C. M. Harbaugh1,5, J. S. Lee1, K. Chua3,6, B. Kenney5, T. J. Iwashyna2,5, M. J. Englesbe1,5, C. M. Brummett5,7, A. S. Bohnert4,5, J. F. Waljee1,5 1University Of Michigan,Department Of Surgery,Ann Arbor, MI, USA 2University Of Michigan,Department Of Internal Medicine, Division Of Pulmonary Critical Care Medicine,Ann Arbor, MI, USA 3University Of Michigan,Department Of Pediatrics And Communicable Diseases,Ann Arbor, MI, USA 4University Of Michigan,Department Of Psychiatry,Ann Arbor, MI, USA 5University Of Michigan,Institute For Healthcare Policy And Innovation,Ann Arbor, MI, USA 6University Of Michigan,Child’s Health Evaluation And Research Center,Ann Arbor, MI, USA 7University Of Michigan,Department Of Anesthesia, Division Of Pain Medicine,Ann Arbor, MI, USA
Introduction: Opioid-naïve teens prescribed opioids after surgery have a substantial risk of developing prolonged opioid use. Prior work has shown that this risk is associated with individual-level factors such as mental health disorders and prior substance use. It is unknown whether the risk of persistent use varies with exposure to chronic opioid use among family members.
Methods: We performed a retrospective analysis of 2010-2016 Truven MarketScan Commercial Claims including opioid-naïve patients aged 13–21 years who had surgery without subsequent procedure or anesthesia, perioperative opioid prescription, and dependent status with ≥1 family member on the same insurance plan (N=257,550). The dependent variable was new persistent opioid use (NPOU, ≥1 prescription 91–180 days after surgery). The main independent variable was household chronic opioid use (≥120 days or ≥3 prescriptions within 90 days in the year before surgery among any family member). We used generalized estimating equations with robust standard errors at the family level to model NPOU as a function of household chronic use, controlling for patient demographics; patient mental health and chronic pain conditions; and household mental health and chronic pain conditions. Sensitivity analysis evaluated the association with the number of adult (primary insurance holder/spouse) and youth (other dependent) family members with chronic opioid use. Average marginal effect for the primary outcome was calculated using observed values.
Results: In this cohort, 4.3% of patients (11,087) have a household member with chronic opioid use with an unadjusted NPOU rate of 2.5% overall. Patients and family members with household chronic opioid use were more likely to have chronic pain, mental health, and opioid use disorders compared to families without any chronic opioid use. The adjusted odds of NPOU were 57% higher among patients with any household chronic use (aOR 1.57, 95% CI 1.42 – 1.74). The adjusted rate of NPOU was 2.4% for patients with no household chronic use compared to 4.1% for patients with household chronic use (Figure). On sensitivity analysis, NPOU was significantly associated with chronic opioid use among adults in a dose-dependent manner (one adult: aOR 1.55, 95% CI 1.40 – 1.72; two adults: aOR 1.63, 95% CI 1.12 – 2.37), but not other youth (aOR 1.49, 95% CI 0.89 – 2.49).
Conclusion: The risk of persistent opioid use after surgery was significantly higher among adolescents who had family members with chronic opioid use. This suggests that household opioid use patterns should be assessed to determine which patients might require closer monitoring and heightened anticipatory guidance when opioids are prescribed.
