Dr. Yanos is Associate Professor, Department of Psychology, John Jay College of Criminal Justice, City University of New York, New York, New York; Ms. Vreeland is Program Coordinator, University of Medicine and Dentistry of New Jersey-University Behavioral HealthCare (UMDNJ-UBHC), Center for Excellence in Psychiatry, Clinical Assistant Professor of Nursing, UMDNJ School of Nursing, and Clinical Assistant Professor of Psychiatry & Family Medicine, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey; Dr. Minsky is Director of Quality Improvement, UMDNJ-UBHC, and Associate Professor, Department of Psychiatry, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey; Dr. Fuller is Director, Counselling Services, University of New Brunswick, Fredericton, New Brunswick, Canada; and Dr. Roe is Chair, Department of Community Mental Health, Faculty of Social Welfare and Health Studies, University of Haifa, Mount Carmel, Haifa, Israel. The authors disclose that they have no significant financial interests in any product or class of products discussed directly or indirectly in this activity, including research support.
The authors thank the staff and clients of the University of Medicine and Dentistry of New Jersey-University Behavioral HealthCare Newark Adult Partial Hospital Program for their time and support during this study.
Address correspondence to Philip T. Yanos, PhD, Associate Professor, Department of Psychology, John Jay College of Criminal Justice, City University of New York, 445 W. 59th Street, New York, NY 10019; e-mail: email@example.com.
It is now widely agreed that the public mental health system must move toward providing recovery-oriented services (President’s New Freedom Commission on Mental Health, 2003). A parallel initiative has been an emphasis on moving routine mental health settings toward the implementation of evidence-based services (Drake et al., 2001; Torrey et al., 2001). Although, in the past, some have suggested that these two principles represented possibly incompatible imperatives in mental health care (Anthony, Rogers, & Farkas, 2003; Frese, Stanley, Kress, & Vogel-Scibilia, 2001), there is now beginning to be agreement that the two models can coexist (Bond, Salyers, Rollins, Rapp, & Zipple, 2004) and that together they reflect “essential” ingredients of services improvement in the public mental health system (Torrey, Rapp, Van Tosh, McNabb, & Ralph, 2005).
Despite general agreement on the need to implement both evidence-based and recovery-oriented services, there is as yet no consensus on how this implementation can be applied in the diverse service models that exist in the public mental health system (e.g., assertive community treatment, outpatient services). Partial hospitalization, alternately referred to as day treatment (the two terms will be used interchangeably in this article), is a major service modality that exists in the public systems of many states. Such programs typically provide intensive, primarily group-based services during weekdays (lasting from morning to early afternoon).
In the past, some suggested that this service modality is inherently incompatible with both evidence-based and recovery-oriented treatment, and that it should be replaced with other service modalities such as supported employment or intensive outpatient treatment (Becker et al., 2001; Hoge, Davidson, Hill, Turner, & Ameli, 1992). These critiques reflect the belief that partial hospitalization programs have unending lengths of stay and do not work to move mental health consumers toward vocational or other recovery-oriented goals. Indeterminate lengths of stay are believed to be particularly problematic, as they communicate the expectation that consumers should resign themselves to a lifetime of attending day treatment, rather than pursuing employment, education, and other activities in the community (Becker et al., 2001; Hoge et al., 1992). In a discussion of an effort to replace day treatment programs with supported employment programs in Rhode Island, Becker et al. (2001) made a pointed indictment of the partial hospitalization model, suggesting that this kind of treatment has the “unintended effect of socializing [mental health consumers] into disability” (p. 356) and therefore impedes, rather than promotes, recovery.
Despite such criticisms, day treatment continues to exist and to be supported by both researchers and consumer groups who believe it can be successfully transformed (Consumer Advocacy Partnership, 2006; Reger, Wong-McDonald, & Liberman, 2003). For example, Reger et al. (2003) discussed how a day treatment program that was modified to include evidence-based skills training services showed promise in promoting relatively rapid discharge to less intensive forms of care and, ultimately, employment. This suggests that some programs may help consumers to be more successful when they are referred to supported employment services. In a discussion of what is needed to transform partial hospitalizations in the state of New Jersey, a mental health consumer group (Consumer Advocacy Partnership, 2006) called for shorter lengths of stay, better staff training to ensure services are recovery oriented, and the implementation of evidence-based practices. While acknowledging considerable problems with many day treatment programs, this discussion did not suggest that the service model be discontinued.
