Commentary & Perspective by
Frederick A. Matsen, III, MD*,
Department of Orthopaedics and Sports Medicine, University of Washington, Seattle, Washington
The authors have provided us with a most provocative article, deserving of substantial reflection. This commentary will focus on three key words in the article: "high volume," "outcomes," and "policies."
The authors chose surgeon and hospital volume categories to obtain approximately similar percentages of procedures in each category. This led them to designate surgeons who performed five (or more) procedures per year as "high-volume surgeons" and hospitals in which ten (or more) procedures per year are performed as "high-volume hospitals." While we appreciate the statistical convenience of these designations, they create some serious concerns. Doing something once every ten weeks is hardly "high volume." Would you want your kidney transplant, your taxes, your flight qualification, or even your haircut done by someone who does five per year? In their landmark article, Katz et al. defined a "high-volume" surgeon as one who did fifty (or more) total hips per year and pointed out that complications were threefold higher for surgeons who performed five (or fewer) of those procedures per year1. How can one reconcile this order-of-magnitude discrepancy in the definition of "high-volume?" Is a shoulder arthroplasty ten times easier to perform than a total hip arthroplasty, so that only one-tenth the experience is necessary to qualify as a "high-volume surgeon?" It seems doubtful.
With the definitions and the data as presented in this article, we derived the following relationships between total shoulder outcome and surgeon volume (Fig. 1).
Fig. 1. Mortality, complication, and disposition rates by surgeon case volume per year.
It is of interest that there is no evidence that the volume effect reaches a plateau with increasing surgeon case volume. In fact, the increment in quality of outcome tends to be greater when comparing "high" with "medium" volumes than when comparing "medium" with "low" volumes. What would this graph look like if the abscissa were extended out to ten, twenty, or even fifty cases per year? How much might the outcome of the procedure for the patient be improved?
What is the "outcome" of a shoulder arthroplasty? As important as are the data on prevalence of complications, nonroutine disposition, and mortality, these metrics seem no more important than the excellence and durability of function following the procedure. What would be of at least equal interest to prospective patients would be the relationship between surgeon volume and the functional improvement after shoulder arthroplasty. Although complications, nonroutine disposition, and mortality are critical, they would seem less related to excellence of technique that is acquired by regular performance of shoulder arthroplasty than they are to comfort, range of motion, the ability to perform regular activities, and freedom from longer-term complications, such as glenoid component loosening. Does this paper really determine the degree to which patients treated by high-volume surgeons are more likely to have better outcomes? Might not the effect of surgeon case volume be even greater for functional outcome than it is for mortality?
One of the most serious deficiencies in the literature concerning shoulder arthroplasty is that our impressions regarding the effectiveness of this procedure are formed by reading publications by surgeons who perform this procedure every week—"high-volume" surgeons, by any definition. In contrast, most shoulder replacements are performed by surgeons who perform ten or fewer of these procedures each year2. Unfortunately we have no information that would reflect the quality of results achieved by lower-volume surgeons.
There is no simple solution to this problem and, as Codman's preface points out, it is not in the interest of the health-care delivery system to collect end-result data3. Would it be worthwhile for Medicare to send a Simple Shoulder Test end-result form two years after surgery to each patient who undergoes shoulder arthroplasty? Without such a step, it seems likely that we will only be able to speculate on the fate of the typical patient who has a shoulder replacement.
In reading the excellent observations of Jain et al. about the strong influence of surgeon and hospital volume on the quality of the result for the patient, it is irresistible to skip to the end of the paper to see what the authors propose that we do about it. This paper concludes with the statements, "Our study showed that better outcomes can be achieved for shoulder arthroplasty when patients are referred to high-volume surgeons and hospitals. This additional evidence may help in the formulation of health policies to encourage better outcomes." The readers of The Journal should be convinced of the first statement, but what can we do about the second? What health policies should be put in place to achieve better outcomes? Recognizing that it is not in the economic self-interest of general orthopaedic surgeons to have their patients "referred to high-volume surgeons," how might such a referral process be implemented?
It would seem that there are at least two policies that should be considered: information and incentivization. Information might require the Centers for Medicare and Medicaid Services to make public the surgical volumes for surgeons and medical centers in the United States as well as the data on the relation of volume to outcome (following the lead of New York's Center for Medical Consumers [www.medicalconsumers.org]). Incentivization suggests that Medicare might cover a higher percentage of the patient's cost of a shoulder arthroplasty if the procedure is performed at a higher-volume center and by a higher-volume surgeon. The forum for the consideration of these and other ideas should be distanced from the economic interests of individual surgeons and medical centers. The editorial pages of The Journal might provide the appropriately protected space for the discussion.
These comments are not intended to detract from the fine data collection and analysis carried out by the authors. Rather, they are intended to illuminate how provocative these data are and how they can lead us to ask some really important, challenging questions regarding the principles that should guide the practice of shoulder arthroplasty thirty years after its introduction and the methods by which these principles might be put into practice.
(For more conversation on this key topic, readers are referred to the JBJS Letter to The Editor written by Matsen et al.4.)
*The author did not receive grants or outside funding in support of their research or preparation of this manuscript. The author received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity (DePuy). In addition, a commercial entity (DePuy) paid or directed, or agreed to pay or direct, benefits to a research fund, foundation, educational institution, or other charitable or nonprofit organization with which the author is affiliated or associated.
1. Katz JN, Losina E, Barrett J, Phillips CB, Mahomed NN, Lew RA, Guadagnoli
E, Harris WH, Poss R, Baron JA. Association between hospital and surgeon
procedure volume and outcomes of total hip replacement in the United States
Medicare population. J Bone Joint Surg Am. 2001;83:1622-9.
2. Hasan SS, Leith JM, Smith KL, Matsen FA 3rd. The distribution of shoulder replacement among surgeons and hospitals is significantly different than that of hip or knee replacement. J Shoulder Elbow Surg. 2003;12:164-9.
3. Codman EA. The shoulder. Rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa. Boston: privately printed; 1934.
4. Matsen FA 3rd, Katz JN, Losina E, Phillips CB, Mahomed NN, Lew RA, Harris WH, Poss R, Baron JA, Fossel AH, Maher N, Barrett J, Tullar J. The relationship of surgical volume to quality of care: scientific considerations and policy implications. J Bone Joint Surg Am. 2002;84:1482-5.