Figures
Abstract
Background
The relationship between hospital volumes and short-term patients’ outcomes of colon cancer (CC) surgery is not well established in the literature. Moreover, evidence about short-term outcomes of urgent compared with elective CC procedures is scanty. The aims of this study are 1) to determine whether caseloads and other hospital characteristics are associated with short-term outcomes of CC surgery; 2) to compare the outcomes of urgent and elective CC surgery.
Methods
A total of 14,200 patients undergoing CC surgery between 2005 and 2010 in the General Surgery Units (GSUs) of the hospitals of Emilia-Romagna region, Northern Italy, were identified from the hospital discharge records database. The outcomes of interest were 30-day in-hospital mortality, re-intervention and 30-day re-admission. Using multilevel analysis, we analyzed the relationship of GSU volumes and focused practice, defined as the percentage of CC operations over total operations, with the three outcomes.
Results
High procedure volumes were associated with a lower risk of 30-day in-hospital mortality, after adjusting for patients’ characteristics [aOR (95% CI) = 0.51 (0.33–0.81)]. Stratified analyses for elective and urgent surgery showed that high volumes were associated with a lower 30-day mortality for elective patients [aOR (95% CI) = 0.35 (0.17–0.71)], but not for urgent patients [aOR (95% CI) = 0.72 (0.42–1.24)]. Focused practice was an independent predictor of re-intervention [aOR (95% CI) = 0.67 (0.47–0.97)] and re-admission [aRR (95% CI) = 0.88 (0.78–0.98)].
Citation: Lenzi J, Lombardi R, Gori D, Zanini N, Tedesco D, Masetti M, et al. (2013) Impact of Procedure Volumes and Focused Practice on Short-Term Outcomes of Elective and Urgent Colon Cancer Resection in Italy. PLoS ONE 8(5): e64245. https://doi.org/10.1371/journal.pone.0064245
Editor: Antonio Moschetta, University of Bari & Consorzio Mario Negri Sud, Italy
Received: March 1, 2013; Accepted: April 11, 2013; Published: May 16, 2013
Copyright: © 2013 Lenzi et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In Western countries, colorectal cancer is the third most commonly diagnosed cancer in males and the second in females. About two-thirds of colorectal cancers occur in the colon [1] and early diagnosis and radical resection may represent the only chance of cure for patients [2]. This has led many Western countries, including Italy, to introduce colon cancer (CC) screening programs. In 2005 Emilia-Romagna region, in Northern Italy, launched a screening program for early detection of colorectal cancer targeted to people aged 50–74 years, with a compliance of 46.7% in 2007 that increased to 53.7% in 2008 [3].
Despite substantial advances in surgical techniques and peri-operative care during the last decades, morbidity and mortality after surgery remain considerable, ranging from 18% to 35% and 1% to 11%, respectively [4]–[8]. However, it is well known that the risk of adverse events after colorectal surgery depends on patient-, disease-, and treatment-related characteristics, some of which are modifiable [9], [10]. Moreover, identification of outcome predictors liable to preventive measures is crucial for improving surgical care quality.
Since late 1970s, several authors analyzed the relationship between hospital volume and short- and long-term outcomes, and found a positive correlation for complex surgical procedures [11]–[17]. A recent Cochrane review and meta-analysis based on studies carried out in USA, UK and Northern Europe showed that higher surgeon volumes were associated with better outcomes of CC surgery, while hospital volumes were unrelated with these outcomes [18].
To our knowledge, no study has investigated the relationship between caseloads and outcomes of CC surgery in Italy, where colorectal surgery is performed in General Surgery Units (GSUs). Moreover, little is known about the outcomes of CC surgery in elective and urgent patients. In a recent study carried out in Denmark, the authors found a significant variation in mortality between low- and high-volume hospitals for urgent surgery, but not for elective surgery [19].
The aims of the present study are: 1) to determine whether caseloads and other hospital characteristics are associated with short-term outcomes of CC surgery; 2) to compare the outcomes of urgent and elective CC surgery.
