Social Workers' Propensity to Engage in Recovery Oriented Services: A Randomized Factorial Design

Background and Purpose: The mental health service delivery system has undergone a transformation to adopt a recovery orientation. After more than ten years, systems and policies have substantially changed, and the language of recovery pervades mental health organizations; however, front-line social workers’ behavior may not have changed commensurately. This study seeks to identify key factors that predict social workers’ likelihood of providing services consistent with a recovery orientation and hypothesizes that client characteristics and providers’ understanding of recovery predict their likelihood of endorsing recovery-oriented service provision for adults with serious mental illness.

Methods: Data were collected through an online survey employing a randomized factorial design and two standardized instruments measuring recovery knowledge and expectations. Respondents, obtained through snowball sampling, were social workers (N=107) who primarily serve adults with serious mental illness. The survey presented four vignettes (n=398), each with randomly inserted client characteristics followed by a set of 16 items measuring endorsement of recovery-oriented services. Backwards deletion identified the variables predicting a social worker’s likelihood for endorsing recovery-oriented services. Significant variables were entered into a model, and multiple regression was used, controlling for the clustering effect of participants responding to multiple vignettes, to determine the overall significance of the model and coefficients of relevant variables.

Results: The model was significant (p<.0001) and accounted for 58% of the variance (R² .577) of the total score measuring endorsement of recovery-oriented services. Predictors (with an indirect relationship) included overall recovery knowledge and the following dichotomous variables: psychotic symptoms, homelessness, living in a residential facility, Bipolar Disorder, poor adherence to treatment plan, and active use of substances. Recovery knowledge (p<.001) accounted for about half of the total variance while the remaining client characteristics explained about 11%.

Conclusions and Implications: Although more than ten years have passed since the President’s New Freedom Commission on Mental Health, the actual behaviors of social workers providing services to adults with serious mental illness may be incompatible with a recovery orientation. Data suggest that having a robust understanding of recovery is the most important predictor of endorsement of recovery-oriented service provision. Policies and literature indicate that recovery is relevant to all people at all stages of treatment. The finding that client characteristics predict endorsement of recovery-oriented services underscores a fundamental misunderstanding of the concept, suggesting that providers differentially endorse recovery based on client presentation. Despite the saturation of recovery language within service organizations, providers may need further training to develop a robust understanding of the concept and ensure that front-line services embody the spirit of recovery.