Skip to main content

Mortality From Winter Storm Uri and Texas Power Outages Among Medicare Patients

In brief

Winter Storm Uri linked to 3 excess deaths per 10,000 Medicare patients weekly

Among Texas Medicare beneficiaries, mortality rose by about 3 deaths per 10,000 per week during the four weeks after the storm began. Hospitalizations also increased over that period, and low-income beneficiaries had the largest relative rise in deaths; counties with the fewest power outages saw no clear mortality increase. The findings underscore the risks of outages, but do not establish that they caused the excess deaths.

Journal
JAMA network open (Q1)
Published
2 November 2026
Study design
Non-randomized / quasi-experimental trial
Evidence level
Level 2, Moderate (CEBM 2b)
Authors
Nicholas L Berlin, Jessica Phelan, Joshua Calianos, E John Orav, Thomas C Tsai, Renee N Salas, et al.
PMID
43062828
DOI
10.1001/jamanetworkopen.2026.45376

Why clinicians should know about it

  • Picked for Epidemiology (top studies of the week, 22 November 2026): Quasi-experimental natural disaster and power outage mortality study

Abstract

IMPORTANCE: In February 2021, a severe freeze event and winter storm, together named Winter Storm Uri, caused widespread power outages across Texas, leaving millions vulnerable to record-breaking cold temperatures. The association of these events and the related power outages with population-level health outcomes is unclear. OBJECTIVE: To estimate the association of Winter Storm Uri and related power outages with mortality and hospitalizations among Medicare beneficiaries in Texas through a quasi-experimental design. DESIGN, SETTING, AND PARTICIPANTS: This cohort study used data on a 100% sample of Texas Medicare beneficiaries from 2019 to 2023. The week of Winter Storm Uri was defined as February 14 to February 20, 2021. Statistical analyses were carried out between August 2025 and June 2026. EXPOSURES: Winter Storm Uri and related power outages in February 2021. MAIN OUTCOMES AND MEASURES: The primary outcome was weekly county-level all-cause mortality in Texas counties among Medicare beneficiaries, compared with the change observed in the same counties in the following year. The secondary outcome was county-level all-cause hospitalizations, measured identically. Difference-in-differences regression models estimated excess mortality and hospitalization rates 4 weeks before and during the week of the storm. Analyses were repeated for the 4 weeks from storm onset. For both outcomes, subgroup analyses were stratified by power outage quartile and by characteristics for vulnerable populations. RESULTS: The main study sample comprised Texas counties with Medicare beneficiaries in 2021 (n=3 955 190; 50.2% aged 65-74 years; 54.1% female) and the main control sample comprised the same counties in 2022 (n=4 053 788; 49.9% aged 65-74 years; 54.1% female). During the week of Winter Storm Uri, weekly mortality increased by 2.94 (95% CI, 2.12-3.76) deaths per 10 000 beneficiaries. During the 4-week period from storm onset, excess mortality persisted, with 3.08 (95% CI, 2.72-3.44) excess deaths per 10 000 beneficiaries per week. Hospitalizations were not higher during the storm week; however, over the 4-week period from storm onset, hospitalizations were 2.04 (95% CI, 1.30-2.78) higher per 10 000 beneficiaries per week. Although there was widespread storm activity throughout the state, no increase in mortality was observed in counties with the fewest power outages, with 0.66 (95% CI, -2.16 to 3.48) excess deaths per 10 000 beneficiaries per week. Medicare beneficiaries with low income had the highest relative increase in deaths during the storm week, with 5.40 (95% CI, 2.82-7.99) excess deaths per 10 000 beneficiaries. CONCLUSIONS AND RELEVANCE: In this cohort study of Texas Medicare beneficiaries, Winter Storm Uri was associated with excess mortality and hospitalizations, particularly in counties with power outages and among Medicare beneficiaries with low income. These findings highlight the need for targeted public health and policy efforts to protect vulnerable Medicare populations from the growing health consequences of weather-related events.

Abstract as published, via PubMed.

View on PubMedFull text at the publisherOpen in the app

For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.