Evaluation of Time Trends of Weekly Number of Diseases

Overview

In routine systems investigating the morbidity according to diagnosis it is very useful to analysethe development in time (for example the development of weekly reports). This paper is concernedwith the methodology of such analyses. In practice it appears that the number of cases depends onseason. It stands to reason, that it is necessary to consider also long-therm trends. In this paper twodifferent approaches are discussed – the Box-Jenkins analysis, which describes the random errorand the Method of Trend Decomposition which spread the number of cases into the systematiccomponent (long term trend and seasonal effect) and random variability. The authors describe themethod of smoothing the estimate of the time series by kernel estimate. In both approaches theyuse weekly reports from the whole Czech Republic of diagnoses viral hepatitis A, rubella andsalmonellosis.

Key words:
time series – kernel estimate – analysis of number of cases.