a proposition that could lengthen the amount of time some parolees stay in the program, which over time will increase caseloads. We attempted to assess the current caseloads of the clinicians. However, even though we were able to obtain average caseloads for its clinicians, because the program has not established caseload standards for its clinicians and we were not able to find mental health organiza- tions that delivered services in a similar way, we could not adequately assess the size of the clinicians’ current caseloads. We believe this highlights the need for the program to develop caseload standards for its clinicians so that it can assess the adequacy of its resources. Moreover, the impending increase in caseloads is another reason for the program to implement its new data management system as soon as possible. Changes That Have Occurred Are Likely to Increase the Clinics’ Caseloads Recent changes inside and outside the program will likely increase the clinics’ caseloads. As Table 2 on page 17 indicates, the program fails to serve many mentally ill parolees. In re- sponse, the program has implemented its continuum process to If the program is better identify and serve parolees who are diagnosed with successful in identifying mental illness. If the program is successful in achieving this goal, and providing services it will significantly increase the number of parolees it serves. For to mentally ill parolees example, if the program is successful in identifying and provid- it failed to serve in the ing services to all the mentally ill parolees it has failed to serve past, its caseload will in the past, it will increase its caseload by more than 2,400 cases. increase by more than 2,400 cases. Other changes are also likely to affect the clinics’ caseloads. In November 2000 voters approved Proposition 36, which changed the way the State treats drug offenders. In general, the proposition requires the State to provide treatment instead of incarceration to individuals, including parolees, convicted of using or possessing illegal drugs when no violent acts were involved. Although this measure will not directly affect the program, it may have indirect effects. There is no direct effect because the clinics do not treat parolees with drug problems unless they have been diagnosed with mental illness. The 30 department provides separate programs for parolees with drug- dependency treatment needs. It will likely indirectly affect the program because parolees in clinic treatment, who would other- wise have been returned to prison when convicted of a drug violation, may now remain on parole. Over time, this will increase the number of parolees receiving services from the clinics because some parolees will be receiving services longer while new parolees continue to enter the program. Although we could not determine the actual effect that the passage of Proposition 36 will have on the program, the impact could be significant because so many of the parolees in the program have prior drug convictions. We looked at all active cases in the program on March 31, 2001, and determined that nearly 24 percent included prior drug offenses. Given that a 1991 department study concluded that 94 percent of mentally ill parolees return to custody within 24 months of parole, the number of mentally ill drug offenders who would otherwise return to prison could be significant. Although a prior drug offense does not necessarily mean the parolee will have a future violation for drug use or possession, we think this indicator has value for estimating the impact that Proposition 36 could have on the program. One recent change will reduce the number of patients the clinics will have to serve. In fiscal year 2000-01 the department began Once its contractors the Specialized Sex Offender Containment Program (contain- begin providing services, ment program) to treat high-risk sex offenders. The containment a new department program, operated by contractors, will not affect mentally ill program for sex offenders parolees but will provide focused services to parolees with prior could reduce the number serious sex offenses who have been identified as being at high of parolees served by the risk for recidivism. The department advised us that it will place clinics by 2,900. up to 2,900 parolees in this containment program but that, as of June 2001, its contractors had not yet begun providing services. When the contractors do begin providing services, the program will cease treating parolees in the containment program unless they also have a mental illness. Although this will reduce the number of sex offenders that the program would otherwise have treated, many will remain. Because the program estimated that approximately 9,000 sex offenders need treatment, even when the containment program begins, the program will still have more than 6,000 sex offenders who will need treatment. 