Background/Purpose: Over a third of older Americans experience chronic pain, typically occupying multiple musculoskeletal body sites. While number of pain sites (NPS) is often touted as a stronger predictor of quality of life and physical functioning than pain location, studies identifying commonly co-occurring pain locations included younger individuals, whose pain location patterns likely vary from older adults with comorbid pain conditions and activity limitations. This study seeks to identify pain location patterns in older adults to elucidate the role of pain location on age-related pain burden.

Methods: We used data from a community-based sample of older adults from an ancillary study of the Framingham Heart Study/Omni1. We collected data on number and distribution of pain sites (articular, non-articular, and axial) on validated body and joint homunculi. We used an agnostic approach to identify patterns of pain locations. Specifically, we conducted a latent class analysis and used traditional model fit statistics (e.g., BIC, AIC, G2) to select the optimal number of classes. After identifying the optimal model, we examined mental and physical health-related quality of life (HRQoL), physical functioning, and pain interference by pain location pattern (i.e., class). For function, we assessed the Nagi functional impairment scale, Rosow-Breslau mobility scale, and Katz activities of daily living (ADLs) index; for pain interference, we used the Brief Pain Inventory, and for HRQOL, we used the mental and physical component scores from the Short Form-12.

Results: We included 1610 participants (mean age, 76.1 ± 7.5 years; 56% Female). The model with 5 classes offered the best model fit. The 5 pain patterns identified in this model included: 1) minimal pain sites (MIN; 53.2%); 2. predominantly hand/wrist pain (Hand; 21.2%); 3. multisite joint and axial pain (JT/AX; 12.8%); 4. predominantly lower extremity pain (Predom LE; 7.3%); and 5. extensive articular and non-articular multisite pain (Extensive; 5.5%). All 5 classes reported some frequency of low back pain, ranging from 21% in the MIN class to 77% in the Extensive class with the Hand and Predom LE class reporting the same frequency (38%). The Extensive class reported the lowest (i.e., worst) HRQoL and physical functioning (Table 1), while the Predom LE class reported both the second-highest average NPS and the second-lowest scores on measures of function. Despite the Hand and JT/AX classes having similar NPS and pain interference scores, the JT/AX class reported worse physical HRQoL and functioning.

Conclusion: We identified pain location patterns with differential patterns of physical functioning, pain, and HRQoL. Overall, higher average NPS was associated with lower (worse) scores on measures of function and HRQoL, but JT/AX reported worse outcomes than the Hand class despite reporting the same mean NPS. Our data suggest that pain location contributes to pain burden beyond NPS. More specifically, lower extremity and/or lower back pain appears to contribute to poor physical HRQoL and functioning outcomes more so than predominantly hand pain.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/chronic-pain-phenotypes-in-older-adults-identifying-patterns-of-musculoskeletal-pain-and-associated-burden/