Annals of Surgical Oncology

, Volume 16, Issue 7, pp 1799–1808

Effect of Surgeon Training, Specialization, and Experience on Outcomes for Cancer Surgery: A Systematic Review of the Literature

Authors

    • Northwestern Memorial Hospital, Department of SurgeryFeinberg School of Medicine, Northwestern University
  • Joseph D. Phillips
    • Northwestern Memorial Hospital, Department of SurgeryFeinberg School of Medicine, Northwestern University
  • Colin E. Rock
    • Northwestern Memorial Hospital, Department of SurgeryFeinberg School of Medicine, Northwestern University
  • Amanda Hayman
    • Northwestern Memorial Hospital, Department of SurgeryFeinberg School of Medicine, Northwestern University
  • Jay B. Prystowsky
    • Northwestern Memorial Hospital, Department of SurgeryFeinberg School of Medicine, Northwestern University
    • Jesse Brown VA Medical Center
  • David J. Bentrem
    • Northwestern Memorial Hospital, Department of SurgeryFeinberg School of Medicine, Northwestern University
    • Jesse Brown VA Medical Center
Healthcare Policy and Outcomes

DOI: 10.1245/s10434-009-0467-8

Cite this article as:
Bilimoria, K.Y., Phillips, J.D., Rock, C.E. et al. Ann Surg Oncol (2009) 16: 1799. doi:10.1245/s10434-009-0467-8

Abstract

Background

Outcomes after cancer resections have been shown to be better for high-volume surgeons compared with low-volume surgeons; however, reasons for this relationship have been difficult to identify. The objective of this study was to assess studies examining the effect of surgeon training and experience on outcomes in surgical oncology.

Methods

A systematic review of the literature was performed to assess articles examining the impact of surgeon training, certification, and experience on outcomes. Studies were included if they examined cancer resections and performed multivariable analyses adjusting for relevant confounding variables.

Results

An extensive literature search identified 29 studies: 27 examined surgeon training/specialization, 1 assessed surgeon certification, and 4 evaluated surgeon experience. Of the 27 studies examining training/specialization, 25 found that specialized surgeons had better outcomes than nonspecialized surgeons. One study found that American Board of Surgery (ABS)-certified surgeons had better outcomes than noncertified surgeons. Of the two studies examining time since ABS certification, both found that increasing time was associated with better outcomes. Of the four studies that examined experience, three studies found that increasing surgeon experience was associated with improved outcomes.

Conclusions

Although numerous studies have examined the impact of surgeon factors on outcomes, only a few cancers have been examined, and outcome measures are inconsistent. Most studies do not appear robust enough to support major policy decisions. There is a need for better data sources and consistent analyses which assess the impact of surgeon factors on a broad range of cancers and help to uncover the underlying reasons for the volume–outcome association.

Considerable variability has been demonstrated in outcomes after cancer surgery.14 Numerous studies have demonstrated a difference in outcomes after cancer resections according to surgeon volume, where higher volume is associated with lower perioperative mortality, recurrence, and complication rates, as well as higher long-term survival rates.57 However, identification of specific factors underlying this volume–outcome association has been challenging, and volume has been shown not to be the most important predictor of outcomes.8,9

Surgical specialization via additional training beyond residency may contribute to differences in outcomes according to surgeon volume. Fellowship-trained surgeons may be more likely to focus on particular disease processes or operations, thus resulting in a more in-depth knowledge of the disease and increased technical expertise, as well as a high-volume practice. Alternatively, it may be that the time in practice since completing training provides surgeons with the experience in surgical technique and postoperative management which can only be gained with time. Moreover, the volume of the hospital may also affect surgeon outcomes, particularly short-term morbidity and mortality; and it must be considered that the system of care, i.e., anesthesia and intensive care unit (ICU) care, is as important as the surgeon performing the operation, particularly for short-term outcomes.10,11

The majority of cancer patients in the USA are treated by relatively low-volume surgeons.5 Thus, it is of particular interest to understand the factors which lead to better outcomes for patients treated by high-volume surgeons and attempt to transfer these characteristics to low-volume surgeons. However, studies have offered conflicting information using inconsistent methodology regarding the importance of surgeon training and experience. Thus, we performed a systematic review of the literature to assess studies examining the effect of surgeon training and experience on outcomes after adjusting for differences in case mix.

