Extreme Heat Exposure and Dementia-Related Mortality: A Systematic Review and Meta-Analysis
- 1Department of Counseling Psychology, Shinhan University, Uijeongbu 11644, Gyeonggi-do, Republic of Korea
- 2Department of Big Data Management, Shinhan University, Uijeongbu 11644, Gyeonggi-do, Republic of Korea
- 3Department of Dental Hygiene, Shinhan University, Uijeongbu 11644, Gyeonggi-do, Republic of Korea
Abstract
Background/Objectives: The world is undergoing rapid population aging while experiencing more frequent and intense heatwaves driven by climate change. People living with dementia—who can have impaired thermoregulation, a reduced ability to perceive and respond to heat, and high care dependency—may be especially susceptible to heat, but the evidence on mortality has not been synthesized. We conducted a systematic review and meta-analysis to quantify the association between heatwave exposure and dementia-related mortality. Methods: PubMed, Web of Science Core Collection, ProQuest Dissertations & Theses Global, and APA PsycINFO were searched from inception to 15 June 2026; the searches were rerun on 15 October 2026 and supplemented by citation searching and by a PubMed search without the older-adult concept block. The review is reported according to PRISMA 2020 (PROSPERO CRD420261436457, registered after the searches had been run). Observational studies reporting relative effect estimates for mortality during defined heatwave or extreme heat events, among people with diagnosed dementia or for deaths attributed to dementia in mortality records, were eligible. Risk of bias was assessed with an adapted ROBINS-E approach and the certainty of evidence with GRADE. Odds ratios (ORs) from time-stratified case-crossover analyses, which estimate rate ratios, were pooled in a random-effects model with restricted maximum-likelihood estimation and Hartung–Knapp 95% confidence intervals (CIs); heterogeneity was described with Cochran’s Q, I2, and a 95% prediction interval. The primary analysis was restricted to studies with a dementia-specific outcome definition and an estimate of the total heatwave association and used one study from each pair of studies based on the same mortality data. Results: Six studies conducted in China and Australia met the eligibility criteria, one of them identified only by the supplementary search; three contributed to the primary analysis, and three were used in sensitivity analyses only. All study-level estimates exceeded unity, but the pooled association was imprecise and not statistically significant (OR 1.84, 95% CI 0.54–6.25), with high heterogeneity (I2 = 94.9%; 95% prediction interval 0.21–16.31). In sensitivity analyses that retained at least three studies, pooled ORs ranged from 1.50 to 1.84; analyses reduced to two studies had uninformative CIs. The CI excluded the null when an alternative national study was substituted (OR 1.76, 95% CI 1.13–2.72), when all six estimates were pooled without regard to overlapping data (OR 1.57, 95% CI 1.15–2.15), or when the conventional DerSimonian–Laird method was used. No included study compared dementia with non-dementia deaths, and only one was restricted to adults aged 60 years or older; the others were predominantly older or did not report age. The certainty of evidence was very low. Conclusions: The evidence is inconclusive regarding the magnitude and consistency of an association between heatwaves and dementia-related mortality. The study-level estimates point in the same direction, but the pooled estimate is an average across highly heterogeneous settings, is too imprecise to establish an effect, and cannot show whether the relative increase exceeds that for other deaths; evidence specific to older adults with dementia is largely indirect. Including people with dementia in heat–health action plans remains a reasonable precaution, and independent, methodologically harmonized primary studies are needed.
Related
- Prenatal extreme heat exposure and neonatal mortality in Bangladesh, 2015 to 2025: a national birth-history study
- Outdoor extreme heat-related dementia risk and the role of socioeconomic disparities: evidence from 20 years of data on US older adults.
- Trends in reported HIV diagnoses, exposure patterns, and AIDS-related mortality in the Eastern Mediterranean Region, 2014–2023
- Short-Term Mortality and Its Predictors in Alcohol-Related Acute-on-Chronic Liver Failure: A Systematic Review and Meta-Analysis.
