A Longitudinal Study Testing the Effectiveness of Supervisory Support on Turnover Intention through Reduced Burnout and Improved Job Satisfaction

Background: Staff turnover negatively affects both workers (e.g., well-being, financial stability) and the agency (e.g., cost for new hires, productivity). The rate of turnover in behavioral health has been reported as much higher than in other health care, varying from 30 to 60%. Research indicates that the quality of supervisory support could alleviate some of the staff turnover by decreasing burnout as well as increasing job satisfaction (Kadushin & Harkness, 2002). However, the long-term mediation mechanisms of reduced burnout and improved job satisfaction facilitated by supervisory support, which leads to reduced turnover intention, are not clear. Understanding these mechanisms could promote more effective and strategic supervisory practices.

Methods: The current research conducted a secondary data analysis. The original data was collected from 193 direct clinical care workers at two community mental health centers (CMHCs) longitudinally at three time points (baseline, 3 mo., and 6 mo.). The majority of participants were White (86%) and female (80%), and the average age was 40 (±12) years old. About half of the workers held degrees in either social work or psychology. The average work experience within the agencies was 5 (±6) years. The variables used in this research include the perceptions of supervisory support scale (Fukui et al, 2014), burnout (emotional exhaustion -- a subscale of the Maslach Burnout Inventory), overall job satisfaction (Nagy, 2002), and turnover intention (Salyers et al, 2015). Structural equation modeling was used to examine the cross-sectional and long-term effects of supervision on turnover intention with burnout and job satisfaction as mediators (i.e., predicting turnover intention at 6 mo. by the baseline supervisory support scores through the mediators at 3 mo.). Mplus 7.1 was used for the analysis.

Results: The longitudinal mediation model fit the data well (χ²₍₂₆₎=40.32, p<.05, RMSEA=.053 _{[90% CL .014 - .084]}, CFI=.988, TLI=.971). The results indicate that supervision had a concurrent effect (effect at each time point) on turnover intention through both burnout and job satisfaction at any of the time points. However, supervision had a longer term effect (effect across time points) on turnover intention through job satisfaction (p=.03) but not burnout (p>.05).

Conclusions and implications: The current study suggests that support provided by supervisors to reduce burnout and to increase job satisfaction is critical when alleviating workers’ on-going/recurring turnover intention. However, supervisory support promoting workers’ job satisfaction seems to be more effective in terms of reducing future turnover intention. Most CMHCs face financial challenges and increased social demands (e.g., productivity) which may inhibit incentivizing workers at structural levels. Supervisors’ roles in facilitating staff morale seem to be a key strategy to retain workers in the long run. Implications for supervisory strategies that may prevent future staff turnover will be discussed.