Prescription opioid use and opioid use disorder among older adults with HIV in the USA from 2008 to 2021: a retrospective repeated cross-sectional study.

Background: There is longstanding concern that people with HIV receive prescription opioids at higher rates and have a disproportionate burden of opioid use disorder (OUD) compared with their counterparts without HIV. We aimed to evaluate trends of opioid prescriptions and indicators of OUD in an understudied but growing population of older adults with HIV. Methods: For this retrospective repeated cross-sectional study, we constructed annual cohorts through a nationally representative sample of fee-for-service Medicare beneficiaries aged 65 years and older in the USA with Part D coverage (ie, prescription drug) enrolled between Jan 1, 2008, and Dec 31, 2021. Beneficiaries were eligible for inclusion in each cross-sectional cohort if they had reached the age of 65 years by Jan 1 of the calendar year and had 1 year of continuous Medicare enrolment in Part A (inpatient hospital care), B (outpatient care), and D. Beneficiaries with HIV were matched in a 1:3 ratio to beneficiaries without HIV on age, sex, race or ethnicity, US state, and dual eligibility status (Medicare and Medicaid). The main outcomes were receipt of at least one opioid prescription and any indicator of OUD (ie, formal diagnosis, medication for OUD, or opioid-related or emergency department visits) during each calendar year. Generalised estimating equations were used to estimate odds ratios (ORs) of each outcome, comparing matched beneficiaries with or without HIV. Due to data availability, our analysis of indicators of OUD was restricted to 2008-16. Findings: Across all years, 163 429 beneficiaries with HIV and 490 287 beneficiaries without HIV were included (475 516 [72·7%] were male, 178 200 [27·3%] were female; 305 776 [46·8%] were non-Hispanic White, 238 172 [36·4%] were Black [or African American], 84 128 [12·9%] were Hispanic, 8964 [1·4%] were Asian or Pacific Islander, and 16 676 [2·6%] were other races or ethnicities). During 2008-21, 57 373 (35·1%) of 163 429 people with HIV and 138 547 (28·3%) of 490 287 people without HIV received at least one opioid prescription. During 2008-16, 2408 (3·1%) of 76 637 people with HIV and 2831 (1·2%) of 229 911 people without HIV had any indicator of OUD. Across all analysed years, beneficiaries with HIV had significantly increased odds of receiving at least one opioid prescription (OR 1·38, 95% CI 1·36-1·39) and having indicators of OUD (2·61, 2·47-2·76) compared with their matched counterparts without HIV. Interpretation: Medicare beneficiaries aged 65 years and older with HIV in the USA were more likely to receive opioid prescriptions and have OUD indicators than matched beneficiaries without HIV. Our findings could help guide clinical opioid prescription guidelines and public health surveillance among this vulnerable ageing population. Funding: National Institute on Drug Abuse, New Jersey Alliance for Clinical and Translational Science, and National Center For Advancing Translational Sciences of the National Institutes of Health.