Problems in determining occurrence rates of multimorbidity.

This article describes methodological decisions that have to be made when studying multiple pathology and presents appropriate analytical techniques. The main question of this article is: how can comorbidity and multimorbidity be operationalized with respect to the number and type of diseases studied, and which analytic approaches are available for the evaluation of multiple pathology? Choices regarding the number and type of diseases studied have great impact on the observed incidence and prevalence rates of comorbidity and multimorbidity. These rates are largely dependent on age, sex, and other determinants. In addition to crude descriptive measures, odds ratios and relative risks can be used to study comorbidity, whereas multimorbidity can be studied using observed/expected ratios. While basic analyses of comorbidity can be performed using standard statistical packages, two additional programs were developed for the analysis of the distribution of multimorbidity and statistically unexpected comorbidity, respectively. As some analyses are addressing multicomparisons, external validity testing is recommended.