Special Article

Hospital Volume and Surgical Mortality in the United States

List of authors.
  • John D. Birkmeyer, M.D.,
  • Andrea E. Siewers, M.P.H.,
  • Emily V.A. Finlayson, M.D.,
  • Therese A. Stukel, Ph.D.,
  • F. Lee Lucas, Ph.D.,
  • Ida Batista, B.A.,
  • H. Gilbert Welch, M.D., M.P.H.,
  • and David E. Wennberg, M.D., M.P.H.

Abstract

Background

Although numerous studies suggest that there is an inverse relation between hospital volume of surgical procedures and surgical mortality, the relative importance of hospital volume in various surgical procedures is disputed.

Methods

Using information from the national Medicare claims data base and the Nationwide Inpatient Sample, we examined the mortality associated with six different types of cardiovascular procedures and eight types of major cancer resections between 1994 and 1999 (total number of procedures, 2.5 million). Regression techniques were used to describe relations between hospital volume (total number of procedures performed per year) and mortality (in-hospital or within 30 days), with adjustment for characteristics of the patients.

Results

Mortality decreased as volume increased for all 14 types of procedures, but the relative importance of volume varied markedly according to the type of procedure. Absolute differences in adjusted mortality rates between very-low-volume hospitals and very-high-volume hospitals ranged from over 12 percent (for pancreatic resection, 16.3 percent vs. 3.8 percent) to only 0.2 percent (for carotid endarterectomy, 1.7 percent vs. 1.5 percent). The absolute differences in adjusted mortality rates between very-low-volume hospitals and very-high-volume hospitals were greater than 5 percent for esophagectomy and pneumonectomy, 2 to 5 percent for gastrectomy, cystectomy, repair of a nonruptured abdominal aneurysm, and replacement of an aortic or mitral valve, and less than 2 percent for coronary-artery bypass grafting, lower-extremity bypass, colectomy, lobectomy, and nephrectomy.

Conclusions

In the absence of other information about the quality of surgery at the hospitals near them, Medicare patients undergoing selected cardiovascular or cancer procedures can significantly reduce their risk of operative death by selecting a high-volume hospital.

Introduction

Over the past three decades, numerous studies have described higher rates of operative mortality with selected surgical procedures at hospitals where few such procedures are performed (low-volume hospitals).1-4 Several recent reviews suggest that thousands of preventable surgical deaths occur each year in the United States because elective but high-risk surgery is performed in hospitals that have inadequate experience with the surgical procedures involved.5-7 As part of a broader initiative aimed at improving hospital safety, a large coalition of private and public purchasers of health insurance — the Leapfrog Group — is encouraging patients undergoing one of five high-risk procedures to seek care at high-volume hospitals.8 In the lay media, there has been an emphasis on the importance of experience with particular procedures,9,10 and several consumer-oriented Web sites (e.g., http://www.healthscope.org) have begun providing patients with information about volume at hospitals near them.

Despite the recent interest in surgical volume, many question the applicability of previous research on volume and outcome to current practice.11,12 First, many studies of volume and outcome are outdated. Given that the surgical mortality associated with many procedures has fallen considerably since these studies were conducted,13,14 the relative importance of the volume of procedures performed may be declining. Second, most published studies on volume and outcome have used state-level data bases or regional populations that are served by a small number of high-volume centers.6 Whether their results are broadly generalizable is uncertain. And finally, although some procedures (e.g., cardiac surgery) have been studied extensively, the relative importance of hospital volume to mortality with many other high-risk procedures either has not been explored or has been studied in samples that were too small to permit assessment of performance at all meaningful levels of hospital volume.

To address many of these limitations, we studied surgical mortality in the Medicare population, which accounts for the majority of all patients in the United States who undergo high-risk surgery and an even larger majority of those who die after surgery.15 Using current national data (from 1994 through 1999), we studied the importance of hospital volume to the operative mortality associated with six types of cardiovascular procedures and eight types of major cancer resections.

