Sociodemographic and Clinical Predictors of Prescription Opioid Use in a Longitudinal Community-Based Cohort Study of Middle-Aged and Older Adults

Objectives: Identifying factors associated with opioid use in middle-aged and older adults is a fundamental step in the mitigation of potentially unnecessary opioid consumption and opioid-related harms. Methods: Using longitudinal data on a community-based cohort of adults aged 50–90 years residing in Johnston County, North Carolina, we examined sociodemographic and clinical factors in non-opioid users (n = 786) at baseline (2006–2010) as predictors of opioid use at follow-up (2013–2015). Variables included age, sex, race, obesity, educational attainment, employment status, household poverty rate, marital status, depressive symptoms, social support, pain catastrophizing, pain sensitivity, insurance status, polypharmacy, and smoking status. Results: At follow-up, 13% of participants were using prescription opioids. In the multivariable model, high pain catastrophizing (adjusted odds ratio; 95% confidence interval = 2.14; 1.33–3.46), polypharmacy (2.08; 1.23–3.53), and history of depressive symptoms (2.00; 1.19–3.38) were independent markers of opioid use. Discussion: Findings support the assessment of these modifiable factors during clinical encounters in patients ≥ 50 years old with chronic pain