Working Paper: NBER ID: w25791
Authors: Justine S. Hastings; Mark Howison; Sarah E. Inman
Abstract: Misuse of prescription opioids is a leading cause of premature death in the United States. We use new state government administrative data and machine learning methods to examine whether the risk of future opioid dependence, abuse, or poisoning can be predicted in advance of an initial opioid prescription. Our models accurately predict these outcomes and identify particular prior non-opioid prescriptions, medical history, incarceration, and demographics as strong predictors. Using our model estimates, we simulate a hypothetical policy which restricts new opioid prescriptions to only those with low predicted risk. The policy’s potential benefits likely outweigh costs across demographic subgroups, even for lenient definitions of “high risk.” Our findings suggest new avenues for prevention using state administrative data, which could aid providers in making better, data-informed decisions when weighing the medical benefits of opioid therapy against the risks.
Keywords: opioid prescriptions; machine learning; predictive modeling; public health; policy implications
JEL Codes: D61; I1; I12; I18; Z18
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| prior non-opioid prescriptions (L42) | risk of future opioid dependence (I12) |
| medical history (N30) | risk of future opioid dependence (I12) |
| incarceration status (K14) | risk of future opioid dependence (I12) |
| demographic information (J10) | risk of future opioid dependence (I12) |
| prior non-opioid prescriptions (L42) | risk of opioid abuse (I12) |
| medical history (N30) | risk of opioid abuse (I12) |
| incarceration status (K14) | risk of opioid abuse (I12) |
| demographic information (J10) | risk of opioid abuse (I12) |
| prior non-opioid prescriptions (L42) | risk of opioid poisoning (I12) |
| medical history (N30) | risk of opioid poisoning (I12) |
| incarceration status (K14) | risk of opioid poisoning (I12) |
| demographic information (J10) | risk of opioid poisoning (I12) |
| predictive models (C52) | opioid prescription policy outcomes (L42) |