Methods

A systematic review of published primary research examining the association of surgeon training/experience and outcomes was performed. Surgeon training was defined as residency, fellowship or self-declared specialty interest. Surgeon experience was defined as years in practice since certification or completion of residency. The PubMed database was searched for all published articles prior to August 1, 2008 using all combinations of the keywords: surgeon, training, post-residency, fellowship, cancer, outcome, volume, experience, certification, as well as the various specialties and specialty accreditation boards.

Abstracts were searched for all papers that compared surgical specialty/training or that compared surgeons based on years of experience. This produced 79 relevant articles. Studies were excluded if there were updates over time based on the same dataset or patient population, not in the English language, published before 1980, focused on learning curves (i.e., laparoscopic techniques), were letters or opinion articles, reviews or single-surgeon experiences. The bibliographies of these studies were also reviewed for articles not produced in the initial literature search. This search methodology produced a total of 44 articles. Next, only articles examining the association between surgeon training/experience and outcome using multivariable methods were included. Included studies were required to make some attempt to adjust for case-mix variables. Whether studies included adjustment for hospital volume was assessed but was not required for inclusion in the study. Outcomes examined included perioperative morbidity and mortality, recurrence, and long-term survival. Twenty-nine articles were used as the basis for this review. A study’s results were considered statistically significant if P < 0.05 for the association between the explanatory variable and the outcome.

Results

The 29 articles that met the inclusion criteria examined nine different types of cancers including colorectal, ovarian, melanoma, breast, bladder, lung, esophageal, gastric, and hepatocellular.1236 Fifteen different physician specialties were assessed, including colorectal surgery, transplant surgery, vascular surgery, dermatology, general practitioners, plastic surgery, surgical oncology, hepatopancreaticobiliary, urology, urologic oncology, thoracic surgery, cardiothoracic surgery, obstetrics and gynecology, gynecologic oncology, and general surgery. Four articles examined outcomes based on the surgeon’s years of experience. The main outcome measures assessed were perioperative morbidity, mortality, recurrence, and long-term survival. Characteristics and results of each study are summarized in Table 1.
Table 1

Studies examining the effect of surgeon training on outcomes

Study no.

Author, year

Site

Comparison*

Data source

Setting (number of hospitals)

Number of patients

Morbidity

Postoperative mortality

Recurrence

Long-term survival

Gastrointestinal

1

Dimick 2005

Esophageal

Thoracic vs. general surgeon

Medicare provider analysis review file

Multicenter

1,946

 

Improved

  

2

McArdle 2004

Colorectal

Colorectal vs. general surgeon

Case notes

Multicenter (11)

3,200

 

No significant difference

 

Improved

3

Callahan 2003

Colon and Gastric

Specialist (surgical oncologist/colorectal) vs. nonspecialist (general surgeon)

New York State Department of Health’s Statewide Planning and Research Cooperative System

Multicenter (223-colon and 213-gastric)

55,016 (48,582-colon and 6,434-gastric)

 

Improved

  

4

Smith 2003

Colorectal

Colorectal vs. general surgeon

Patient records

Multicenter (11)

4,562

 

Improved

Improved

Improved

5

Wrigley 2003

Colorectal

Colorectal vs. general surgeon

Wessex Cancer Registry, patient records and follow-ups

Multicenter

5,176 (4,419 complete)

   

Improved

6

Prystowsky 2002

Colon

Colorectal vs. general surgeon, ABS-certified vs. noncertified, university vs. nonuniversity trained

Illinois Hospitals and Health Systems Association Compdata

Multicenter (76)

15,427

Improved for ABS-certification

Improved for ABS-certification

  

7

Read 2002

Rectal

Colorectal vs. general surgeon

Patient records

Single center

384

  

Improved

Improved

8

Dorrance 2000

Colorectal

Colorectal vs. vascular/transplant vs. general surgeon

University teaching hospital pathology records

Single center

378

  

Improved

 

9

Wigmore 1999

Colorectal

Colorectal vs. general surgeon

Patient records and follow-ups

Single center

230

  

Improved

Improved

10

Porter 1998

Rectal

Colorectal vs. general surgeon

Patient Records

Multicenter (5)

683

  

Improved

Improved

11

Bokey 1997

Rectal

Colorectal vs. general surgeon

Patient records and follow-ups

Single center

992

   

