Study objective: To investigate the association of alcohol use and night driving with traumatic snowmobile fatalities.Design: Case-control study.Participants: Traumatic deaths occurring while driving a snowmobile during the years 1985 to 1990 were reviewed. A sample of 1989 to 1990 fatal motor vehicle driver and motorcycle driver accidents were used as controls. Records were obtained from the provincial coroner.Results: One hundred eight snowmobile fatalities, 432 motor vehicle fatalities, and 108 motorcycle fatalities were included. Young men (mean age, 30 years) made up the snowmobile fatalities population, with weekend fatalities predominating (67%). Snowmobile fatalities were associated with use during times of suboptimal lighting (crude odds ratio, 1.9 [95% confidence interval, 1.1-3.3]; P<.01). Blood alcohol concentration exceeded provincial limits in 64% of cases. When snowmobile fatalities were adjusted for occurrence during suboptimal lighting conditions, only alcohol use was associated independently with fatal outcome (adjusted odds ratio, 4.3 [95% confidence interval, 2.5-7.0]; P<.0001).Conclusion: Drivers in snowmobile fatalities are associated with an approximately fourfold greater use of alcohol than are age- and sex-matched drivers in automobile and motorcycle fatalities. Preventive strategies should be targeted at reducing the use of alcohol while snowmobile driving in young men.

OBJECTIVES:To investigate the demographic characteristics and circumstances surrounding fatal snowmobile accidents in Ontario, to examine the risk factors and to observe any fatality trends over the study period.DESIGN:Case series.PATIENTS:All 131 people who died accidentally while operating a snowmobile in Ontario from 1985-86 to 1989-90. Records were obtained from the chief coroner's office; registration data were obtained from the Ministry of Transportation.RESULTS:Although the absolute number of deaths increased each year, owing to a rapid increase in the number of registered snowmobiles, the risk of death from snowmobile accidents remained relatively constant. Deaths occurred most frequently in northeastern Ontario. Youths and men predominated among the victims. Fatal accidents occurred more often on lakes (in 66% of the cases in which this information was known) than on roads (in 26%) or trails (in 8%). Weekend fatalities predominated, and deaths occurred most often during times of suboptimal lighting (from 4 pm to 8 am). The driver was killed in 84% of the cases in which the person's role was known. Alcohol use before death was implicated in 69% of the cases, the level exceeding the Ontario legal limit in 59%.CONCLUSION:Snowmobile-related deaths result from factors that are generally avoidable. Strategies need to be instituted to reduce the rate of these events.

Rate based analyses such as per member per month (PMPM) and units of measure per 1,000 are important for understanding cost and utilization of a pen population. While retrospective claims data is rich with information regarding utilizers of health care services (patients), it does not provide the necessary information for population based analyses. For population based analyses one must have enrollment data as well, whch often is not as ready avadable. This paper describes the process by which Phdetrics, Inc. determines a total population in the absence of detailed, high quality, enrollment data.