Epidemiological studies on the transmission of bovine spongiforme encephalopathy (BSE) - Some remarks from a biometrical point of view

This contribution discusses biometrical aspects of the design and analysis of epidemiological case-control studies. The study chosen as an example was undertaken to investigate the transmission of bovine spongiform encephalopathy (BSE) to calves and the role of the inclusion of meat and bonemeal in proprietary feedstuffs. The hypothesis was examined by means of a case-control study which is summarized with emphasis on its study design and the collection of data. The analysis included the estimation of raw odds-ratios for the inclusion of meat and bonemeal in proprietary feedstuffs for calves as well as adjusted odds-ratios using logistic regression. This strategy is discussed and the chosen regression model is compared with an alternative one. Results show that problems occurred when recruiting herds for the study leading to uneven proportions of cases and controls. Because of these misproportions and due to different data quality for cases and controls the interpretation of the study results is difficult.