Improved

Ovarian

12

Chan 2007

Ovarian

Gynecologic oncologist vs. other

California Cancer Registry, medical records

Multicenter

1,491

   

Improved for gynecologic oncologists

13

Earle 2006

Ovarian

Gynecologic oncologist vs. gynecologist vs. general surgeon

SEER-Medicare

Multicenter

3,067

   

Improved for gynecologic oncologists and gynecologists

14

Engelen 2005

Ovarian

Gynecologic oncologist vs. gynecologist

Regional Cancer Registry of the Comprehensive Cancer Center North

Multicenter (16)

512

   

Improved for gynecologic oncologists

15

Tingulstad 2003

Ovarian

Gynecologic oncologist vs. gynecologist

Patient records

Multicenter (8)

82

   

Improved

16

Elit 2002

Ovarian

Gynecologic oncologist vs. gynecologist vs. general surgeon

Patient records and the Canadian Institute for Health information

Multicenter

3,585

 

No significant difference

 

Improved for gynecologic oncologists and gynecologists

17

Junor 1999

Ovarian

Gynecologic oncologist vs. gynecologist vs. general surgeon

Scottish National Cancer Registry and case notes

Multicenter

1,866

   

Improved for stage III disease for gynecologic- oncologists and general surgeons (gynecologist is reference)

18

Woodman 1997

Ovarian

Gynecologist vs. general surgeon (±oncologist)

North Western Regional Cancer Registry

Multicenter

691

   

Improved

19

Kehoe 1994

Ovarian

Gynecologist vs. general surgeon

West Midlands Regional Cancer Registry and patient records

Multicenter

1,184

   

Improved

20

Eisenkop 1992

Ovarian

Gynecologic oncologist vs. other

Patient records

Multicenter (14)

263

No significant difference

Improved

 

Improved

Other

21

McKay 2008

Hepatic

Surgical oncology vs. HPB vs. other specialist vs. general surgeon

Patient records

Multicenter (9)

1,107

Improved for HPB

No significant difference

  

22

Goodney 2005

Lung

Thoracic vs. cardiothoracic vs. general surgeon

National Medicare Database

Multicenter

25,545

 

Improved

  

23

Herr 2004

Bladder

Urologic oncology vs. urology

Patient records

Multicenter (109)

268

  

No significant difference

No significant difference

24

McKenna 2004

Melanoma

Dermatologist vs. plastic surgeon vs. general practitioner vs. general surgeon

Scottish Melanoma Group database

Multicenter

1,536

  

Improved for dermatologists

Improved for dermatologists compared with plastic surgeons, general practitioners, and general surgeons

25

Silvestri 1998

Lung

Thoracic vs. general surgeon

South Carolina’s State Budget and Control Board’s Office of Research and Statistics

Multicenter

1,583

 

Improved for lobectomy but not pneumonectomy

  

26

Gillis 1996

Breast

Specialist vs. general surgeon

West of Scotland Cancer Registry

Multicenter (10)

3,786

   

Improved

27

Gulliford 1991

Bladder

Urologist vs. general surgeon

Thames Cancer Registry

Multicenter (71)

574

   

No significant difference

ABS American Board of Surgery, SEERSurveillance Epidemiology and End Results, HPB hepatopancreaticobiliary

* The first specialization group listed is the one being examined. The second specialization group listed is the reference category

Specialization

Twenty-seven articles investigated the role of specialization on outcomes for cancer surgery (Table 1). Twelve specialties were compared, with 22 of the studies comparing a surgical subspecialty to general surgeons. Colorectal surgeons had the most comparisons (n = 10), followed by gynecologists (n = 7), gynecologic oncologists (n = 7), thoracic surgeons (n = 3), urologists (n = 2), surgical oncologists (n = 2), plastic surgeons (n = 1), urologic oncologists (n = 1), hepatopancreaticobiliary (n = 1), dermatologists (n = 1), and general practitioners (n = 1). The cancer that was most studied was colorectal (n = 10).

