Research That Matters (January 17 - 20, 2008)

Background and Purpose: Factors that facilitate or hinder implementation of evidence-based practices (EBPs) in community mental health settings are both an area of emerging interest and of limited knowledge. Several authors have suggested that the relative instability of the mental health workforce may impede delivery of evidence-based mental health services. This study explored the relationship between workforce instability and implementation outcomes using data from a national EBP implementation demonstration project. This study had three specific aims: 1) to describe workforce characteristics of mental health workers implementing psychosocial EBPs, 2) to examine team-level staffing turnover rates over the 24-month implementation period, and 3) to explore the role of staffing stability in implementation. Methods: A mixed-method design was employed. The project involved 52 sites implementing one of five psychosocial EBPs. Workforce characteristics (N=365) were collected at baseline. Quantitative staffing data were available for 36 of the sites. Team turnover rates for the 24 month project were calculated as the number of team departures divided by the number of positions on the team at baseline. The authors examined the relationship between staffing challenges and two measures of successful implementation: fidelity to the practice model and penetration, or proportion of eligible clients served. Qualitative analyses were conducted using observational and interview data collected throughout the project. Results: Study participants were demographically similar to the U.S. human services workforce. Forty-five percent held a bachelor's degree as their highest degree. Forty-five percent were at the agency less than one year at the start of the project. Excluding departures for program closures, interns, and retirees, 44% (n=161) of staff members left the teams during the implementation period. Approximately 54% (n=97) of these departures were due to staff leaving the agency. About a quarter (n=41) were a result of intra-agency transfers. Twelve percent (n=20) were the result of involuntary termination Over the 24 month implementation period, teams experienced a mean (SD) turnover rate of 80% (48%). Exploratory Spearman correlations suggested that penetration of the EBP practice model was inversely correlated with turnover rates (N=34 teams, Spearman Rho=-.40, p=0.025). No statistically significant association was observed between program fidelity and turnover rates. Several qualitative themes relevant to workforce issues and implementation emerged. Qualitative data supported the quantitative finding suggesting that when staffing difficulties occurred, penetration, and not fidelity, was affected. Second, when sites experienced a minimum of approximately 33% turnover over 24 months, stakeholders began noting staffing as a factor in implementation. All sites with turnover rates above 100% noted staffing as a significant factor. Finally, 46% (11/24) of teams noting turnover as an implementation factor considered it to be a facilitator to implementation, as it was used to assemble a team better equipped to implement the EBP. Conclusions and Implications: This study documented high turnover rates among staff in mental health services agencies. Staffing turnover was viewed as a facilitative factor in implementation nearly as often as it was a barrier. Understanding the complexity in the relationship between turnover and implementation outcomes may facilitate development of improved implementation models.