Materials and Methods
Ethics Statement
The study was carried out in conformity with the regulations on data management of the Regional Health Authority of Emilia-Romagna, and with the Italian law on privacy (Art. 20–21, DL 196/2003) (http://www.garanteprivacy.it/web/guest/home/docweb/-/docweb-display/docweb/1115480, published in the Official Journal no. 190 of August 14, 2004) which explicitly exempts the need of ethical approval for anonymous data (Preamble #8).
Data were anonymized prior to the analysis at the regional statistical office, where each patient was assigned a unique identifier. This identifier does not allow to trace the patient’s identity and other sensitive data. As anonymized administrative data are used routinely for health-care management no specific written informed consent was needed to use patient information.
The data set will be made freely available upon request.
Population and Data
Data were extracted from the Hospital Discharge Records (HDRs) database, that includes all discharges from the 86 GSUs of the 66 hospitals in Emilia-Romagna region (4.4 million inhabitants, 42% aged >50 years) [20]. GSUs provide, in addition to gastrointestinal surgery, abdominal, thyroid and breast surgery. Large hospitals may have more than one GSU.
For each GSU, volume was defined as the mean annual number of CC procedures carried out over 6 years, and focused practice as the percentage of CC operations over total operations. A tertile split was used to classify GSUs into three volume categories: low-volume (<40 CC cases/year), intermediate-volume (40–64 CC cases/year), or high-volume (≥65 CC cases/year). A median split was used to classify GSUs as non-focused (<5% CC cases over total operations) or focused (≥5% CC cases).
Hospitals were categorized as private or public and teaching or non-teaching. Public hospitals are owned by the regional government, while private hospitals are privately owned. In the presence of an agreement with the Regional Health Authority, private hospitals supply services for the regional health care system and receive public funding. Teaching hospitals are public hospitals affiliated with a medical school.
ICD-9-CM codes were used to identify patients with a primary diagnosis of carcinoma in situ or malignant neoplasm of the colon (codes 230.3 and 153.x, respectively) and an operation in the digestive system (codes 42–54) as the primary procedure. This methodology decreases the risk of excluding from the analyses patients undergoing multi-visceral resections for locally advanced colonic tumors. 14,809 HDRs were extracted for the period 1/1/2005–12/31/2010. 609 transfers from other hospitals were excluded.
Independent variables used for case mix-adjusted analyses were: age, sex, length of stay of the index admission, comorbidities, presence/absence of metastases, type of resection and type of admission (urgent/elective). Comorbidity was assessed using secondary diagnoses at the index admission and in the two previous years. Tumor spread was determined using diagnostic codes that signaled the involvement of other organs (197.x and 198.89). In the absence of these codes, it was assumed that no metastasis was present. Interventions were categorized as partial colectomies (code 45.7) or total colectomies (code 45.8). The remaining interventions were classified as “other”.
Outcome Measures
The outcomes considered were: 30-day mortality (death within 30 days of surgery related to the index or any subsequent hospitalization), 30-day re-admission (admission occurring for any reason within 30 days of index discharge) and re-intervention in the index hospitalization, identified by means of a specific algorithm that combines surgical procedures and complications occurring in the days after CC surgery (Text S1).
Statistical Analysis
Student’s t-test, χ2 test and Fisher’s exact test were used, where appropriate, to analyze the relationship between patient characteristics and each of the three outcomes. We also analyzed the relationship between GSU volume and focused practice using Spearman’s rho.
In order to allow for the hierarchical structure of the data, in which patients are clustered into GSUs and GSUs into hospitals, we analyzed the relationship of GSU and hospital characteristics with outcomes using multilevel logistic regression analyses. For each outcome, the multilevel analysis was carried out in two steps. In the first step, a three-level model (M1) was built including patient characteristics significantly (p<0.05) associated with the outcome and random intercepts for GSUs and hospitals. In the second step, significant GSU and hospital characteristics were added to the model (M2) to determine the variability in outcome associated with these variables after controlling for patient case mix. In this model, we also tested the presence of interactions between GSU and hospital characteristics and the admission status (elective/urgent).