There were ten studies which investigated perioperative mortality. Six studies showed a significantly lower risk of death when operated on by a surgeon with specialty training. The six studies that showed a difference examined colon, breast, lung, and gastric cancer. The four studies that did not show a difference examined colon, hepatic, ovarian, and esophageal cancer. Prystowsky et al. found that perioperative mortality for colectomy was not significantly associated with certification by the American Board of Colon and Rectal Surgery (ABCRS).14 Another study by Dimick et al. found that perioperative mortality was lower for esophageal resection preformed by a thoracic surgeon compared with a general surgeon after adjusting for hospital volume but not after adjusting for surgeon volume.35 The third study which did not show a difference was an ovarian cancer study by Elit et al. that found that perioperative mortality for ovarian cancer surgery was not significantly different between gynecologic oncologists, gynecologists, and general surgeons.24 Mckay et al. found no significant difference in mortality when comparing surgical oncologists, hepatobiliary surgeons, “other” subspecialty training, and general surgeons performing hepatic resections for hepatocellular carcinoma, metastatic lesions, and other various indications.36 Eisenkop et al. found that gynecological oncologists had a higher morbidity rate compared with gynecologists, general surgeons, and surgical oncologists; however, gynecologic oncologists had the lowest mortality rate.23

Seven studies looked at the association of surgeon specialization and recurrence rates for three cancers (five colorectal, one melanoma, and one bladder). Six of the seven studies found a significantly lower risk of local recurrence when operated on by a surgeon with specialty training. Five studies compared colorectal surgeons with general surgeons, with patients operated on by the latter having higher recurrence rates for colorectal cancer. McKenna et al. showed that, for melanoma, dermatologists have the lowest recurrence rates 5 years post excision compared with general and plastic surgeons (benefits were only suggestive compared with general practitioners), even after adjusting for age, sex, tumor depth, histology, site, and ulceration.30 It should be noted that a greater percentage of patients in the dermatology or general practitioner groups underwent a subsequent wide local excision when compared with general surgeons (90% versus 43%), and general surgeons performed the second procedure in 64% of cases. Herr et al. reported a difference in recurrence for patients with bladder cancer treated by a urologic oncologist when compared with a urologist, but the finding was not significant.37

Twenty of the 27 studies examined long-term survival for five cancers, including colorectal, melanoma, breast, bladder, and ovarian (Fig. 1). All but two studies found that patients who were operated on by specialty surgeons had significantly better long-term survival rates, and a third found a significant difference for gynecologic oncologists and gynecologists when compared with general surgeons only in stage III ovarian cancer.26 The studies that did find a significant difference in survival examined ovarian (n = 9), colorectal (n = 7), melanoma (n = 1), and breast (n = 1) cancers.
https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0467-8/MediaObjects/10434_2009_467_Fig1_HTML.gif
Fig. 1

Effect estimates from the studies examining the impact of surgeon training on long-term survival. Estimates less than 1.0 indicate a lower risk death long term

Of the 20 studies examining long-term survival, two studies did not find that surgeon specialization was associated with long-term survival. Gulliford et al. examined the difference between urologists and general surgeons, while Herr et al. compared survival between urologic oncologists and general urologists for the treatment of bladder cancer.32,37 Both studies were suggestive of better survival when the more specialized surgeon performed the procedure, but this finding did not reach significance. Chan et al. found a significant survival advantage for being treated by a gynecologic oncologist for ovarian cancer but the magnitude of the difference diminished after controlling for both primary surgery (diagnosis, staging, and tumor cytoreduction) and chemotherapy usage.21 Although Bokey et al. found improved long-term survival after rectal cancer resection by colorectal surgeons as compared with general surgeons, the difference was not statistically significant.12 Finally, there was one study by Elit et al. that showed better overall survival for patients that were operated on by gynecologists compared with gynecologic oncologists.24 However, both gynecologists and gynecologic oncologists still had better long-term survival rates compared with general surgeons. This was the only study to find that gynecologists outperformed gynecologic oncologists.

