Yeo H, Viola K, Berg D, Lin Z, Nunez-Smith M, Cammann C, Bell RH, Sosa JA, Krumholz HM, Curry LA. Attitudes, Training Experiences, and Professional Expectations of US General Surgery ResidentsA National Survey. JAMA. 2009;302(12):1301–1308. doi:10.1001/jama.2009.1386
Author Affiliations: Division of Surgery (Dr Yeo); Robert Wood Johnson Clinical Scholars Program (Drs Viola, Nunez-Smith, Krumholz, and Curry); Section of General Internal Medicine (Dr Nunez-Smith); Divisions of Endocrine Surgery and Surgical Oncology (Dr Sosa); Section of Cardiovascular Medicine (Dr Krumholz); Division of Health Policy and Administration, School of Public Health (Dr Curry); Department of Psychiatry (Dr Berg), Yale University School of Medicine, New Haven, Connecticut; Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, Connecticut (Drs Krumholz and Lin); Capella University, Minneapolis, Minnesota (Dr Cammann); and American Board of Surgery Inc, Philadelphia, Pennsylvania (Dr Bell).
Context General surgery residency programs are facing multiple pressures, including attracting and retaining residents. Despite the importance of resident perspectives in designing effective responses to these pressures, understanding of residents' views is limited.
Objective To profile US general surgery residents; characterize resident attitudes, experiences, and expectations regarding training; and examine differences by sex and training year.
Design, Setting, and Participants Cross-sectional study of all general surgery residents completing a survey in January 2008 following administration of the American Board of Surgery In-Training Examination.
Main Outcome Measures Resident satisfaction; perceived supports, strains and concern; career motivations; and professional expectations.
Results Of 5345 categorical general surgery residents, 4402 (82.4%) responded, representing 248 of 249 surgical residency programs. Most respondents expressed satisfaction with training (3686 [85.2%]; 95% confidence interval [CI], 84.1%-86.3%) and supportive peer relationships (3433 [84.2%]; 95% CI, 83.1%-85.3%). However, residents also reported unmet needs and apprehensions about training and careers. Worry that they will not feel confident performing procedures independently was reported by 1185 (27.5%; 95% CI, 26.2%-28.8%), while 2681 (63.8%; 95% CI, 62.4%-65.3%) reported that they must complete specialty training to be competitive. Perceptions of program support differ, with men more likely than women to report that their program provides support (2188 [74.5%] vs 895 [65.6%]; P < .001), and that they can turn to faculty when having difficulties (2193 [74.5%] vs 901 [66.4%]; P < .001). Reports of having considered leaving training in the prior year differed significantly across years (P < .001), highest in postgraduate year 2 (19.2%) and lowest in postgraduate year 5 (7.2%).
Conclusions General surgery residents' attitudes, experiences, and expectations regarding training reflect both high levels of satisfaction and sources of strain. These factors vary by sex and training year.
General surgery residency training is facing formidable pressures1,2 and is likely to undergo considerable changes in the coming decade.3,4 These pressures include diminished attraction to surgery as a profession,4 increasing interest in surgical subspecialization with 70% of residents completing fellowships in 2004,5,6 and estimated attrition rates of 17% to 26% among categorical general surgery residents (residents who have guaranteed positions in a 5-year training program).3,7,8 These attrition rates are higher than other medical residencies2 and have persisted9 despite major reforms such as elimination of the pyramidal system of residency training.1 At the same time, a substantial shortage of general surgeons is predicted.5,10- 12 Strategies responding to these complex and competing challenges can be informed by understanding general surgery residents' attitudes and experiences regarding training, and their association with attrition.
Studies have examined resident satisfaction with training experience and working conditions,13,14 views on the 80-hour work week,15- 18 motivations for19,20 or deterrents to21 pursuing a career in general surgery, and perceptions of the future of general surgery.22 Available evidence indicates that attrition is associated with preferences for controllable lifestyle,7,23 academic and nonacademic factors identifiable on residency applications,24 and female sex,7,25 although findings on sex are inconsistent.9 Studies on the effect of the 80-hour work week on attrition have also yielded mixed findings.23,26,27 Although these studies provide important insights, most were limited by small samples, low response rates, reliance on limited program-level data, or a narrow focus on specific aspects of residents' attitudes and experiences.
Accordingly, we designed a longitudinal study of all US general surgery residents (the National Study of Expectations and Attitudes of Residents in Surgery [NEARS]) with 3 objectives: (1) to provide a national demographic profile of surgical residents; (2) to characterize a comprehensive range of resident attitudes, experiences, and expectations regarding residency training; and (3) to identify predictors of attrition from residency training. This article presents baseline cross-sectional findings.
