Adjusted or unadjusted outcomes

Surgeons have long maintained an interest in assessing the outcomes of their therapeutic efforts. Previously, such goals were supported by mortality and morbidity conferences or surgical audits. The latter typically focus on unadjusted rates of adverse events. While comparisons of these outcome metrics between individuals or institutions are held to portray quality of surgical care, such comparisons are compromised by the divergent risk profiles associated with the patient population in question, underscoring the import of adequate and validated risk adjustment measures. Numerous risk adjustment strategies have been developed, based either on more readily available, and thus less expensive, administrative claims data or on more reliable and robust, but more expensive, clinical data. The National Surgical Quality Improvement Program (NSQIP), which is requires abstraction of clinical data, stands as the "gold standard" for assessment of surgical quality.

While risk adjustment seems essential to truly assess surgical quality, alternatives to use of clinical data continue to be explored. Some alternatives from previous works published by the author and others are briefly reviewed in this presentation.

When assessing the quality of surgical outcomes, risk adjustment seems superior. Ongoing research is needed to help achieve balance between the benefit of clinically based models and less expensive alternatives based on administrative data.