Labor Market Outcomes of Individuals in Recovery from Addiction to Alcohol

Aldridge, Arnie

Labor Market Outcomes of Individuals in Recovery from Addiction to Alcohol Public Deposited

Analytics

Download PDF

Request Accessible Version

Last Modified

March 22, 2019

Creator

Aldridge, Arnie
- Affiliation: College of Arts and Sciences, Department of Economics

Abstract

The majority of the cost burden of Alcohol Use Disorders (AUDs) is due to alcohol's adverse impact on the labor market in the form of lost wages for those not employed and decreased productivity for those employed. In this study, I develop a model of employment, drinking, and treatment-seeking that is based on an economic model of individual behavior. The model is estimated using longitudinal data on individuals from COMBINE, a National Institute of Alcoholism and Alcohol Abuse and Drug Abuse (NIAAA) randomized control trial of two pharmacotherapies and a cognitive behavioral intervention for dependence. The first aim of this study is to estimate the causal effects of AUD outcomes on employment over a three-year period following the COMBINE trial. The second aim is to estimate the effects of employment outcomes on subsequent drinking. The third aim is to evaluate the role of ongoing therapy for AUDs. To this end, I develop a dynamic model that attempts to control for time varying and permanent individual heterogeneity and uses an identification strategy to reduce any bias from the endogenous relationships across these outcomes. Within this framework, I also evaluate several policy experiments related to the price of consumption goods and treatment as well as policies around treatment dosage. The results indicate that drinking behavior during treatment (in this case, the COMBINE trial) have large and lasting effects on subsequent drinking, though the effects stabilize over the long term. Full time employment over 90% of the period leads to a slight increase in problem drinking. A ten percent increase in gasoline prices leads to a 3.6 percentage points increase in the probability of abstinence (p=.003). A simulated experiment of doubling the prescription period for pharmacotherapy has an unequivocally positive effect on drinking outcomes. Here, abstinence increases by 4.6 percentage points (p50% of the period (PDH) decreased by 2.6 pp (p=.005). These results improve our understanding of how trials for treatment of alcohol use disorders can be evaluated, particularly to understand how changes in alcohol consumption translate into employment outcomes and then used to inform policy decisions.

Date of publication