A body of literature has demonstrated an association between higher hospital and surgeon procedural volume and decreased morbidity and mortality.1–3 The association between surgical volume and outcomes is most pronounced for procedures associated with significant morbidity and mortality, such as cardiovascular operations and high-risk oncologic surgeries.4–6 Outcomes for hysterectomy, for both benign and malignant disease, are improved when the procedure is performed at high-volume hospitals and by high-volume surgeons; however, the magnitude of this association is lower than for other procedures.7–9
Despite the benefits of treatment by high-volume surgeons, there are significant disparities in access to such care. Black patients, those without commercial insurance, and patients with lower socioeconomic status are less likely to be referred to high-volume hospitals for cancer and other complex surgeries.10–12 For hysterectomy, Black and Hispanic patients are more likely to undergo open hysterectomies at smaller hospitals, resulting in increased perioperative complications.13 A recent study examining survival in endometrial cancer found that the effect of race on mortality was mitigated, albeit not eliminated, by treatment at a higher volume hospital.14 An important goal of regionalization of surgical procedures to high-volume centers is to reduce disparities and improve care for all patients. To date, data describing whether disparities in care and outcomes exist within high-volume hospitals are limited.
The objective of our study was to examine disparities in access to care and outcomes for patients who are undergoing hysterectomy at high-volume hospitals. Specifically, we examined access to high-volume surgeons in comparison with low-volume surgeons within high-volume hospitals, and we compared perioperative morbidity and mortality between high and low-volume surgeons within these high-volume centers.
For this analysis, we used data from SPARCS (the Statewide Planning and Research Cooperative System), which is maintained by the New York State Department of Health.15 SPARCS is an all-payer database that captures patient characteristics, diagnoses, services, and charges for hospital inpatient admissions and outpatient visits. Encrypted physician and hospital identifiers, and limited hospital information are included. Data quality is ensured through periodic reviews and by comparing SPARCS data with data from other Department of Health databases. This study used deidentified data and was classified as nonhuman subject research by the Columbia University Institutional Review Board.
Women who underwent hysterectomy (abdominal, robot-assisted, laparoscopic, and vaginal) from 2000 to 2014 were selected. Annualized hospital procedural volume was then estimated for each hospital. The annualized hysterectomy volume was calculated as the number of hysterectomies a given center performed divided by the total number of years in which the hospital performed at least one hysterectomy. Hospital volume was then visually inspected, and the top quartile by volume hospitals were selected as high-volume centers for further analysis. Among the high-volume hospitals, patients were excluded if they had a hysterectomy before admission or if they had an obstetric-related hysterectomy (Fig. 1). The attending surgeon for each hysterectomy was then identified. Within SPARCS, each physician is assigned a unique identification number that can be tracked across hospitals. Patients with an invalid or missing physician identifier also were excluded. The surgeons at the high-volume hospitals were then selected, and their annualized hysterectomy volume across all hospitals in New York State was estimated. The data were visually inspected, and the surgeons were categorized into four quartiles based on the annualized hysterectomy volume. The lowest-volume quartile performed a mean of one procedure per year, the second quartile 1.5 per year (range 1.1–1.9), the third quartile 3.7 per year (range 2.0–6.8), and the high-volume quartile 20.7 per year (range 6.9–248.9).
The outcomes of interest were perioperative morbidity and mortality. Perioperative morbidity was defined as the occurrence of an intraoperative complication, surgical site complication, or medical complications as previously defined during the index admission.16 Intraoperative complications included vascular, gastrointestinal, genitourinary, or neurologic injury, or reoperation during the index admission. Surgical site complications included wound complications, abscess, hemorrhage, gastrointestinal bleeding, hemorrhage, bowel obstruction, and ileus. Medical complications included myocardial infarct, cardiopulmonary arrest, respiratory failure, renal failure, stroke, sepsis, pneumonia, and arrhythmia. We defined inpatient mortality as death during the index admission. Blood transfusion during the index admission was also examined.
Patient demographic characteristics included year of surgery, age at surgery (younger than 40 years, 40–49, 50–59, 60–69, 70 or older), race and ethnicity (White, Black, Hispanic, other, unknown), and insurance type (none, Medicare, Medicaid, private insurance, other). Comorbidity score was estimated using the Elixhauser Comorbidity Index and categorized as 0, 1, or 2 or higher.17 Each procedure was classified as elective or emergent–urgent. Indications for surgery included leiomyoma, endometriosis, abnormal bleeding, ovarian cysts, pelvic organ prolapse, uterine cancer, cervical cancer, and ovarian cancer (including fallopian tube and primary peritoneal). These indications were not mutually exclusive. The route of hysterectomy was categorized as abdominal, laparoscopic, robotic-assisted, or vaginal. Performance of concomitant oophorectomy was noted for each patient.
