Patients commonly ask the attending surgeon, “Who will be performing this operation?”6 While most patients indicate a willingness to take part in resident education, one-third expressly disallow resident participation in their procedures, and others express concern over the role that a resident physician will play in their care.3 The notion that resident participation may lead to higher rates of morbidity and mortality has been studied through-out many surgical fields.5,9–14 In general, staff surgeons perform operations faster, but the influence of resident participation on complication rates varies across disciplines and has been deemed clinically insignificant.5,9
In neurosurgery, scant literature exists on outcomes associated with resident participation.11 Therefore, we sought to determine the effect of resident surgeon participation on overall 30-day morbidity and mortality following neurosurgical procedures by using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.
Methods
Data Source and Patients
The ACS NSQIP database was queried for all patients who had undergone elective or emergent neurosurgical procedures between 2006 and 2012. Including over 250 participating international academic and nonacademic hospitals, the NSQIP is a nationally validated, risk-adjusted, prospectively collected database.8 Each hospital assigns a trained surgical clinical reviewer to collect perioperative clinical data on randomly assigned patients. Each hospital also assigns a “surgeon champion” to oversee accurate program implementation and ensure data reliability. To date, data on more than 2 million patients have been prospectively collected, including over 60,000 neurosurgical cases.
Definition of Morbidity and Mortality
The primary outcome was short-term (30-day) overall postoperative morbidity, which was an aggregation of all available complications in the NSQIP database. These complications include wound infection, systemic infection, cardiac, respiratory, renal, neurological, thromboembolic events, and unplanned returns to the operating room. Mortality was also defined within a 30-day postoperative window. The operating surgeon(s), whether an attending only or an attending plus a resident, was assessed for his or her influence on morbidity and mortality.
Statistical Analyses
General summary statistics were used to describe the study population with continuous variables shown as the mean ± standard deviation, and categorical variables are shown as the number of patients with the percentage of the cohort. Group comparisons were made using the t-test or Wilcoxon rank-sum test for continuous variables or the chi-square test for categorical variables, as appropriate. Results were considered significant if the observed p value was less than 0.05.
Logistic Regression Analyses
Multivariate logistic regression was used to estimate odds ratios for 30-day postoperative morbidity and mortality outcomes for the attending-only versus attending plus resident cohorts (attending group vs attending+resident group, respectively). A forward stepwise approach was used for modeling with adjustment for both clinically and statistically relevant confounders. The following 20 variables were adjusted for: age, sex, body mass index (BMI), smoking status, operation year, work relative value units, wound classification, current wound infection, transfusion < 72 hours prior to surgery, American Society of Anesthesiologists classification, cardiovascular, neurological, respiratory, renal, hepatobiliary, and hemato-oncological comorbidity, diabetic status, steroid use for chronic condition, length of operation, and history of operation within 30 days of the surgery.
Data management and statistical analyses were done with Stata/SE 12. In accordance with Johns Hopkins guidelines, which follow the US Code of Federal Regulations for the Protection of Human Subjects, institutional review board approval was not needed or sought for the present study. Data were collected as an ACS NSQIP quality assurance endeavor, and only de-identified data were received.
Results
Demographic and Preoperative Comorbidity Data
In this study we analyzed all 61,973 neurosurgical patients available from the ACS NSQIP database for the period between 2006 and 2012. Among these patients, 40,939 had available data indicating whether they were elective or trauma cases. The study population included patients with surgeon identification, either attending alone or attending plus resident. The final cohort, including cases with both elective or emergent designation and primary surgeon information, consisted of 16,098 patients total (15,038 elective and 1060 emergent). The top 12 diagnoses were displacement of lumbar intervertebral disc without myelopathy (13.44%), lumbar spinal stenosis (11.46%), malignant neoplasm of the brain (4.81%), cervical spondylosis with myelopathy (4.70%), lumbosacral spondylosis without myelopathy (3.11%), degeneration of lumbar or lumbosacral intervertebral disc (2.75%), secondary malignant neoplasm of brain and spinal cord (2.73%), benign neoplasm of cerebral meninges (2.73%), cervical spondylosis without myelopathy (2.48%), acquired spondylolisthesis (2.42%), cervical intervertebral disc disorder with myelopathy (2.18%), and nonruptured cerebral aneurysm (1.48%). The mean patient age was 56.8 ± 15.0 years, and 49.8% of patients were women (Table 1).
