Venue: The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27708-0120
Presentation
Measuring and Explaining Management Practices in Addiction Treatment Programs
Variations in the performance of organizations is common in many fields. Economic research has focused primarily on identifying productivity differences through better measures of inputs (capital, labor, skills, materials, technology). A potentially critical component - managerial quality - has typically been treated as a fixed, unobserved effect.
However, recent work by Bloom and Van Reenen (QJE, 2007) suggests that management practices can be quantified and used to explain organizational performance. Using an innovative survey tool, they scored management practices for 732 industrial firms, and demonstrated that better scores were strongly associated with firm-level productivity.
We adapted this survey tool for the field of addiction treatment and conducted telephone interviews with 137 addiction treatment programs in 4 states. To minimize interviewee response bias, we used a 'double-blind' methodology based on open questions, with interviewees unaware they are being scored, and interviewers unaware of actual program performance. The survey scored programs on 14 management practices, including, e.g., client intake and retention, quality improvement, goal setting, and incentives for employees. Each of the 14 practices was scored from 1 (worst practice) to 5 (best practice) and ultimately aggregated into a single 'average management practice score.'
We then assess the association between management practice scores and program performance. We examine two financial outcomes and one patient outcome: days until first available appointment. This 'time-to-treatment' measure was collected through monthly pseudopatient phone calls to our 137 programs. For financial outcomes, we use OLS regressions on the management practice z-score and the log of employee size, with interviewer and state fixed effects, with standard errors clustered by the organization. We find that management practice score is weakly associated with revenues (p = 0.13) and profit margins (p=0.05). In negative binomial regressions, better management practice scores are associated with shorter times-to-treatment (p = 0.03).
We also find a wide distribution of management practice scores. Thus, we assess the affects of competition on management practices. We find higher management practices are associated with catchment areas with larger numbers of addiction treatment programs (p <0.001).
These results parallel the results of Bloom and Van Reenen. The similar findings in a different field (consisting primarily of not-for-profit and government organizations) suggest the importance of understanding how management practices affect program behavior and patient treatment.
Our work also responds to a call from the addiction services field for better understanding of the 'business' of addiction treatment. In addition to our general findings, the survey also provides insight into the organization of addiction treatment programs. For example, a substantial number of programs have practices that have focused on intake and time-to-treatment, but relatively few paid attention to improving client retention. Many programs appear to have difficulty in aligning employee incentives with program goals, with a number expressing difficulty in firing poor performers. Bonus and incentive programs for high performers do exist but at a relatively small proportion of programs.