In 2008, 248 of the 249 surgical residency training programs in the United States distributed optical scan surveys to all general surgery residents immediately following the administration of the American Board of Surgery In-Training Examination (ABSITE). Residency program coordinators at each test setting provided surveys to all residents at the completion of the ABSITE. Program coordinators collected both completed and blank surveys, and placed them in a sealed envelope and into a preaddressed, prepaid package shipped directly to the study principal investigator.
There were 7458 general surgery residents included in the American Board of Surgery (ABS) 2007-2008 Resident Roster. Of these, 6112 were categorical general surgery residents, 5345 of whom were residents in postgraduate year (GY) levels 1 through 5 and therefore eligible for inclusion in the survey. Because we were interested in the experiences of categorical residents, results from other types of residents, including designated preliminary, nondesignated preliminary, or research year residents, were excluded from this analysis. An information sheet indicating that participation was voluntary and ensuring confidentiality was provided with each survey. Completion of the survey constituted implied consent. The protocol was approved by the institutional review board at the Yale School of Medicine.
We used a comprehensive, multistage method to develop the survey.28 The first phase used qualitative methods to identify de novo factors perceived by residents as central to their training experiences by interviewing key informants29 who had direct experience regarding the phenomenon of interest (surgical residency training). We identified all residents who completed the ABSITE examination in 2005 but not in 2006 (n = 155) and contacted programs to confirm the status of these residents (n = 117 left voluntarily, 10 left involuntarily, 9 unspecified, and 1 not specified). We used an SPSS random number generator to determine the order in which we contacted male and female former surgical residents and used purposeful sampling strategies to ensure representation across sex and program year of attrition.29,30 None of the individuals contacted refused to participate.
We conducted in-depth interviews31 to generate narrative data regarding experiences with residency training generally and factors that contributed to the decision to leave training. We interviewed residents until we achieved theoretical saturation,29 that is, until no new concepts emerged from additional interviews. This occurred after 15 interviews. Data were collected by gender-concordant surgical residents. The interviews explored training experiences and reasons for the respondent leaving his or her training program, using open-ended questions such as “what was it like to be a surgical resident?” and “please tell me about your decision to leave X program.”
A 6-member multidisciplinary team32 (H.Y., D.B., M.N.S., J.A.S., L.C., and MR) with diverse backgrounds in surgery, internal medicine, pediatrics, organizational psychology, health services research, and qualitative methods used standard qualitative analysis procedures32,33 and the constant comparative method34 facilitated by Atlas.ti software version 5.0.67 (Scientific Software Development, GmbH, Berlin). The constant comparative method involves classifying verbatim quotations into their essential concepts with the use of codes developed iteratively to reflect the data. This constant comparison of data refines the properties and dimensions of existing codes and identifies new codes to fit the concepts emerging from the data. Using these coded data, we summarized recurrent concepts that described the residents' training experiences and the reasons for their attrition.
We developed a 73-item survey addressing aspects of training described in the literature (eg, motivations for pursuing surgery and interest in fellowship) and concepts that emerged from the qualitative data. The survey was pilot tested with 57 residents at 3 residency training programs. We performed factor analysis on the pilot test data and used the results to identify measures of key concepts and items that could be eliminated from the survey.
The result was a 52-item survey (eTable 1), which sought information on motivations for pursuing surgery as a career, views on specialty training, self-assessed performance, ethical dimensions of training, and perceptions of the future of general surgery. Questions were structured as statements to which the respondent indicated level of agreement on a 5-point Likert response scale, ranging from strongly agree to strongly disagree. Demographic items include sex, race/ethnicity, marital status, number of children, and age. A race/ethnicity variable was included because we wanted to describe the sociodemographic characteristics of residents using US Census categories for race/ethnicity; qualitative interview participants raised the issue and importance of racial/ethnic identity to their professional experiences; and prior research indicates that race/ethnicity is associated with professional outcomes.
Results were summarized by reporting responses on all survey items by all 5 response categories. Fifteen illustrative items were selected for more detailed analyses on characteristics of interest (sex and GY). For ease of presentation, the 5 response categories were collapsed into 3 categories (strongly agree and agree, neutral, strongly disagree and disagree). We focused on the strongly agree/agree category and reported proportions of participants' responses on selected items, with corresponding confidence intervals (CIs). χ2 Tests were used to assess the association between selected respondents' characteristics and their responses. Statistical significance was defined as P < .05, all tests were 2-sided, and CIs were defined as 95%. Analyses were conducted using SAS software, version 9.1 (SAS Institute Inc, Cary, North Carolina).
