Use of different comorbidity scores for risk-adjustment in the evaluation of quality of colorectal cancer surgery: does it matter?

Abstract

Aims: Comorbidity affects outcomes after colorectal cancer surgery. However, it's importance in risk adjustment is unclear and different measures are being used. This study aims to assess its impact on post-operative outcomes.

Methods: All 2204 patients who were operated on for stage I-III colorectal cancer in the Midwestern region of the Netherlands between January 1, 2006 and December 31, 2008 were analyzed. A multivariate two-step enter-model was used to evaluate the effect of the American Society of Anaesthesiologists Physical Status classification (ASA) score, the sum of diseased organ systems (SDOS), the Charlson Comorbidity Index (CCI) and a combination of specific comorbidities on 30-day mortality, surgical complications and a prolonged length of stay (LOS). For each retrieved model, and for a model without comorbidity, a ROC curve was made.

Results: High ASA score, SDOS, CCI, pulmonary disease and previous malignancy were all strongly associated with 30-day mortality and a prolonged LOS. High ASA score and gastro-intestinal comorbidity were risk factors for surgical complications. Predictive values for all comorbidity measures were similar with regard to all adverse post-operative outcomes. Omitting comorbidity only had a marginal effect on the predictive value of the model.

Conclusion: Irrespective of the measure used, comorbidity is an independent risk factor for adverse outcome after colorectal surgery. However, the importance of comorbidity in risk-adjustment models is limited. Probably the work and costs of data collection for auditing can be reduced, without compromising risk-adjustment.