Consumer/Provider Agreement On Needs Assessment within Programs Serving Homeless Adults with Serious Mental Illness

Background and Purpose: In this era of mental health recovery, effective treatment planning is optimal when consumers' perceived needs align with their providers' assessments of those needs. Studies that have directly compared consumer and case manager perceptions of service needs, however, indicate low levels of agreement (Gibbons et al., 2005). For individuals with serious mental illness who also experience homelessness, case managers have a more difficult task reaching consensus with their clients in order to address multiple inter-related needs. This study compares consumer and case manager perceptions of consumer needs for individuals newly enrolled in services for homeless adults with serious mental illnesses. Research questions are: What is the extent of agreement? Are there categories of need that produce more (or less) agreement? Methods: 72 consumer-case manager dyads were recruited from four residential programs in New York City serving homeless adults with serious mental illness. In-depth interviews were conducted with clients and their case managers that included a 20-item client needs checklist with a “yes/no” response. Five categories of needs included: clinical, social, financial, vocational, and other. Descriptive statistics, including percent agreement, were calculated and the AC1 coefficient--a more stable alternative to the kappa coefficient (Gwet, 2008)--was used to test for levels of agreement/disagreement controlling for chance. Study protocols were approved by the authors' Institutional Review Board. Results: Percentage of agreement between clients and providers ranged from 42% to 85% across the 20 items. Case managers assessed greater consumer needs in the clinical and social domains, while consumers reported greater vocational and independent living needs. Fair to high levels of agreement beyond chance were observed for needing help with housing (AC1=.80), legal issues (AC1=.66) and not needing help finding a partner (AC1=.62). Moderate agreement was found with regard to needing help with mental health problems (AC1=.44), substance abuse (AC1=.44), and the independent living task of preparing meals (AC1=.46). Conclusions and Implications: Clients and case managers agreed on clients' needs (beyond chance) on 6 of 20 needs assessments items. While low levels of agreement are consistent with the literature, it is notable that members of the dyads agreed on clinical categories of need (e.g., for addressing mental health and substance abuse problems) since engagement in treatment has been identified as highly challenging. Further research is needed to identify how agreement/ disagreement affect engagement in services as well as how best to address occasions when there is little or no agreement and the risk of service dropout runs high. Implications for practice draw on these findings and the mental health recovery paradigm in arguing for greater attention to consumer preferences and balancing treatment for clinical problems with assistance in basic needs such as housing and social services.