Comparison of Temperature-Mortality Associations Across the Middle East Using Different Exposure Estimation...

Area of Science:

• Environmental epidemiology
• Climate change and health
• Public health research

Background:

• Traditional temperature-health studies often use station data or gridded averages.
• Population-weighted averages may improve accuracy in regions with variable temperatures and populations.
• Few studies have examined this in hot, understudied regions outside North America and Europe.

Purpose of the Study:

• To demonstrate the utility of population-weighted temperature exposures in the Middle East, a climate hotspot.
• To assess temperature-mortality associations using different exposure measurement approaches.

Main Methods:

• Utilized a daily 1km resolution temperature dataset across 152 administrative regions in 12 Middle Eastern countries (2003-2020).
• Calculated daily minimum and maximum population-weighted and unweighted spatial average temperatures.
• Employed distributed lag non-linear models to analyze temperature-mortality associations in Kuwait and Jordan, comparing station, unweighted, and population-weighted temperatures.

Main Results:

• Population-weighted and unweighted spatial averages yielded similar exposure-response curves, differing from station temperatures.
• Minimum mortality temperatures varied: Kuwait (station: 30.2°C, unweighted: 28.6°C, weighted: 28.3°C); Jordan (station: 20.6°C, unweighted: 20.9°C, weighted: 20°C).
• Heat-attributable mortality increased by 15% using population-weighted temperatures compared to station temperatures in Kuwait and Jordan.

Conclusions:

• Spatial averaging (weighted or unweighted) is valuable for estimating heat-attributable mortality, especially in data-limited regions like the Middle East.
• Population-weighted temperatures may better represent localized heat exposures in areas with clustered populations.
• Further research is needed to fully ascertain the added benefit of population-weighted over unweighted spatial averages.