Methods

Subjects and Data Bases

We obtained the Medicare Provider Analysis and Review (MEDPAR) files and the denominator files from the Center for Medicare and Medicaid Services for the years 1994 through 1999. These files contain hospital-discharge abstracts for the acute care hospitalizations of all Medicare recipients covered by the hospital care program (Part A). Only patients covered by fee-for-service arrangements are included in the MEDPAR file; thus, our sample excludes the approximately 10 percent of Medicare patients who were enrolled in risk-bearing health maintenance organizations during this period. We excluded patients who were under 65 years of age or over 99 years of age. The study protocol was approved by the institutional review board of Dartmouth Medical School.

Hospital Volume

Patients undergoing each of the 14 procedures examined in our analysis were identified with the use of appropriate procedure codes from the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM).16 These procedures were selected because they are relatively complex, are associated with a nontrivial risk of operative mortality, and are most often performed on an elective basis.

We focused on the total number of each type of procedure performed at a given hospital (hospital volume), not the total number of procedures involving Medicare recipients (Medicare volume), in order to place our results in the context of the volume standards suggested by the Leapfrog Group8 and others. To estimate total volumes, we examined data from the all-payer 1997 Nationwide Inpatient Sample. We determined the proportion of all patients undergoing each procedure who were covered by Medicare; the proportion ranged from 43 percent (for nephrectomy) to 75 percent (for carotid endarterectomy). To estimate the total volume at individual hospitals, we divided the observed Medicare volume (the total number of each type of procedure performed on Medicare patients during the six-year study period) by these procedure-specific proportions.

Hospital volume, expressed as the average number of procedures per year, was first evaluated as a continuous variable. To simplify the presentation of our results, however, we also created categorical variables, defining five categories of hospital volume: very low, low, medium, high, and very high. For each procedure, the hospitals were ranked in order of increasing total hospital volume, and then five volume groups were defined by the selection of whole-number cutoff points for annual volume that most closely sorted the patients into five groups of equal size (quintiles). The cutoff points were established before mortality was examined in order to avoid selecting cutoff points that could maximize the associations between volume and outcome.17 To reflect most accurately the overall institutional experience with each type of operation, we combined the replacement of aortic and mitral valves (into the single category of heart-valve replacement) and lobectomy and pneumonectomy (into the category of lung resection) in determining hospital volume. However, the outcomes of these procedures were assessed separately.

Assessment of Outcomes

In creating cohorts for the analysis of outcomes, we applied several restrictions in order to increase the homogeneity of the study samples and thus minimize the potential for confounding by case mix. For the eight types of major cancer resections, we excluded patients without an accompanying cancer-diagnosis code (related to the index procedure). Patients undergoing repair of an abdominal aortic aneurysm were excluded if they had a diagnosis or procedure code suggesting rupture of the aneurysm, thoracoabdominal aneurysm, or both. Patients undergoing coronary-artery bypass grafting were excluded if they simultaneously underwent valve replacement.

Our primary outcome measure was operative mortality, defined as the rate of death before hospital discharge or within 30 days after the index procedure. Because a large proportion of surgical deaths before discharge occurred more than 30 days after surgery, we decided that 30-day mortality alone would not adequately reflect true operative mortality. Because the length of stay did not vary systematically according to hospital volume, the inclusion of late, in-hospital deaths would not be expected to bias our results. Moreover, associations between volume and outcome were largely unchanged when we repeated our analyses using 30-day mortality alone.

Statistical Analysis

We used multiple logistic regression to examine relations between hospital volume and operative mortality, with adjustment for characteristics of the patients.18 We used the patient as the unit of analysis, with volume measured at the hospital level. We first fitted separate models for each procedure against the logarithm of volume to establish the general form of the relation. We then fitted models against the quintiles of volume for each procedure.

