Association of disability with COVID-19 outcomes in older adults: a prospective analysis of the US health and retirement study

Wenjie Zeng, Inyoung Jun, Rosha A Loach, Aryaman Sriram, Nichole E. Stetten, Panayiotis V. Benos, Nicole M. MarlowBMJ 2025 · DOI

Abstract

Comprehensive strategies beyond vaccination are needed to decrease infections and prevent post-COVID conditions (PCC) for vulnerable ageing adults with functional disability. As the first study to prospectively investigate associations between functional disability, vaccination status, and COVID-19 outcomes using nationally representative data, these findings provide a robust foundation for future research and policy decisions to protect vulnerable populations during public health emergencies.

Introduction

While it has been widely observed that persons with disabilities (PWD) suffered disproportionately during the pandemic, most early research was descriptive or cross-sectional. This made it difficult to determine if disability itself was the primary driver of poor outcomes or if other factors like age, socioeconomic status, or vaccine access were responsible.

This prospective study of over 9,000 older US adults reveals that functional disability is a potent, independent risk factor for both COVID-19 infection and post-COVID conditions (PCC). While vaccination remains highly effective at preventing initial infection, the research highlights a sobering reality: for those with disabilities, vaccination does not appear to reduce the significantly elevated risk of developing long-term "Long COVID" symptoms.

Methods and Data

The researchers utilized data from the Health and Retirement Study (HRS), a nationally representative longitudinal survey of US adults aged 50 and older. The study tracked outcomes from the 2018 baseline through the 2021 Perspectives on the Pandemic Mail-in Survey.

Study flow diagram illustrating participant selection and analysis stages.

Functional disability was defined as having difficulty with Activities of Daily Living (ADLs) or Instrumental ADLs (IADLs). To move beyond simple correlations, the team employed:

  • Multivariate Logistic Regression: To adjust for age, race, income, and chronic conditions.
  • Propensity-Score (PS) Weighting: To balance characteristics between PWD and persons without disabilities (PWoD).
  • Targeted Maximum Likelihood Estimation (TMLE): To assess associations within specific vaccination strata.
  • Causal Mediation Analysis: To quantify how much of the disability effect was mediated by vaccination status.

Key Insights

  • Disability as an Independent Driver: Even after controlling for age and comorbidities, PWD faced a 28% to 39% higher risk of infection.
  • The Vaccination Paradox: Vaccination was a powerful shield against initial infection (reducing risk by nearly half), but it failed to protect against PCC. For those infected, being vaccinated did not statistically lower the odds of lingering symptoms if they had a baseline disability.
  • Minimal Mediation: Only a tiny fraction of the risk associated with disability was explained by vaccination status (11.5% for infection and a negligible 1.4% for PCC). This suggests that the vulnerability of PWD is rooted in biological or structural factors—such as higher viral load exposure or different immune responses—that vaccines alone cannot address.
Causal mediation analysis of the relationship between baseline disability and post-COVID conditions (PCC).

Key Results

  • Infection Risk: Baseline functional disability was associated with an Odds Ratio (OR) of 1.28 (95% CI 1.01 to 1.64) in adjusted models, rising to 1.39 (95% CI 1.03 to 1.88) with propensity-score weighting.
  • PCC Risk: The association was even stronger for post-COVID conditions, with an OR of 2.24 (95% CI 1.05 to 4.77), increasing to 2.73 (95% CI 1.31 to 5.71) after PS weighting.
  • Vaccine Efficacy: Vaccination was strongly protective against infection (OR: 0.53), but showed no significant association with a reduction in PCC (OR: 0.89, p > 0.05).
  • Acute Severity: Interestingly, disability was not significantly associated with increased ER visits or hospitalizations in this cohort, suggesting the primary "disability penalty" was the risk of infection and subsequent chronic conditions.

Limitations and Future Directions

The study relies on self-reported data, which may introduce recall bias. Additionally, breaking down disability by specific types resulted in small sub-groups with wide confidence intervals.

The authors emphasize that public health policy must move "beyond the needle" to include better environmental protections and clinical monitoring for PWD. Future research should utilize clinical records and longer-term follow-ups to better understand the trajectory of "Long COVID" in aging populations with functional limitations.

Summary updated

Abstract

Comprehensive strategies beyond vaccination are needed to decrease infections and prevent PCC for vulnerable ageing adults with functional disability. As the first study to prospectively investigate associations between functional disability, vaccination status and COVID-19 outcomes using nationally representative data, these findings provide a robust foundation for future research and policy decisions to protect vulnerable populations during public health emergencies.
BibTeX
@inproceedings{zeng2025association,
  author = {Wenjie Zeng and Inyoung Jun and Rosha A Loach and Aryaman Sriram and Nichole E. Stetten and Panayiotis V. Benos and Nicole M. Marlow},
  title = {{Association of disability with COVID-19 outcomes in older adults: a prospective analysis of the US health and retirement study}},
  booktitle = {BMJ},
  year = {2025},
  doi = {10.1136/bmjph-2025-003116},
}
Figure