General characteristics of patients who underwent neurosurgical procedures, stratified by operating surgeon status*
| Variable | All Neurosurgical Patients | Attending+Resident Group | Attending Group | p Value |
|---|---|---|---|---|
| Total no. of patients | 16,098 | 7843 | 8255 | |
| Mean age in yrs | 56.8 ± 15.0 | 56.0 ± 15.1 | 57.7 ± 14.9 | 0.271 |
| Mean BMI in kg/m2 | 29.5 ± 6.7 | 29.3 ± 6.6 | 29.8 ± 6.8 | 0.027 |
| % Female sex | 49.75 | 49.85 | 49.64 | 0.185 |
| Race | <0.001 | |||
| % White | 69.70 | 71.99 | 67.28 | |
| % African American | 7.29 | 7.46 | 7.10 | |
| % Latino | 19.19 | 17.40 | 21.08 | |
| % Asian/American Indian/Native Hawaiian/Pacific Islander | 1.43 | 1.25 | 1.62 | |
| % Unknown | 2.40 | 1.90 | 2.92 | |
| % Current smokers | 24.22 | 23.42 | 24.98 | 0.021 |
| % w/ Diabetes | 14.65 | 13.27 | 15.95 | <0.001 |
| Non–insulin dependent | 9.62 | 8.27 | 10.89 | |
| Insulin dependent | 5.03 | 5.00 | 5.06 | |
| Mean op time in mins | 165.0 ± 117.0 | 200.4 ± 128.0 | 131.4 ± 94.0 | <0.001 |
| % w/ Alcohol intake in prior 2 wks†‡ | 3.73 | 3.55 | 3.91 | <0.001 |
| % w/ Steroid use for chronic condition | 6.42 | 8.72 | 4.24 | <0.001 |
| % w/ Op w/in prior 30 days‡ | 2.36 | 3.35 | 1.30 | <0.001 |
| Composite system morbidity variable | ||||
| % Cardiovascular | 14.00 | 12.25 | 15.65 | <0.001 |
| % Neurological§ | 20.03 | 28.74 | 11.76 | <0.001 |
| % Respiratory | 6.14 | 6.94 | 5.39 | <0.001 |
| % Hepatobiliary | 0.09 | 0.17 | 0.02 | 0.003 |
| % Renal | 0.53 | 0.60 | 0.46 | 0.224 |
| % Hemato-oncological | 8.34 | 11.19 | 5.62 | <0.001 |
Values expressed as mean ± standard deviations, unless indicated otherwise. Values in boldface type indicate statistical significance.
More than 2 drinks/day.
Value reflects percentage of patients with available data rather than percentage of column total.
Composite neurological morbidity other than main lesion.
Between the attending and attending+resident groups, the latter had slightly lower BMIs, a greater percentage of whites and African Americans than Latino and Asian patients, and a greater percentage of patients on steroids for a chronic condition. The attending+resident cohort also had more operations within the previous 30 days, longer operation times, and lower rates of diabetes, was less likely to smoke, and consumed less alcohol within 2 weeks of surgery. The two cohorts were similar in age and sex distributions. Patients in the attending+resident group had higher rates of preoperative composite respiratory, hepatobiliary, neurological, and hemato-oncological comorbidities. The attending cohort had a higher rate of cardiovascular comorbidity. Renal comorbidity was similar between the two cohorts (Table 1).