Experience

Four studies investigated the role of surgeon experience on outcome (Table 2). Experience was either measured in years after residency or after medical school. Of the four studies, three focused on colorectal cancer, and one focused on all cancers of the alimentary tract. All four studies examined perioperative outcomes. Two studies demonstrated a decrease in perioperative mortality rates with increasing surgeon experience, and two studies did not show a difference in outcomes by years of experience. The two that showed a decrease were by the same primary author and were large, state-wide studies that were not cancer specific, but were operation specific.14,38 One examined segmental colon resection while the following paper expanded to all alimentary tract surgeries. In both studies, the surgeons were stratified into groups based on years practicing since American Board of Surgery (ABS) certification and showed decreasing mortality with increasing years since certification. Conversely, two studies did not show a difference in outcomes based on surgeon experience. McArdle et al. separated attending surgeons into groups based on when they completed their fellowship.39 They found no difference in outcome after controlling for volume and case-related variables. The study by Holm et al. examining colorectal cancer was also suggestive of a difference in mortality between surgeons with more than 10 years of experience compared with those with less than 10 years, but the finding did not reach statistical significance.40
Table 2

Effect of experience on surgical outcomes

Study no

Author, year

Malignancy

Comparison

Data source

Setting (number of hospitals)

Number of patients

Morbidity

Postoperative mortality

Recurrence

Survival

1

Prystowsky 2005

Alimentary tract

Experience (time since ABS certification)

Illinois Hospitals and Health Systems Association Compdata database

Multicenter (205)

120,160

Improved in highly complex cases

Improved in higher complexity cases

  

2

McArdle 2004

Colorectal

Years of experience

Case notes

Multicenter (11)

3,200

 

No significant difference

 

No significant difference

3

Prystowsky 2002

Colon

Experience (time since ABS certification)

Illinois Hospitals and Health Systems Association Compdata

Multicenter (76)

15,427

No significant difference

Improved for longer time since ABS certification

  

4

Holm 1997

Rectal

Experience (>10 years vs. <10 years)

Patient records

Multicenter (14)

1,399

No significant difference

No significant difference

Improved

Improved

ABS American Board of Surgery

Three articles looked at the association of years of experience and morbidity rates. Prystowsky et al. grouped surgeons into three different groups:<5 years, 5–20 years, and >20 years since ABS certification in both of their studies.14,38 Despite controlling for differences in case mix, they found that surgeons with fewer than 5 years of experience had higher morbidity and mortality rates after high-complexity alimentary tract surgery. No difference in morbidity and mortality rates according to experience was seen with low-complexity alimentary tract surgery. No difference was seen between the 5–20 years and >20 years groups for either high- or low-complexity cases. Conversely, Holm et al. did not find a difference in complication rates by surgeon experience.40 Two studies investigated the association between surgeon experience and survival. McArdle et al. found no significant difference by experience.39 The sole study to find a difference in long-term outcomes by surgeon experience was by Holm et al. that found a lower risk of recurrence and death from rectal cancer when operated on by a surgeon with more than 10 years of experience.

Discussion

Numerous studies have shown considerable variability in outcomes after cancer resections.14 The surgeon is one factor that may partially explain the differences in outcomes. Most studies thus far have focused on surgeon volume.41 However, volume is a surrogate measure of other factors and may not be the most important factor associated with outcomes.9 Surgeon training and experience may underlie this variability in postsurgical outcomes. Our systematic review of the literature found that 27 of 29 published studies meeting our inclusion criteria demonstrated that surgeon specialty training and increased experience resulted in better outcomes.

Surgeon Specialization

Numerous surgical oncology specialties were represented among the studies that focused on surgical training, including gynecology, urology, general surgery, thoracic surgery, and dermatology. Overall, seven of ten studies demonstrated improvement for short-term outcomes (perioperative morbidity or mortality) when the operation was done by a more specialized surgeon. Six of 7 studies showed a difference in recurrence, and 18 of 20 showed a difference in long-term survival when a more specialized surgeon performed the surgery.

The effect of surgeon training on quality and outcomes remains controversial. The definition of specialization varied widely in these studies, ranging from training levels, practice designations, and self-reporting. It remains unclear whether additional training or a specialized practice is really what affects outcomes. Is it more important to do an additional year of training in a fellowship program, or to focus your practice for decades on a specific disease process and perform a limited number of operations, or both? Surgeons with fellowship training in a particular area are certainly more likely to have higher case volumes, but it is unclear whether it is case volume or other aspects of the fellowship that result in better outcomes. Dimick et al. found that the effect of surgeon specialty on perioperative mortality persisted for esophageal resections after adjusting for hospital volume, but became insignificant when controlling for surgeon volume.35 Thus, for this one cancer, it appears that a surgeon’s operative volume and experience with a specific disease entity may be more important than fellowship training. Moreover, this report demonstrates that a well-done study with consideration of important confounding factors can help to inform the debate. Risk-adjusted outcomes data such as that from the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) may also offer an additional opportunity to study individual surgeon outcomes with adequate risk adjustment for case mix.