31 Current Caseloads Are Difficult to Assess Because the Program Has Not Developed Standards To assess the current caseloads of program clinicians, we obtained caseload reports from each of the four regions as of June 30, 2001. We grouped the clinicians into three groups: psychiatrists, psychologists, and social workers. From the reports, we determined that the average caseload for each category was as follows: Psychiatrists 170 Psychologists 128 Social workers 41 We then attempted to identify a benchmark by which we could assess whether these levels were reasonable. We first asked the program whether it has established a caseload standard so that it could assess its clinicians’ caseloads; we learned that the program has not established such standards. We then tried to identify comparable agencies to determine whether they had caseload standards. Most states that we contacted either did not have a similar function or contracted the services out. We then surveyed the 10 largest county mental health departments within the State to develop comparable caseloads. However, we found that these organizations are not truly comparable because the counties often deliver services in a different way than the clinics do, which may affect the comparability of caseload numbers. Further, the program explained that at the time of our analysis, many of its social workers were not providing treat- ment to parolees in the clinics. Rather, they were completing prerelease assessments in the prisons as part of the program’s continuum process. Additionally, the program told us that often its psychologists were performing services that the social workers would otherwise do. Therefore, we could not make meaningful comparisons. However, our review does illustrate the need for the program to develop caseload standards so that it can adequately monitor and assess the caseloads of its clinicians. Further, the program could use standards to assess and justify the need for changes to its staffing as its workload changes. 32 The Program Should Use Its New System to Identify Treatment Costs The program is financed by funds appropriated to the depart- ment for casework services. Figure 5 presents the program’s budgeted funding for the last five fiscal years. FIGURE 5 Budgeted Program Funding for the Last Five Fiscal Years (In Millions) $25 22.4 20 15 15.5 13.2 10 8.7 8.8 5 0 1997-98 1998-99 1999-2000 2000-01 2001-02 Source: Department of Corrections. The increased funding the program received in fiscal year 1999-2000 was for its continuum process. As discussed earlier, the program received an additional $6 million to enhance the services it provides to mentally ill parolees. The treatment the clinics provide to parolees with problems other than mental illness is paid for out of its existing funds. However, the program’s current data management system is not able to identify the level of effort—and related expense—that it incurs in treating the various types of parolees in its program. For example, a clinician may treat several different types of parolees: the mentally ill, serious sex offenders, and violent criminals. Because the program has not tracked the time clinicians spend providing services, it is not able to track how much of its resources it uses on the various types of parolees receiving treatment. 33 Although its current system does not collect this information, The program has an the program has an opportunity to use its new data manage- opportunity to use its ment system to begin collecting the data it needs to determine new system to begin to the costs of services it provides to the different types of parolees. identify its cost of To accomplish this, the program would have to establish a providing services to the unique designator for each type of parolee it serves, record the different types of parolees amount of time that clinicians spend with different types of it serves. parolees, and include all of its parolees on the system. Currently, the program is planning to use its new data management system to track mentally ill parolees; other types of parolees will continue to be tracked on its existing system. However, if the program established an indicator in its new system that identi- fied the parolee type, it could begin to determine the costs of providing treatment to the various types of parolees. The program could also use this information as it evaluates how effectively it is meeting its goals of reducing the symptoms of mental illness among parolees, lowering the rate of recidivism, and improving public safety. To adequately assess its effective- ness, the program should also consider its costs in pursuing these goals. The program could also use the information acquired through its new system to justify additional resources it may need to serve all the types of parolees it has been asked to treat. RECOMMENDATIONS To help ensure that the public is adequately protected