Relationship between surgical volume and patient outcomes

Transcription

1 7 Relationship between surgical volume and patient outcomes MASSIMILIANO SPALIVIERO AND JAMES A. EASTHAM The authors discuss the inverse relationship between hospital volume of surgical procedures and rates of operative mortality, looking in particular at surgeons experience of performing radical prostatectomy and patient outcomes. In recent years, numerous studies have examined the inverse relation between hospital volume of surgical procedures and rates of operative mortality, which for selected surgical procedures was found to be higher at low-volume hospitals. 1, As a result, a significant amount of surgical deaths in patients undergoing elective but high-risk surgery in hospitals with limited experience could be prevented each year. 3 Differences between high- and low-volume hospital outcomes may be a reflection of several factors, either alone or in combination, including the skill and experience of the individual surgeon, better organised postoperative care, intensive care units staffed by intensivists, and greater resources for dealing with postoperative complications available at high-volume hospitals. 1, These factors may also result in lower rates of failure to rescue, defined as death in a patient after a major complication, and ultimately influence postoperative mortality. Measures of surgical quality, such as hospital and surgeon volumes, have the potential to help patients select the best provider and hospital; to help payers to create incentives that direct patients to centres with the best outcomes; and to encourage providers to develop and implement quality improvement protocols.,7 Adam Gault/Science Photo Library Massimiliano Spaliviero, MD, Postdoctoral Fellow; James A. Eastham, MD, Chief, Urology Service, Sidney Kimmel Center for Prostate and Urologic Cancers, Memorial Sloan-Kettering Cancer Center, New York, NY, USA TRENDS IN UROLOGY & MEN S HEALTH MARCH/APRIL 1

2 HOSPITAL VOLUME AND OUTCOMES Birkmeyer et al. 1 examined the mortality associated with 1 cardiovascular or cancer-related procedures between 199 and 1999 (total number of procedures,. million) using data from the national Medicare claims database and the Nationwide Inpatient Sample. Hospital volume, defined as the average number of procedures per year, was first evaluated as a continuous variable, and then stratified into five categories of hospital volume (eg very low, low, medium, high, very high). After adjusting for patient characteristics, the relationship between hospital volume (total number of procedures/year) and mortality (in-hospital or within 3 days of surgery) was described. Patient age and gender were similar among strata of hospital volume; however, Charlson score in patients managed at higher-volume hospitals was somewhat higher, and rates of non-elective admissions were higher at lower-volume hospitals. Also, black patients were more likely to receive surgical treatment at lower-volume hospitals. For all 1 procedures, higher hospital volume was related to lower observed and adjusted mortality, although the relative importance of volume varied markedly according to the type of procedure. When comparing very-low-volume hospitals to very-high-volume hospitals, the absolute difference in adjusted mortality rates was over 1% for pancreatic resection (1.3 versus 3.%, respectively); greater than % for oesophagectomy and pneumonectomy; % for gastrectomy, cystectomy, repair of a non-ruptured abdominal aneurysm, and replacement of an aortic or mitral valve; less than % for coronary-artery bypass grafting, lowerextremity bypass, colectomy, lobectomy and nephrectomy; and only.% for carotid endarterectomy, 1.7 versus 1.%, respectively. SURGEON VOLUME AND OUTCOMES Surgeon volume is a strong independent predictor of operative mortality. The impact of the surgeon s experience on the observed effects of hospital volume and on the operative risk associated with eight cardiovascular procedures or cancer resections was evaluated by Birkmeyer et al. in 7 1 patients using information from the national Medicare claims database for Operative mortality was defined as death before hospital discharge or within 3 days after the index procedure. Surgeon volume was first evaluated as a continuous variable, and then stratified into three categories (low, medium and high). After adjusting for patient and hospital characteristics, surgeon volume (Figure 1) was inversely related to operative mortality for all eight procedures (p=.3 for lung resection; p<.1 for all other procedures). For patients treated by a low-volume surgeon the adjusted odds ratio for increase in operative mortality ranged from 1. for lung resection to 3.1 for pancreatic resection higher than a high-volume surgeon. Surgeon volume accounted for 1% of the effect of the hospital volume for aortic-valve replacement; 7% for elective repair of an abdominal aortic aneurysm; % for pancreatic resection; 9% for coronary-artery bypass grafting; % for oesophagectomy; 39% for cystectomy; and % for lung resection. For most procedures, mortality rate was higher among patients of low-volume surgeons, regardless of the surgical volume of the hospital in which they practised (Figure ). Hospital volume was inversely related (p<.1) to operative mortality for all procedures, with the exception (p=.) of carotid endarterectomy, when evaluated as a continuous variable. However, it remained a significant predictor of decreased mortality for only four procedures (repair of an abdominal aortic aneurysm, cystectomy, lung resection and pancreatic resection) after being adjusted for surgeon volume. The authors concluded that surgeon volume mediates the associations between hospital volume and operative mortality and that patients of high-volume surgeons may have better chances of survival, even at high-volume hospitals. For some procedures, eg outpatient surgeries, the technical skill of the surgeon is the predominant determinant of operative outcomes. In contrast, complications such as multisystem failures and death resulting from major surgeries might not be directly related to the procedures themselves. In those cases, the quality of the perioperative care provided by better hospital-based services becomes the most important determinant of outcomes. HOSPITAL VOLUME AND COSTS The body of evidence showing associations between hospital and surgeon volumes and outcomes has led to increased referral of high-risk procedures to high-volume centres and, in turn, to market concentration. This, however, raised the concern that better outcomes could have been achieved by high-volume hospitals at significantly increased costs secondary to an increased utilisation of resources. On the other hand, high-volume centres might have been able to reduce total costs through greater economies of scale and by preventing complications and readmissions. 9 Using the Surveillance, Epidemiology, and End Results Medicare database, Nathan et al. 1 analysed the risk-adjusted 3-day episode Medicare payments for the index hospitalisation, readmissions, physician services, emergency room visits and post-discharge ancillary care for colectomies, 7 cystectomies, 11 pancreatectomies, 7 proctectomies, 1 prostatectomies and 1 11 pulmonary lobectomies performed in elderly patients with non-metastatic disease. The majority (7 7%) of the total episode payment for each procedure consisted of the payment for the index admission, followed by payment for physician services ( %), readmission ( 11%) and other services ( 1%). No meaningful associations between total risk-adjusted payments and TRENDS IN UROLOGY & MEN S HEALTH MARCH/APRIL 1

