This population-based study examines patterns of fatal motor vehicle traffic crashes (MVTCs) by age group (16-24, 25-64, 65+) among Canadian drivers. The Canadian Traffic Accident Information Databank (TRAID) provided information about fata MVTCs between 1984 and 1993. Distribution of risk factors was compared by age group. Crude odds ratios and 95% CIs were calculated for both young and elderly drivers compared with middle-aged drivers. The middle-age group was selected as reference population since it demonstrates the lowest risk of fatality. Compared to the middle-aged group, young drivers demonstrated excess risk for (1) risk-taking behaviours and conditions, specifically alcohol and illicit drug use, speeding, non use of seat belts, fatigue and falling asleep, and inexperience; (2) crashes during the summer, during weekends and at night; and (3) single-vehicle collisions and on performing overtaking manoeuvres. Excess risk among elderly drivers was noted for (1) medical and physical conditions, inattention and inexperience; (2) driver actions, for example, improper turning, failure to yield right-to-way; (3) occurrence on weekdays and during the day; and (4) collisions at intersection and vehicle--vehicle sideswipes. The results show notable differences in risk factors by age group and confirm the need for preventive efforts that incorporate age-specific strategies.

This study examined histology-specific incidence trends of ovarian cancer in Canada, 1969–1993. The impact of age, period and cohort effects on these trends were studied by means of age-period-cohort analysis. Age-standardized incidence rates of serous, endometrioid, clear cell and germ cell tumours increased significantly and the rates of sex cord-stromal and other classified epithelial ovarian tumours decreased considerably. The rates of mucinous and NOS/unclassified tumours remained unchanged. Cohort effect has a major impact on incidence trends of serous, endometrioid, germ cell, sex cord-stromal and other classified epithelial ovarian tumours but no meaningful impact on trends of mucinous, clear cell, or NOS/unclassified ovarian tumours. Various cohort patterns by histology subtypes were observed: the risk of developing serious tumours increased markedly among birth cohorts of 1895–1930, stabilized thereafter and decreased among young cohorts of 1950–1960; the risk of germ cell tumours increased significantly among young cohorts of 1965–1980; and the risk of sex cord-stromal tumours dropped constantly among cohorts 1910–1950. Various period patterns by histology subtypes observed in this study suggested changes in histology classification criteria over the period. Further studies need to consider the various etiologies and the classification criteria changes according to histology subtypes. © 1999 Cancer Research Campaign