Premature Mortality in the Kingdom of Saudi Arabia Associated with Particulate Matter Air Pollution from the 1991 Gulf War

ABSTRACT

The State of Kuwait oil fires and military operations associated with the 1991 Gulf War resulted in substantially increased levels of airborne particulate matter (PM) in the Kingdom of Saudi Arabia (KSA) during 1991 and 1992. Using quantitative risk assessment methodology, this article estimates the increase in premature deaths in citizens of the KSA associated with the Gulf War–related increase in PM air pollution levels. Meta-analysis of daily time-series studies of non-accidental mortality associated with increased PM₁₀ levels using two alternative methodologies yielded exposure-response relative risk functions of 2.7% and 3.5% per 50 μ g/m³ increase in PM₁₀ concentration. Combining these exposure-response functions with estimates of the magnitude and duration of the increased PM₁₀ exposure, the size of the exposed population and baseline mortality rates provided an estimate of approximately 1,080 to 1,370 excess non-accidental deaths of Saudi citizens during 1991–1992 associated with the Gulf War–related increase in PM levels.

Key Words:

• particulate matter
• Gulf War
• risk assessment
• mortality
• Saudi Arabia

ACKNOWLEDGMENTS

The authors thank Rebecca Nachman for her assistance in the preparation of this article. We are also grateful to the three anonymous referees for their helpful comments. Funding for this project was provided by the Kingdom of Saudi Arabia Presidency of Meteorology and Environment.

J. Morel Symons is currently at Haskell Laboratory for Health and Environmental Sciences, E.I. DuPont de Nemours and Company, 1090 Elkton Road, P.O. Box 50, Newark, DE 19714, USA.

Notes

¹2000 Demographic Survey 2 (by age, sex, nationality, and religion); 1992 Population and Housing Census 3 (by age, sex, and nationality);

²Includes only deaths that occurred in Ministry of Health hospitals. Tables 1-17, 1-35, 1-34, and 1-33, respectively, from each yearly report summarize the data by cause of death, nationality, and gender. Tables 1-37, 1-35, and 1-34, available in the reports published in 1999 through 2001, provide data on mortality trends by disease group for the current and two preceding years. Table 1-36, which was only included in the 1999 report (for the period May 1998 to April 1999), provides data on the distribution of deaths by region and nationality.

¹The equivalent Hijri calendar date is 1421 A.H.

²The equivalent Hijri calendar date is 1419 A.H.

**All RR percent estimates are for standardized change in percent of relative risk per 50 μ g/m³ increase in PM₁₀.

^∧Formula for crude relative risk % average = (Σ RR%)/(Number of Studies).

^†Formula for weighted relative risk % average = (Σ weighted RR%/Σ weights)

³The supplemental material is available at www.jhsph.edu/RiskSciences/Research or upon request from the corresponding author.

*Estimated values in italics.