As a recent survey study of senior general surgery residents by Borman et al. showed that 77% of residents are electing to go on for more specialty training, it is also important to examine the effect of additional training, as considerable time and resources are spent on post-residency fellowships.42 Additional studies are needed to determine precisely which factors of the fellowship training relate to improved outcomes. Transference of these factors to residency training may help to improve outcomes as these surgeons enter practice and reduce the need for such a large number of residents to undergo additional training after already completing 5 years of residency.

Experience

Although only four studies assessed the effect of surgeon experience, they all examined gastrointestinal cancers. Three examined colon or rectal cancer, while the fourth also included esophageal, biliopancreatic, and foregut cancers. These analyses were all large studies with adjustment for relevant confounding factors. However, as they did not focus on cancers beyond gastrointestinal malignancies, their conclusions cannot be generalized to other organ systems and operations.

The effect of experience is greatly dependent on how it is categorized. If one compares surgeons in their first year of practice to those with ten or more years of experience, then there will likely be a large difference in outcomes. However, a comparison of those with less than five years to those with five or more years may not show an appreciable difference. Examination of experience using a continuous variable or objective, a priori categorizations of time thresholds are needed in future studies.

The current literature available regarding surgeon training and experience has some limitations. First, studies have used various study designs and methodologies. Many of the datasets used to address these issues do not contain enough information to adjust for important confounding factors such as hospital volume, surgeon volume, and comorbidities. Eisenkop et al. found that gynecologic oncologists had higher morbidity but actually lower overall mortality than nonspecialists, and suggested that academic or specialty trained physicians may be referred sicker patients or those with more advanced disease, though true differences in the quality of care cannot be excluded.23 Thus, it is important to control for comorbidities and disease stage when examining the impact of surgeon training and experience. Differences in case mix may have important effects when comparing outcomes based on provider specialization. Secondly, studies have focused on a relatively small number of malignancies, thus the results may not be broadly generalizable. Thirdly, there is undoubtedly a publication bias. Most studies are performed by research teams who are clinically focused on a particular disease process. Though often not intentional, these teams are motivated to show that their care is superior to that of nonspecialists, thus if a study shows no difference in outcomes between specialists and nonspecialists, these studies may not be published. Similarly, some of the studies included in this report may have conclusions that are not well supported by their results. Thus, the limited scope and inherent flaws in many of these studies should be weighed cautiously before policy decisions are made based on these reports.

Finally, administrative datasets often contain information on surgeons and perioperative outcomes, thus as a matter of convenience, most studies have focused on short-term results. These short-term outcomes may be influenced by the surgeon, but are more likely dependent on the hospital (i.e., specialized intensive care units, nursing, and anesthesia). The real benefit of an experienced oncologic surgeon may be seen in examining long-term outcomes such as recurrence and survival, as well as quality of life and functional status. Specialists may perform a more complete resection and better staging (i.e., number of lymph nodes examined), and they may also be more inclined to perform less radical procedures (i.e., sphincter-preserving surgery and lumpectomy).4346

In the end, we should keep the surgeon’s role in perspective. As eloquently stated by Dr. Blake Cady, “In the field of surgical oncology, tumor biology is king, patient selection is queen, and technical maneuvers are the prince and princess who try, but usually fail, to usurp the throne.”47 Technical capacity may relate to both training and experience. These factors may be mutable in comparison with tumor biology. However, the surgeon is the one who generally selects the patient, and specialty training and experience may help in this aspect as well.

Conclusion

Payers and oversight agencies are beginning to use structural characteristics such as surgeon training, experience, and volume as a basis for referral decisions. Moreover, the majority of surgery residents are prolonging their training to gain additional experience. Thus, there is a need to understand specific factors which underlie the better outcomes for specialty-trained, experienced, high-volume surgeons. Further development of guidelines and quality measures addressing these factors can help to identify issues that inexperienced, nonspecialty, low-volume surgeons can use to improve patient care.

Acknowledgment

Supported in part by a VA Career Development Award (DJB) and by a grant from the Illinois Division of the American Cancer Society (DJB).

Copyright information

© Society of Surgical Oncology 2009