from parolees diagnosed as mentally ill while in prison, the program should fully implement its new data management system. After implementing the system, the program should do the following: • Develop a process to monitor its contractors to ensure that they complete prerelease assessments on all mentally ill inmates scheduled for parole. • Ensure that clinics see mentally ill parolees within required time frames. • Identify parolees whom it failed to identify as needing services and ensure that they receive the treatment they need. 34 To more effectively identify all the parolees the program will serve, including those with problems other than mental illness, the program should link its new system to other department computer systems containing the information needed to do so. To better identify its costs of treating parolees and to better justify additional resources it may require, the program should track the amount of time and resources it spends treating the different types of parolees. To identify and serve parolees with problems other than mental illness, the department should ensure that the program has adequate processes and resources to do so. To ensure that the public is protected from the more severely mentally ill parolees, the program should implement its plan to include in its continuum process those parolees designated while in prison to have been in the Department of Mental Health inpatient and Crisis Beds programs. To appropriately assess its clinicians’ workloads and evaluate the need for additional resources, the program should develop caseload standards for its clinicians. To determine the progress the program has made in identifying and serving mentally ill and other parolees, the department should reassess the program one year after implementing the new system. The department should submit the completed assessment to the Youth and Adult Correctional Agency. 35 We conducted this review under the authority vested in the California State Auditor by Section 8543 et seq. of the California Government Code and according to generally accepted government auditing standards. We limited our review to those areas specified in the audit scope section of this report. Respectfully submitted, ELAINE M. HOWLE State Auditor Date: August 29, 2001 Staff: Nancy C. Woodward, CPA, Audit Principal David E. Biggs, CPA Joe Azevedo Laneia Grindle Roberta Kennedy Kenneth Louie 36 APPENDIX Caseload Information for the Parole Outpatient Clinic Program T he Joint Legislative Audit Committee asked us to provide information on the parolees being referred to the Parole Outpatient Clinic Program (program) during fiscal years 1998-99 to 2000-01. Specifically, it asked us to provide informa- tion on the number of cases that are active, successfully closed, or referred to other mental health providers, as well as the number of parolees returned to jail or prison. We were not able to develop information on cases referred to other mental health providers because the program does not keep data on referrals. Additionally, we could not determine the number of cases the clinics had successfully closed because the program also does not gather that information. However, we do present cases that were closed and the corresponding reasons recorded by the clinics. Table 8 shows the parolee cases that were closed because the parolee returned to prison (parole revoked) or had been incarcer- ated in a local jail. However, because the program does not track the parolees it serves after they have been discharged from the program, this table may not present a complete picture of parolees who returned to jail or prison. For example, if parolees whose cases were closed because they either met their treatment goals or were discharged from parole subsequently committed other offenses resulting in incarceration, the table would not indicate this. Although these and other data in Table 8 might lead one to conclude that the program is not very successful in achieving its goals of reducing recidivism, that may not necessarily be the case. The Department of Corrections (department) reports that the recidivism rate for all prisoners is almost 68 percent. Additionally, a study conducted by the department in 1999 concluded that mentally ill parolees were likely to have increased rates of recidivism. Therefore, the recidivism rates presented in this table are not inconsistent with those of the general parolee population. 