Patterns of pain medication usage and self-reported pain in older Irish adults with osteoarthritis: A latent class analysis of data from the Irish Longitudinal Study on Ageing

BackgroundThis study aimed to identify and describe links between pain medication use and self-reported pain among people aged >= 50 years with osteoarthritis (OA) in an Irish population, and to examine the relationships between pain, medication usage and socioeconomic and clinical characteristics.MethodsSecondary data analysis of wave 1 cross-sectional data from The Irish Longitudinal Study on Ageing (TILDA) was undertaken of 1042 people with self-reported doctor-diagnosed OA. We examined use of medications typically included in OA clinical guidelines, including non-opioid analgesics (e.g. paracetamol), topical and oral non-steroidal anti-inflammatory drugs (NSAIDs), opioids and nutraceuticals. Latent Class Analysis (LCA) was used to identify underlying clinical subgroups based on medication usage patterns, and self-reported pain severity. Multinomial logistic regression was used to explore sociodemographic and clinical characteristic links to latent class membership.ResultsA total of 358 (34.4%) of the 1042 people in this analysis were taking pain medications including oral NSAIDs (17.5%), analgesics (11.4%) and opioids (8.7%). Nutraceutical (glucosamine/chondroitin) use was reported by 8.6% and topical NSAID use reported by 1.4%. Three latent classes were identified: (1) Low medication use/no pain (n = 382, 37%), (2) low medication use/moderate pain (n = 523, 50%) and (3) moderate medication use/high pain (n = 137, 13%). Poorer self-rated health and greater sleep disturbance were associated with classes 2 and 3; depressive symptoms and female gender were associated with class 2, and retirement associated with class 3.ConclusionsWhilst pain medication use varied with pain severity, different medication types reported broadly aligned with OA guidelines. The two subgroups exhibiting higher pain levels demonstrated poorer self-rated health and greater sleep disturbance. Latent class analysis identified different subgroups based on pain severity and pain medication use, which may help to identify more targeted individualised pain management interventions.