Gonorrhea case reports to the Alaska Department of Health and Social Services were used to study the contribution of reinfection to rates of gonorrhea infection in Alaska. The case reports of 13,910 infections among 11,132 persons who had laboratory-proven gonorrhea between 1983 and 1987 were examined. Among 1,886 persons who had multiple infections, the average number of infections per person was 2.5 (range = 2-11). These persons accounted for 33.5% of all infections and 16.9% of all patients with gonorrhea from 1983 to 1987. Compared to persons with one infection, those having multiple infections were more likely to be Alaska Natives (relative risk = 1.8, 95% confidence interval = 1.6-1.9) and less than 21 years of age (relative risk = 1.3, 95% confidence interval = 1.2-1.4). There was no difference in risk between men and women. Two thirds of the reinfections occurred within 12 months of the initial infection. If gonorrhea incidence were calculated based on the number of people infected rather than as a "case rate," the mean annual rate (per 100,000) from 1983 to 1987 decreased from 1,644 to 1,228 (a 25.3% decrease) for Alaska Natives and from 316 to 267 (a 15.5% decrease) for non-Natives. Reporting gonorrhea incidence rates by number of persons infected rather than by total number of cases more accurately measures gonorrhea morbidity in a population and will allow prevention efforts to be directed to those persons who contribute the most to perpetuating the disease.