Aging
Evaluating time-varying pain status and memory: Evidence from the Health and Retirement Study Yulin Yang* Yulin Yang Yang Yang Yang Yang Yang Yang Yang University of California, San Francisco
The relationship between pain and cognitive decline is unclear, in part because of the fluctuating nature of pain. We evaluated time-varying high impact pain (HIP) on cognitive, accounting for time-varying confounding.
We analyzed data from 14,559 Health and Retirement Study participants over 12 biennial waves (1996-2018). The analytic sample consisted of participants aged 60 years or older at the end of follow-up. Baseline was defined as each participant’s first wave with a valid cognitive assessment using the 10-word immediate- and delayed-recall tests. At each wave, HIP was defined as pain that limits every day activities (e.g., work or household chores) with the reference group of no pain/low impact pain. Memory was operationalized as the sum of immediate and delayed 10-word recall (range: 0-20). We fit linear mixed-effect models to estimate the association between lagged HIP and memory scores two years later, adjusted for baseline age, gender, race and ethnicity, and education. Stabilized inverse probability (IP) weights were estimated using logistic regression to account for baseline and time-varying confounding.
The average baseline age was 57.6 [SD=7.9] years (57.8 for those without HIP, and 57.2 with HIP). In unweighted models, HIP was associated with worse memory two years later (β= -0.24; 95% CI: -0.30,-0.18). After accounting for time-varying confounding, estimates for recent symptoms were attenuated (β= -0.05; 95% CI: -0.09, 0.05).
High-impact pain was associated with worse memory performance in unweighted models, but this association was substantially attenuated after accounting for time-varying confounding. These findings suggest that previously observed links between pain and memory loss may partly reflect dynamic health and social processes that co-evolve with pain over time. Future research should continue to model pain as a dynamic process and examine how cumulative exposure, severity, and specific pain progression types play a role in cognitive aging.