Of the 5345 categorical general surgery residents in GY 1 through 5 in all 249 programs, completed surveys were received from 4402 (82.4%). A complete survey was defined in accordance with Council of American Survey Research Organizations standards with >80% items completed. There was 92% item completeness of data. All but one of the 249 residency training programs participated. Table 1 summarizes the demographic characteristics of the respondents.
Respondents reported their attitudes, training experiences, and professional expectations on a range of aspects of residency training (Table 2; eTable 2). To illustrate the scope and nature of the findings, response distributions for selected items in each domain are summarized, combining strongly agree/agree (Table 3).
A series of questions examined satisfaction with didactic and operative experiences. The majority of respondents (3686, 85.2%; 95% CI, 84.1%-86.3%) expressed high levels of satisfaction with training. However, 883 (20.5%; 95% CI, 19.3%-21.7%) reported that surgery training is too long, and 661 (15.3%; 95% CI, 14.3%-16.4%) agreed with the statement “I have considered leaving my program in the last year.” This latter item was intentionally broad in nature and therefore reflects a range of potential reasons for having considered leaving, including transfer to another categorical residency slot, reasons not directly related to training, and dissatisfaction with the program in particular or surgery in general.
Another series of questions explored formal program support mechanisms, as well as informal types of supports derived from relationships with faculty and peers. The majority of respondents (3089, 71.6%; 95% CI, 70.3%-73%) reported that their program has support structures for residents who are struggling and that they can turn to the faculty when having difficulties in the program (3101, 71.9%; 95% CI, 70.6%-73.3%). Residents generally reported very positive collaborative relationships with peers, with 3433 (84.2%; 95% CI, 83.1%-85.3%) indicating that they can count on other residents to help them out when they are having a problem. While relationships with attending physicians were reported as generally positive, 671 (15.6%; 95% CI, 14.5%-16.7%) reported that their attending physicians will think worse of them if they ask for help with a procedure, and 348 (8.1%; 95% CI, 7.3%-8.9%) reported that they do not feel respected by their attending physicians.
A series of items examined various sources of stress and concern among residents. Notable proportions of residents reported feeling uneasy or troubled by aspects of training and skill development. Of the respondents, 1312 (30.7%; 95% CI, 29.3–32.1%) reported that the stress of work is causing strain on their family life. Also, 1185 (27.5%, 95% CI, 26.2%-28.8%) expressed apprehension about their clinical skills, worrying that they will not feel confident enough to perform procedures by themselves before they finish training, while 2688 (63.7%; 95% CI, 62.2%-65.1%) reported worry about hurting patients.
Residents reported multifaceted motivations for pursuing surgery as a career. Of the respondents, 3796 (89.6%; 95% CI, 88.7%-90.5%) indicated that they get a tremendous amount of satisfaction working with patients; 1378 (32.8%; 95% CI, 31.4%-34.2%) reported that 1 of the factors that influenced their decision to become a surgeon was the expectation of good financial compensation. Residents also reported their expectations for the future of general surgery, with 2573 (61.0%; 95% CI, 59.5%-62.5%) reporting being concerned that other professionals will take over some of the procedures they perform. Specialty training was widely perceived as necessary, with 2681 (63.8%; 95% CI, 62.4%-65.3%) indicating they will need to complete additional specialty training in order to be competitive in the job market.
The selected items were analyzed by sex (Table 3). Men were more likely to report being satisfied with residency training than women (2553 [86.6%] vs 1123 [82.2%]; P < .001), and women were more likely to have considered leaving residency during the prior year (260 [19.1%] vs 398 [13.6%]; P < .001). Regarding program support, men were more likely to feel their training program would provide them with someone to turn to when they are struggling (2188 [74.5%] vs 895 [65.6%]; P < .001) and to feel they can turn to members of the faculty when having difficulties in the program (2193 [74.5%] vs 901 [66.4%]; P < .001). Women were more likely to express worry that they will not feel confident to perform procedures by themselves before they finish training (518 [37.9%] vs 661 [22.5%]; P < .001). There were also differences in terms of motivations for pursuing a surgical career, with men more likely to indicate that the expectation of good financial compensation influenced their decision (1088 [38.0%] vs 290 [21.7%]; P < .001). When considering specialty training, men were more likely to worry that other professionals will take over some of the procedures that they perform (1824 [63.6%] vs 743 [55.5%]; P < .001).
