We thank Drs Lee and Weed for their insightful comments. Dr Lee expresses concerns about the influence of poorly measured and unmeasured confounders on the results of our study.1 We support his call for additional rigorous multicenter studies to determine if our findings are confirmed in other settings and thereby help expand the evidence base for improving preoperative medical consultations. However, several of our analyses specifically address the methodological concerns that he raises. For example, differences in illness severity were partially accounted for through a sensitivity analysis that incorporated the number of recent acute care hospitalizations in the propensity score (eTable 2 of our study1). Our findings were unaltered by this alternative analysis. In addition, we accounted for differences in the number of important comorbidities in a sensitivity analysis that, while matching on the propensity score, simultaneously matched on age, Charlson comorbidity index score,2 ischemic heart disease, pulmonary disease, and Revised Cardiac Risk Index score.3Again, our findings were unchanged by this more exact matching method (eTable 21). Finally, although smoking and obesity are important perioperative risk factors that are not captured by administrative health care databases, their influence was partially addressed through a “tracer” outcome, namely surgical site infections. Specifically, both obesity and smoking are associated with an increased risk of surgical site infection,4,5 yet rates of surgical site infections were essentially identical in the matched consultation and no-consultation arms. Importantly, these sensitivity analyses for unmeasured confounding do not prove that there was no residual confounding. Rather, they suggest that the magnitude of any residual confounding was unlikely to have completely explained our results.
Wijeysundera DN, Beattie WS, Austin PC, Hux JE, Laupacis A. Is There Value in a Preoperative Medical Consultation?—Reply. Arch Intern Med. 2011;171(4):365–369. doi:10.1001/archinternmed.2011.8
Artificial Intelligence Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.