The Angioplasty Numbers Game
The Angioplasty Numbers Game
abstract & commentary
Synopsis: The disparity in acute outcomes of angioplasty in low- vs. high-volume hospitals has narrowed as outcomes in general have improved. Thus, minimum-volume standards need to be reevaluated, especially in sparsely populated regions of the country.
Source: Ho V. Circulation 2000;101:1806-1811.
Previous cross-sectional studies relating procedure volume to outcomes have suggested minimum volume limits for performing angioplasty. Such numbers, when embraced by national organizations and health care payers, can have profound influences on hospitals and the health care of sparsely populated regions of the United States. Yet, outcomes over time have not been studied. Thus, Ho, from the H.M. Olin School of Business at Washington University in St. Louis, Mo., examined an administrative patient database of all hospital discharges in California from 1984 to 1996. The 355,673 admissions for angioplasty were evaluated for the relationship between the number of procedures performed at the hospitals vs. the bypass surgery and death rates during the same hospitalization. Three time periods were assessed—1984-87, 1988-92, and 1993-96. The number of angioplasties performed at each of the 129 hospitals ranged from 1 to 1247, but over the three time periods the median annual hospital volume increased from 89 to 200 to 272. In the early period, 93% of the hospitals performed fewer than 200 angioplasties a year; whereas in the latest period only 44% fell into this category. More than one-third of the patients were treated in low-volume hospitals in the earliest period, and by the latest period this was about 10%. In-hospital mortality and bypass surgery rates in the early period were 2.5% and 12.4% for low-volume hospitals and 1.3% and 6.9% for high-volume hospitals. By the latest period, these numbers had fallen to 1.7% and 4.6% and 1.3% and 3.3%. Patient characteristics changed over time: during the later period they were more likely women, older, acute myocardial infarction (MI) victims subjected to multivessel angioplasty, and sicker overall. A logistic regression model demonstrated that the reductions in mortality and bypass surgery rates over time in low-volume hospitals were greater in magnitude than moving the average patient from a low-volume to a high-volume hospital in the same time period. Ho concluded that over time the disparity in acute outcomes of angioplasty in low- vs. high-volume hospitals has narrowed as outcomes in general have improved. Thus, minimum volume standards need to be reevaluated, especially in sparsely populated regions of the country.
Comment by Michael H. Crawford, MD
As a resident of a sparsely populated state, I have always believed that volume minimums for procedures such as angioplasty do not tell the whole story and have cautioned against their codification in guidelines. However, the believers in numbers, most often physicians from large-volume centers, have usually prevailed in these debates. Thus, the ACC/AHA guidelines suggest centers do at least 200 angioplasties per year. The data in this paper suggest that this number may be lower based upon a longitudinal look at a California database that contains 18 hospitals performing fewer than 100 angioplasties per year. Although large-volume hospitals still had lower acute mortality and a need for bypass surgery rates, the difference narrowed appreciably as complication rates in general decreased over a decade. Thus, the performance advantage of high-volume hospitals was small by 1996 and would not justify the reduced patient access in less populated areas of regionalizing this procedure.
Although there was a decrease in the numbers of low-volume hospitals over the period studied, the results were not explained by an increase in volume at low-volume hospitals. There were real improvements in outcomes in low-volume hospitals. The reasons for such a large improvement in the low-volume hospitals cannot be ascertained from this administrative database. There are no individual physician data, but previous studies have suggested the individual physician volume is not a major factor in hospital outcomes. Perhaps the guideline assault on low-volume hospitals stimulated them to improve outcomes at a greater rate than their high-volume brethren. Perhaps the latest equipment innovations became available first at the high-volume centers and only later at the low-volume centers. So as the field matured, the equipment-related gains in outcome narrowed between the two types of hospitals.
Despite these encouraging results for low-volume hospitals, there are considerable limitations to these administrative databases. Physician, patient, and hospital characteristics that affect patient care are largely unknown and may be important (i.e., physician training, the patient’s ejection fraction, and availability of intra- aortic balloon pumping at a hospital). Also, hospitals may code admissions and diagnoses differently to maximize diagnosis-related groups. Despite these limitations, Dr. Ho’s data should stimulate a reevaluation of the numbers game in cardiovascular procedures.
Angioplasty outcomes in California:
a. have worsened over a decade.
b. are related to hospital volume.
c. are uniformly poor in low-volume hospitals.
d. suggest a hospital minimum of 600 per year.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.