We adjusted for age group (65 to 69 years, 70 to 74 years, 75 to 79 years, 80 to 84 years, or 85 to 99 years), sex, race (black or nonblack), and their interactions, as well as the year of the procedure, the relative urgency of the index admission (elective, urgent, or emergency), the presence of coexisting conditions, and mean income from Social Security.18 This last measure was assessed at the ZIP Code level (on the basis of the 1990 Census file) because patient-level information on socioeconomic status is not available.

Coexisting conditions were identified with the use of information from the index admission and any other admissions that had occurred within the preceding six months. Relative to low-volume hospitals, high-volume hospitals treat a larger number of patients who have been transferred or referred from other centers. To minimize the possibility of bias due to the identification of more previous admissions (and thus more coexisting conditions) at high-volume centers, we excluded information on coexisting conditions identified at previous admissions that occurred within two weeks before the index hospitalization. For the purposes of risk adjustment, coexisting conditions (identified by their appropriate ICD-9-CM codes) were compiled into a Charlson score (the number of coexisting conditions, weighted according to their relative effects on mortality),19,20 which was modified to exclude conditions that were likely to reflect either the primary indication for surgery or postoperative complications.21,22 We also explored two alternative approaches to incorporating coexisting conditions into our risk-adjustment models: including Charlson scores with weights derived empirically for each procedure and including coexisting conditions individually by inserting into each model each condition that was present in at least 2 percent of the patients. Because all three approaches yielded virtually identical results, we report only those from the model that used the Charlson score with published weights.19

We used overdispersed binary logistic models to adjust for clustering of deaths within hospitals.23 The net effect was to increase the width of the confidence intervals between 2 percent (cystectomy) and 44 percent (lobectomy), with a mean increase of 25 percent. We computed adjusted mortality rates on the basis of the average values of the characteristics of the patients by back-transforming predicted mortality from the logistic model. Our final risk-adjustment models had intermediate discriminative ability, with C statistics ranging from 0.60 (for pneumonectomy) to 0.71 (for nephrectomy). All P values are two-tailed.

Because the Medicare files used for this analysis reflect the use of procedures among patients with fee-for-service arrangements for health care, we may have underestimated hospital volume in regions of the country that had a high penetration of Medicare managed care during the study period (mainly southern California and the Southwest). For this reason, we repeated our analyses after restricting our data set to hospital-referral regions with a penetration of Medicare managed care of less than 10 percent. Because the adjusted odds ratios for death associated with hospital volume changed negligibly as a result of this restriction, these data are not presented.

Results

Table 1. Table 1. Distribution of Patients and Hospitals among Quintiles of Volume for the 14 Procedures.

Between 1994 and 1999, approximately 2.5 million Medicare patients underwent 1 of the 14 cardiovascular or cancer-related procedures that we studied. The criteria used to define the five strata of hospital volume varied markedly according to procedure, reflecting the relative frequency with which each is performed (Table 1). Medicare volume and total volume for the 14 procedures were highly correlated at the hospital level (overall correlation coefficient, 0.97).

Table 2. Table 2. Characteristics of the Patients According to Hospital Volume.

The age and sex of the patients did not vary consistently among strata of hospital volume (Table 2). However, for most of the 14 procedures, black patients were more likely to undergo surgery at a lower-volume hospital. For most procedures, Charlson scores tended to be slightly higher at higher-volume hospitals. However, patients were more likely to be admitted nonelectively at lower-volume hospitals. This trend was more apparent with respect to several cancer-related resections (e.g., esophagectomy) than with respect to cardiovascular procedures.

Table 3. Table 3. Operative Mortality Rates and Their Association with Hospital Volume.

When it was assessed as a continuous (logarithmic) variable, hospital volume was related to both observed and adjusted operative mortality rates for all 14 procedures (P<0.001 for all trends). In terms of odds ratios for death, adjustment for characteristics of the patients attenuated the associations between volume and outcome moderately for carotid endarterectomy, colectomy, gastrectomy, esophagectomy, and pulmonary lobectomy (Table 3). Risk adjustment had negligible effect with respect to the other procedures.