Postoperative Complications
Overall, 15.80% of patients had at least one postoperative complication. The attending+resident cohort demonstrated a complication rate of 20.12%, while patients with an attending-only surgeon had a statistically significantly lower complication rate at 11.70% (p < 0.001; Table 2). Superficial surgical site infection (SSI; 1.02% vs 0.90%) and wound dehiscence (0.19% vs 0.15%) were the same (p > 0.05) between the attending+resident and attending cohorts. All morbidity complications that significantly differed between the two cohorts had higher rates in the attending+resident cohort, including deep incisional SSI (0.71% vs 0.45%), organ space SSI (0.77% vs 0.24%), pneumonia (1.96% vs 1.03%), unplanned intubation (1.85% vs 0.80%), pulmonary embolism (0.83% vs 0.39%), ventilator use > 48 hours (2.64% vs 1.03%), urinary tract infection (2.27% vs 1.78%), cerebrovascular accident/stroke with neurological deficit (1.13% vs 0.36%), coma longer than 24 hours (0.60% vs 0.30%), myocardial infarction (0.37% vs 0.17%), blood transfusions (10.17% vs 4.49%), deep venous thrombosis/thrombophlebitis requiring treatment (1.59% vs 0.75%), sepsis (1.71% vs 0.80%), septic shock (0.56% vs 0.25%), and unplanned returns to the operating room (5.01% vs 3.19%). In the total population, 263 patients (1.63%) died within 30 days of their surgery. Stratified by operating surgeon status, 162 patients (2.07%) in the attending+resident cohort died versus 101 (1.22%) in the attending group, which was statistically significantly different (p < 0.001).
Thirty-day mortality and morbidity complications in patients who underwent neurosurgical procedures, stratified by operating surgeon status
| Variable | Attending+Resident Group | Attending Group | p Value |
|---|---|---|---|
| Total no. of patients | 7843 | 8255 | |
| % Deceased | 2.07 | 1.22 | <0.001 |
| Morbidity | |||
| % w/ Superficial SSI | 1.02 | 0.90 | 0.421 |
| % w/ Deep incisional SSI | 0.71 | 0.45 | 0.026 |
| % w/ Organ space SSI | 0.77 | 0.24 | <0.001 |
| % w/ Wound dehiscence | 0.19 | 0.15 | 0.477 |
| % w/ Pneumonia | 1.96 | 1.03 | <0.001 |
| % w/ Unplanned intubation | 1.85 | 0.80 | <0.001 |
| % w/ Pulmonary embolism | 0.83 | 0.39 | <0.001 |
| % On ventilator >48 hrs | 2.64 | 1.03 | <0.001 |
| % w/ Progressive renal insufficiency | 0.15 | 0.12 | 0.584 |
| % w/ Acute renal failure | 0.15 | 0.06 | 0.071 |
| % w/ Urinary tract infection | 2.27 | 1.78 | 0.028 |
| % w/ CVA/stroke w/ neurological deficit | 1.13 | 0.36 | <0.001 |
| % w/ Coma longer than 24 hrs | 0.60 | 0.30 | 0.005 |
| % w/ Cardiac arrest requiring CPR | 0.32 | 0.23 | 0.282 |
| % w/ Myocardial infarction | 0.37 | 0.17 | 0.014 |
| % w/ Blood transfusions | 10.17 | 4.49 | <0.001 |
| % w/ Graft/prosthesis failure | 0.03 | 0.01 | 0.534 |
| % w/ DVT/thrombophlebitis requiring treatment | 1.59 | 0.75 | <0.001 |
| % w/ Sepsis | 1.71 | 0.80 | <0.001 |
| % w/ Septic shock | 0.56 | 0.25 | 0.002 |
| % w/ Unplanned return to operating room* | 5.01 | 3.19 | <0.001 |
| Total no. morbidity events | 2659 | 1443 | |
| Total % patients w/ ≥1 morbidity event | 20.12 | 11.70 | <0.001 |
CPR = cardiopulmonary resuscitation; CVA = cerebrovascular accident; DVT = deep vein thrombosis.
