AbstractThe main purpose of this paper was to assess the effects of age, period, and cohort on stroke mortality among the urban Lithuanian population. Routine stroke mortality data among the Lithuanian urban population aged 25–64 years (1041 men and 724 women) between 1980 and 2004 were obtained from the official Kaunas region mortality register and classified by codes 430–438 and 160–169 in the 9th and 10th revisions of the International Classifications of Diseases (ICD), respectively. Mortality rates per 100,000 persons for men and women were age-adjusted using the age distribution of the European Standard Population. Goodness of fit of the Poisson regression models was evaluated using the Pearson and Freeman-Tukey residuals. During the study period, mortality rates decreased from 46.8 to 33.0 per 100,000 for men, and from 20.2 to 18.1 per 100,000 for women (average annual decrease of −1.3%, p<0.1 for men, and −1.6%, p<0.03 for women). An age effect was present in both sexes. The definite upward period effect was observed from 1990 to 1994 both among men and women, and was followed by a sharp fall during 2000–2004. Cohort and period effects have contained relevant information that partially explained trends in stroke mortality among 25–64 year-olds in the Lithuanian urban population. The Poisson regression models could be applied for the examination and explanation of the different causes of the population mortality.
Fitting Parametric and Semi-parametric Conditional Poisson Regression Models with Cox’s Partial Likelihood in Self-controlled Case Series and Matched Cohort Studies
