Unemployment rate, opioids misuse and other substance abuse: quasi-experimental evidence from treatment admissions data

Abstract

Abstract Background The relationship between economic conditions and substance abuse is unclear, with few studies reporting drug-specific substance abuse. The present study examined the association between economic conditions and drug-specific substance abuse admissions. Methods State annual administrative data were drawn from the 1993–2016 Treatment Episode Data Set. The outcome variable was state-level aggregate number of treatment admissions for six categories of primary substance abuse (alcohol, marijuana/hashish, opiates, cocaine, stimulants, and other drugs). Additionally, we used a broader outcome for the number of treatment admissions, including primary, secondary, and tertiary diagnoses. We used a quasi-experimental approach -difference-in-difference model- to estimate the association between changes in economic conditions and substance abuse treatment admissions, adjusting for state characteristics. In addition, we performed two additional analyses to investigate (1) whether economic conditions have an asymmetric effect on the number of substance use admissions during economic downturns and upturns, and (2) the moderation effects of economic recessions (2001, 2008–09) on the relationship between economic conditions and substance use treatment. Results The baseline model showed that unemployment rate was significantly associated with substance abuse treatment admissions. A unit increase in state unemployment rate was associated with a 9% increase in treatment admissions for opiates (β = 0.087, p < .001). Similar results were found for other substance abuse treatment admissions (cocaine (β = 0.081, p < .001), alcohol (β = 0.050, p < .001), marijuana (β = 0.036, p < .01), and other drugs (β = 0.095, p < .001). Unemployment rate was negatively associated with treatment admissions for stimulants (β = − 0.081, p < .001). The relationship between unemployment rate and opioids treatment admissions was not statistically significant in models that adjusted for state fixed effects and allowed for a state- unique time trend. We found that the association between state unemployment rates and annual substance abuse admissions has the same direction during economic downturns and upturns. During the economic recession, the negative association between unemployment rate and treatment admissions for stimulants was weakened. Conclusion These findings suggest that economic hardship may have increased substance abuse. Treatment for substance use of certain drugs and alcohol should remain a priority even during economic downturns.