Aggregate difference-in-differences estimates
| . | (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . |
|---|---|---|---|---|---|---|---|---|
| OxyContin Misuse | 1.642* | 1.672** | 1.444* | 1.442* | 1.558* | 1.283* | 1.763** | 1.628* |
| (0.651) | (0.627) | (0.694) | (0.647) | (0.606) | (0.573) | (0.649) | (0.661) | |
| Pain Reliever Misuse | 0.163 | 0.206 | ||||||
| (0.184) | (0.175) | |||||||
| Pre-Period | 2006–2009 | 2006–2009 | 2006–2009 | 2006–2009 | 2006–2010 | 1999–2009 | 2006–2009 | 2006–2009 |
| States | All | All | All | All | All | All | No FL | No WV |
| Time-Varying Controls | No | Yes | No | Yes | Yes | Yes | Yes | Yes |
| . | (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . |
|---|---|---|---|---|---|---|---|---|
| OxyContin Misuse | 1.642* | 1.672** | 1.444* | 1.442* | 1.558* | 1.283* | 1.763** | 1.628* |
| (0.651) | (0.627) | (0.694) | (0.647) | (0.606) | (0.573) | (0.649) | (0.661) | |
| Pain Reliever Misuse | 0.163 | 0.206 | ||||||
| (0.184) | (0.175) | |||||||
| Pre-Period | 2006–2009 | 2006–2009 | 2006–2009 | 2006–2009 | 2006–2010 | 1999–2009 | 2006–2009 | 2006–2009 |
| States | All | All | All | All | All | All | No FL | No WV |
| Time-Varying Controls | No | Yes | No | Yes | Yes | Yes | Yes | Yes |
Notes: Standard errors in parentheses are adjusted for clustering at the state level. The outcome is suicides per 100,000 for ages 10–17 by demographic group. The post-period is 2011–2020. The pre-reformulation suicide rate (2006–2009) is 2.794 per 100,000. The estimated specification included state × race × age × sex fixed effects and year × race × age × sex × census region fixed effects and was weighted by population size. Time-varying controls included fraction of the population that is White, fraction that is Black, the minimum wage, anti-bullying laws, PDMPs, and “mandatory access” PDMPs. Columns 3 and 4 include an additional control for the pre-reformulation pain reliever misuse rate (excluding OxyContin) interacted with an indicator equal to 1 for 2011–2020. Each point estimate represents the coefficient on the nonmedical OxyContin misuse (%) variable for 2011–2020, except for the point estimate related to the pain reliever misuse rate. FL = Florida, WV = West Virginia.
p < .05; **p < .01