Statistics in medicine
-
Statistics in medicine · May 1997
Improved odds ratio estimation by post hoc stratification of case-control data.
We propose a logistic regression analysis of unmatched or frequency matched case-control studies with conditional maximum likelihood estimation through post hoc stratification. In this model fewer parameters have to be estimated. ⋯ A more refined post hoc stratification reduces computing time, but to the cost of a larger bias and a loss in efficiency. The model was also applied to data of unmatched case-control studies on laryngeal cancer, oesophageal cancer and lung cancer.
-
Multi-centre databases are making an increasing contribution to medical understanding. While the statistical handling of randomized experimental studies is well documented in the medical literature, the analysis of observational studies requires the addressing of additional important issues relating to the timing of entry to the study and the effect of potential explanatory variables not introduced until after that time. A series of analyses is illustrated on a small data set. ⋯ The aim of each analysis, the choice of data used, the essentials of the methodology, the interpretation of the results and the limitations and underlying assumptions are discussed. It is emphasized that, in contrast to randomized studies, the basis for selection and timing of interventions in observational studies is not precisely specified so that attribution of a survival effect to an intervention must be tentative. A glossary of terms is provided.