These analyses suggest that the problems that have been associated with day treatment (e.g., that groups exist simply to pass the time, keep consumers from engaging in more productive activity, and retain consumers indefinitely) are in need of reform but are not endemic to the model. They also reflect what we believe to be a pragmatic perspective with regard to mental health service transformation, wherein existing service models are changed from within, rather than eliminated and replaced with other models (a position which Corrigan and Boyle  described as the “evolutionary” approach to systems change).
In this article, we examine a real-world example of the effects of an effort to make fundamental changes in the nature of an existing partial hospitalization program by initiating a move toward an evidence-based, recovery-oriented service model. The program, located in the disadvantaged urban environment of Newark, New Jersey, began a change process in the early 2000s to reduce lengths of stay by working with mental health consumers with severe mental illness to move more rapidly toward educational and vocational goals. Simultaneously, it focused on hiring staff who were trained in psychiatric rehabilitation and began implementing evidence-based practices on a number of levels.
The purpose of the current study was to examine the accuracy of the view that partial hospitalization programs by definition have unnecessary lengths of stay and provide services that are incompatible with recovery-oriented and evidence-based practice. Using three distinct kinds of information (i.e., administrative data, self-report data from program staff, and fidelity ratings from independent assessors), we evaluated the services provided by the Newark program. We report cross-sectional data on discharge and length of stay within a 2-year, 9-month period, the implementation of an evidence-based practice (integrated dual disorder treatment), and the use of recovery-oriented principles and practices by program staff.
The Newark Adult Partial Hospitalization program is a service of the Newark campus of the University of Medicine and Dentistry of New Jersey-University Behavioral HealthCare (UMDNJ-UBHC). It provides full-day (9 a.m. to 3:30 p.m.) psychiatric rehabilitation services designed to meet the special requirements of adults with severe mental illness from Newark and the surrounding community. While at the program, individuals are provided with a variety of services depending on their needs, including psychoeducational group treatment, nursing services, psychiatric medication evaluation and monitoring, pre-vocational services, case management services, and individual psychotherapy. Many groups at the program follow treatment manuals that have a basis in research evidence, including mental health psychoeducation, coping skills training, social skills training, social anxiety treatment, anger management, health and wellness psychoeducation, and dialectical behavior therapy. The stated goal of the partial hospitalization program is to reduce the likelihood of rehospitalization and facilitate successful community integration. The program maintains an informal relationship with two local, consumer-run drop-in centers and encourages consumers to attend these programs during evenings and weekends.
At the time of this study, the program had a total census of approximately 150 individuals and typically served between 80 and 100 individuals on a daily basis. The major referral source for the program is the inpatient unit of the University Hospital, although clients are also referred through the University’s crisis clinic, as well as other sources such as homeless shelters, parole officers, and state and county hospitals. The program has a subprogram geared toward the treatment of individuals with co-occurring substance abuse problems (i.e., the Mentally Ill Chemical Abusers [MICA] unit).
Approval was received from the UMDNJ-Newark campus Institutional Review Board for all aspects of the study. The study involved three elements: administrative data analysis, assessment of recovery-oriented practices among program staff, and an assessment of fidelity to integrated dual disorder treatment (IDDT).
Administrative Data Analysis. Data from UMDNJ-UBHC’s administrative database were analyzed to examine the characteristics of all new admissions between January 1, 2004 and September 25, 2006 (the ending date was selected arbitrarily as the date when the data were extracted). Variables studied included the mean length of stay for all new program admissions, insurance source (Medicaid, hospital care payment assistance program, or other) for all new program admissions, and the racial, gender, and diagnostic makeup of all new program admissions.
Assessment of Recovery-Oriented Practices Among Program Staff. The program’s recovery orientation was assessed by administering the provider version of the Recovery Self-Assessment Scale (RSA) (O’Connell, Tondora, Croog, Evans, & Davidson, 2005) to all 31 program staff and supervisors. The RSA is a validated scale that assesses perceptions of several domains of recovery-oriented service provision, including life goals, consumer involvement, treatment options, client choice, and individually tailored services. After discussing the rationale for the study in a regular staff meeting, copies of the scale were left for staff in their mailboxes, along with a letter of introduction (signed consent forms were not used to maintain anonymity). Staff completed the scales anonymously and placed completed questionnaires in the principal investigator’s mailbox. All scales were returned between June 15, 2006 and August 15, 2006.