Unplanned reoperations only recorded post-2011.
Regression Analyses
In regression analyses we compared the attending+resident group with the attending cohort, the referent group. Exact adjustments dependent on multivariate models are listed in Table 3. After univariate regression analysis, the attending+resident cohort demonstrated increased odds of 30-day morbidity (OR = 1.85, 95% CI 1.68–2.04) and mortality (OR = 1.79, 95% CI 1.24–2.60) compared with the attending group following elective surgery. For emergent cases, univariate regression demonstrated increased odds of 30-day morbidity (OR = 1.64, 95% CI 1.27–2.10) but not mortality (OR = 1.08, 95% CI 0.75–1.55). Following adjustment for preoperative patient characteristics and comorbidities in a multivariate regression, the odds of morbidity among elective surgeries with resident participation were equivalent to those without resident participation (OR = 1.05, 95% CI 0.93–1.18). Furthermore, the odds of mortality among elective surgeries with resident participation were equivalent to those for the attendingonly counterparts (OR = 1.01, 95% CI 0.64–1.58). Emergent surgeries demonstrated the same trend of equivalent odds of either morbidity (OR = 1.04, 95% CI 0.72–1.50) or mortality (OR = 0.84, 95% CI 0.49–1.41) between the attending+resident and the attending cohorts.
Logistic models for 30-day overall morbidity and mortality for patients who underwent neurosurgical procedures
| Variable | Univariate Regression | Multivariate Regression* | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p Value | OR | 95% CI | p Value | |
| Elective surgery | ||||||
| 30-Day morbidity | ||||||
| Attending only | 1.0 | 1.0 | ||||
| Attending+resident | 1.85 | 1.68–2.04 | <0.001 | 1.05 | 0.93–1.18 | 0.442 |
| 30-Day mortality | ||||||
| Attending only | 1.0 | 1.0 | ||||
| Attending+resident | 1.79 | 1.24–2.60 | 0.002 | 1.01 | 0.64–1.58 | 0.979 |
| Emergent surgery | ||||||
| 30-Day morbidity | ||||||
| Attending only | 1.0 | 1.0 | ||||
| Attending+resident | 1.64 | 1.27–2.10 | <0.001 | 1.04 | 0.72–1.50 | 0.827 |
| 30-Day mortality | ||||||
| Attending only | 1.0 | 1.0 | ||||
| Attending+resident | 1.08 | 0.75–1.55 | 0.677 | 0.84 | 0.49–1.41 | 0.500 |
Multivariate analysis for 30-day morbidity after adjusting for the following variables: 1) age, 2) sex, 3) BMI, 4) smoking status, 5) American Society of Anesthesiologists classification, 6) work relative value units, 7) wound classification, 8) current wound infection, 9) transfusion <72 hours prior to surgery, 10) previous cardiovascular morbidity, 11) previous neurological morbidity, 12) previous respiratory morbidity, 13) previous renal morbidity, 14) diabetic status, 15) steroid use for chronic condition, 16) length of operation, and 17) history of previous operation within 30 days of surgery. The multivariable analysis for 30-day mortality after adjusting for the following variables: 1–13 above and 14) previous hepatobiliary morbidity, 15) previous hemato-oncological morbidity, 16) diabetic status, 17) steroid use for chronic condition, and 18) length of operation.
Subanalyses Based on Procedure
Univariate and multivariate analyses were performed to obtain adjusted odds ratios of morbidity and mortality for each group of procedures (Table 4). The subanalysis findings were the same as demonstrated for all procedures, revealing no significant increase in morbidity and mortality for cases involving residents. However, there was a trend toward increased morbidity in spinal surgery cases involving residents (OR 1.17, 95% CI 0.99–1.39), but this did not reach statistical significance (p = 0.055).
