B. Anderson

Colorado State University

Author Of 6 Presentations

Assessing exposure to tropical cyclones for epidemiological research: Existing resources and future needs (ID 2523)

Q&A (ID 2627)

Webcast

[session]
[presentation]
[presenter]
Hide

P-0387 - Association between heat warnings and mortality among older adults in the United States (ID 1284)

Date
08/24/2020
Room
Not Assigned
Session Name
E-POSTER GALLERY (ID 409)
Lecture Time
07:40 AM - 08:00 AM
Presenter

Presenter of 5 Presentations

Q&A (ID 2627)

Webcast

[session]
[presentation]
[presenter]
Hide

Assessing exposure to tropical cyclones for epidemiological research: Existing resources and future needs (ID 2523)

Poster Author Of 1 e-Poster

E-POSTER GALLERY (ID 409)

P-0387 - Association between heat warnings and mortality among older adults in the United States

Abstract Control Number
1736
Abstract Body
Background: Heat warnings are issued in advance of forecast periods of extreme heat in order to protect the public’s health. However, little evidence is available regarding their effectiveness in reducing mortality. We investigated the association between heat warnings and mortality among older adults living in populous counties across the United States, 2006-2016. Methods: In each county, we matched days with heat warnings to days without heat warnings on month and maximum heat index (within 2 degrees Fahrenheit). We additionally restricted the pool of potential days without heat warnings to those more than three days before or after a day with a heat warning. We used a log-linear mixed effect Poisson model with a random intercept for county to estimate the association between heat warnings and daily all-cause mortality counts among Medicare beneficiaries age 65 and older, adjusting for year, day of week, and federal holidays. Results: We identified at least one pair of days with and without heat warnings in 249 populous counties (defined as those with a 2010 population size greater than 250,000). Across these counties, we did not observe evidence of a reduction in mortality on heat warning days (percent change in risk of mortality comparing days with versus without heat warnings: -0.06 [95% confidence interval: -1.06, 0.95]). However, in a sub-analysis of counties with data on heat response plans, results varied by heat response plan presence and time period. Conclusions: Overall across the study counties we did not observe a lower risk of mortality on days with heat warnings, but our results provide initial evidence that the effectiveness of heat warnings may vary by county characteristics. The identification of factors associated with differing heat warning effectiveness is critical in light of projected future increases in temperature due to climate change.