We present the associations of GSUs and hospital characteristics with outcomes deriving from the model M2 in terms of odds ratios (ORs) or risk ratios (RRs) with 95% confidence intervals (95% CIs) [21]. We also provide GSU- and hospital-level variance of the model M2, and how much of this variability is attributable to GSU and hospital characteristics. This last measure is calculated as the proportional change in variance between M1 and M2.
Statistical analyses were carried out using the procedure xtmelogit of Stata software, version 12 (StataCorp. 2011. Stata Statistical Software: Release 12. College Station, TX: StataCorp LP).
Results
Baseline Characteristics and Population Case-mix
The study cohort consisted of 14,200 patients: 7,722 men (54.4%), with a mean age of 70 years. A total of 10,831 patients underwent elective operation (76.27%) and 3,369 patients urgent operation (23.73%) (Table 1).
Of the 66 hospitals included in the present analyses, 4 were teaching and 62 non-teaching hospitals (93.9% of hospitals), 23 were private and 43 public (65.2% of hospitals). Twelve hospitals had more than one GSU. Of these, eight hospitals had two GSUs, and four hospitals had more than two GSUs.
Of the nine high-volume GSUs, six operated in non-teaching public hospitals and three in teaching hospitals; private hospitals had only low-volume GSUs. Of the twenty-two focused GSUs, more than half (twelve) operated in non-teaching public hospitals (Table 2).
GSU volume and focused practice were moderately correlated (Spearman’s rho = 0.49, p<0.001), suggesting that the two variables are not interchangeable.
Outcomes
The prevalence of 30-day in-hospital mortality, 30- day re-admission, and re-intervention was 1.9% (range, 0.0%–16.7%), 28.1% (range, 0.0%–60.0%), and 3.3% (range, 0.0%–14.3%), respectively.
Crude Associations of Patient Characteristics with Outcomes
Crude associations of patient characteristics with outcomes are shown in Table 3. 30-day mortality was significantly higher among patients with at least one comorbidity [2.5% vs. 1.2%, p<0.001; OR (95% CI) = 2.05 (1.57–2.67)], and among those who underwent urgent procedures [5.2% vs. 0.9%, p<0.001; OR (95% CI) = 5.84 (4.56–7.50)]. The same associations were found for re-interventions; re-admission was more likely among younger patients and among those undergoing urgent surgery.
Adjusted Associations of GSU and Hospital Characteristics with Outcomes
After adjusting for patient characteristics in multilevel logistic regression analysis (models M2), GSU volume predicted only 30-day mortality. Specifically, patients who underwent surgery at high-volume GSUs had a significant reduction in the mortality risk [aOR (95% CI) = 0.51 (0.33–0.81)] compared with patients undergoing surgery at low-volume GSUs. The mortality risk did not differ significantly between patients who underwent surgery at low- and intermediate-volume GSUs [aOR (95% CI) = 0.83 (0.54–1.27)]. Because of the interaction between admission status (elective/urgent) and GSU volume, a stratified analysis by admission status was carried out: high volumes were associated with a lower 30-day mortality for elective patients [aOR (95% CI) = 0.35 (0.17–0.71)], but not for urgent patients [aOR (95% CI) = 0.72 (0.42–1.24)] (Table 4).
GSU focused practice was an independent predictor of re-intervention and re-admission. In particular, patients who underwent surgery at focused GSUs had a significant reduction in the risk of re-intervention [aOR (95% CI) = 0.67 (0.47–0.97)] and re-admission [aRR (95% CI) = 0.88 (0.78–0.98)] (Table 4). There was no evidence of an interaction between GSU focused practice and the admission status.
Hospital characteristics (teaching/non-teaching, public/private) were unrelated to the three outcomes.
Variations among GSUs in 30-day Mortality, Re-intervention and 30-day Re-admission
The random part of the models M2 is shown in Table 5. We found significant variations among GSUs in 30-day mortality after elective surgery (GSU- and hospital-level variance = 0.471; p = 0.002) and no significant variations in mortality after urgent surgery (GSU- and hospital-level variance = 0.209; p = 0.067). We also found significant variations among GSUs in re-intervention and 30-day re-admission.