Clinical and demographic characteristics of the patients were compared across the surgeon volume quartiles using chi square tests. A multinomial logistic regression model using generalized estimation equation to account for hospital clustering was used to estimate clinical and demographic characteristics associated with treatment by a low-volume surgeon in comparison with a high-volume surgeon.
The associations between surgeon volume, and morbidity and mortality were examined using marginal log-linear regression models with Poisson distribution and the log link function, adjusting for hospital clustering. Covariates in the model included age, race, health insurance, year of surgery, Elixhauser comorbidity score, admission type, indication for surgery, and performance of oophorectomy. Results are reported as adjusted risk ratios.
Given the association between surgeon volume and urgency of the procedure and presence of cancer, sensitivity analyses were performed stratified by surgical urgency (emergent–urgent vs elective) and presence of cancer. All analyses were conducted with SAS 9.4. All statistical tests were two-sided. P<.05 was considered statistically significant.
We identified a total of 300,586 patients who underwent hysterectomy by 5,505 surgeons at 59 high-volume hospitals in New York State between 2000 and 2014 (Table 1). There were 2,105 (38.2%) low-volume surgeons who treated 2,900 (1.0%) patients. In contrast, 1,377 (25.1%) high-volume surgeons operated on 262,005 (87.2%) of the patients.
The majority of cases were abdominal hysterectomies (57.5%), followed by laparoscopic (23.9%), vaginal (13.2%), and robotic-assisted (5.3%) procedures. A minority of cases were performed for gynecologic cancer (19.1%), with the remainder performed for benign indications. The majority of patients (64.4%) were 40–59 years of age. White patients accounted for 63.7% of the cohort, 15.1% of patients were Black, and 8.5% were Hispanic. Most patients (77.4%) had commercial insurance, and 28.9% of patients had a comorbidity score of 2 or higher.
Women treated by low-volume surgeons, in comparison with high-volume surgeons, were more often Black (19.4% vs 14.3%; adjusted odds ratio [aOR] 1.26; 95% CI 1.09–1.46) and had Medicare insurance (20.6% vs 14.5%; aOR 1.22; 95% CI 1.04–1.42) (Table 2). Low-volume surgeons were more likely to perform both emergent–urgent procedures (26.1% vs 6.4%; aOR 3.91; 95% CI 3.26–4.69) and abdominal hysterectomy, in comparison with minimally invasive hysterectomy (77.8% vs 54.7%; aOR 1.91; 95% CI 1.62–2.24). Lower volume surgeons were also more likely to perform surgery on patients with greater comorbidity (comorbidity score higher than 2; 43.4% vs 29.2%; aOR 1.67; 95% CI 1.47–1.89).
The overall complication rate was 31.0% for patients treated by low-volume surgeons, compared with 10.3% for those treated by high-volume surgeons (P<.001) (adjusted risk ratios [aRR] 1.84; 95% CI 1.71–1.98) (Table 3). Intraoperative complications (10.2% vs 3.3%; aRR 2.15; 95% CI 1.90–2.44), surgical site complications (14.9% vs 4.3%; aRR 1.85; 95% CI 1.67–2.04), medical complications (19.6% vs 5.1%; aRR 2.09; 95% CI 1.88–2.31), and transfusion (37.1% vs 11.6%; aRR 1.68; 95% CI 1.52–1.86) were all more common in patients operated on by low-volume surgeons. The perioperative mortality rate was 2.2% for patients treated by low-volume surgeons, compared with 0.2% for those operated on by high-volume surgeons (P<.001) (aRR 3.04; 95% CI 2.20–4.21).
Because low-volume surgeons were more likely to perform urgent and emergent procedures, we performed sensitivity analyses stratified by urgency of the procedure. Among patients who underwent procedures classified as emergent–urgent, those operated on by low-volume surgeons were more likely to experience any complications (53.2% vs 22.1%; aRR 1.62; 95% CI 1.47–1.78) and had a higher inpatient mortality rate (6.7% vs 1.1%; aRR 2.76; 95% CI 1.91–3.99) (Table 4). Similar findings were noted for elective procedures; any complications (24.2% vs 10.7%; aRR 1.97; 95% CI 1.81–2.14), and inpatient mortality (0.6% vs 0.1%; aRR 2.76; 95% CI 1.91–3.99) were higher for low-volume surgeons, in comparison with high-volume surgeons.