Figure 1. Figure 1. Adjusted In-Hospital or 30-Day Mortality among Medicare Patients (1994 through 1999), According to Quintile of Total Hospital Volume for Peripheral Vascular Procedures (Panel A) and Cardiac Procedures (Panel B).

P<0.001 for all procedures. The outcomes for aortic-valve and mitral-valve replacement were stratified according to the total volume of heart-valve replacements. Values above the bars are the percent mortality.

Figure 2. Figure 2. Adjusted In-Hospital or 30-Day Mortality among Medicare Patients (1994 through 1999), According to Quintile of Total Hospital Volume for Resections of Gastrointestinal Cancer (Panel A) and Resections of Other Cancers (Panel B).

P<0.001 for all procedures. The outcomes for lobectomy and pneumonectomy were stratified according to the total volume of lung resections. Values above the bars are the percent mortality.

In terms of absolute differences in adjusted mortality rates, the importance of hospital volume varied markedly according to the type of procedure (Figure 1 and Figure 2). For example, for pancreatic resection, adjusted mortality rates at very-low-volume hospitals were 12.5 percent higher than at very-high-volume hospitals (16.3 percent vs. 3.8 percent) (Figure 2A). Relatively large differences in risk were also observed for esophagectomy (11.9 percent) and pneumonectomy (5.4 percent). Absolute differences in adjusted mortality rates between very-low-volume and very-high-volume hospitals were between 2 percent and 5 percent for gastrectomy, cystectomy, repair of a nonruptured abdominal aortic aneurysm, and aortic- and mitral-valve replacement, and the differences were less than 2 percent for coronary-artery bypass grafting, lower-extremity bypass, colectomy, lobectomy, and nephrectomy. The absolute difference in mortality between very-low-volume and very-high-volume hospitals was smallest for carotid endarterectomy (1.7 percent vs. 1.5 percent).

Relations between volume and outcome in the intermediate strata of hospital volume also varied widely according to the type of procedure (Figure 1 and Figure 2). For several types of procedure (including coronary-artery bypass grafting, valve replacement, and pancreatic resection), mortality declined monotonically with each stratum of increasing hospital volume. For others (including elective repair of an abdominal aortic aneurysm, gastrectomy, and pneumonectomy), differences in mortality were most apparent at the extremes of volume, whereas hospitals in the intermediate-volume strata had similar mortality rates.

Discussion

In this large, national study, higher-volume hospitals had lower operative mortality rates for six types of cardiovascular procedures and eight types of major cancer resections. However, the absolute magnitude of the relation between volume and outcome varied markedly among the types of procedures. Dramatic differences in mortality between very-low-volume and very-high-volume hospitals were observed for pancreatic resection and esophagectomy (more than 12 percent, in absolute terms), whereas relatively small differences in mortality (1 percent or less) were found for 3 of the 14 procedures examined in our analysis. These findings suggest the relative importance of hospital volume for individual patients who are considering where to undergo various procedures. From the public health perspective, however, one must also consider the total number of patients who undergo each procedure. For example, in the case of coronary-artery bypass grafting (for which volume had a moderate effect but which is very common), 314 deaths would be averted in the United States each year if the mortality rate at very-low-volume hospitals were reduced to the rate at very-high-volume centers. Conversely, in the case of pancreatic resection (for which volume had a very large effect but which is performed infrequently), lowering the mortality rate at very-low-volume centers to that observed at very-high-volume centers would avert only 32 deaths annually.

We believe that our results reflect real differences in the quality of surgery between high-volume and low-volume hospitals. First, the effect is large. For some procedures, mortality at low-volume centers was several times as high as at high-volume hospitals — a difference that is too great to be attributed to chance or unmeasured confounding. Second, relations between volume and outcome are remarkably consistent over time and across studies. According to one recent structured review of the literature, 123 of 128 analyses involving 40 different procedures (96 percent) found lower mortality at high-volume hospitals (differences were statistically significant in 80 percent of these analyses).5 Only 4 of the 128 (3 percent) found higher mortality rates at high-volume hospitals, but none of these findings were statistically significant. And finally, the link between surgical volume and mortality is clinically plausible. Although the mechanisms underlying the relations between volume and outcome have not been fully characterized, high-volume hospitals may have more surgeons who specialize in specific procedures, more consistent processes for postoperative care, better-staffed intensive care units, and greater resources, in general, for dealing with postoperative complications.

