Association of opioid dose and use frequency with 12-month pain severity and function trajectories

Numerous risk factors for persistent pain severity and pain related interference have been identified but opioid use characteristics associated with pain trajectories are understudied. We determined if opioid dose and frequency of use were associated with non-cancer chronic pain trajectories.

Participants starting a new period of prescription opioid use lasting 30-90 days were recruited from 2 health care systems that deliver care in Michigan, Illinois, Missouri, Oklahoma, and Wisconsin. Participants completed structured surveys at baseline and completed up to 12 monthly surveys for PEG-assessed pain severity and interference. Growth mixture modeling was used to compute trajectories. Multinomial, logistic regression models estimated the association between patient opioid use characteristics and trajectory membership, after adjusting for covariates.

The sample (n = 842) was, on average, 53.8 ± 11.8 years of age, 68.8% female and 29.0% non-White race. The best fitting model was a 3-class solution with trajectories characterized as "mild-improving," "severe-improving," and "stable-severe." Opioid dose ≥50 morphine milligram equivalent and daily vs non-daily opioid use were not associated with these trajectories. More pain sites was significantly associated with severe improving and stable-severe vs the mild-improving trajectory. Higher income and ability to participate in social roles were inversely associated with membership in the -stable-severe trajectory.

In this sample with 30-90 days of opioid use at baseline, we identified 3 pain trajectories. Relatively few patient characteristics were associated with pain severity and interference trajectories. Opioid dose and frequency of use were not predictors of the course of non-cancer pain.