Multivariate analyses of morbidity and mortality in patients who underwent surgery by attendings only versus attendings plus residents, stratified by procedure type*
| Type of Surgery | Morbidity | Mortality | ||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Spine surgery | 1.17 | 0.99–1.39 | 2.7 | 0.88–8.42 |
| Brain tumor surgery | 0.88 | 0.63–1.21 | 0.49 | 0.23–1.04 |
| Vascular neurosurgery | 1.44 | 0.41–5.00 | 0.94 | 0.00–100 |
| Functional neurosurgery | 5.18 | 0.53–50.6 | — | — |
| Brain trauma | 1.04 | 0.51–2.12 | 1.13 | 0.53–2.42 |
| Other | 0.26 | 0.06–1.18 | — | — |
The attending-only group was used as the reference (1.0) for both morbidity and mortality. All analyses were adjusted for significant covariates after univariate analyses, mirroring Table 3. “Other” procedures include nerve decompression for pain and decompressions for Chiari malformation. Calculations for mortality following functional neurosurgery and other procedures were not possible given the small number of observations.
Discussion
Dr. William Halsted of Johns Hopkins Hospital established the contemporary system of surgical residency in the late 19th century.2 The Halstedian surgical training regimen remains in place today (albeit with alterations) despite enormous societal, regulatory, and medico-legal pressures.7,12 In recent years, attending physicians and residents have expressed concern relating to work-hour restrictions, breadth of surgical volume, perioperative continuity of care, and surgical autonomy.1,7,12,14 In this paper, we examined another concern regarding resident education: impact of resident participation on surgical outcomes.
When attending surgeons operated with residents, 30-day morbidity and mortality were no different from those after attending physicians operated without residents. However, many morbidity variables, along with mortality, were increased with resident participation. Chi-square testing revealed that the incidence of complications and mortality was greater in any surgeries with resident participation, but multivariate regression analysis demonstrated that confounding factors influenced this increase. After adjustment, the odds of 30-day morbidity and mortality were no different between the two cohorts, suggesting that resident participation is not an independent risk factor for the increased morbidity and mortality observed in these cases. As expected, operation times were longer during attending+resident operations, a potential source for increased rates of postoperative complications. Additionally, patient's preoperative health comorbidities were generally higher in the attending+resident cohort, indicating patients were in worse preoperative health. The increased complexity of cases in the attending+resident cohort may be a result of hospital teaching status. Resident participation generally occurs in teaching hospitals, whereas most attending-only surgeries may have occurred in nonteaching hospitals.
To our knowledge, neurosurgical resident surgeon participation and its influence on patient outcomes has been analyzed in two prior studies.11,15 Morgan et al. reviewed craniotomies for berry aneurysms ≤ 10 mm in size in the internal carotid artery (beyond the paraclinoid segment) and middle cerebral artery. Resident participation was defined as a resident scrubbed into the procedure, and whether the resident was within his or her last 3 years of residency was noted. Of 355 cases, 196 involved resident participation and 159 did not. Cases with participation from residents within their first 3 years of residency had a complication rate of 10.7%. Among cases with participation from residents in their last 3 years (an advanced resident), the complication rate decreased to 2.4%, which was statistically significant. When cases with participation from advanced residents were compared with cases handled by attending surgeons, there was no difference in adverse outcome rates. Morgan and associates concluded that a learning curve is seen as residents progress through their program years, and they suggested that senior resident participation in the repair of small unruptured aneurysms does not compromise patient outcomes. In the second study, a general surgery group assessed only the laparoscopic aspect of ventriculoperitoneal shunt placement, finding no increase in adverse events with resident participation.15
Many surgical fields have studied the influence of resident participation on surgical outcomes, none more so than general surgery. Hernández-Irizarry et al. studied 6223 laparoscopic inguinal hernia repairs utilizing the NSQIP database, finding that resident participation increased surgical time but did not affect surgical morbidity outcomes.9 Relles et al. studied 686 pancreaticoduodenectomy (PD) outcomes at a single high-volume center13 and established a relationship between increasing resident experience and improved outcomes. This relationship mirrors the volume relationships found among both attending surgeons and hospitals in terms of decreasing adverse events as more procedures are done per annum. Schoenfield et al. used the NSQIP database to analyze 43,343 cases of 12 different orthopedic procedures.14 They found a minimal increase in the risk of complications in joint arthroplasty surgeries if a resident had participated, but no relationship was found for the 11 remaining procedures. Mortality was no different among resident cases versus attending-only cases in any orthopedic procedure.
