Background/Purpose: Osteoarthritis (OA) of the knee is a painful condition affecting ~13% of persons >65 years of age. OA is often accompanied by comorbidities. Effective pain control is important for managing this chronic disease. In the last decade, physicians have become increasingly aware of contraindications to NSAIDs due to comorbid conditions. We sought to examine whether physicians have begun to use opioids more frequently.

Methods:  We assembled national cohorts of individuals with knee OA using data from waves of the Medicare Beneficiary Survey (MCBS) administered in the years 2003, 2006, and 2009. We examined temporal trends in opioid prescriptions in these cohorts over time. The MCBS survey consists of a nationally representative sample of Medicare beneficiaries. The MCBS survey data can be linked to Medicare claims. Survey respondents provided data on demographics, health status and prescribed medications. We selected knee OA cohorts from each of the three years from community-dwelling MCBS respondents aged 65 and older who had at least one outpatient visit in Medicare Parts A or B with an ICD-9-CM diagnostic code of a) 715.x6 (OA of knee) or b) ICD-9 code of 719.46 (knee pain) plus 715.x8, 715.x9 or 715.x0 (OA at other specified sites, at multiple sites and at unspecified sites, respectively). Functional status was categorized as poor if the survey respondent indicated “a lot of difficulty” or “inability” to walk ¼ mile or 2-3 blocks. The following six comorbidity categories were assigned from Medicare Parts A or B claims: cancer, cardiovascular disease, musculoskeletal disease other than OA, diabetes, depression and chronic obstructive pulmonary disease (COPD). We included all prescribed medication records classified as opioids under the MCBS antiarthritic or analgesic categories. The outcome was defined as receiving at least one prescription for an opioid in the year of study. We used multivariate logistic regression to establish whether opioid use changed over time and identify factors leading to greater utilization of opioids in elderly persons with OA.

Results: 488 (5%) subjects from the 2003 MCBS survey, 477 (5%) from the 2006 MCBS and 422 (5%) from the 2009 MCBS were identified with knee OA. Mean age and sex were similar across years (77 years, 70% females). We found a significant increase in opioid prescribing between 2003 and 2009, with 31% of the patients receiving opioids in 2003, 39% in 2006 and 40% in 2009 (OR 1.5, 95% CI 1.1, 2.0 for both years 2006 and 2009 compared to  2003). Across all time periods, independent correlates of opioids use included: female sex (OR 1.5, 95% CI 1.2, 2.0), functional limitations (OR 2.1, 95% CI 1.7, 2.7), poor self-reported health status (OR 1.6, 95% CI 1.2, 2.0), COPD (OR 1.4, 95% CI 1.0, 1.8) and musculoskeletal disease in addition to OA (OR 1.9, 95% CI 1.2, 2.8).   

Conclusion: Opioid use in elderly persons with knee OA increased substantially between 2003 and 2009. The more frequent use of opioids is likely due to low efficacy and to toxicity of other classes of analgesics drugs currently used for pain relief in knee OA patients. As prevalence and incidence of knee OA continues to grow the public health impact of greater use of opioids should be monitored carefully.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/trends-in-prescription-of-opioids-from-2003-2009-in-persons-with-knee-osteoarthritis/