More than 40% of the variability in 30-day mortality after elective surgery was attributable to GSU volume, and GSU focused practice accounted for about 7% and 6% of the differences among GSUs in re-intervention and 30-day re-admission, respectively.
Discussion
Our results indicate that patients undergoing CC surgery at higher volumes GSUs had a decreased risk of post-operative mortality. This adds to the growing body of evidence (including the recent Cochrane review and meta-analysis [18]) showing a relationship between care provider volume and post-operative CC mortality [22]–[25], and are in contrast with other studies that failed to demonstrate such a relationship [26]–[28].
Few studies examined the relationship of care provider volume with re-intervention and re-admission after CC surgery, and evidence on this topic is mixed [25], [28], although these outcomes have been advocated as potentially useful targets for measurement of the quality of surgical care [17], [29]–[32]. In the present study, we did not observe any relationship of re-intervention and re-admission with surgical volumes.The re-admission rate in our study was 28%, which exceeds the range of rates from the literature (11–27%) [30]–[34]. However, this should be interpreted keeping in mind that our post-operative mortality is at the lower boundary of the mortality rate range reported in other studies (1–11%) [4]–[8], [13], [23], [25], [26], [28]. High rates of re-hospitalization may reflect a care policy favoring early detection and treatment of surgical complications with the aim of reducing the prevalence of post-operative deaths [35]. However, there is also evidence that unplanned 30-day re-admission may be associated with increased post-operative mortality [31].
Our results concerning the relationship between focused practice and outcomes indicate that GSUs with ≥5% CC over total surgery had significantly lower re-admission and re-intervention rates, but did not differ from GSUs with <5% CC on post-operative mortality rates. This suggests that studies on CC surgery outcomes should examine both the effects of volumes and focused practice, because these two variables have a different pattern of association with outcomes. Although evidence from the literature on the effect of focused practice is not available, a recent Spanish study carried out on patients undergoing emergency colorectal resection showed that being operated by a colorectal surgeon compared with a general surgeon was associated with a lower 30-day mortality, after adjusting for patients’ gender, age, ASA score and type of operation [36].
We found that urgent procedures were associated with a higher 30-day mortality. Specifically, urgent patients were about six times as likely to die within 30 days compared with elective patients. This finding is consistent with the literature, indicating that urgent procedures are strongly associated with adverse outcomes after colorectal resection [19], [37], [38], although these authors had a broader focus on colorectal surgery for any reason, and not only for CC. Furthermore, separate analyses carried out in urgent and elective patients revealed that the adjusted risk of post-operative mortality was increased in low-volume GSUs for elective CC surgery, but not for urgent surgery. We also found that more than 40% of the variability in 30-day mortality for elective surgery was accounted for by the GSU volume, whereas no significant variation was found among GSUs for urgent surgery. This is in contrast with the results of a recent study in Denmark, in which a significant variation in mortality between low- and high-volume hospitals was found for urgent (but not elective) surgery [19].
Our findings of better outcomes in high-volume settings bear directly to the question of whether GSU volume is a proxy of other variables such as availability of sophisticated clinical services (e.g., intensive care units (ICUs) and advanced diagnostic/interventional services) and high quality of nursing care. These variables have been proposed as explanatory variables of better outcomes, in particular mortality, in high-volume centers [17], [39]–[41].
The relationship between volumes and outcomes has substantial clinical and organizational implications. In fact, unlike the case of less frequent complex procedures in which the overall effect of higher GSU volumes makes centralization desirable, unintended negative consequences of centralizing colonic resection for cancer must be considered [15]. Referring a large number of cases to a limited number of centers might decrease accessibility for patients and their families [42], and threaten continuity of care after surgery.
However, because our results suggest a relationship between GSU volumes and outcomes in elective patients, we argue that centralization may facilitate the quality of surgery for these patients, including for screen-detected ones, to avoid exposure of apparently healthy people to unnecessary harmful treatments [43].