3 9 a b Carotid endarterectomy < >. Resection for lung cancer < >17. Aortic valve replacement <..-. < <..-3. >3. hospital volume were detected. Surgical mortality was low (.% for prostatectomy to 3.% for pancreatectomy). Complication rates, which did not correlate with hospital volume, ranged from 1% (prostatectomy) to % (lobectomy). Coronary-artery bypass grafting..3. < >1. Annual surgeon volume <..-. >. Elective repair of an abdominal aortic aneurysm < >17. Cystectomy Oesophagectomy Pancreatic resection Annual surgeon volume <..-. >. Figure 1. Adjusted operative mortality among Medicare patients in 199 and 1999, according to surgeon-volume stratum, for a) four cardiovascular procedures and b) four cancer resections. Operative mortality was defined as the rate of death before hospital discharge or within 3 days after the index procedure. Surgeon volume was determined on the basis of the total number of procedures performed in both Medicare and non-medicare patients. p<.1 for all procedures except resection for lung cancer; p=.3 for lung resection; p values reflect associations between operative mortality and volume assessed as a continuous variable (reproduced with permission from Birkmeyer et al. ) Occurrence of complications was associated with 7 7% higher costs. The authors concluded that improvements in short-term mortality at high-volume hospitals do not come at higher cost. VOLUME OUTCOME RELATIONSHIP FOR RADICAL PROSTATECTOMY Increased surgeon s experience with radical prostatectomy (RP) was found to have a strong correlation with cancer control, as suggested by a risk of recurrence at years of 11% in patients operated by surgeons who had completed their learning curve ( prior RPs) compared to a risk of 1% for patients treated by surgeons who had performed only 1 prior RPs. 11 A slower learning curve was noted for laparoscopic RPs. 1 Using the Nationwide Inpatient Sample and the Healthcare Research and Quality databases, Savage et al. 13 found that in the USA more than % of surgeons performed only one RP in year and more than % of surgeons had an annual volume of 1 or fewer procedures. Thus, the vast majority of surgeons appeared to be unable to complete within their career the learning curve required to provide optimum cancer control and likely provide patients with poorer oncologic outcomes than highvolume surgeons. The relationship between hospital and surgeon volume and perioperative, oncologic and functional outcomes after RP has been examined in several studies that have been systematically reviewed by Trinh et al. 1 Increasing hospital volume was found to be inversely associated with mortality; the risk of serious complications after RP (7 versus 3% complication rate in high- versus low-volume hospitals, respectively); 1 mean length of hospital stay (1% lower in the highest than the lowest quartile of hospital volume); 1 rates of late urinary complications (bladder neck obstruction, strictures, fistulas); 1 and the need for salvage hormone-ablative or radiation therapy > months after surgery. 1,17 It also improved recurrence-free survival. 1 Surgeon volume was inversely associated with perioperative adverse events and oncologic outcomes. 1 A twofold decrease in overall complications (odds ratio:.3; range:.3.9) was observed in TRENDS IN UROLOGY & MEN S HEALTH MARCH/APRIL 1

