Older Adults With ADRD More Likely Hospitalized After Extreme Heat Exposure

Extreme heat exposure could pose a growing risk for hospitalization among individuals with Alzheimer disease and related dementias (ADRD), according to a study published in JAMA Internal Medicine.

Researchers conducted a population-based cohort study with a time-stratified case-crossover design to investigate the association between extreme heat and hospitalization with ADRD. Participants aged 65 years and older enrolled in Medicare Part A fee-for-service insurance from 2000 through 2018 were included; each case was defined as a beneficiary’s first known hospitalization that included a diagnosis code for AD, related disorders, or senile dementia.

Eligible hospitalizations included participants admitted from May to September of the study years; admissions from a non–health-care facility point of origin, via a clinic referral, or via the emergency department were included. High-resolution temperature and humidity data were collected and aggregated to zip codes to match the Medicare data. Cases were self-matched, and control days were matched to case days using the same day of the week, month, and year of the case day, both before and after the case day. Conditional logistic regression models were employed for statistical analysis.

A total of 3,329,977 first hospitalizations with ADRD were identified (mean age, 82.52 years [SD, 7.48]; age 85 years and older, 42.4%; women 63.9%; White race, 85.0%) and analyzed together with a mean of 3.43 control days (SD, 0.51) per case day. Compared with the overall sample, a larger proportion of beneficiaries in temperate climates were of Black race (14.0% vs 10.7%).

High heat index exposure was associated with a significantly increased risk for hospitalization with ADRD, particularly on days in the 99th vs 50th percentile of the heat index distribution (odds ratio [OR], 1.02 [95% CI, 1.01-1.02). This corresponded to 0.8 additional hospitalizations (95% CI, 0.5-1.1) with ADRD per 1000 beneficiaries. Estimated effects of extreme heat on both current and prior days suggests that extreme heat associations persist for 3 days beyond the initial day; these estimated effects accumulate after sustained heat exposure.

The OR of hospitalization with ADRD after 4 days of continuous exposure to heat indexes at the 99th vs 50th percentile was 1.04 (95% CI, 1.03-1.04), corresponding to 1.7 additional hospitalizations (95% CI, 1.3-2.0) with ADRD per 1000 beneficiaries.

Sustained exposure to moderately increased temperatures was also associated with an increased risk for hospitalization with ADRD. When extrapolated to the 6.7 million adults currently living with ADRD, results suggest that each day of extreme heat could contribute to at least 5360 hospitalizations with ADRD nationwide.

In arid regions, results indicated that the estimated effects lasted 3 days (3-day cumulative OR in 99th vs 50th percentile, 1.05 [95% CI, 1.02-1.08]). In tropical regions, the corresponding 3-day cumulative OR was 1.03 (95% CI, 0.99-1.08).

The point estimates for the increased risk for ADRD-related hospitalization associated with 4 consecutive days of extreme heat exposure (99th vs 50th percentile of the heat index) were higher among Asian (OR, 1.09; 95% CI, 1.02–1.17), Black (OR, 1.07; 95% CI, 1.05–1.10), and Hispanic (OR, 1.08; 95% CI, 1.03–1.13) beneficiaries, and were 2.6 to 3.2 times greater than the estimate for White beneficiaries (OR, 1.03; 95% CI, 1.00–1.06).

Age, urbanicity, and Medicaid eligibility did not modify estimated effects of extreme heat; however, estimates may have been somewhat larger in men vs women, in more expensive vs more affordable housing price to income ratio areas, and in more advantaged vs some more disadvantaged areas. 

Study limitations included a lack of indoor and outdoor activity patterns and potential outcome misclassification.

“As our society ages and our climate changes, clinicians should consider counseling patients living with ADRD on the risks of extreme heat, while policymakers should devise programs to mitigate these risks,” the researchers concluded.

One study author declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.