Revista Española de Salud Pública
Print version ISSN 1135-5727
ORTEGA CALVO, Manuel and CAYUELA DOMINGUEZ, Aurelio. Unconditioned logistic regression and sample size: a reference source review. Rev. Esp. Salud Publica [online]. 2002, vol.76, n.2, pp. 85-93. ISSN 1135-5727. http://dx.doi.org/10.1590/S1135-57272002000200002.
Unconditioned logistic regression is a highly useful risk prediction method in epidemiology. This article reviews the different solutions provided by different authors concerning the interface between the calculation of the sample size and the use of logistics regression. Based on the knowledge of the information initially provided, a review is made of the customized regression and predictive constriction phenomenon, the design of an ordinal exposition with a binary output, the event of interest per variable concept, the indicator variables, the classic Freeman equation, etc. Some skeptical ideas regarding this subject are also included.
Keywords : Logistic regression; Sample size; Research Design; Epidemiology.