Use of a clustered model to identify factors affecting hospital length of stay

Predictive models have been used to identify factors that may prolong hospi tal length of stay (LOS). However, because predictors of LOS are collinear, the proportion of variance associated with each factor in a multivariate s tepwise regression model may not reflect its mathematical contribution in e xplaining LOS. In an attempt to model factor contribution to LOS more reali stically, we evaluated a clinically based clustered model. This model uses classes of candidate predictors, that is, patient attributes, adverse event s, treatment modality, and health provider identity. Clusters of variables are permitted to enter into the model in a theoretically based predetermine d sequence, so that the additional contribution of each cluster of factors can be assessed while the contribution of preceding factors is preserved. T he clustered model was tested and compared with a free stepwise multivariat e analysis in a cohort of patients undergoing prostatectomy for benign pros tatic hypertrophy. We found that both models explained a similar proportion of the variance in LOS (56%-57%). However, some important differences were evident. Prostate size, associated with 12% of the variance in the cluster ed model, was not an independent predictor in the free model. A higher prop ortion of variance was associated with process variables, such as treatment modality in the free model. We conclude that use of a clustered model may facilitate more realistic assessment of the relative contribution of factor s to LOS. J CLIN EPIDEMIOL 52;11:1031-1036, 1999. (C) 1999 Elsevier Science Inc.