AIMS: To consider the effects of alcohol prohibition and police presence upon serious injury in isolated Alaska Native villages. DESIGN: We compared rates of injury attributed to assault, self-harm, motor vehicle collisions and 'other causes' between villages with or without local prohibition and between villages with or without local police. Negative binomial regression was used to assess the relative effects of prohibition and police presence upon serious injury rates net of potential confounders. PARTICIPANTS: A total of 132 isolated Alaska Native villages between the years 1991 through 2000. MEASUREMENTS; Serious injury was measured using Alaska Trauma Registry and Alaska Bureau of Vital Statistics death certificate records. Local option election records were used to classify cases as occurring in wet or dry villages and police deployment records were used to classify cases as occurring in villages with or without local police. Village-level statistics from the 1990 and 2000 US censuses were used in the negative binomial regression analyses. FINDINGS: Villages that prohibited alcohol had lower age-adjusted rates of serious injury resulting from assault, motor vehicle collisions and 'other causes'. Dry villages with a local police presence had a lower age-adjusted rate of serious injury caused by assault. Controlling for the relative effects of village isolation, access to alcohol markets and local demographic structures, local prohibition was associated with lower rates of assault injuries and 'other causes' injuries while local police presence was associated with lower rates of assault injuries. CONCLUSIONS: Residents of isolated Alaska Native villages are safer when they prohibit alcohol. A local police presence in dry villages provides further reduction of the incidence of assault.