37 TABLE 6 Caseload as of March 31, 2001 Less Severely More Severely Serious Active Cases Mentally Ill Mentally Ill Sex Offenders* Other† Region I 2,653 1,112 162 952 427 Region II 3,722‡ 1,406 326 1,186 804 Region III 3,371 1,586 293 694 798 Region IV 3,810‡ 1,740 321 706 1,043 Totals 13,556 5,844 1,102 3,538 3,072 * Not identified as mentally ill when paroled. † Not identified as mentally ill or serious sex offender when paroled. ‡ These regions included in their caseloads prerelease assessments completed by their social workers for inmates being released to other regions as part of the new continuum process. TABLE 7 New Cases Opened During Fiscal Years 1998-99 to 2000-01 2000-01 2000-01 Percent Percent (7/1/00 to Percent (10/1/00 to Percent Region 1998-99 of Total 1999-2000 of Total 9/30/00)* of Total 3/31/01)* of Total Region I 2,517 19 2,333 18 726 16 1,857 18 Region II 2,775 22 2,744 21 1,287† 28 2,089† 20 Region III 3,901 30 3,875 30 890 20 2,489 24 Region IV 3,709 29 3,999 31 1,642† 36 3,912† 38 Totals 12,902 100 12,951 100 4,545 100 10,347 100 * Fiscal year 2000-01 is split in order to show the number of cases opened since the parole clinic program began its Mental Health Services Continuum Program in October 2000. † Regions II and IV included in their cases opened the prerelease assessments that their social workers completed for parolees being released to other regions as part of the continuum process. As a result, the numbers for these regions are inflated and should not be compared to Regions I and III. 38 TABLE 8 Cases Closed During Fiscal Years 1998-99 to 2000-01 2000-01 Closure Percent Percent (7/1/00 Percent Reason 1998-99 of Total 1999-2000 of Total to 3/31/01) of Total Treatment goals met or evaluation only 1,813 16 1,575 13 1,605 14 Discharged from parole 1,440 12 1,706 14 1,451 13 Parole revoked 3,889 34 3,942 32 3,154 28 Local jail 1,201 10 1,150 9 845 8 Did not keep appointments 1,002 9 1,264 10 1,063 10 Unable to locate (parolee at large) 882 7 847 7 717 6 Incomplete referral 646 6 996 8 385 4 Not amenable to treatment 224 2 122 1 114 1 Transfer to another region 207 2 180 2 1,358* 12 Other 233 2 466 4 429 4 Totals 11,537 100 12,248 100 11,121 100 * Number includes prerelease assessments that were completed by parole clinic social workers as part of the new Mental Health Services Continuum Program. After assessments were completed, the cases were transferred to the regional clinic where the parolee was being paroled. 39 Blank page inserted for reproduction purposes only. 40 Agency’s comments provided as text only. Department of Corrections P.O. Box 942883 Sacramento, CA 94283-0001 August 10, 2001 Elaine M. Howle, State Auditor* 555 Capitol Mall, Suite 3000 Sacramento, CA 95814 Dear Ms. Howle: Enclosed you will find our response to the Audit Report entitled, "Department of Corrections: Though Improving, the Department Still Does Not Identify and Serve All Parolees Needing Outpatient Clinic Program Services, but Increased Caseloads Might Strain Clinic Resources," requested by the Joint Legislative Audit Committee. Also enclosed for inclusion in our response are flow charts and a list of departmental acronyms used in the flow charts. If you have any questions, please contact Millicent Gomes, Health Administrator at (916) 327-4612. Sincerely, (Signed by: Teresa Rocha) TERESA ROCHA Director (A) Department of Corrections Enclosures *California State Auditor’s comments appear on page 51. 41 INTRODUCTION MENTAL HEALTH SERVICES CONTINUUM PROGRAM OVERVIEW In July 2000, the California Department of Corrections (CDC), Parole and Community Services Division (P&CSD) began implementation of the Mental Health Services Continuum Program (MHSCP). The new program is an expansion and enhancement of mental health treatment services delivered by P&CSD's existing Parole Outpatient Clinics (POC) throughout the state. The mission of the MHSCP is to reduce the symptoms of mental illness among parolees by providing timely, cost-effective mental health services that optimizes their level of individual functioning in the community thereby reducing recidivism and improving public safety. The MHSCP provides pre-release needs assessment, benefits eligibility and application assistance to paroling mentally ill inmates. The MHSCP increases the prospect of the parolee's successful reintegration into the community by providing expanded and enhanced post-release mental health treatment combined with improved continuity of care from the institution's mental health service delivery system. TARGET POPULATION The MHSCP target population consists of parolees who were receiving mental health treatment in the institutions under the Mental Health Services Delivery System prior to release to parole. The criteria for admission to both the institution's and parole's mental health treatment programs is a diagnosis of one or more of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) psychiatric disorders. The two mental health designations used to determine the level of treatment need for both inmates/parolees are Correctional Clinical Case Management System (CCCMS) and Enhanced Outpatient Program (EOP). The EOP parolee is lower functioning than the CCCMS due to an acute onset or significant deterioration of a serious mental disorder characterized by a definitive impairment of reality testing and/or judgment which creates dysfunctional or