These items were also examined by postgraduate training year (eTable 3). Residents' overall satisfaction with their training programs varied significantly across training years (P = .001), with the lowest reported levels in GY2 and GY3 (82.8% and 83.2%, respectively) and the highest level in GY5 (89.7%). There were significant differences (P < .001) in residents' reports of having considered leaving training in the prior year (19.2% in GY2 and 7.2% in GY5). Reports of feeling they could turn to faculty when having difficulties in their program differed across training years (P = .02), with the lowest levels in GY2 (68.3%) and the highest levels in GY5 (76.1%) residents.
Concerns that attending physicians would think worse of them if they asked for help when they did not know how to do a procedure also varied significantly (P < .001), with lowest levels among GY1 (11.6%) and highest levels among GY4 (18.2%) and GY5 (17.0%) residents. Worries about hurting patients also differed significantly (P = .003) across training years, with 68.3% of GY1 residents reporting this worry; lowest levels were among GY4 residents (59.2%). Views about the need to complete specialty training differed across years (P < .001), with the least concern among GY5 residents (58.4%) and the greatest concern among GY2 (67.3%) and GY3 (67.3%) residents.
This study characterized general surgical trainees from 248 US residency training programs in terms of demographics, attitudes, training experiences, and professional expectations. It adds to the literature in several ways. First, we extend previous efforts to describe views and experiences of surgical residents by characterizing dimensions of lifestyle issues in depth and by assessing multiple strains, concerns, and formal and informal supports. These include strains in relationships with attendings, the influence of the practice of surgery on emotional health, performance anxieties, and concerns about their professional futures including financial compensation. There are notable apprehensions and unmet needs among residents of both sexes and across training years. Only 75% of male residents and 66% female residents reported feeling that they can turn to program support structures or faculty during difficult times in their training, suggesting the importance of improving outreach for certain residents. Furthermore, residents reported feeling vulnerable at multiple levels, including a lack of confidence to perform independently upon completion of training, fears about hurting patients, and concerns that other professionals will expand their scope of procedures into general surgery.
Second, our findings validate common perceptions that postgraduate years 2 and 3 are a difficult time during the training trajectory, as many of these residents are preparing to complete research years, with residents in these years expressing the lowest levels of satisfaction, greatest likelihood to have considered leaving, and the greatest worries about their ability to perform procedures upon completion of training. An important caution is that changes in reported levels of satisfaction over time are multifactorial. Because residents who choose to leave training each year are likely to have been unhappy with the experience, those remaining in the later years may be subject to self-selection in terms of satisfaction. The differences in views and attitudes reported by residents early in their postgraduate training compared with residents further along in training suggest that retention strategies may vary by postgraduate year. In cases in which a resident is not well suited for surgery, early attrition is appropriate and prevents waste of resources by the resident or program. However, our results suggest that residency programs may need to tailor supportive structures to meet the changing needs and expectations of surgical residents as they progress through residency training.
Third, we found that attitudes and experiences vary significantly by sex. As the numbers of women entering the medical profession35 and surgical training36 continue to increase, it will be important to improve understanding the degree to which their training needs or expectations differ from men and how best to attract women to pursue surgical careers and to support those who do enter residency. One caution is the risk of stereotyping by these characteristics (for instance, that all women contemplate leaving residency) because residency training is both an individual and a community experience. However, these findings may be of value to surgical residency training program directors (who face increasingly complex challenges and limited resources) in developing and prioritizing strategies to address the needs of specific cohorts and the potentially distinct experiences of female residents.
Our findings are consistent with prior smaller studies that demonstrate surgical residents' generally high degree of satisfaction with training13,14 and professional relationships,14 although a substantial minority of residents in this study reported feeling dissatisfied. We also found positive perceptions of and interest in specialty training,6 potentially different motivations and training experiences for women compared with men,20 increasing interest in controllable lifestyles,22 and a notable proportion (14%) who considered leaving training.2,3,7- 9,37
The study design has several strengths that support the validity of the findings and suggest the potential for future analyses. First, while there have been national surveys of residency program directors in surgery,3,38,39 to our knowledge very few have gathered primary data from a national sample of residents, and these were restricted to graduating chief residents,6,14 or examined demographics and specialty plans without distinguishing among preliminary and categorical residents.37 The scope of our study, including all categorical surgery residents in all 5 training years, with its high response rate and representation from all but 1 of the 249 training programs, offers a comprehensive national picture of surgical residency training.