Our results should be interpreted keeping in mind some important limitations. First, administrative databases have a limited ability to capture illness severity. To minimize this bias, in the absence of information on cancer staging, we classified cancers as metastatic/non-metastatic. Moreover, we considered comorbidities in the index hospitalization and those of the two previous years, as suggested in Davoli et al. [44]. In this way associated relevant medical illnesses, that most likely affect outcomes, were taken into consideration. Second, the potential for inaccurate coding exists in administrative databases such as the hospital discharge records database. However, one study using administrative data in our region showed that hospital discharge records have good specificity, sensitivity and positive predictive value (84.8%, 99.0% and 90.6%), compared with cancer registries [45]. Moreover, the lack of ambiguity regarding diagnoses and procedures for CC, coupled with the fact that outcomes of interest are well defined and not particularly subject to misinterpretation, minimizes this potential bias. Third, we might have not captured the full spectrum of post-operative morbidity. However, many adverse events occurring after colonic resection are recognized during the index hospitalization. In this regard, we searched re-interventions related to index procedures using a specific subset of severe surgical complications. Moreover, we used the 30-day re-admission rate, that may be considered as a fairly good surrogate of surgical complications occurring after hospital discharge [31], [34]. Fourth, information on the individual surgeon volume is not available from administrative databases. Higher surgeon volumes were associated with better outcomes in several studies, including for instance Birkmeyer and Chang [46], [17]. Lastly, we could not examine the relationship of GSU volumes and other provider characteristics with other outcomes of CC surgery (e.g. radical nature of the resection, number of retrieved lymph nodes, local recurrence rate and disease free survival) which are decisive to monitor quality of care and focus improvement initiatives in CC surgery [18] because this information is not available in routine databases.
Conclusions
The present study provided further evidence of the beneficial effect of GSU volume on mortality for elective CC surgery and of focused practice on re-intervention and re-admission. This indicates that clinicians, policy makers and hospital administrators should consider the opportunity to centralize CC surgery keeping in mind their pros and cons, and establish audit of current practice and outcomes to ensure that the benefits of high-volume and focused practice care can be translated into service organization.
Supporting Information
Text S1.
ICD-9-CM diagnostic codes identifying surgical complications.
https://doi.org/10.1371/journal.pone.0064245.s001
(DOC)
Author Contributions
Conceived and designed the experiments: JL MPF. Performed the experiments: RL NZ MM EJ. Analyzed the data: JL DT NZ. Wrote the paper: JL RL DG DT MPF EJ. Adapted ICD-9-CM algorithms for comorbidities and surgical complications: NZ MM.
References
- 1. Jemal A, Bray F (2011) Center MM, Ferlay J, Ward E, et al (2011) Global cancer statistics. CA Cancer J Clin 61: 69–90.
- 2. Spruce LR, Sanford JT (2012) An intervention to change the approach to colorectal cancer screening in primary care. J Am Acad Nurse Pract 24: 167–174.
- 3. Zorzi M, Baracco S, Fedato C, Grazzini G, Sassoli de’ Bianchi P, et al. (2009) Screening for colorectal cancer in Italy, 2009 survey. Epidemiol Prev 35: 55–77.
- 4. Ansari MZ, Collopy BT, Hart WG, Carson NJ, Chandaray EJ (2000) In-hospital mortality and associated complications after bowel surgery in Victorian public hospitals. Aust N Z J Surg 70: 6–10.
- 5. Alves A, Panis Y, Mathieu P, Mantion G, Kwiatkowski F, et al. (2005) Postoperative mortality and morbidity in French patients undergoing colorectal surgery: results of a prospective multicenter study. Arch Surg 140: 278–283.
- 6. Hendren S, Birkmeyer JD, Yin H, Banerjee M, Sonnenday C, et al. (2010) Surgical complications are associated with omission of chemotherapy for stage III colorectal cancer. Dis Colon Rectum 53: 1587–1593.
- 7. Benedix F, Kube R, Meyer F, Gastinger I, Lippert H (2010) Comparison of 17,641 patients with right- and leftsided colon cancer: differences in epidemiology, perioperative course, histology, and survival. Dis Colon Rectum 53: 57–64.