disruptive social interaction or severe impairment of activities of daily living. The MHSCP target population also consists of those parolees designated while in prison to have been in a Mental Health Crisis Bed and those releasing from any Department of Mental Health facility. The P&CSD is currently amending its pre-release contracts to include these populations in the MHSCP. PROGRAM COMPONENTS Contract providers, at ninety (90) days and thirty (30) days prior to the inmate's release, complete comprehensive computerized pre-release needs assessments for EOP and CCCMS designated inmates. The MHSCP's pre-release component is modeled after the pre-release services provided through P&CSD's existing Transitional Case Management Program (TCMP) for inmates/parolees who have been diagnosed with Human Immunodeficiency Virus (HIV) and/or Acquired Immunodeficiency Syndrome (AIDS). The TCMP-HIV/AIDS has been in existence since 1993, and has proven to be an effective case model. 42 Because of their outstanding performance with the TCMP-HIV/AIDS, Kern County Department of Public Health and San Diego County Department of Health Services were selected to provide the MHSCP's pre-release services component. Both contractors bring to the MHSCP a wealth of experience with pre-release services including comprehensive assessments, service plans and benefit eligibility assistance. This process is known as the Transitional Case Management Program-Mentally Ill (TCMP-MI). These comprehensive pre-release assessments are used to provide services that include: benefits application assistance; coordination of transportation and housing; provision of medications upon release; assurance of arrival at first parole agent and MHSCP appointments; and development of clinical treatment plans. This allows both the MHSCP clinical team (psychiatrist, psychologist and psychiatric social worker) and the parole agent to prepare for their initial meeting with the inmate upon release to parole. Upon release, a variety of clinical therapies are combined and tailored to meet the individual needs of each MHSCP parolee. An initial clinical evaluation including, in most cases, psychological testing is completed and a parolee criminal history is reviewed prior to the mental health treatment plan development. Clinics offer individual and group psychotherapies specific to parolee needs such as anger management, social integration, community re-entry, and release preparation. The frequency and duration of therapy is determined by the primary clinician. On-going clinical treatment in both individual and group therapy settings is combined with Interdisciplinary Treatment Team (IDTT) meetings to effectively monitor the treatment of EOP parolees. The IDTT evaluates treatment progress and updates the treatment and service plans for the parolee. The IDTT consists of the parolee's MHSCP clinical team, parole agent, and appropriate community resource providers. One purpose of the IDTT is to enhance the interaction between the MHSCP clinicians and the parole agent by assisting the parole agent with these high service cases (see attached flow charts). When the parolee is within 120 days of discharge from parole, MHSCP discharge planning is provided. It includes review of the parolee's file and treatment plan, assessment of parolees functioning with activities of daily living, completion of a written release plan which includes any community treatment and transitional resources, and facilitation of the actual transition to the community by MHSCP psychiatric social workers. There is also an evaluation component built into the MHSCP. P&CSD is currently developing the Scope of Services for an independent evaluation of the program. The evaluation contract will be for a period of three years. The contractor will submit progress reports to the Department throughout the contact period. The Department, in turn, will submit the program evaluation to the Youth and Adult Correctional Agency when com- pleted. 43 AUDIT RESULTS: THE DEPARTMENT HAS FAILED TO IDENTIFY ALL PAROLEES REQUIRING MENTAL HEALTH TREATMENT Prior to receiving a $6 million augmentation in Fiscal Year (FY) 1999/2000 the POC operated with a base budget of $8.7 million that, in part, funded 52 clinicians who were responsible for treating an increasing service population. As noted in the audit report, there were a total of 23,814 mentally ill parolees and 10,376 sex offenders released during the period from July 1998 through September 2000 who "technically" would require POC services Despite its limited resources, during the above mentioned time period the POC serviced 12,203 mentally ill parolees and 6,436 sex offenders. In order to supplement the program, parole agents also would directly access treatment services for the mentally ill through available county resources. This practice is consistent with the Departmental Operations Manual and the California