| Summary: | Abstract Background Disease mapping aims at identifying geographic patterns in disease. This may provide a better understanding of disease aetiology and risk factors as well as enable targeted prevention and allocation of resources. Joint mapping of multiple diseases may lead to improved insights since e.g. similarities and differences between geographic patterns may reflect shared and disease-specific determinants of disease. The objective of this study was to compare the geographic patterns in incident acute myocardial infarction (AMI), stroke and atrial fibrillation (AF) using the unique, population-based Danish register data. Methods Incident AMI, stroke and AF was modelled by a multivariate Poisson model including a disease-specific random effect of municipality modelled by a multivariate conditionally autoregressive (MCAR) structure. Analyses were adjusted for age, sex and income. Results The study included 3.5 million adults contributing 6.8 million person-years. In total, 18,349 incident cases of AMI, 28,006 incident cases of stroke, and 39,040 incident cases of AF occurred. Estimated municipality-specific standardized incidence rates ranged from 0.76 to 1.35 for AMI, from 0.79 to 1.38 for stroke, and from 0.85 to 1.24 for AF. In all diseases, geographic variation with clusters of high or low risk of disease after adjustment was seen. The geographic patterns displayed overall similarities between the diseases, with stroke and AF having the strongest resemblances. The most notable difference was observed in Copenhagen (high risk of stroke and AF, low risk of AMI). AF showed the least geographic variation. Conclusion Using multiple-disease mapping, this study adds to the results of previous studies by enabling joint evaluation and comparison of the geographic patterns in AMI, stroke and AF. The simultaneous mapping of diseases displayed similarities and differences in occurrence that are non-assessable in traditional single-disease mapping studies. In addition to reflecting the fact that AF is a strong risk factor for stroke, the results suggested that AMI, stroke and AF share some, but not all environmental risk factors after accounting for age, sex and income (indicator of lifestyle and health behaviour).
|
|---|
