Research That Matters (January 17 - 20, 2008)

Purpose: People with severe and persistent mental illnesses are at higher risk for substance use disorders and Post-Traumatic Stress Disorder (PTSD) than people in the general population. Preliminary research findings have suggested that social supports may mediate the relationship between symptoms of mental illness and both substance abuse and PTSD symptom severity. For many of these clients, perceived social support provided by line staff may be essential in helping them reduce high risk behaviors (e.g., drinking) and PTSD symptoms. The current study of 275 community mental health center clients with either a schizophrenia spectrum or major mood disorder tests the hypotheses that perceived social support provided by treatment staff will show direct and indirect effects, respectively, on reducing client substance use and PTSD symptoms. Methods: Survey data were collected by mental health staff (social workers, case managers, nurses) in face-to-face interviews with 275 community mental health center clients. Several measures of known reliability and validity were employed: The Psychosocial Wellbeing scale (PSWS), drinking to cope with negative emotions (a subscale of the Drinking Context Scale), the Post-Traumatic Stress Disorder Scale (PSS-I), and a brief scale that measured perceived staff social support, a scale which was validated as part of the current study. Exploratory factor analysis revealed high factor loadings for the three-item 5-point Likert scale: ---“I can change things for the better” (.79), “staff helped my self-esteem” (.80), and “staff helped improve my relationships” (.80). Internal consistency was good (Cronbach's alpha=.78). Results: A combined measurement-path structural equation model revealed that six of seven indices confirmed a high degree of congruence between model and data: (ײ (215) = 297.46, p < .01); NFI = 92, RFI = .90, IFI = .98, TLI = .97, CFI = .98, RMSEA = .04. Perceived social support from the staff showed substantial direct effects on both psychological wellbeing (B = .30) and social wellbeing (B = .22). In addition, staff social support appeared to have a very large impact on the likelihood of a client drinking to cope with negative emotions (B = -.51), and an even larger effect on overall client satisfaction with services (B = .69). Staff social support also had the largest indirect effect (B = -.16) on PTSD symptoms of all factors for which indirect effects were measured. Implications: Given that people with severe mental illnesses receive much of their social support from agency staff, results of this study suggest that this day-to-day impact on client wellbeing may be a critical factor in good client outcomes. The working relationship has been demonstrated to be a critical factor in effective psychosocial interventions across a wide array of disorders and problems-in-living. Practitioners, insurance payers, and policy makers should not give short-shrift to the essential role of a consistent working relationship as they press for more cost-effective treatment methods.