In another series of sensitivity analyses limited to cancer patients, both complications (43.2% vs 20.6%; aRR 1.36; 95% CI 1.22–1.51) and mortality (5.1% vs 0.6%; aRR 2.59; 95% CI 1.63–4.12) were higher for low-volume surgeons compared with high-volume surgeons (Table 4). Similar findings were seen for noncancer surgery; complications (28.4% vs 7.7%; aRR 2.15; 95% CI 1.98–2.34) and mortality (1.6% vs 0.04%: aRR 3.84; 95% CI 2.41–6.14) were higher for the low-volume surgeons compared with high-volume surgeons.
These findings suggest that within high-volume hospitals, disparities in access to high-volume surgeons as well as outcomes remain. Within high-volume hospitals, women treated by low-volume surgeons were more often Black, had Medicare insurance, had more comorbidities, and were more likely to undergo abdominal hysterectomy. Of concern, those patients who underwent surgery with low-volume surgeons were at substantially greater risk for perioperative morbidity and mortality.
The relative importance of hospital and surgeon volume on outcome varies across procedures.2 Recently, a number of studies have demonstrated that individual surgeon experience is more important in achieving better outcomes than performance of the procedure at a high-volume hospital. An analysis of patients undergoing pancreaticoduodenectomy noted that the benefits of a high-volume hospital can accompany high-volume surgeons when they relocate to a low-volume hospital; the study concluded that optimal outcomes are more a result of the “who” not “where.”18 In another study, within a single high-volume center, acute type A aortic dissection repair by low-volume surgeons had a nearly fourfold increase in in-hospital mortality compared with procedures performed by high-volume teams, emphasizing the critical importance of surgeon experience for high-risk surgeries. For hysterectomy, our findings also suggest that the benefits of treatment at a high-volume center may be blunted when the procedure is performed by a low-volume gynecologic surgeon. Both morbidity and mortality were substantially higher for low-volume surgeons in comparison with high-volume surgeons.
The current study also highlights the racial and socioeconomic disparities in access to high-volume surgeons for hysterectomy. Underinsured patients and non-White patients have consistently been noted to be less likely to receive care at either a high-volume hospital or by a high-volume surgeon.19 This disparity in access can be explained in part by systematic barriers including geography and financial incentives, as higher-quality health care professionals tend to attract patients with commercial insurance coverage, who are disproportionately White.20 A survey of Medicare patients who underwent high-risk operations reported that their referring physician was more likely to be the main decision maker about where and with whom the patient would have surgery21; as such, White patients and those with private insurance may have access to better-informed referral networks. Additionally, underinsured patients may be more likely to reside in rural areas with limited access to high-volume surgeons.22,23 However, our data are disturbing in that, even when patients receive care at a high-volume center, these disparities in access to high-volume surgeons remain. Paradoxically, regionalization of surgical care for disadvantaged patient groups may have little effect on improving outcomes if these patients still do not have access to high-volume surgeons.1
Even within high-volume hospitals, a significant number of gynecologic surgeons who perform hysterectomy had a very low volume. Within our cohort, 38% of the surgeons were classified in the lowest volume quartile, which was defined as an annualized volume of one hysterectomy per year. We have previously demonstrated that outcomes for very low-volume surgeons who perform hysterectomy are poor.8 The low-volume surgeons in our cohort performed a disproportionate number of urgent and emergent surgeries, suggesting that at least a proportion of these physicians may simply be providing emergent coverage rather than performing elective procedures. However, even among the lowest volume surgeons, 70% of these cases were elective and outcomes were inferior to higher volume surgeons for both elective and nonelective procedures. It is unclear why such a large number of surgeons performing hysterectomy at high-volume centers have such a low-volume, and this phenomenon clearly warrants investigation in other procedures.
In this study, low-volume surgeons were more likely to perform abdominal hysterectomy compared with a minimally invasive approach. Minimally invasive hysterectomy, when feasible, has been shown to have several advantages over abdominal hysterectomy including shorter length of stay, decreased hospital costs, and more rapid recovery.24 Concordant with our data, previous research has demonstrated racial and socioeconomic disparities in access to minimally invasive hysterectomy, even after accounting for clinical differences.13,25–29 Based on the results of this study, it is plausible that these disparities in access to minimally invasive hysterectomy may be mitigated, at least in part, by surgeon volume.