4 1 a Carotid endarterectomy Aortic valve replacement 3 1 Surgeon volume < > < > Surgeon volume <..-. > < >13. Coronary-artery bypass grafting Elective repair of an abdominal aortic aneurysm 1 Surgeon volume < > < >. 1 Surgeon volume < > < >. b 3 1 Resection for lung cancer Surgeon volume < > < > Cystectomy Surgeon volume <..-3. > <..-1. >1.. Oesophagectomy Pancreatic resection Surgeon volume <..-. > < > Surgeon volume <..-. > < >13. Figure. Adjusted operative mortality among Medicare patients in 199 and 1999, according to hospital-volume stratum and surgeonvolume stratum for a) four cardiovascular procedures and b) four cancer resections (reproduced with permission from Birkmeyer et al. ) TRENDS IN UROLOGY & MEN S HEALTH MARCH/APRIL 1

5 11 high-volume ( cases/year) compared with low-volume (< cases/year) surgeons. 19 Increasing surgeon volume is inversely associated with positive surgical margin rates, need for adjuvant therapy and biochemical recurrence. 1 Patients treated by high-volume surgeons (eg 1 cases/year) were also found to have a 7% functional (urinary and erectile) recovery rate, while patients treated by a low-volume surgeon (eg cases/year) had a 1% functional recovery rate. The effects of surgeon variability on oncologic and functional outcomes after RP were examined by Carlsson et al. 1 in 1 patients with prostate cancer treated with RP by one of nine surgeons practising in a Swedish academic centre between 1 and. Potency and continence outcomes were measured preoperatively and 1 months postoperatively by patient-administered questionnaires. Biochemical recurrence (BCR) was defined as a PSA value >.ng/ml with at least one confirmatory rise. After adjusting for age, PSA at diagnosis, pathologic stage, pathologic Gleason score, year of surgery and surgical experience, a statistically significant (p=.1) heterogeneity among surgeons was detected, with postoperative 1-month continence rates varying from 7 to 93%. No differences among surgeons were found in terms of potency outcomes, adjusted probabilities of functional recovery and -year probability of freedom from BCR. Based on their results, the authors suggested the need for qualityassurance measures involving performance feedback to help surgeons improving outcomes once aware of their own results. Bianco et al. investigated the variations among experienced surgeons in cancer control after open RP. Types of variation included the surgeon volume (measured) and the surgeon s technique (unmeasured). The study cohort consisted of 77 patients with clinically localised prostate cancer treated with open RP at four major American academic medical centres from 197 to 3 by 1 of surgeons. BCR was defined as PSA.ng/ml followed by a higher level. After adjusting for case mix (baseline PSA, pathological stage and grade), surgery year and surgeon experience, the authors found a statistically significant heterogeneity in the prostate cancer recurrence rate independent of surgeon experience (p=.). Adjusted -year prostate cancer recurrence rate was less than 1% in seven and greater than % in another five surgeons, even though the surgeons had similar levels of experience. The authors concluded that both measured and unmeasured characteristics of the treating surgeon had an impact on the oncologic results of RP. Vickers et al. examined heterogeneity for urinary and erectile outcomes in 191 patients who underwent RP by one of 11 surgeons at Memorial Sloan Kettering Cancer Center between January 1999 and July 7. After adjustment for case mix (baseline age and PSA, pathologic stage and grade, comorbidities) and year of surgery, significant heterogeneity in functional outcomes at 1 year after RP (p<.1 for both urinary and erectile function) was detected. Adjusted rates of full continence were <7% for four surgeons and >% for three surgeons. Adjusted rates of erectile function recovery were <% for two surgeons and >% for another two. Of note, better functional outcomes were associated with lower rates of BCR. The authors concluded that higher surgical quality resulted in better functional outcomes in the absence of poorer cancer control. In an attempt to reduce heterogeneity among surgeons and uniformly offer patients high-quality care provided by urologic surgeons, a performance feedback system was piloted at Memorial Sloan Kettering Cancer Center in 9. 3 Surgeons who had performed a minimum of 3 procedures received their adjusted rates for recurrence, erectile dysfunction and continence obtained directly from the patient-filled electronic questionnaires using a previously validated interface. Surgeons were able to explore their oncologic and functional outcomes, both case-mix adjusted and in comparison to their peers, anonymously graphed on the report. Surgeons showed positive reactions to the feedback system, which prompted educational activities including refinements of the surgical technique. The monitoring of surgical outcomes and the development of strategies for their improvement after the implementation of the performance feedback system is ongoing at this institution. CONTEMPORARY TRENDS Reames et al. evaluated whether changes in the relationship between hospital volume and operative mortality could be detected in 317 patients undergoing one of eight complex gastrointestinal, cardiac or vascular procedures at hospitals in the lowest and highest quintiles of operative volume in the recent decade ( 9) as a result of new health policies and quality improvement initiatives developed to improve surgical quality and standards of care among poorly performing hospitals. After adjusting for patient characteristics, the relationship between hospital volume and mortality, and changes over time were examined. A significant inverse relationship was observed in all procedures throughout the 1-year period in five of the eight procedures. Although only data from a subset of US hospitals in 1 year were included in the study, the strong inverse relationship between hospital volume and mortality appeared to have persisted despite all the changes stimulated by the body of evidence reported above. These recent data suggest that the efforts aimed at the identification and implementation of measures for the improvement of the quality of surgical care delivered by lowvolume hospitals and surgeons should be re-evaluated, with continued efforts to decrease the quality gap between high- and low-volume centres. TRENDS IN UROLOGY & MEN S HEALTH MARCH/APRIL 1

