Skip to main content
. Author manuscript; available in PMC: 2021 May 1.
Published in final edited form as: J Health Econ. 2020 Mar 4;71:102286. doi: 10.1016/j.jhealeco.2019.102286

Table 5:

Opioid-Related Substance Abuse Treatments, Ages 12–54

Outcome: Opioid-Related Treatment Admissions Per 100,000
% Elderly2003 × Post (1) (2) (3) (4) (5) (6)
11.540**
(4.699)		 11.925***
(2.679)		 10.918***
(2.880)		 9.852**
(4.825)		 9.849***
(3.375)		 8.571***
(3.098)
State time-varying controls × Year Fixed Effects No Yes Yes Yes Yes Yes
No No No Yes No No
Policy Variables No No Yes Yes Yes Yes
Sample Full Full Full Full Balanced Balanced
Population All All All All All No SSDI
Mean Outcome: 86.69 86.69 86.69 86.69 87.91 82.13
N 587 587 587 587 516 516

Notes:

***

Significance 1%,

**

Significance 5%,

*

Significance 10%.

Standard errors in parentheses adjusted for clustering at state level. All regressions weighted by population. Controls included in all models but not shown: state fixed effects and year fixed effects. State time-varying controls include the unemployment rate, % white, 6 age group shares, % no college, and % some college (but no degree). When these covariates are interacted with year indicators, the age group shares are not included due to collinearity concerns (given the interaction term of interest). Instead, we also include the 2003 share ages 25–44 interacted with year indicators. Policy variables include whether the state has a PDMP, a medical marijuana law, legal and operational medical marijuana dispensaries, and pain clinic regulations. “Balanced” uses the sample of states reporting to TEDS in all years 2000–2011. The “No SSDI” population excludes individuals reporting labor force participation of “Retired/Disabled.”.