Code of Regulations Title15. Crime Prevention and Corrections Section 3706. While the numbers of those referrals were not tracked in the old database, any referrals made to outside treatment providers under the new data management system will be tracked effective August 2001. In addition to the offender lists generated from the Department's Offender Information Services Branch (OISB) the MHSCP has been obtaining updated information and additional referrals from individual institutional medical staff in order to identify those in need of continuum services. The P&CSD will continue to explore alternatives to improve its access to complete listings of mentally ill inmates who will be paroling so that appropriate treatment services can be provided to this population. This change and the overall implementation of the MHSCP will enable CDC to improve its processes for identifying and treating mentally ill parolees. It is the intent of the program to continue treatment for those parolees previously identified as mentally ill and needing services. The P&CSD will identify and treat, as appropriate, those parolees who were released to parole from the Mental Health Services Delivery System and who are not currently receiving services from the MHSCP. Progress in treating all parolees requiring mental health treatment will be chronicled in future Audit Report updates. THE CLINICS DO NOT ALWAYS PERFORM NEEDED PRE-RELEASE ASSESSMENTS OR PROVIDE TIMELY SERVICES Implementation of the MHSCP will enable the P&CSD to perform more pre-release assessments and provide more timely services. Two million dollars of the initial six million dollars allocated to the program are used for outside contract providers to perform the pre-release assessment component of the MHSCP. As of May 2001, the P&CSD had successfully finalized contracts for the provision of the MHSCP's pre-release component. The contracts provide the P&CSD with dedicated resources to perform the pre-release assessments. Prior to the execution of the contracts, the P&CSD had to use newly hired psychiatric social workers to accomplish the pre-release assessments. 44 Despite difficulties in recruiting and hiring the psychiatric social workers and the fact that they were new to the correctional system, during the first five months of the new program operation, 61.5% (3,853) of all newly released mentally ill parolees were seen at a POC. As noted above, in May 2001, the contractors began performing pre-release services thereby allowing the POC psychiatric social workers, formerly providing pre-release services, to be deployed to the field parole units to provide post-release treatment services. Furthermore, the P&CSD was successful in the FY 2000/2001 budget process in establishing a population driven staffing ratio. In conjunction with additional staff, the POC contract dollars are also subject to increase as the mentally ill population grows. The P&CSD is in the process of amending our pre-release contracts to ensure there are sufficient staff to meet the needs of the program's escalating mentally ill population. A NEW DATA MANAGEMENT SYSTEM MAY ADDRESS SOME PROGRAM WEAKNESSES BUT COULD BE MORE EFFECTIVE IF LINKED TO OTHER DEPARTMENT SYSTEMS The new data management system (system), to begin operation by the end of August 2001, would provide POC with the latest technology and an automated data collection system to evaluate and measure the program's effectiveness. For the first time since its inception, the POC's will be able to communicate with one another on a real time basis, and will be able to receive updated information on mentally ill parolees at the click of a mouse. The new POC system has been meticulously created to provide the most comprehensive mental health database ever used by the P&CSD. It will provide conclusive statistical analysis of any and all of the "entry fields" it utilizes. The new system is designed to notify and flag contractors and clinicians of: 1. The date and location a pre-release assessment is due to be completed. 2. The name of the clinician who preformed the pre-release assessment. 3. When a change is entered on an inmate's Earliest Possible Release Date (EPRD), the system notifies the POC that the appointment date needs to be changed to conform to program guidelines. 