Using careful meta-analysis methods, the authors determined that about 4% of patients continued to use prescription opioids beyond 2 months after surgery or trauma. They also identified the following risk factors as being “among the most important predictors of prolonged opioid use” in these patients:
- Prior use of opioids or benzodiazepines
- Depression
- Long-duration hospital stay
- History of back pain
Mohamadi et al. also calculated a “number needed to harm” (NNH) from their data. NNH indicates the number of patients with a certain risk factor that is necessary to result in 1 person with prolonged opioid use beyond that of a patient population without that risk factor. They found that for every 3 patients with a history of opioid use, every 23 patients with a history of back pain, every 40 with depression, or every 62 with a history of benzodiazepine use, 1 patient will continue to use prescribed opioids for an extended time period.
Because this meta-analysis was derived from observational studies, the authors caution that “causal inferences could not be drawn for the proposed risk factors.” But they do offer a practical piece of advice gleaned from prior research: Provide patients with an opioid-tapering plan at the time of discharge to significantly reduce the likelihood of prolonged opioid use.