Our analysis has several limitations. First, because we studied only Medicare patients, our results may not be generalizable to patients under 65 years of age. However, there is no evidence that age affects the relations between volume and outcome. Second, our measure of volume was imperfect. We estimated total hospital volume by extrapolating from Medicare volume, not by direct measurement. Although Medicare and total volumes are highly correlated at the hospital level, there probably remains some degree of misclassification of hospital-volume status, which would tend to bias our analysis toward the null hypothesis (no effect of volume on outcome). Third, because our primary goal was to estimate the potential effect of referral policies that focus exclusively on volume, we did not attempt to adjust for characteristics of the provider that are likely to be highly correlated with volume. Analyses that aimed to assess the independent effect of hospital volume would need to account for other variables that may influence mortality, including hospital size and teaching status, the volume of procedures performed by a particular surgeon, and staffing patterns in the intensive care unit.24-27

Finally, because we relied on administrative data, we may not have accounted adequately for differences in case mix among strata of hospital volume. Administrative data are limited in their ability to differentiate patients according to the severity of illness.21,22,28,29 Age and the prevalence of coexisting conditions did not vary substantially according to hospital volume in our data set. However, even for conditions for which the procedure itself is almost always elective, patients at lower-volume hospitals were more likely to have been admitted nonelectively. Conversely, patients at higher-volume hospitals were more likely to have had recent nonelective admissions elsewhere. Although these findings raise the possibility of unmeasured differences in case mix among hospitals, we do not believe that confounding is a likely explanation for our main findings.

Although relations between volume and outcome have long been recognized, large-scale efforts to reduce surgical mortality by concentrating selected procedures in high-volume hospitals are only now beginning to gain momentum. The most visible of these efforts is being directed by the nonprofit Leapfrog Group, a coalition of more than 80 large public and private purchasers that insure more than 25 million persons. The coalition is encouraging both patients and payers to select hospitals that meet minimal volume standards for coronary-artery bypass surgery (500 procedures per year), coronary angioplasty (400 per year), carotid endarterectomy (100 per year), repair of abdominal aortic aneurysm (30 per year), and esophagectomy for cancer (6 per year). Although our analysis does not indicate that these specific volume thresholds are better than other alternatives, it does confirm that the proposed standards could reduce the surgical mortality associated with several of these procedures.

Many may object to such initiatives aimed at concentrating selected surgical procedures in high-volume hospitals. They may rightly point out that procedure volume is an imperfect proxy for quality — that some low-volume hospitals have excellent outcomes, whereas some high-volume hospitals have poor outcomes. Unfortunately, most patients facing high-risk surgery have no way of knowing the relative quality of the hospitals near them. Although several states currently have public reporting systems in place,30,31 these efforts are largely restricted to reporting on cardiac surgery. Most other procedures are not performed frequently enough to allow assessment of procedure-specific mortality at the level of the individual hospital. Thus, in the absence of better information about surgical quality, patients undergoing many types of procedures can substantially improve their odds of survival by selecting a high-volume hospital near them.

Funding and Disclosures

Supported by a grant (R01 HS10141-01) from the Agency for Healthcare Research and Quality. Dr. Birkmeyer is also supported by a Career Development Award from the Veterans Affairs Health Services Research and Development program. The views expressed herein do not necessarily represent the views of the Department of Veterans Affairs or the United States Government.