- The Role of Greenness in Heat‑Related Mortality: A Systematic Review and Meta‑Analysis
BibTeX
@article{lee2026extreme,
title = {{Extreme Heat Exposure and Dementia-Related Mortality: A Systematic Review and Meta-Analysis}},
author = {Jung-Eun Lee and Gyu Tae Lee and Han-A Cho},
journal = {Journal of Clinical Medicine},
year = {2026},
month = nov,
doi = {10.3390/jcm15218655},
url = {https://www.mdpi.com/2077-0383/15/21/8655},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Extreme Heat Exposure and Dementia-Related Mortality: A Systematic Review and Meta-Analysis
AU - Jung-Eun Lee
AU - Gyu Tae Lee
AU - Han-A Cho
PY - 2026
DA - 2026/11/06
JO - Journal of Clinical Medicine
DO - 10.3390/jcm15218655
UR - https://www.mdpi.com/2077-0383/15/21/8655
AB - Background/Objectives: The world is undergoing rapid population aging while experiencing more frequent and intense heatwaves driven by climate change. People living with dementia—who can have impaired thermoregulation, a reduced ability to perceive and respond to heat, and high care dependency—may be especially susceptible to heat, but the evidence on mortality has not been synthesized. We conducted a systematic review and meta-analysis to quantify the association between heatwave exposure and dementia-related mortality. Methods: PubMed, Web of Science Core Collection, ProQuest Dissertations & Theses Global, and APA PsycINFO were searched from inception to 15 June 2026; the searches were rerun on 15 October 2026 and supplemented by citation searching and by a PubMed search without the older-adult concept block. The review is reported according to PRISMA 2020 (PROSPERO CRD420261436457, registered after the searches had been run). Observational studies reporting relative effect estimates for mortality during defined heatwave or extreme heat events, among people with diagnosed dementia or for deaths attributed to dementia in mortality records, were eligible. Risk of bias was assessed with an adapted ROBINS-E approach and the certainty of evidence with GRADE. Odds ratios (ORs) from time-stratified case-crossover analyses, which estimate rate ratios, were pooled in a random-effects model with restricted maximum-likelihood estimation and Hartung–Knapp 95% confidence intervals (CIs); heterogeneity was described with Cochran’s Q, I2, and a 95% prediction interval. The primary analysis was restricted to studies with a dementia-specific outcome definition and an estimate of the total heatwave association and used one study from each pair of studies based on the same mortality data. Results: Six studies conducted in China and Australia met the eligibility criteria, one of them identified only by the supplementary search; three contributed to the primary analysis, and three were used in sensitivity analyses only. All study-level estimates exceeded unity, but the pooled association was imprecise and not statistically significant (OR 1.84, 95% CI 0.54–6.25), with high heterogeneity (I2 = 94.9%; 95% prediction interval 0.21–16.31). In sensitivity analyses that retained at least three studies, pooled ORs ranged from 1.50 to 1.84; analyses reduced to two studies had uninformative CIs. The CI excluded the null when an alternative national study was substituted (OR 1.76, 95% CI 1.13–2.72), when all six estimates were pooled without regard to overlapping data (OR 1.57, 95% CI 1.15–2.15), or when the conventional DerSimonian–Laird method was used. No included study compared dementia with non-dementia deaths, and only one was restricted to adults aged 60 years or older; the others were predominantly older or did not report age. The certainty of evidence was very low. Conclusions: The evidence is inconclusive regarding the magnitude and consistency of an association between heatwaves and dementia-related mortality. The study-level estimates point in the same direction, but the pooled estimate is an average across highly heterogeneous settings, is too imprecise to establish an effect, and cannot show whether the relative increase exceeds that for other deaths; evidence specific to older adults with dementia is largely indirect. Including people with dementia in heat–health action plans remains a reasonable precaution, and independent, methodologically harmonized primary studies are needed.
SN - 2077-0383
ER -