Assessment of Fidelity to IDDT. Two outside assessors who were not affiliated with the Newark Adult Partial Hospitalization program conducted an assessment of the MICA unit’s fidelity to IDDT using the Integrated Dual Disorders Treatment Fidelity Scale (Substance Abuse and Mental Health Services Administration [SAMHSA], 2003). Both individuals had experience in dual disorder mental health services, and one had considerable experience conducting fidelity ratings. Following the procedure outlined in the guidelines for the scale and as discussed by others (Bond, Evans, Salyers, Williams, & Kim, 2000), this assessment involved a 1-day site visit (in July 2006) during which unstructured interviews were conducted with administrators and staff members, service documents were reviewed, and service groups were observed. The two assessors conducted ratings independently and then met together to discuss areas of discrepancy. Areas in which there was a discrepancy were discussed, and consensus ratings were established.
The Table presents findings from the analysis of administrative data. The program served a predominantly African American population, primarily with psychotic and major affective disorders, approximately one third of whom lacked insurance. Approximately one fourth of the new admissions had secondary substance use diagnoses. Of 428 individuals admitted to the program between January 1, 2004 and September 25, 2006, approximately three fourths had been discharged by September 25, 2006. Among those who had been discharged during the 2-year, 9-month period, the mean length of stay was approximately 5 months. The finding that one third were covered by the hospital care payment assistance program is consistent with the view that many individuals admitted for services were new to the mental health system and lacked entitlements, as individuals who are covered by emergency entitlements such as welfare do not receive full Medicaid coverage in New Jersey.
Table: Demographic and Service Use Characteristics of Individuals Admitted to the Newark Adult Partial Hospitalization Program (N = 428)
Twenty-one of 31 providers working at the agency returned completed RSAs. Mean scores for the RSA (possible range of 1 = strongly disagree to 5 = strongly agree for all scales) were as follows: Summary Score = 3.73 (SD = 0.62), Goals Stressed Over Symptoms = 4.12 (SD = 0.7), Rights and Respect = 4.16 (SD = 0.61), Individually Tailored Services = 3.72 (SD = 0.53), Diversity of Treatment Options = 3.67 (SD = 0.8), and Consumer Involvement = 3.23 (SD = 0.71).
Mean scores above 4 on the first two subscales indicate that providers at the agency perceive that recovery-oriented practices are used in direct interactions with clients. Scores were lowest (although still above 3) for the subscales that deal with programmatic practices where the partial hospitalization model may have less flexibility (i.e., Consumer Involvement and Diversity of Treatment Options). When compared with scores obtained from 344 providers in a survey of providers in the state of Connecticut (O’Connell et al., 2005), where day treatment programs no longer operate, similar mean scores on all scales were found (mean scores ranged from 3.39 for Consumer Involvement to 4.1 for Goals Stressed Over Symptoms, and the mean Summary Score was 3.87). Standard deviations for all subscales were also comparable with those observed by O’Connell et al. (2005).
Ratings of the IDDT Fidelity Scale range from 1 (no implementation) to 5 (full implementation). Fidelity ratings on the IDDT scale were as follows: Multidisciplinary Team = 5, Integrated Substance Abuse Specialists = 5, Stage-wise Interventions = 1, Access to Comprehensive Services = 1, Time-unlimited Services = 5, Outreach = 3, Motivational Interventions = 3, Substance Abuse Counseling = 5, Group Dual Disorder Treatment = 5, Family Psychoeducation = 1, Participation in Self-Help Groups = 3, Pharmacological Treatment = 5, Interventions to Promote Health = 4, and Secondary Interventions for Nonresponders = 3.
Overall, the program was found to be moderately successful in its implementation of IDDT (overall mean rating = 3.5, indicating moderate overall fidelity). An examination of the individual item ratings on the scale revealed specific areas where the program had failed to implement IDDT practices, which reduced its overall score—specifically its lack of implementation of family interventions and stage-based interventions. However, the program was found to have full fidelity in its implementation of key aspects of the IDDT model, such as substance abuse counseling, group dual disorder treatment, and pharmacological treatment.
