Controlling for Continuous Confounders in Epidemiologic Research

Abstract
This case-control study used the National Crime Victimization Survey database (a national sample of housing addresses) to examine sociodemographic risk factors for becoming a victim of work-related robbery and assault. Cases (N = 267) reported having been violently victimized in the previous 6 months. Controls (N = 1,783) were chosen from all nonvictims of violent crime at the end of the 6-month period. Risk factors varied by type of victimization, and differences were evident between men and women. Men less than 45 years of age had an increased risk for assault [odds ratio (OR) = 2.0–2.7], compared with those 55 years of age and older; and those with a family income of less than $40,000 had an increased risk for assault (OR = 1.7–1.9), compared with those having a family income of $50,000 or more. We found a decreased risk for those with a high school education (OR = 0.6), compared with those with some college education. For women, an increased risk was seen for ages 16–18 years (OR = 3.3) and 25–34 years (OR = 2.3), compared with those 55 years of age or older. Women who were divorced or separated (OR = 4.4) and never-married (OR = 2.1) were at higher risk than women who were married. We found a decreased risk for nonwhites (OR = 0.5), compared with whites.