TO THE EDITOR:
Given the Health Care Financing Administration's proposal to designate so-called centers of excellence as part of its total joint demonstration project, the article "Relationship between the Volume of Total Hip Replacements Performed by Providers and the Rates of Postoperative Complications in the State of Washington" (79-A: 485—494, April 1997), by Kreder et al., is a very interesting and timely piece of scholarship.
I register these comments about the paper.
First, in my experience as a medical student at Cornell and as a resident and attending physician at the University of Pennsylvania, I have seen that surgeons with a high-volume practice tend to attract patients from a very wide geographic area. Accordingly, a patient who is managed by such a surgeon in New York, for example, may be treated for a medical complication at a hospital in New Jersey (or New Mexico or even New Delhi). Thus, scanning of the readmission list in the state where the original operation was performed may lead one to underestimate the complications related to the surgeons with a higher-volume practice. Admittedly, both the New York City and Philadelphia metropolitan areas comprise three states, so this may be more of an issue there than in Washington. Still, I am curious to hear the authors' thoughts on this topic.
Second, if surgeons with a high-volume practice are more likely to treat patients who travel a great distance for expert care, then it is likely that their patients are more affluent. In my experience as a resident working at a veterans' hospital, I saw that less affluent patients have more difficulty complying with their assigned postoperative regimens and thus may be more liable to complications.
Finally, the true benefits of high volume may be attained only at levels above those of the upper tier of this study. It is my guess that the Health Care Financing Administration intends to have some centers performing hundreds, if not thousands, of joint replacements annually. (Without any stimulus from this Administration, such centers already exist in New York and Philadelphia.) Within the cohort of hospitals studied, were there any that could be described as ultra-high-volume? If so, did their performance differ from the other institutions in the study group?
The authors are to be commended for reporting their conclusions, which, given the position of The American Academy of Orthopaedic Surgeons on centers of excellence, may be viewed as politically incorrect. Their paper provides a strong hint that the Health Care Financing Administration is on the right track. I would be very interested in The Academy's response.
Joseph Bernstein, M.D.: Department of Orthopaedic Surgery, University of Pennsylvania Health System, 39th and Market Street-1 Cupp Pavilion, Philadelphia, Pennsylvania 19104
Dr. Kreder, Dr. Deyo, Dr. Koepsell, Dr. Swiontkowski, and Mr. Kreuter reply:
We welcome Dr. Bernstein's interest and make the following comments regarding the three issues that he addressed.
First, Dr. Bernstein has correctly pointed out that even population data sets may not completely capture all numerator events of interest. He states that some patients may have a primary replacement in one state and treatment of subsequent complications in another. He suggests that the patients of surgeons with a high-volume practice may be more likely to travel out of state. Thus, rates of complications for such surgeons might be underestimated.
While this assumption may be correct, there are no data to support it. We could find no significant association between the volume of procedures performed by a surgeon and the percentage of non-Washington residents whom he or she treated (5.7 per cent for those performing a low volume, 4.3 per cent for those performing a medium volume, and 4.5 per cent for those performing a high volume; p > 0.05). (These data did not appear in the manuscript.) It may be that such travel would be more of an issue in regions with densely populated states of relatively small geographic size. Finally, the assumed differential in patient travel would not apply to in-hospital complications, duration of stay, or hospital charges.
Second, we agree that patient education and motivation as well as other factors (perhaps partly captured by income as a surrogate) affect the functional outcome and probably the rates of postoperative complications also. These variables are not available on an individual patient basis in most hospital-discharge data sets; however, we did attempt to model complications as a function of mean incomes by patient-residence census tract. With use of this rather crude measure, we could not demonstrate a difference in income across provider volume categories and we were unable to find an association between income and complications (p > 0.05).
Third, we agree that ten hip replacements a year does not represent a particularly high volume, especially since all hip replacements (revision and primary, including bipolar replacements for fractures and so on) were counted in the volume computations. There were no ultra-high-volume providers in the state of Washington. The highest five-year average annual volumes were seventy-eight and 329 hip replacements for the surgeons and hospitals, respectively. After addition of an extra volume category of greater than the ninety-fifth percentile, we could find no difference in outcome between this category and the eightieth to ninety-fifth-percentile category (p > 0.05).
Please note that there was an error in Table I. The average hospital volume per year for the low, medium, and high-volume surgeon categories should read 79.5, 79.1, and 97.6, respectively. The average surgeon volume per year for the low, medium, and high-volume hospital categories should read 6.5, 11.8, and 17.3, respectively. We regret this error.
Hans J. Kreder, M.D.: Musculoskeletal Health Status Working Group, Division of Orthopaedics, Sunnybrook Health Science Center, Suite A-315, 2075 Bayview Avenue, North York, Ontario M4N 3M5, Canada
Richard A. Deyo, M.D.; Thomas Koepsell, M.D.; Marc F. Swiontkowski, M.D.: Department of Orthopaedics, University of Washington, Harborview Medical Center, Box 359798, 325 Ninth Avenue, Seattle, Washington 98104
William Kreuter, M.P.A.: Department of Health Services, University of Washington, Box 354807, Seattle, Washington 98195-4807