Understanding the Association between Access to Non-Pharmacologic Pain Providers and High-Risk Opioid Prescription Patterns Among Medicare Beneficiaries with Chronic Musculoskeletal Pain

Introduction: Chronic pain is a prevalent and costly condition with a major impact on the lives of older adults. The Centers for Disease Control’s Guideline for Prescribing Opioids for Chronic Pain recommends non-pharmacologic treatments, such as physical therapy and psychotherapy, over pharmacologic treatments, like opioids, for chronic pain management. Non-pharmacologic treatments are underused, and opioids are overused. Among older adults, high-risk opioid use patterns such as long-term use and high doses have emerged. Individual risk factors such as demographics, comorbidities, and initial opioid prescription characteristics only partly explain high-risk opioid use. Contextual factors, defined as the characteristics of the groups that individuals belong to (e.g. their providers, care setting, or community) may explain opioid use. Access to non-pharmacologic treatments is an understudied but potential contextual risk factor for high-risk opioid use. This study examines the association between the supply of non-pharmacologic providers and high-risk opioid prescription patterns among older adults with chronic musculoskeletal pain.

Methods: In this retrospective cohort study, we used individual level data from Medicare claims from a 5% sample of fee for service beneficiaries with Part D and county level data from the Area Health Resource File (2006-2014). Adults older than 65 with chronic musculoskeletal pain were included. We defined access to non-pharmacologic pain management as the supply of physical therapists and mental health providers in a county. We defined high-risk prescription patterns as 1) long-term use (≥90 days supply of opioids) and 2) high-dose of opioids (≥ 90 milligrams morphine equivalent). We measured the outcomes 1 year after the index date. We used multi-level models to estimate the association between the supply of non-pharmacologic providers and high-risk opioid prescription patterns, and adjusted for individual and county characteristics.

Results: We report the demographic characteristics of older adults with chronic musculoskeletal pain and the estimated associations between the supply of non-pharmacologic providers and high-risk opioid prescription patterns.

Conclusions: The study’s findings improve the understanding of how the structure of the health care system may influence individual prescription opioid use. As the US population ages and the burden of chronic pain increases, policies aimed at reducing opioid use without providing access to comprehensive pain care may ineffectively prevent high-risk opioid use and leave individuals vulnerable to undertreated pain. The findings may inform policies which address the health care system’s capacity to support individuals with chronic pain by providing alternatives to opioids.