Incentives in public addiction treatment systems: Context and effects on access, retention, selection and outcomes

Abstract

Aims: Value-based purchasing (VBP) aims to align program incentives and purchaser goals, often to improve quality of care, yet is uncommon in addiction treatment systems. Maine is the only state currently using an incentivized contract (IC) in its public addiction treatment system. We aim to understand the context in which the IC was implemented and determine whether rewarded measures (access and retention), client selection, and outcomes changed under the IC. Methods: We use state administrative data (N=37,470 admissions) and multivariate multilevel modeling with a difference-indifference approach and a nonIC comparison group. Propensity score techniques adjusted for pre-period differences in IC and nonIC admissions. Program director interviews collected information on activities implemented in response to the IC and more broadly to understand the context for any potential changes. Results: The probability of receiving 4 ormore outpatient treatment sessions did not change significantly between the 2005–07 pre-period and the 2008–11 post-period and was not significantly different in IC agencies and non-IC agencies (p= .43). Neither waiting time nor probability of remaining in treatment for over 90 days was significant. Admission rates of people with diagnosed mental disorders did not indicate selection effects. Early substance use andcriminal justiceoutcomes showed littledifferenceacross IC and non-IC groups over time. While most program directors acknowledged making changes to respond to the IC system, they quickly became acclimatized to the IC, with incentives not a large driver in changing how they approach care or do business. Non-IC programs are less likely to think of quality and outcomes measurement in a precise way. Conclusions: Introduction of an IC is a significant change in payment design that may affect addiction treatment service delivery, thus increasing our understanding of its effects is critical. We find the overall net effect of incentivized contract is very small and not significant. Financial support: NIDA R01 DA033402.