Vitality Watch

Dossier pubmed observational 28-sep-2026
Human cohort, case-control, or epidemiological association

Identifying older adults with heightened stress responsivity before public health crises: a retrospective cohort study of the US and England.

Key claim: Accessible pre-pandemic indicators—including poor self-rated health, loneliness, functional impairment, chronic disease burden, and lower physical activity—identify older adults with relatively elevated multidimensional stress responses during a public health crisis, with a broadly similar vulnerability pattern in US HRS and English ELSA cohorts.

Abstract

OBJECTIVE: Using COVID-19 as a natural experiment, this study aimed to develop a pre-crisis approach to identify older adults with relatively elevated stress responses and evaluate the robustness of the identified vulnerability profile across the HRS and ELSA cohorts. METHODS: Harmonized pre-pandemic (2018) longitudinal data of 11,533 seniors from U.S. HRS and English ELSA cohorts were analyzed. The primary outcome was the Pandemic Stress Response Index (PSRI), constructed separately within each cohort as a conceptually harmonized measure of relative multidimensional stress response. Weighted LASSO with stability selection screened stable predictors, followed by a pooled weighted multivariable logistic regression adjusted for cohort/country fixed effect. Weighted AUC, Brier score, and diagnostic indices assessed overall model performance, with descriptive cohort-specific performance evaluated using predictions from the same final model. RESULTS: The final sample included 11,533 participants (6,185 HRS; 5,348 ELSA), of whom 2,931 (25.4%) were classified into the relative high-stress group. Fourteen predictors were retained by stability selection. Poor self-rated health, perceived loneliness, functional impairment, chronic disease burden, visual impairment, higher BMI, and female sex/gender were associated with higher odds of high stress response, whereas older age, higher life satisfaction, greater household wealth, better cognitive function, and greater physical activity were associated with lower odds. The pooled AUC was 0.729 (0.703 in HRS and 0.764 in ELSA). CONCLUSIONS: Accessible pre-emergency indicators may help identify older adults with relatively elevated stress responses. The broadly similar vulnerability pattern observed across HRS and ELSA supports further evaluation of multidimensional pre-crisis vulnerability profiling for public health preparedness.

Why this matters

Most chronic-stress and loneliness evidence in older adults is cross-sectional or links exposures to distal mortality endpoints; this analysis instead tests whether cheap, routinely collected pre-crisis measures prospectively identify who will mount a heightened stress response when a population-wide stressor hits. Replication of the same vulnerability pattern across HRS (US) and ELSA (England) strengthens the signal that the loneliness/functional-health/chronic-disease cluster is a cross-nationally stable marker of stress-responsivity risk, not a cohort artifact. It remains observational and the stress-response outcome is crisis-period and multidimensional rather than a hard aging endpoint.

Source

Link

Briefing card

Identifying older adults with heightened stress responsivity before public health crises: a retrospective cohort study of the US and England.
Stage: observational · pubmed · 28-sep-2026

If you are an older adult entering a public health crisis with pre-existing loneliness, poor self-rated health, functional impairment, chronic disease, or low physical activity, this analysis found you were more likely to show elevated stress responses during the pandemic in both US and English cohorts.

Accessible pre-pandemic indicators—including poor self-rated health, loneliness, functional impairment, chronic disease burden, and lower physical activity—identify older adults with relatively elevated multidimensional stress responses during a public health crisis, with a broadly similar vulnerability pattern in US HRS and English ELSA cohorts.

Cross-references (0)

None recorded — doctrine links and citations appear here as scans and citation sweeps find them.

External: doi:10.1080/13607863.2026.2735468

build build 880 · 2bcef25 · 2026-11-18