4. The date and location a clinical appointment is scheduled for the parolee. The new system is also capable of identifying the costs of treatment associated with the mentally ill population. The system tracks the number of contacts and the amount of time by clinician discipline spent with each type of mentally ill parolee. The system is designed to account for all populations receiving services through the POC. During the initial phase of implementation, the system will capture information on all parolees who were released to parole from the Mental Health Services Delivery System. In subsequent phases, the P&CSD intends to track the time and resources used to treat all parolees in the POC. 45 It is possible to link other departmental systems to the POC system. The barriers to such an effort include data conversion that may be problematic and the additional costs and work associated with adding this functionality. The P&CSD remains committed to the development of a system that will enable multiple systems to communicate with one another. The program will continue to develop the best possible, cost-effective means of gathering information critical to program success. The future deployment of the Strategic Offender Management System (SOMS) will consider facilitating the data exchange between this system and other departmental systems. THE CLINICS PROVIDE SERVICES TO MANY PAROLEES WITH PROBLEMS OTHER THAN MENTAL ILLNESS While an individual may not have a diagnosed mental disorder, historically the POC's have been an available resource for parolees who are in acute emotional and/or psychological distress. This is consistent with the program guidelines used to treat such individuals. In addition, the POC provides the parole agent with a tool to deal with those parolees who are presenting the greatest risk to public safety. With the implementation of the MHSCP, the P&CSD will refine its use of uniform screening assessment tools and diagnostic evaluations to better determine which sexual or violent offenders can most benefit from treatment. The P&CSD currently has a contract out to bid for Relapse Prevention Programs for sex offenders. Once those contracts are secured and are operational, it is the intent of the P&CSD to review the resources needed to serve the remaining sex offenders and violent offenders. The P&CSD will utilize the annual budget process to request necessary resources for the populations POC must serve. THE PROGRAM NEEDS TO TAKE ADDITIONAL ACTIONS TO MANAGE EXPECTED CASELOAD INCREASES As mentioned previously, the program funding is now population driven. As the mentally ill population increases, staffing and contract dollars also increase to manage caseload work. The P&CSD is in the process of recruiting and hiring 38 new clinicians to work in conjunction with the existing 54 clinicians hired with the six million dollar augmentation in FY 99/00 for a total of 156 clinicians statewide. With these new resources, the program is currently restructuring its staffing compliment in order to meet the needs of the mentally ill. Prior to the new program, each parolee was assigned one clinician. The new program is structured in a way to have a treatment team approach for the delivery of services to the mentally ill. 46 For instance, the psychiatric social worker assigned to the case will provide individual and group therapy and will work to ensure all benefit eligibility assistance has been provided. The psychologist will provide needed psychological testing and individual and group therapy. The psychiatrist will provide medication management, individual and group therapy and other services as necessary. As new staff is being hired and trained, the program is in the process of establishing treatment teams. It is the goal of the program to establish clinician caseload standards. CONCLUSION The Department appreciates the observations and recommendations contained in the Audit Report. This audit is especially helpful for the infancy phase of this new statewide program, as the findings will assist in operationalizing this very valuable public safety program. 47 48 1 49 Acronyms in MHSCP Pre and Post Release Flow Charts 1. AOR - Agent of Record 2. CCCMS - Correctional Clinical Case Management System 3. EOP - Enhanced Outpatient Program 4. EPRD - Earliest Possible Release Date 5. IDTT - Interdisciplinary Treatment Team 6. MHSCP - Mental Health Services Continuum Program 7. OIS - Offender Information Services 8. POC - Parole Outpatient Clinic 9. PSW - Psychiatric Social Worker 10. TCMP-MI - Transitional Case Management Program - Mentally Ill 50 COMMENTS California State Auditor’s Comments on the Response From the Department of Corrections T o provide clarity and perspective, we are commenting on the Department of Corrections’ (department) response to our audit report. The number below corresponds to the number we placed in the margin of the department’s response. 1 The Parole Outpatient Clinic Program (program) has reduced its standard for seeing less severely mentally ill parolees from 10 business days to 7 days. During our audit, the 10 business day standard was in effect and was the basis we used to determine whether the program provided timely services. 51 cc: Members of the Legislature Office of the Lieutenant Governor Milton Marks Commission on California State Government Organization and Economy Department of Finance Attorney General State Controller State Treasurer Legislative Analyst Senate Office of Research California Research Bureau Capitol Press 52