The comment by Zahl  illustrates clearly the two different approaches in estimating overdiagnosis in the Norwegian Breast Cancer Screening Program. Zahl uses grouped data often named ecological data and mathematical models of underlying changes in risk to assess the association between participation in the screening programme and the potential for overdiagnoses. Owing to the potential for ecological fallacy, this kind of analyses is generally considered useful for generating hypotheses but not for testing hypotheses because of the large potential for biases or the ecological fallacy.
Re: No overdiagnosis in the Norwegian Breast Cancer Screening Program estimated by combining record linkage and questionnaire information in the Norwegian Women and Cancer Study
Leggi l'articolo originale