- 8. Richards CH, Leitch FE, Horgan PG, McMillan DC (2010) A systematic review of POSSUM and its related models as predictors of post-operative mortality and morbidity in patients undergoing surgery for colorectal cancer. J Gastrointest Surg 14: 1511–1520.
- 9. Longo WE, Virgo KS, Johnson FE, Oprian CA, Vernava AM, et al. (2000) Risk factors for morbidity and mortality after colectomy for colon cancer. Dis Colon Rectum 43: 83–91.
- 10. Ragg JL, Watters DA, Guest GD (2009) Preoperative risk stratification for mortality and major morbidity in major colorectal surgery. Dis Colon Rectum 52: 1296–1303.
- 11. Luft HS, Bunker JP, Enthoven AC (1979) Should operations be regionalized? The empirical relation between surgical volume and mortality. N Engl J Med 301: 1364–1369.
- 12. Begg CB, Cramer LD, Hoskins WJ, Brennan MF (1998) Impact of hospital volume on operative mortality for major cancer surgery. JAMA 280: 1747–1751.
- 13. Birkmeyer JD, Siewers AE, Finlayson EV, Stukel TA, Lucas FL, et al. (2002) Hospital volume and surgical mortality in the United States. N Engl J Med 346: 1128–1137.
- 14. Finlayson EV, Goodney PP, Birkmeyer JD (2003) Hospital volume and operative mortality in cancer surgery: a national study. Arch Surg 138: 721–725.
- 15. Gooiker GA, van Gijn W, Wouters MW, Post PN, van de Velde CJ, et al. (2011) Systematic review and meta-analysis of the volume-outcome relationship in pancreatic surgery. Br J Surg 98: 485–494.
- 16. Marusch F, Koch A, Schmidt U, Pross M, Gastinger I, et al. (2001) Hospital caseload and the results achieved in patients with rectal cancer. Br J Surg 88: 1397–1402.
- 17. Chang CM, Huang KY, Hsu TW, Su YC, Yang WZ, et al. (2012) Multivariate analyses to assess the effects of surgeon and hospital volume on cancer survival rates: A nationwide population-based study in Taiwan. PLoS One 7: e40590.
- 18. Archampong D, Borowski D, Wille-Jørgensen P, Iversen LH (2012) Workload and surgeon's specialty for outcome after colorectal cancer surgery. Cochrane Database Syst Rev 3: CD005391.
- 19. Osler M, Iversen LH, Borglykke A, Mårtensson S, Daugbjerg S, et al. (2011) Hospital variation in 30-day mortality after colorectal cancer surgery in Denmark: the contribution of hospital volume and patient characteristics. Ann Surg 253: 733–738.
- 20.
Statistica Emilia-Romagna (2011). Popolazione residente per province, comuni e classi di età al 1 gennaio 2011. Available: http://sasweb.regione.emilia-romagna.it/cgi-bin/broker.exe?imap=TRPPRES11T. Accessed 30 September 2011.
- 21. Flanders WD, Rhodes PH (1987) Large sample confidence intervals for regression standardized risks, risk ratios, and risk differences. J Chronic Dis 40: 697–704.
- 22. Schrag D, Cramer LD, Bach PB, Cohen AM, Warren JL, et al. (2000) Influence of hospital procedure volume on outcomes following surgery for colon cancer. JAMA 284: 3028–3035.
- 23. Bilimoria KY, Bentrem DJ, Feinglass JM, Stewart AK, Winchester DP, et al. (2008) Directing surgical quality improvement initiatives: comparison of perioperative mortality and long-term survival for cancer surgery. J Clin Oncol 26: 4626–4633.
- 24. Rogers SO Jr, Wolf RE, Zaslavsky AM, Wright WE, Ayanian JZ (2006) Relation of surgeon and hospital volume to processes and outcomes of colorectal cancer surgery. Ann Surg 244: 1003–1011.
- 25. Billingsley KG, Morris AM, Dominitz JA, Matthews B, Dobie S, et al. (2007) Surgeon and hospital characteristics as predictors of major adverse outcomes following colon cancer surgery: understanding the volume-outcome relationship. Arch Surg 142: 23–31.