This study has a number of important limitations. First, as with any study of observational data, there may have been under coding of some of the outcomes of interest. To mitigate this bias, we only included major perioperative outcomes that were likely to generate a billing claim. Second, we lack data on some unmeasured confounders that undoubtedly influenced outcomes including clinical characteristics, such as surgical history and complexity. Third, a unique strength of SPARCS is the ability to track surgeons across hospitals; however, we lack more detailed information on surgeon characteristics that may have influenced outcomes. Fourth, our study spanned 15 years, a relatively long period in which practice patterns evolved. Additionally, as some outcomes, such as mortality, were relatively rare, our power to detect differences is limited in some cohorts, and our models are subject to overfitting. Lastly, our analysis was limited to New York State. Although the cohort includes a diverse sociodemographic makeup, these findings may not be generalizable to other states.
In conclusion, this study demonstrates increased perioperative morbidity and mortality for patients who underwent hysterectomy by low-volume surgeons, in comparison with high-volume surgeons, at high-volume hospitals. Importantly, patients operated on by lower volume surgeons were more likely to be Black or to have Medicare insurance, have more comorbidities, and undergo abdominal hysterectomy. Although centralization of complex surgical care to higher volume hospitals may have benefit, there are additional surgeon-level factors that must be considered to address disparities in access to high-quality care for patients undergoing hysterectomy. Further work is needed to implement initiatives to ensure all women who are undergoing hysterectomy receive guideline-adherent care and to investigate referral patterns to identify areas for intervention to reduce disparities.
1. Birkmeyer JD, Siewers AE, Finlayson EV, Stukel TA, Lucas FL, Batista I, et al. Hospital volume and surgical mortality in the United States. New Engl J Med 2002;346:1128–37. doi: 10.1056/NEJMsa012337
2. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. New Engl J Med 2003;349:2117–27. doi: 10.1056/NEJMsa035205
3. Maruthappu M, Gilbert BJ, El-Harasis MA, Nagendran M, McCulloch P, Duclos A, et al. The influence of volume and experience on individual surgical performance: a systematic review. Ann Surg 2015;261:642–7. doi: 10.1097/SLA.0000000000000852
4. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. Jama 1998;280:1747–51. doi: 10.1001/jama.280.20.1747
5. Gonzalez AA, Dimick JB, Birkmeyer JD, Ghaferi AA. Understanding the volume-outcome effect in cardiovascular surgery: the role of failure to rescue. JAMA Surg 2014;149:119–23. doi: 10.1001/jamasurg.2013.3649
6. Lapar DJ, Mery CM, Kozower BD, Kern JA, Kron IL, Stukenborg GJ, et al. The effect of surgeon volume on mortality for off-pump coronary artery bypass grafting. J Thorac Cardiovasc Surg 2012;143:854–63. doi: 10.1016/j.jtcvs.2011.12.048
7. Wright JD, Ruiz MP, Chen L, Gabor LR, Tergas AI, St Clair CM, et al. Changes in surgical volume and outcomes over time for women undergoing hysterectomy for endometrial cancer. Obstet Gynecol 2018;132:59–69. doi: 10.1097/AOG.0000000000002691
8. Ruiz MP, Chen L, Hou JY, Tergas AI, St Clair CM, Ananth CV, et al. Outcomes of hysterectomy performed by very low-volume surgeons. Obstet Gynecol 2018;131:981–90. doi: 10.1097/AOG.0000000000002597
9. Wallenstein MR, Ananth CV, Kim JH, Burke WM, Hershman DL, Lewin SN, et al. Effect of surgical volume on outcomes for laparoscopic hysterectomy for benign indications. Obstet Gynecol 2012;119:709–16. doi: 10.1097/AOG.0b013e318248f7a8
10. Wasif N, Etzioni D, Habermann EB, Mathur A, Pockaj BA, Gray RJ, et al. Racial and socioeconomic differences in the use of high-volume commission on cancer-accredited hospitals for cancer surgery in the United States. Ann Surg Oncol 2018;25:1116–25. doi: 10.1245/s10434-018-6374-0
