Mental Health Intensive Case Management (MHICM) is an effective, evidence-based intervention for seriously mentally ill veterans implemented as part of the VA Uniform Mental Health Services Handbook. We know veterans make gains in MHICM, but very little is known about how well veterans do when transitioned to lower levels of care. The proposed research responds to a programmatic change at West Los Angeles VA MHICM programs. Local leadership asked these MHICM programs to transition veterans who have made gains in MHICM into lower levels of care. This observation study gathered data on the discharge process and post-discharge outcomes. This study is unique because little process-oriented data is available to describe transitions from MHICM. Supporting veterans' transitions to lower levels of care is critical to implementing a high-quality, evidence-based, recovery-oriented continuum of care for veterans with severe mental illness.
We followed outcomes for all veterans identified by MHICM teams for discharge. Veterans were assessed when the teams identified them for discharge (baseline) and followed every 3 months (wave 1-3) over 9 months after discharge. We also used qualitative approaches with MHICM providers and veterans to understand the discharge process. We pursued 3 aims:
(1) To compare veterans' status on three domains (illness management, community integration, functioning) before MHICM discharge and across 9 months after discharge
(2) To test predictors of psychiatric hospitalization, and to explore other predictors of clinically-significant negative outcomes, over the follow up period
(3) To complete a process evaluation with veterans and providers to identify additional factors that impact the discharge process and/or subsequent outcomes
This study used a quasi-experimental, before-after, observational design on two MHICM teams that are, as a result of local VA priorities, aiming to create capacity through discharge of stable veterans. The study enrolled veterans selected by the MHICM teams for discharge to a lower level of care. We estimated that 35 veterans would be discharged and enrolled from the Greater Los Angeles (GLA) MHICM program. The GLA team discharged considerably more slowly than anticipated. To improve sample size, a second MHICM team from the Long Beach VA was recruited. This team also discharged slowly. Overall, team 1 discharged approximately 18 veterans and team 2 discharged 4 veterans to lower levels of care over the study period. Two veterans could not be enrolled due to conservatorships; five veterans refused participation as a result of severe symptoms, including paranoia or anger at the MHICM team; four other veterans could not be reached for consenting. Not all enrolled veterans completed all follow up assessments, but no veteran withdrew from the study. All enrolled veterans gave permission for their case manager to be enrolled in the study, and all case managers agreed to participate. Due to low discharge rates, additional case managers who had not discharged veterans were recruited to further examine barriers to discharge. The final sample size includes 14 providers (18 interviews) and 11 veterans.
Clinicians who both had (n=6) and had not (n=8) discharged veterans were uniformly in favor of improving their ability to transition stable veterans to lower levels of care. Barriers to transition include: 1) an inability to predict that stability would be maintained; 2) ambivalence about whether disengaged Veterans were appropriate for discharge; and 3) questions about who should make the decision. These results were published as E Bromley, L Mikesell, NP Armstrong, AS Young. "You might lose him through the cracks:" Clinicians' Views on Discharge from Assertive Community Treatment. Administration and Policy in mental health and mental health Services Research, 2014, ePub Apr2.
Clinicians discharged far fewer Veterans than anticipated. Six of 11 veterans experienced poor outcomes at 9 months, including incarceration and relapse to substance use. Among 11 enrolled Veterans, reason for discharge, as designated by the Veteran's primary case manager at the time of discharge, predicted outcomes (p=0.02) at 9 months, with disengaged Veterans experiencing poorer outcomes at 9 months than stable Veterans. These results suggest that client engagement in the transition process may indicate higher likelihood of successful outcome after transition from ACT to a less intensive clinical program. These results will be submitted as E Bromley, L Mikesell, F Whelan, G Hellemann, M Hunt, G Cuddeback, D Bradford, A.S. Young, "Clinical Factors Associated with Successful Discharge from Assertive Community Treatment," Psychiatric Services, Brief Report.
Analysis of qualitative data from veterans shows that all but one transitioned veteran described some feelings of abandonment. At baseline or follow up, most all veterans described missing the program or the case manager. About half of the veterans could not articulate the reason for the discharge; a minority recognized that they were discharged to redirect program resources to veterans in more need. A minority of veterans was neutral about the discharge and 4 veterans were actively angry at the program or the case manager. Resources that facilitated success after discharge primarily included natural supports rather than clinical supports.
This small-scale study demonstrates the significant barriers to transitioning improved veterans from intensive case management services to routine mental health care. Results show that team commitment is insufficient to effect transitions. Data also demonstrate that providers face considerable uncertainty in determining when and for whom transition is appropriate. Transition readiness scales alone may not address providers' concerns. Nonetheless, results indicate that supporting providers in transition decision-making would increase the efficiency and value of the MHICM intervention and support the recovery-oriented continuum of care.
The study has resulted in several programmatic improvements in the management of transitions from MHICM. Historically, discharge readiness, discharge rates, and reason for departure (e.g., move from area versus discharge to less intensive care) could be tracked only through chart review and gaps-in-care analyses. For the past 18 months, discharge readiness has been tracked, using the ACT Transition Readiness Scale developed by co-Investigator Cuddeback with QUERI support. In addition, this research team developed a new Discharge Form for use in program evaluation for MHICM. Systematic data on discharge has been gathered by the Northeast Program Evaluation Center within OMHO since April 2014 using a new Discharge Form as part of the program evaluation. The Discharge Form elicits information about any Veteran leaving the MHICM program, including structured data on reason for discharge. Reason for discharge data are structured in accord with findings from this RRP to differentiate Veterans discharged because all treatment goals have been met (i.e., stability), Veterans discharged because the Veteran chooses to do so despite the fact that not all treatment goals have been met (i.e., disengagement), Veterans discharged because of a move out of area, and Veterans discharged to more intensive treatment such as a nursing home. Previously, this critical information could only be gathered through chart review.
- Bromley E, Mikesell L, Armstrong NP, Young AS. "You might lose him through the cracks": clinicians' views on discharge from Assertive Community Treatment. Administration and policy in mental health. 2015 Jan 1; 42(1):99-110.
- Bromley E, Adams GF, Brekke JS. A video ethnography approach for linking naturalistic behaviors to research constructs of neurocognition in schizophrenia. The Journal of Neuropsychiatry and Clinical Neurosciences. 2013 Apr 26; 24(2):125-40.
- Mikesell L, Bromley E. Patient centered, nurse averse? Nurses' care experiences in a 21st-century hospital. Qualitative Health Research. 2012 Dec 1; 22(12):1659-71.
- Bromley E. Building patient-centeredness: hospital design as an interpretive act. Social science & medicine (1982). 2012 Sep 1; 75(6):1057-66.