- 26. Engel AF, Oomen JL, Knol DL, Cuesta MA (2005) Operative mortality after colorectal resection in the Netherlands. Br J Surg 92: 1526–1532.
- 27. Karanicolas PJ, Dubois L, Colquhoun PH, Swallow CJ, Walter SD, et al. (2009) The more the better?: the impact of surgeon and hospital volume on in-hospital mortality following colorectal resection. Ann Surg 249: 954–959.
- 28. Drolet S, MacLean AR, Myers RP, Shaheen AA, Dixon E, et al. (2011) Elective resection of colon cancer by high volume surgeons is associated with decreased morbidity and mortality. J Gastrointest Surg 15: 541–550.
- 29. Morris AM, Baldwin LM, Matthews B, Dominitz JA, Barlow WE, et al. (2007) Reoperation as a quality indicator in colorectal surgery: a population-based analysis. Ann Surg 245: 73–79.
- 30. Guinier D, Mantion GA, Alves A, Kwiatkowski F, Slim K, et al. (2007) Risk factors of unplanned readmission after colorectal surgery: a prospective, multicenter study. Dis Colon Rectum 50: 1316–1323.
- 31. Greenblatt DY, Weber SM, O'Connor ES, LoConte NK, Liou JI, et al. (2010) Readmission after colectomy for cancer predicts one-year mortality. Ann Surg 251: 659–669.
- 32. van Westreenen HL, Ijpma FF, Wevers KP, Afzali H, Patijn GA (2011) Reoperation after colorectal surgery is an independent predictor of the 1-year mortality rate. Dis Colon Rectum 54: 1438–1442.
- 33. Goodney PP, Stukel TA, Lucas FL, Finlayson EV, Birkmeyer JD (2003) Hospital volume, length of stay, and readmission rates in high-risk surgery. Ann Surg 238: 161–167.
- 34. Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 360: 1418–1428.
- 35. Joynt KE, Jha AK (2012) Thirty-day readmissions-truth and consequences. N Engl J Med 366: 1366–1369.
- 36. Biondo S, Kreisler E, Millan M, Fraccalvieri D, Golda T, et al. (2010) Impact of surgical specialization on emergency colorectal surgery outcomes. Arch Surg 145: 79–86.
- 37. Morris EJ, Taylor EF, Thomas JD, Quirke P, Finan PJ, et al. (2011) Thirty-day postoperative mortality after colorectal cancer surgery in England. Gut 60: 806–813.
- 38. Ingraham AM, Cohen ME, Bilimoria KY, Feinglass JM, Richards KE, et al. (2010) Comparison of hospital performance in nonemergency versus emergency colorectal operations at 142 hospitals. J Am Coll Surg 210: 155–165.
- 39. Ghaferi AA, Birkmeyer JD, Dimick JB (2009) Variation in hospital mortality associated with inpatient surgery. N Engl J Med 361: 1368–1375.
- 40. Pronovost PJ, Angus DC, Dorman T, Robinson KA, Dremsizov TT, et al. (2002) Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. JAMA 288: 2151–2162.
- 41. Khuri SF, Henderson WG (2005) The case against volume as a measure of quality of surgical care. World J Surg 29: 1222–1229.
- 42. Stitzenberg KB, Sigurdson ER, Egleston BL, Starkey RB, Meropol NJ (2009) Centralization of cancer surgery: implications for patient access to optimal care. J Clin Oncol 27: 4671–4678.
- 43. Grimes DA, Schulz KF (2002) Uses and abuses of screening tests. Lancet 359: 881–884.
- 44. Davoli M, Amato L, Minozzi S, Bargagli AM, Vecchi S, et al. (2005) [Volume and health outcomes: an overview of systematic reviews]. Epidemiol Prev 29: 3–63.
- 45. Yuen E, Louis D, Cisbani L, Rabinowitz C, De Palma R, et al. (2011) Using administrative data to identify and stage breast cancer cases: implications for assessing quality of care. Tumori 2011 97: 428–435.
- 46. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, et al. (2003) Surgeon volume and operative mortality in the United States. N Engl J Med 349: 2117–2127.