Authors’ Reply to Luft: “Measuring the Volume-Outcome Relation for Complex Hospital Surgery”

Excerpt

We agree that the literature should aim to measure the underlying causes of the volume-outcome effect, rather than just noticing it is there. In addition, we agree with Dr. Luft [1] that the best studies should be measuring the effects of both surgeon and hospital volume. The senior author for this paper and others have published studies that include surgeon volume; and in each case, both surgeon and hospital volume matter for patient outcomes [2‐5]. As we explain in Section 3.3 of our paper, most published studies only account for hospital volume when measuring the volume-outcome effect. Because our primary aim is to illustrate the how the random-versus fixed-effect specification can affect results, we focus on a specification that accounts for only hospital volume as well. …