ENVIRONMENTAL RESEARCH, SECTION A 79, 71 — 81 (1998) ARTICLE NO. ER983873 COMMENTARY Gulf War Veterans’ Illnesses: A Case Study in Causal Inference 1 Lois M. Joellenbeck,* ,2 Philip J. Landrigan,- and Elaine L. Larson‡ *Division of Assessment and Surveillance, Metropolitan Health Department of Nashville and Davidson County, Nashville, Tennessee; -Department of Community Medicine, Mount Sinai School of Medicine, New York, New York; and ‡Georgetown University School of Nursing, Washington, D.C. Received November 18, 1997 The objective of this study was to determine from the published epidemiological literature whether there is evidence for a causal association between service in the Gulf War and illness in U.S. veterans. Eleven published studies were analyzed using standardized epidemiologic criteria for assessing causality. A consistent association was found be- tween deployment to the Gulf and self-reports of symptoms. No consistency was seen in physical find- ings or laboratory results. Strength of association varied with different study designs. Dose-response information is limited, because of lack of quantitat- ive data on exposures. Biological plausibility varies for different risk factors. Specificity of association is not seen. Frequency of self-reported symptoms is increased in U.S. Gulf War veterans compared to other veterans of the same era, but specific causes of illnesses cannot be ascertained. Major gaps in data that impeded this analysis include (1) lack of objec- tive data on specific environmental exposures (2) lack of baseline health assessments, and (3) lack of objective measures of post-deployment health status. In future deployment of U.S. troops, accu- rate exposure and health data will be needed if the causes of subsequent illnesses are to be accurately assessed.  1998 Academic Press 1 Funding Sources: The analysis described in this paper was undertaken without outside funding. Elaine Larson and Philip Landrigan both serve on the Presidential Advisory Committee on Gulf War Veterans’ Illnesses, for which they receive no financial compensation. Lois Joellenbeck served on the paid staff of the same committee until December, 1996, but was employed else- where during the time this manuscript was written. No new studies involving humans or animals were carried out for this analysis. 2 To whom correspondence and reprint requests should be ad- dressed at Institute of Medicine, 2101 Constitution Avenue N. W., Washington DC 20418. Fax (202) 334-2685. E-mail: ljoellen@ nas.edu. Key Words : Persian Gulf Syndrome; Persian Gulf Syndrome/ep [epidemiology]; military personnel, veterans; environmental exposure/ae [adverse ef- fects]. INTRODUCTION The establishment of causality can contribute greatly to disease prevention and is a major goal of epidemiological research. However, causality is diffi- cult to demonstrate through epidemiological studies. The amounts and types of information needed to assert that a risk factor causes a particular health outcome are perennial questions. Beginning with the work of Jakob Henle in 1840, modified by Henle’s student Robert Koch, and further refined by Hill (1965), Doll (1985), Evans (1976), and the Interna- tional Agency for Research on Cancer (1995), elegant criteria have been developed to assist health scien- tists and policy makers in discerning whether an epidemiological study or a body of epidemiological studies can support an inference of causality (Table 1). A strong stimulus to these endeavors was the bitterly contested effort to establish that cigarette smoking causes lung cancer (Hill, 1965; Doll, 1985). Illnesses reported by veterans of the Persian Gulf War provide an opportunity to test the canons of causality. Approximately 697,000 men and women served in Operations Desert Shield and Desert Storm in 1990 and 1991. Since their return, over 100,000 Gulf veterans have participated in govern- ment health registries, indicating widespread health concerns among these veterans. Several epi- demiological studies have now been completed to explore associations between service in the Gulf War and subsequent illnesses. The purpose of this paper 71 0013-9351/98 $25.00 Copyright  1998 by Academic Press All rights of reproduction in any form reserved.