Second, we used rigorous qualitative methods to inform the survey development. Residents described aspects of experience associated with dissatisfaction and attrition that are distinct from those previously identified, including working conditions,13,14 the 80-hour work week,15- 18 lifestyle,7 and motivations and deterrents to surgery19,21 as a career.
Third, although we report cross-sectional data, the overall study is longitudinal and prospective in design, with 2 years of data collection completed and a final wave planned for 2010. This design will permit tracking the changing views of residents as they progress through training, in particular the apparently vulnerable second and third years, and to identify the factors associated with the decision to leave surgical residency. Future analyses should explore the nature of causal relationships among these concepts using longitudinal data and multivariable analyses.
Fourth, experts have called for more comprehensive understanding of correlates and predictors of increasing fellowship rates, which has important implications for reengineering of surgical education and broader workforce planning.7,40 Future linkage with both attrition and fellowship data from the American Board of Surgery could provide a rich database to examine these critical issues in greater depth.
Findings should be considered in light of potential study limitations. The sensitive nature of the topics explored may have led to response bias, including socially desirable response set41,42 and underreporting of negative opinions for fear of reprisal. To minimize these potential biases we provided detailed assurances of confidentiality, pilot tested and refined the instrument, and used balanced keying in item stems. Because we collected data from residents in 248 US training programs, with an 82.4% response rate, potential bias associated with geographic region or training program is unlikely.
Administration of the survey following the demanding ABSITE examination may have influenced residents experiencing fatigue or emotional responses to the examination. We believed that immediately following the ABSITE was the optimal strategy to maximize participation of this typically difficult-to-reach population. In addition, we considered it critical to minimize direct involvement of programs in recruiting participants and conducting follow-up of nonrespondents to ensure confidentiality of the data and enhance participation. Distributing the survey at the ABSITE location permitted us to reach all surgical residents in a single, coordinated administration wave. This approach is consistent with surveys that are conducted at the conclusion of the Internal Medicine In-Training Examination administered by the American College of Physicians, Association of Program Directors in Internal Medicine, and Association of Professors of Medicine. We also believed that providing the survey concurrent with this examination might generate thoughtful responses, because residents may be primed to think about their professional goals. While postexamination fatigue is a legitimate concern, it seemed less influential than giving the survey at a point when respondents knew their results, potentially biasing their perception of their program.
This baseline descriptive study profiles the current population of US surgical residents and presents an overview of illustrative findings to describe resident views toward complex, interrelated aspects of residency training, including satisfaction with didactic and operative training, faculty and peer relationships, strains and concerns regarding the training experience, motivations for pursuing surgical careers, and interest in surgical specialization. It may help inform efforts to respond to the complex pressures facing the surgical profession, including the ability to attract and retain increasingly diverse general surgery residents43,44 and the projected shortage of general surgeons in the coming decades.5
Corresponding Author: Leslie A Curry, PhD, MPH Health Policy and Administration, Yale School of Public Health, 60 College St, New Haven, CT 06520 (firstname.lastname@example.org).
Author Contributions: Dr Yeo had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Yeo, Berg, Nunez-Smith, Cammann, Bell, Sosa, Curry.
Acquisition of data: Yeo, Bell.
Analysis and interpretation of data: Yeo, Viola, Berg, Lin, Nunez-Smith, Cammann, Sosa, Krumholz, Curry.
Drafting of the manuscript: Yeo, Viola, Lin, Curry.
Critical revision of the manuscript for important intellectual content: Yeo, Viola, Berg, Lin, Nunez-Smith, Cammann, Bell, Sosa, Krumholz, Curry.
Statistical analysis: Yeo, Lin, Nunez-Smith, Cammann, Curry.
Obtained funding: Krumholz.
Administrative, technical, or material support: Nunez-Smith, Bell.
Study supervision: Berg, Nunez-Smith, Sosa.
Financial Disclosures: None reported.
Funding/Support: This study was supported by The Robert Wood Johnson Clinical Scholars Program at the Yale School of Medicine.
Role of the Sponsor: The Robert Wood Johnson Foundation had no substantive involvement in any aspect of the study including the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation of the manuscript. The American Board of Surgery assisted with survey implementation (printed the surveys and sent them to examination sites together with the ABSITE examinations) and with design of the study and review of the manuscript (by co-author Dr Bell).
Additional Contributions: Emily Buckholz, MPH, Yale School of Medicine, assisted with data cleaning and management. She received compensation for this work. Marjorie Rosenthal, MD, MPH, Yale School of Medicine, participated in the analysis of the qualitative component of the study; she did not receive compensation for this role.
This article was corrected online for typographical errors on 10/15/2009.