6 1 Declaration of interests: none declared. REFERENCES 1. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med ; 3: Begg CB, Cramer LD, Hoskins WJ, et al. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 199;: Dudley RA, Johansen KL, Brand R, et al. Selective referral to high-volume hospitals: estimating potentially avoidable deaths. JAMA ;3:119.. Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative mortality in the United States. N Engl J Med 3; 39: Gonzalez AA, Dimick JB, Birkmeyer JD, et al. Understanding the volume-outcome effect in cardiovascular surgery: the role of failure to rescue. JAMA Surg 1;19: Dimick JB, Birkmeyer JD, Upchurch GR Jr. Measuring surgical quality: what s the role of provider volume? World J Surg ; 9: Schrag D, Panageas KS, Riedel E, et al. Hospital and surgeon procedure volume as predictors of outcome following rectal cancer resection. Ann Surg ; 3:3 9.. Birkmeyer JD, Skinner JS, Wennberg DE. Will volume-based referral strategies reduce costs or just save lives? Health Aff (Millwood) ;1: Birkmeyer JD, Gust C, Dimick JB, et al. Hospital quality and the cost of inpatient surgery in the United States. Ann Surg 1;:1. 1. Nathan H, Atoria CL, Bach PB, et al. Hospital volume, complications, and cost of cancer surgery in the elderly. J Clin Oncol 1; 33: Vickers AJ, Bianco FJ, Serio AM, et al. The surgical learning curve for prostate cancer control after radical prostatectomy. J Natl Cancer Inst 7;99: Vickers AJ, Savage CJ, Hruza M, et al. The surgical learning curve for laparoscopic radical prostatectomy: a retrospective cohort study. Lancet Oncol 9;1: Savage CJ, Vickers AJ. Low annual caseloads of United States surgeons conducting radical prostatectomy. J Urol 9;1: Trinh QD, Bjartell A, Freedland SJ, et al. A systematic review of the volume-outcome relationship for radical prostatectomy. Eur Urol 13;: Begg CB, Riedel ER, Bach PB, et al. Variations in morbidity after radical prostatectomy. N Engl J Med ;3: Yao SL, Lu-Yao G. Population-based study of relationships between hospital volume of prostatectomies, patient outcomes, and length of hospital stay. J Natl Cancer Inst 1999;91: Ellison LM, Trock BJ, Poe NR, et al. The effect of hospital volume on cancer control after radical prostatectomy. J Urol ; 173:9. 1. Gooden KM, Howard DL, Carpenter WR, et al. The effect of hospital and surgeon volume on racial differences in recurrencefree survival after radical prostatectomy. Med Care ;: Hu JC, Gold KF, Pashos CL, et al. Role of surgeon volume in radical prostatectomy outcomes. J Clin Oncol 3;1:1.. Vickers A, Savage C, Bianco F, et al. Cancer control and functional outcomes after radical prostatectomy as markers of surgical quality: analysis of heterogeneity between surgeons at a single cancer center. Eur Urol 11;9: Carlsson S, Berglund A, Sjoberg D, et al. Effects of surgeon variability on oncologic and functional outcomes in a populationbased setting. BMC Urol 1;1:.. Bianco FJ Jr, Vickers AJ, Cronin AM, et al. Variations among experienced surgeons in cancer control after open radical prostatectomy. J Urol 1;13: Vickers AJ, Sjoberg D, Basch E, et al. How do you know if you are any good? A surgeon performance feedback system for the outcomes of radical prostatectomy. Eur Urol 1;1: 9.. Vickers AJ, Savage CJ, Shouery M, et al. Validation study of a web-based assessment of functional recovery after radical prostatectomy. Health Qual Life Outcomes 1;:.. Reames BN, Ghaferi AA, Birkmeyer JD, et al. Hospital volume and operative mortality in the modern era. Ann Surg 1;: 1. TRENDS IN UROLOGY & MEN S HEALTH MARCH/APRIL 1