Pain, sleep complaints, and depressive symptom clusters and their association with cognitive decline in older Mexican...

Older Mexican Americans, a fast-growing segment of the U.S. population, have high rates of pain, depression, and sleep complaints. Data are lacking on how these symptoms overlap and affect cognition in this population. We examined patterns of pain-sleep-depressive symptom clusters and their association with cognitive decline in older Mexican Americans.

We used data from the Hispanic Established Populations for the Epidemiologic Study of the Elderly (2010-2016), a study of older Mexican Americans in Southwestern U.S. We used latent class analysis to identify clusters based on pain on weight-bearing, depressive symptoms, and sleep complaints. We used linear mixed-effects models to examine association between different patterns of symptom clusters and cognitive decline, measured by the Mini-Mental State Examination.

We identified three latent classes (n=935): pain only (66.5%), pain with depression (21.0%), and pain with depression and sleep complaints (12.5%). The pain with depression class (β=-4.13, p<0.001) and the pain with depression and sleep complaints class (β=-2.17, p<0.001) had significantly lower baseline cognition compared to the pain only class. There were no significant differences in rate of decline across classes. Education was a significant effect modifier: In stratified analyses, among those with zero and one to six years of education, both the pain with depression and pain with depression and sleep complaints classes had significantly lower baseline cognition than the pain-only class. Among those with seven or more years of education, only the pain with depression class showed significantly lower baseline cognition compared to the pain-only class (β=-2.16, p=0.036). There were no significant differences in rate of decline across classes by education.

Co-occurrence of pain with depression or/and sleep complaints is a common and potentially modifiable contribution to poor cognition. Our findings highlight the need for targeted interventions addressing co-occurring conditions to mitigate cognitive decline among older Mexican Americans. Educational attainment modified these associations, with higher schooling attenuating baseline cognitive differences across symptom classes, consistent with a passive cognitive reserve framework, and should be considered when designing targeted cognitive monitoring and intervention strategies.