Background: It has been known for nearly 20 years that, in cardiovascular o
perations, a significant inverse relationship exists between clinical outco
mes and the volume of procedures performed. Interestingly, this relationshi
p persists 2 decades after it was recognized. Objective: The purpose of thi
s study was to examine the relationship between hospital volume and in-hosp
ital deaths in 3 cardiovascular procedures: coronary artery bypass grafting
, elective repair of abdominal aortic aneurysms, and repair of congenital c
ardiac defects. Methods: The database includes all patients who were hospit
alized in New York State during the Sears 1990 to 1995, Using standard logi
stic regression techniques, we analyzed the relationship between hospital v
olume and outcome. Results. No correlation exists between hospital volume a
nd in-hospital deaths in coronary artery bypass grafting. Statewide, 31 hos
pitals performed 97,137 operations over the 6-year period (overall mortalit
y rate, 2.75%), By contrast, most of the hospitals statewide (195 of 230 ho
spitals) performed 9847 elective abdominal aortic aneurysm repairs with an
overall mortality rate of 5.5%, In abdominal aortic aneurysm operations, a
significant inverse relationship between hospital volume and in-hospital de
aths was determined. Sixteen hospitals performed 7199 repairs for congenita
l cardiac defects. A significant inverse relationship (which was most prono
unced for neonates) was found between volume and death. Conclusions: The im
portance of these findings lies in the rather striking difference between t
he volume-outcome relationship found for operations for abdominal aortic an
eurysms and congenital cardiac defects and the lack of such a relationship
for coronary artery bypass grafting. This observation may be largely explai
ned by the quality improvement program in New York State for bypass operati
ons since 1989. If so, these results have important implications for expand
ing the scope of quality improvement efforts in New York State.