Author Affiliations

From the Veterans Affairs Outcomes Group, Department of Veterans Affairs Medical Center, White River Junction, Vt. (J.D.B., E.V.A.F., H.G.W.); the Department of Surgery, Dartmouth–Hitchcock Medical Center, Lebanon, N.H. (J.D.B.); the Center for the Evaluative Clinical Sciences, Dartmouth Medical School, Hanover, N.H. (J.D.B., T.A.S., H.G.W., D.E.W.); the Center for Outcomes Research and Evaluation, Maine Medical Center, Portland (A.E.S., F.L.L., I.B., D.E.W.); and the Department of Surgery, University of California, San Francisco (E.V.A.F.).

Address reprint requests to Dr. Birkmeyer at the Veterans Affairs Outcomes Group (111B), Veterans Affairs Medical Center, White River Junction, VT 05009, or at .

References (31)

  1. 1. Luft HS, Bunker JP, Enthoven AC. Should operations be regionalized? The empirical relation between surgical volume and mortality. N Engl J Med 1979;301:1364-1369

  2. 2. Hannan EL, O'Donnell JF, Kilburn H Jr, Bernard HR, Yazici A. Investigation of the relationship between volume and mortality for surgical procedures performed in New York State hospitals. JAMA 1989;262:503-510

  3. 3. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 1998;280:1747-1751

  4. 4. Flood AB, Scott WR, Ewy W. Does practice make perfect? I. The relation between hospital volume and outcomes for selected diagnostic categories. Med Care 1984;22:98-114

  5. 5. Dudley RA, Johansen KL, Brand R, Rennie DJ, Milstein A. Selective referral to high-volume hospitals: estimating potentially avoidable deaths. JAMA 2000;283:1159-1166

  6. 6. Halm EA, Lee C, Chassin MR. How is volume related to quality in health care? A systematic review of the research literature. Washington, D.C.: Institute of Medicine, 2000.

  7. 7. Birkmeyer JD, Lucas FL, Wennberg DE. Potential benefits of regionalizing major surgery in Medicare patients. Eff Clin Pract 1999;2:277-283

  8. 8. Birkmeyer JD, Finlayson EVA, Birkmeyer CM. Volume standards for high-risk surgical procedures: potential benefits of the Leapfrog initiative. Surgery 2001;130:415-422

  9. 9. Marsa L. Not every hospital is created equal. Los Angeles Times. September 25, 2000:S1.

  10. 10. Comarow A. Higher volume, fewer deaths. U.S. News & World Report. July 17, 2000:68-70.

  11. 11. Russell TR. Volume standards for high-risk questions: an American College of Surgeons' view. Surgery 2001;130:423-424

  12. 12. Khuri SF. Surgeons, not General Motors, should set standards for surgical care. Surgery 2001;130:429-431

  13. 13. Ghali WA, Ash AS, Hall RE, Moskowitz MA. Statewide quality improvement initiatives and mortality after cardiac surgery. JAMA 1997;277:379-382

  14. 14. Katz DJ, Stanley JC, Zelenock GB. Operative mortality rates for intact and ruptured abdominal aortic aneurysms in Michigan: an eleven-year statewide experience. J Vasc Surg 1994;19:804-817

  15. 15. Finlayson EVA, Birkmeyer JD. Operative mortality with elective surgery in older adults. Eff Clin Pract 2001;4:172-177[Erratum, Eff Clin Pract 2001;4:235.]

  16. 16. Department of Health and Human Services. The international classification of diseases, 9th rev., clinical modification: ICD-9-CM. Washington, D.C.: Government Printing Office, 1998.

  17. 17. Laks MP, Cohen T, Hack R. Volume of procedures at transplantation centers and mortality after liver transplantation. N Engl J Med 2000;342:1527-1527

  18. 18. McCullagh P, Nelder JA. Generalized linear models. 2nd ed. New York: Chapman & Hall, 1989.

