Does Recovery Orientation Influence Turnover? Examining Predictors of Staff Turnover in Community Mental Health Clinics

Background and Purpose: Staff turnover has been identified as a key barrier in the implementation of innovative practices within behavioral health settings. Turnover rates in human service agencies can range from 25% to 50% (Aarons et al., 2010), resulting in the loss of a skilled workforce and a lack of continuity for training efforts. While organizational factors such as leadership, culture and climate have been found to predict turnover, there has been little research on how an agency’s recovery orientation relates to turnover. Mental health recovery has become central to reform efforts throughout the U.S., a values-based approach which honors self-determination and supports people’s individual recovery trajectories. This quantitative study examines the relationship between organizational factors, including recovery orientation, and staff turnover.   

Methods: The study recruited a sample of providers (N=273), from 14 community mental health clinics as part of a multi-state NIMH-funded randomized controlled trial. Leaders, supervisors, and direct care staff completed a baseline survey measuring leadership (Multifactor Leadership Questionnaire), organizational climate (Organizational Readiness for Change), and recovery orientation (Recovery Self-Assessment). Race, years at agency, gender, and role were also measured as control variables. Turnover was measured by whether a participant was still employed at the agency 12 months after completing the survey (Yes=1, No=0). Analyses utilized multivariable logistic regression models, calculating average within-group correlation statistics to determine whether to aggregate individual responses within the organization.

Results: A total of 59 (21.6%) participants left the agency within the year, with turnover rates at the agency level ranging from 0 to 75%. Within group correlation statistics on organizational variables ranged from 0.30 to 0.56 and therefore, were analyzed as individual level variables. Multivariable logistic regression found that individuals who perceived their agencies to be less recovery-oriented were more likely to leave the agency within 12 months (OR=0.24, p<.05), when controlling for race, gender, years at agency, and role. Leadership and organizational climate were not significantly associated with turnover.

Conclusions and Implications: Perceptions of leadership, climate, and recovery orientation varied within sites indicating that staff in behavioral health settings experience organizational level phenomena differently even when they are in the same organization.  However, the study found overall that perception of recovery orientation did predict whether people left the agency within a year. This suggests that agencies which integrate recovery values at an organizational level in order to align with their front-line practices may experience greater staff retention.  Further qualitative research is needed to understand differing perceptions of recovery orientation across staff and how this interacts with intentions to stay within an agency.