11. Liu JH, Zingmond DS, McGory ML, SooHoo NF, Ettner SL, Brook RH, et al. Disparities in the utilization of high-volume hospitals for complex surgery. Jama 2006;296:1973–80. doi: 10.1001/jama.296.16.1973
12. Nabi J, Tully KH, Cole AP, Marchese M, Cone EB, Melnitchouk N, et al. Access denied: the relationship between patient insurance status and access to high-volume hospitals. Cancer 2021;127:577–85. doi: 10.1002/cncr.33237
13. Mehta A, Xu T, Hutfless S, Makary MA, Sinno AK, Tanner EJ, et al. Patient, surgeon, and hospital disparities associated with benign hysterectomy approach and perioperative complications. Am J Obstet Gynecol 2017;216:497.e1–10. doi: 10.1016/j.ajog.2016.12.020
14. Buskwofie A, Huang Y, Tergas AI, Hou JY, Ananth CV, Neugut AI, et al. Impact of hospital volume on racial disparities and outcomes for endometrial cancer. Gynecol Oncol 2018;149:329–36. doi: 10.1016/j.ygyno.2018.02.019
15. New York State Department of Health. Statewide Planning and Research Cooperative System (SPARCS). Accessed January 21, 2021. https://www.health.ny.gov/statistics/sparcs/
16. Wright JD, Burke WM, Tergas AI, Hou JY, Huang Y, Hu JC, et al. Comparative effectiveness of minimally invasive hysterectomy for endometrial cancer. J Clin Oncol 2016;34:1087–96. doi: 10.1200/JCO.2015.65.3212
17. van Walraven C, Austin PC, Jennings A, Quan H, Forster AJ. A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care 2009;47:626–33. doi: 10.1097/MLR.0b013e31819432e5
18. Toomey PG, Teta AF, Patel KD, Ross SB, Rosemurgy AS. High-volume surgeons vs high-volume hospitals: are best outcomes more due to who or where? Am J Surg 2016;211:59–63. doi: 10.1016/j.amjsurg.2015.08.021
19. Epstein AJ, Gray BH, Schlesinger M. Racial and ethnic differences in the use of high-volume hospitals and surgeons. Arch Surg 2010;145:179–86. doi: 10.1001/archsurg.2009.268
20. Mukamel DB, Weimer DL, Buchmueller TC, Ladd H, Mushlin AI. Changes in racial disparities in access to coronary artery bypass grafting surgery between the late 1990s and early 2000s. Med Care 2007;45:664–71. doi: 10.1097/MLR.0b013e3180325b81
21. Wilson CT, Woloshin S, Schwartz LM. Choosing where to have major surgery: who makes the decision? Arch Surg 2007;142:242–6. doi: 10.1001/archsurg.142.3.242
22. Chow CJ, Al-Refaie WB, Abraham A, Markin A, Zhong W, Rothenberger DA, et al. Does patient rurality predict quality colon cancer care?: a population-based study. Dis colon rectum 2015;58:415–22. doi: 10.1097/DCR.0000000000000173
23. Wallace AE, Young-Xu Y, Hartley D, Weeks WB. Racial, socioeconomic, and rural-urban disparities in obesity-related bariatric surgery. Obes Surg 2010;20:1354–60. doi: 10.1007/s11695-009-0054-x
24. Aarts JW, Nieboer TE, Johnson N, Tavender E, Garry R, Mol BW, et al. Surgical approach to hysterectomy for benign gynaecological disease. The Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD003677. doi: 10.1002/14651858.CD003677.pub5
25. Price JT, Zimmerman LD, Koelper NC, Sammel MD, Lee S, Butts SF. Social determinants of access to minimally invasive hysterectomy: reevaluating the relationship between race and route of hysterectomy for benign disease. Am J Obstet Gynecol 2017;217:572.e1–10. doi: 10.1016/j.ajog.2017.07.036
26. Ko JS, Suh CH, Huang H, Zhuo H, Harmanli O, Zhang Y. Association of race/ethnicity with surgical route and perioperative outcomes of hysterectomy for leiomyomas. J Minim Invasive Gynecol 2020 Nov 23 [Epub ahead of print].
27. Cohen SL, Vitonis AF, Einarsson JI. Updated hysterectomy surveillance and factors associated with minimally invasive hysterectomy. JSLS 2014;18:e2014.00096. doi: 10.4293/JSLS.2014.00096
28. Sanei-Moghaddam A, Kang C, Edwards RP, Lounder PJ, Ismail N, Goughnour SL, et al. Racial and socioeconomic disparities in hysterectomy route for benign conditions. J racial ethnic Health disparities 2018;5:758–65. doi: 10.1007/s40615-017-0420-7
29. Alexander AL, Strohl AE, Rieder S, Holl J, Barber EL. Examining disparities in route of surgery and postoperative complications in Black race and hysterectomy. Obstet Gynecol 2019;133:6–12. doi: 10.1097/AOG.0000000000002990