Related to resident education is the so-called “July phenomenon,” in which a supposed increase in adverse events is seen with resident turnover and incoming surgical interns as well as new attending surgeons in July. This phenomenon has been studied in neurosurgery, including an in-depth analysis of over 850,000 neurosurgical patients in the Nationwide Inpatient Sample database.16 Stratifying by nontraumatic hemorrhage, CNS trauma, CNS tumor, or hydrocephalus, the researchers observed no “July phenomenon”; however, a publication on spinal metastasis surgery found that surgeries in July had higher rates of in-hospital mortality and postoperative complications.4 Given that July is when new interns, residents, fellows, and attending surgeons begin practicing, it is difficult to attribute this effect to any one group.
Prospective collection of the ACS NSQIP database provides a robust foundation for the present analysis. The detailed perioperative variables, tailored for surgical outcome studies, provide for extensive adjustment of statistically and clinically relevant confounders. Data in the study suggest that resident participation in neurosurgical procedures does not increase overall 30-day morbidity or mortality. However, limitations to this study do exist, including those inherent to analysis of large administrative databases.17 Only short-term (30-day) conclusions can be drawn from the data. No conclusions or extrapolations can be drawn beyond this 30-day postoperative window. Additionally, hospital teaching status could not be accurately assessed, possibly providing a reason for the greater complications seen in the attending+resident cohort. For this cohort, the database did not provide the postgraduate year level or the extent of resident participation in the procedure. Another issue is the limited applicability of the present findings for the years 2006–2012. However, even with higher incidence rates for various complications, multivariate analysis revealed no difference in either morbidity or mortality. Future prospective studies are needed to validate these conclusions beyond the 30-day postoperative time point. Additionally, a wider scope of subanalyses based on procedure could be performed in future studies, such as long-term outcomes, outcomes in teaching versus nonteaching hospitals, and analyses of junior versus senior residents. We also encourage outcomes research of patients with a higher comorbidity burden who undergo more “complex” procedures by attending physicians only, as this could not be assessed with the current data.
Conclusions
Cases with resident participation were associated with higher rates of mortality and morbidity; however, these cases involved patients with more comorbidities initially. On multivariate analysis, resident participation did not independently increase the odds of postoperative 30-day morbidity or mortality over attending-only cases in elective or emergent neurosurgery. Resident education through supervised operative experience is imperative to training, and based on our data, resident involvement does not compromise immediate procedural outcomes in neurosurgical procedures.
Author Contributions
Conception and design: M Bydon, Abt. Acquisition of data: M Bydon, Abt, De la Garza-Ramos. Analysis and interpretation of data: M Bydon, Abt, De la Garza-Ramos, Macki. Drafting the article: M Bydon, Abt, De la Garza-Ramos, Macki. Critically revising the article: M Bydon, Abt, De la Garza-Ramos, Macki. Reviewed submitted version of manuscript: M Bydon, Abt, De la Garza-Ramos. Statistical analysis: Abt, Macki. Administrative/ technical/material support: Huang, Witham, Gokaslan, A Bydon. Study supervision: Huang, Witham, Gokaslan, A Bydon.
