Should We Trust the Results of Multivariate Analysis?


In 66 patients, Shiba et al. studied 11 variables for early recurrence of HCC, ten for disease-free survival and ten for overall survival. Too few patients, too many variables for the guidelines on subject to variable ratio for multivariate statistical analysis. Moreover, using cut-points to derive subgroups is not appropriate when there is a continuous… (More)
DOI: 10.1007/s11605-009-1126-x


Cite this paper

@article{Braillon2009ShouldWT, title={Should We Trust the Results of Multivariate Analysis?}, author={Alain Braillon}, journal={Journal of Gastrointestinal Surgery}, year={2009}, volume={14}, pages={582-582} }