Investigating associations between ICU level and quality of care in the Netherlands: reporting only SMRs is not the whole story

Response to Girbes et al.: Investigating associations between ICU level and quality of care in the Netherlands: reporting only SMRs is not the whole story

“…In response to the comments by Girbes et al [1], we conclude that each comment is adequately answered in our study [2]. Both the problem of underestimating mortality in the sicker and overestimating mortality in the less sick ICU patient as well as a possible referral bias have been addressed adequately.…”

“…Dear Editor, With interest we read the correspondence by Girbes et al [1] regarding our paper ''The association between ICU level of care and mortality in the Netherlands'' [2] stating that SMRs can be confounded by case-mix differences since SMR is calculated using a logistic regression model, thus underestimating mortality in the sicker, and overestimating mortality in the less sick ICU patient [3]. Also the authors claim that the transfer of patients must have contaminated the findings of our study.…”