Objective-We sought to determine whether demographic differences in eye injury rates persist after adjusting for occupational exposure.Methods-On-duty eye injury hospitalizations were linked to occupation among active-duty US Army personnel.Results-Eye injury rates were higher for white solidiers, men, and for younger soldiers, even after adjusting for occupational group and specific job titles using multivariate models.Conclusions-This finding contrasts with studies of other injuries, suggesting that occupation does not fully account for variations in eye injury risk. Because protective eyewear can prevent most serious eye injuries, we hypothesize that differences in protective eyewear use between men and women may contribute to differences in eye injury rates, although follow-up studies are needed to confirm this. Prevention efforts should consider targeting high-risk demographic groups in addition to high-risk occupations.Occupational injuries are important causes of preventable morbidity and long-term disabilities in working adults in the United States. Studies of occupational injuries have shown wide variations in injury rates by type of occupation, gender, and other demographic factors. Excess injury rates typically are seen in men, younger individuals, and certain racial groups. 1-6 Although it is assumed that differences in occupational exposures may explain in part these disparities, it is not clear the extent to which these variations are caused by differences in behavior (eg, men take greater risks) or caused by differences in exposure to hazardous conditions and occupations (eg, men are more likely to work in "high-risk" occupations). Thus, the extent to which specific occupational exposure is a confounder of the demographic patterns of trauma has not been adequately addressed. Understanding these factors will aid in more effective and targeted injury prevention strategies for those in high-risk occupations.Address correspondence to: Gordon S. Smith, MB, ChB, MPH, Liberty Mutual Research Institute for Safety, 71 Frankland Road, Hopkinton, MA 01748; E-mail: gordon.smith@libertymutual.com. Gordon Smith has no commercial interest related to this article. This article was co-written by an officer or employee of the US Government as part of their official duties and is therefore not subject to U.S. copyright. However, a major challenge facing researchers is how to separate out the influence of risk behaviors from the effects of occupational exposures because accurate information on work exposures is difficult to obtain. Studies that examine work injury risks in broad occupational groups (eg, metal industry or automotive industry) often do not have enough specific information on work tasks and occupational risk. As a result, residual confounding may limit the inferences that can be made about demographic and behavioral risk factors. In fact, some studies that were able to adjust for occupational exposure have shown that demographic risk patterns apparent in unadjusted models may even be reversed on...