Skip to main content
. Author manuscript; available in PMC: 2022 Mar 1.
Published in final edited form as: J Health Econ. 2021 Jan 22;76:102430. doi: 10.1016/j.jhealeco.2021.102430

Appendix Table 2:

Average Effect Estimates for Percentage Working and Percentage Applying for Disability Insurance, Conditional on Census Division-Year Interactions

Panel A Outcome: Percentage Working (BEA)
(1) (2) (3)
OxyContin Misuse −0.034** (0.015) −0.040*** (0.015) −2.199** (1.086)
Pain Reliever Misuse 0.007 (0.005) 0.009* (0.005) 0.414 (0.359)
OLS/Poisson Outcome OLS Log Poisson Total OLS Level
Panel B Outcome: % Apply for Disability
(1) (2) (3)
OxyContin Misuse 0.189 (0.144) 0.287* (0.149) 0.209 (0.163)
Pain Reliever Misuse −0.016 (0.039) −0.049 (0.038) −0.023 (0.044)
OLS/Poisson Outcome OLS Log Poisson Total OLS Level

Notes:

***

1% significance,

**

5% significance,

*

10% significance.

All models include state and Census division-time fixed effects plus covariates mentioned in notes of Figure 5. Estimates presented are average effects for each pre-reformulation misuse variable using equation (2). When Poisson estimation is used, the outcome is the total and population is used as an exposure variable. N=765. Standard errors are adjusted state-level clustering.