These findings are generally supportive of the view that the partial hospitalization program provides services that are consistent with the recommended strategies of shorter length of stay, recovery-oriented services, and implementation of evidence-based practices with fidelity. Notably, three fourths of the individuals who were admitted to the program within a 2-year, 9-month period were no longer attending the program at the time the data were extracted. Among those who had been discharged, the average length of stay was 5 months. These data stand in contrast with the classic image of a partial hospitalization setting as maintaining clients for unnecessarily indeterminate lengths of stay. Rather, they are consistent with the impression that the program serves individuals in need of a transitional phase of treatment, after they have recovered from an acute psychiatric episode but before they are ready to immerse themselves in vocational or educational goals. Ratings on the RSA and the IDDT scales showed that the program has achieved promising results with regard to incorporating recovery-oriented and evidence-based approaches into its treatment.
The findings also revealed some barriers that are perhaps more difficult to surmount due to some of the less flexible elements of the partial hospitalization model. For example, although the program has made numerous attempts to initiate family psychoeducational services, staff interviewed for the IDDT rating noted that this effort has repeatedly encountered obstacles related to the 9-to-5 schedule the program operates on, which makes it difficult for employed family members to come in to meet with staff during office hours. As a result, the families of only a small number of clients were found by the IDDT raters to be involved in the treatment process, and a rating of 1 (no implementation) was assigned.
Similarly, one of the lowest scores for program practices on the RSA was in the area of Diversity of Treatment Options, which partially reflected the fact that the program is not able to offer great flexibility in where and when it provides services (because the program operates on a 5-day, 9-to-5 schedule and because New Jersey Medicaid requirements necessitate that the majority of services be provided onsite). These barriers suggest that, in addition to implementing evidence-based approaches and training staff to provide services in a recovery-oriented manner, programs may need to seek changes in some of these structural barriers to allow partial hospitalizations to become vehicles of evidence-based, recovery-oriented services to the greatest extent possible. This may require that they organize to pressure payers to ease restrictions so more flexibility in service provision is possible.
Some limitations of our findings should be noted. Because of the cross-sectional nature of the data collected, we are unable to confirm whether the findings reflected changes in the practices of the program. We are therefore more confident in concluding that they provide a “snapshot” of how the partial hospitalization program is currently functioning, which we believe to be representative of how programs that have been transformed can appear. Data on the recovery orientation of the program came only from self-reports of staff members who completed the scale and did not include the perspectives of all staff or consumers. It is possible that staff ratings were inflated by a desire to provide socially desirable responses. It is also possible that the relatively low response rate of the service providers may have skewed responses in favor of those providers who had a more favorable impression of the program. It should be noted, however, that O’Connell et al.’s (2005) data from Connecticut indicate that staff tend to provide lower ratings on the RSA than consumers and that staff ratings may thus offer a more conservative estimate of recovery orientation. With regard to evidence-based practices, we only assessed fidelity to IDDT and not other important practices recommended by SAMHSA, such as illness management and recovery and family interventions. However, it should be noted that IDDT reflects the implementation of a comprehensive set of practices and, as such, includes some of the other evidence-based practices in its ratings.
We conclude that despite a marred reputation and some structural barriers, partial hospitalization programs do have the potential to evolve into a part of an evidence-based and recovery-oriented mental health system of care. Specifically, partial hospitalization programs that deliver evidence-based services in a recovery-oriented manner can help consumers who are transitioning from an acute episode by facilitating the development of coping and life skills that can help them cope with stressors as they identify and move toward vocational, educational, or other goals that are part of the recovery process. Consumers making this transition can be helped by partial hospitalization programs if the programs offer structured services that are tailored to individual needs and delivered in a recovery-oriented manner (i.e., in a respectful environment that encourages hope) that facilitates the development of coping skills, social skills, illness management skills, and healthy lifestyle skills. These skills and services can help prepare consumers to live productive and rewarding lives in the community as they transition to less intensive service modalities.