  19. 19. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-383

  20. 20. Romano PS, Roos LL, Jollis JG. Adapting a clinical comorbidity index for use with ICD-9-CM administrative data: differing perspectives. J Clin Epidemiol 1993;46:1075-1079

  21. 21. Iezzoni LI, Foley SM, Daley J, Hughes J, Fisher ES, Heeren T. Comorbidities, complications, and coding bias: does the number of diagnosis codes matter in predicting in-hospital mortality? JAMA 1992;267:2197-2203

  22. 22. Jencks SF, Williams DK, Kay TL. Assessing hospital-associated deaths from discharge data: the role of length of stay and comorbidities. JAMA 1988;260:2240-2246

  23. 23. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121-130

  24. 24. Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA 1999;281:1310-1317

  25. 25. Ghorra S, Reinert SE, Cioffi W, Buczko G, Simms HH. Analysis of the effect of conversion from open to closed surgical intensive care unit. Ann Surg 1999;229:163-171

  26. 26. Taylor DH Jr, Whellan DJ, Sloan FA. Effects of admission to a teaching hospital on the cost and quality of care for Medicare beneficiaries. N Engl J Med 1999;340:293-299

  27. 27. Kelly JV, Hellinger FJ. Physician and hospital factors associated with mortality of surgical patients. Med Care 1986;24:785-800

  28. 28. Fisher ES, Whaley FS, Krushat WM, et al. The accuracy of Medicare's hospital claims data: progress has been made, but problems remain. Am J Public Health 1992;82:243-248

  29. 29. Hsia DC, Krushat WM, Fagan AB, Tebbutt JA, Kusserow RP. Accuracy of diagnostic coding for Medicare patients under the prospective-payment system. N Engl J Med 1988;318:352-355[Erratum, N Engl J Med 1990;322:1540.]

  30. 30. Schneider EC, Epstein AM. Use of public performance data reports: a survey of patients undergoing cardiac surgery. JAMA 1998;279:1638-1642

  31. 31. Hannan EL, Kumar D, Racz M, Siu AL, Chassin MR. New York State's Cardiac Surgery Reporting System: four years later. Ann Thorac Surg 1994;58:1852-1857

Citing Articles (3608)

Only the 1000 most recent citing articles are listed here.

    Letters

    Figures/Media

    1. Table 1. Distribution of Patients and Hospitals among Quintiles of Volume for the 14 Procedures.
      Table 1. Distribution of Patients and Hospitals among Quintiles of Volume for the 14 Procedures.
    2. Table 2. Characteristics of the Patients According to Hospital Volume.
      Table 2. Characteristics of the Patients According to Hospital Volume.
    3. Table 3. Operative Mortality Rates and Their Association with Hospital Volume.
      Table 3. Operative Mortality Rates and Their Association with Hospital Volume.
    4. Figure 1. Adjusted In-Hospital or 30-Day Mortality among Medicare Patients (1994 through 1999), According to Quintile of Total Hospital Volume for Peripheral Vascular Procedures (Panel A) and Cardiac Procedures (Panel B).
      Figure 1. Adjusted In-Hospital or 30-Day Mortality among Medicare Patients (1994 through 1999), According to Quintile of Total Hospital Volume for Peripheral Vascular Procedures (Panel A) and Cardiac Procedures (Panel B).

      P<0.001 for all procedures. The outcomes for aortic-valve and mitral-valve replacement were stratified according to the total volume of heart-valve replacements. Values above the bars are the percent mortality.

    5. Figure 2. Adjusted In-Hospital or 30-Day Mortality among Medicare Patients (1994 through 1999), According to Quintile of Total Hospital Volume for Resections of Gastrointestinal Cancer (Panel A) and Resections of Other Cancers (Panel B).
      Figure 2. Adjusted In-Hospital or 30-Day Mortality among Medicare Patients (1994 through 1999), According to Quintile of Total Hospital Volume for Resections of Gastrointestinal Cancer (Panel A) and Resections of Other Cancers (Panel B).

      P<0.001 for all procedures. The outcomes for lobectomy and pneumonectomy were stratified according to the total volume of lung resections. Values above the bars are the percent mortality.