Examination of the Allostatic Load Construct and Its Longitudinal Association With Health Outcomes in the Boston Puerto Rican Health Study

Abstract

Objective: Despite evidence on allostatic load (AL) as a model explaining associations between stress and disease, there is no consensus on its operationalization. This study aimed to contrast various AL constructs and their longitudinal associations with disease and disability.

Methods: Baseline and 5-year follow-up data from 738 adults participating in the Boston Puerto Rican Health Study were used. Five AL scores were created by summing the presence of 21 dysregulated multisystem physiological parameters using the following: a) z scores, b) population-based quartile cutoffs, c) clinical-based cutoffs, d) 10 preselected clinical-based cutoffs (AL-reduced), and e) 12 clinical-based cutoffs selected a posteriori based on association with disease (AL-select). Adjusted logistic regression models examined associations between each AL score at baseline and 5-year incident type 2 diabetes (T2D), cardiovascular disease (CVD), activities (or instrumental activities) of daily living (ADL; IADL) for physical impairment, and cognitive impairment.

Results: AL-quartile was associated with greater odds of T2D (odds ratio [OR] = 1.20; 95% confidence interval [CI] = 1.07-1.35) and CVD (OR = 1.14; 95% CI = 1.06-1.22). AL-reduced was associated with higher odds of IADL (OR = 1.21; 95% CI = 1.07-1.37) and AL-clinical with CVD (OR = 1.14; 95% CI = 1.07-1.21), IADL (OR = 1.11; 95% CI = 1.04-1.19), and ADL (OR = 1.15; 95% CI = 1.04-1.26). AL-select showed associations with T2D (OR = 1.35; 95% CI = 1.14-1.61), CVD (OR = 1.21; 95% CI = 1.11-1.32), IADL (OR = 1.15; 95% CI = 1.04-1.26), and ADL (OR = 1.24; 95% CI = 1.08-1.41). No associations were found with AL z-score.

Conclusions: AL scores computed with clinical-based cutoffs performed robustly in our sample of mainland Puerto Ricans, whereas z scores did not predict disease and disability. AL-select was the most consistent predictor, supporting its use as a disease-predicting model. Future assessment of AL-select in other populations may help operationalize AL.