References
- 1↑
Bukur M, , Singer MB, , Chung R, , Ley EJ, , Malinoski DJ, & Margulies DR, et al.: Influence of resident involvement on trauma care outcomes. Arch Surg 147:856–862, 2012
- 3↑
Cowles RA, , Moyer CA, , Sonnad SS, , Simeone DM, , Knol JA, & Eckhauser FE, et al.: Doctor-patient communication in surgery: attitudes and expectations of general surgery patients about the involvement and education of surgical residents. J Am Coll Surg 193:73–80, 2001
- 4↑
Dasenbrock HH, , Clarke MJ, , Thompson RE, , Gokaslan ZL, & Bydon A: The impact of July hospital admission on outcome after surgery for spinal metastases at academic medical centers in the United States, 2005 to 2008. Cancer 118:1429–1438, 2012
- 5↑
Davis SS Jr, , Husain FA, , Lin E, , Nandipati KC, , Perez S, & Sweeney JF: Resident participation in index laparoscopic general surgical cases: impact of the learning environment on surgical outcomes. J Am Coll Surg 216:96–104, 2013
- 6↑
Dutta S, , Dunnington G, , Blanchard MC, , Spielman B, , DaRosa D, & Joehl RJ: “And doctor, no residents please!”. J Am Coll Surg 197:1012–1017, 2003
- 7↑
Emery SE, , Guss D, , Kuremsky MA, , Hamlin BR, , Herndon JH, & Rubash HE: Resident education versus fellowship training-conflict or synergy? AOA critical issues. J Bone Joint Surg Am 94:e159, 2012
- 8↑
Fink AS, , Campbell DA Jr, , Mentzer RM Jr, , Henderson WG, , Daley J, & Bannister J, et al.: The National Surgical Quality Improvement Program in non-veterans administration hospitals: initial demonstration of feasibility. Ann Surg 236:344–354, 2002
- 9↑
Hernández-Irizarry R, , Zendejas B, , Ali SM, , Lohse CM, & Farley DR: Impact of resident participation on laparoscopic inguinal hernia repairs: are residents slowing us down?. J Surg Educ 69:746–752, 2012
- 10
Krell RW, , Birkmeyer NJ, , Reames BN, , Carlin AM, , Birkmeyer JD, & Finks JF: Effects of resident involvement on complication rates after laparoscopic gastric bypass. J Am Coll Surg 218:253–260, 2014
- 11↑
Morgan MK, , Assaad NN, & Davidson AS: How does the participation of a resident surgeon in procedures for small intracranial aneurysms impact patient outcome?. J Neurosurg 106:961–964, 2007
- 12↑
Raval MV, , Wang X, , Cohen ME, , Ingraham AM, , Bentrem DJ, & Dimick JB, et al.: The influence of resident involvement on surgical outcomes. J Am Coll Surg 212:889–898, 2011
- 13↑
Relles DM, , Burkhart RA, , Pucci MJ, , Sendecki J, , Tholey R, & Drueding R, et al.: Does resident experience affect outcomes in complex abdominal surgery? Pancreaticoduodenectomy as an example. J Gastrointest Surg 18:279–285, 2014
- 14↑
Schoenfeld AJ, , Serrano JA, , Waterman BR, , Bader JO, & Belmont PJ Jr: The impact of resident involvement on postoperative morbidity and mortality following orthopaedic procedures: a study of 43,343 cases. Arch Orthop Trauma Surg 133:1483–1491, 2013
- 15↑
Stoddard T, & Kavic SM: Laparoscopic ventriculoperitoneal shunts: benefits to resident training and patient safety. JSLS 15:38–40, 2011
- 16↑
Weaver KJ, , Neal D, , Hoh DJ, , Mocco J, , Barker FG II, & Hoh BL: The “July phenomenon” for neurosurgical mortality and complications in teaching hospitals: an analysis of more than 850,000 neurosurgical patients in the nationwide inpatient sample database, 1998 to 2008. Neurosurgery 71:562–571, 2012
- 17↑
Woodworth GF, , Baird CJ, , Garces-Ambrossi G, , Tonascia J, & Tamargo RJ: Inaccuracy of the administrative database: comparative analysis of two databases for the diagnosis and treatment of intracranial aneurysms. Neurosurgery 65:251–257, 2009