- Anthony, W, Rogers, ES & Farkas, M2003. Research on evidence-based practices: Future directions in an era of recovery. Community Mental Health Journal, 39, 101–114. doi:10.1023/A:1022601619482 [CrossRef]
- Becker, DR, Bond, GR, McCarthy, D, Thompson, D, Xie, H & McHugo, GJ et al. . 2001. Converting day treatment centers to supported employment program in Rhode Island. Psychiatric Services, 52, 351–357. doi:10.1176/appi.ps.52.3.351 [CrossRef]
- Bond, GR, Evans, L, Salyers, MP, Williams, J & Kim, HW2000. Measurement of fidelity in psychiatric rehabilitation. Mental Health Services Research, 2, 75–87. doi:10.1023/A:1010153020697 [CrossRef]
- Bond, GR, Salyers, MP, Rollins, AL, Rapp, CA & Zipple, AM2004. How evidence-based practices contribute to community integration. Community Mental Health Journal, 40, 569–588. doi:10.1007/s10597-004-6130-8 [CrossRef]
- Consumer Advocacy Partnership. 2006. Response to proposed revisions to partial care regulations in New Jersey. The Consumer Viewpoint, 5(2)
- Corrigan, PW & Boyle, MG2003. What works for mental health system change: Evolution or revolution?Administration and Policy in Mental Health, 30, 379–395. doi:10.1023/A:1024619913592 [CrossRef]
- Drake, RE, Goldman, HH, Leff, HS, Lehman, AF, Dixon, L & Mueser, KT et al. . 2001. Implementing evidence-based practices in routine mental health service settings. Psychiatric Services, 52, 179–182. doi:10.1176/appi.ps.52.2.179 [CrossRef]
- Frese, FJ 3rd. , Stanley, J, Kress, K & Vogel-Scibilia, S2001. Integrating evidence-based practices and the recovery model. Psychiatric Services, 52, 1462–1468. doi:10.1176/appi.ps.52.11.1462 [CrossRef]
- Hoge, MA, Davidson, L, Hill, WL, Turner, VE & Ameli, R1992. The promise of partial hospitalization: A reassessment. Hospital & Community Psychiatry, 43, 345–354.
- O’Connell, M, Tondora, J, Croog, G, Evans, A & Davidson, L2005. From rhetoric to routine: Assessing perceptions of recovery-oriented practices in a state mental health and addiction system. Psychiatric Rehabilitation Journal, 28, 378–386. doi:10.2975/28.2005.378.386 [CrossRef]
- President’s New Freedom Commission on Mental Health. 2003. Achieving the promise: Transforming mental health care in America (Publication No. SMA 03–3832). Retrieved December 2, 2008, from http://www.mentalhealthcommission.gov/reports/FinalReport/toc.html
- Reger, G, Wong-McDonald, A & Liberman, RP2003. Psychiatric rehabilitation in a community mental health center. Psychiatric Services, 54, 1457–1459. doi:10.1176/appi.ps.54.11.1457 [CrossRef]
- Substance Abuse and Mental Health Services Administration. 2003. Integrated Dual Disorders Treatment Fidelity Scale. In Co-occurring disorders: Integrated dual disorders treatment Retrieved December 2, 2008, from http://download.ncadi.samhsa.gov/ken/pdf/toolkits/cooccurring/IDDTFidelityScaleAJ1_04.pdf
- Torrey, WC, Drake, RE, Dixon, L, Burns, BJ, Flynn, L & Rush, AJ et al. . 2001. Implementing evidence-based practices for persons with severe mental illness. Psychiatric Services, 52, 4550. doi:10.1176/appi.ps.52.1.45 [CrossRef]
- Torrey, WC, Rapp, CA, Van Tosh, L, McNabb, CR & Ralph, RO2005. Recovery principles and evidence-based practice: Essential ingredients of service improvement. Community Mental Health Journal, 41, 91–100. doi:10.1007/s10597-005-2608-2 [CrossRef]
Demographic and Service Use Characteristics of Individuals Admitted to the Newark Adult Partial Hospitalization Program (N = 428)
| African American
| European American
| Schizophrenia spectrum
| Bipolar disorder
| Major depression
| Mood disorder not otherwise specified
| No substance use disorder
| Substance use disorder
| Hospital care payment assistance program
|Discharged as of September 25, 2006
|Age (n = 428)
|Mean length of stay in days (n = 322)