Predictors of Research Productivity in General Surgery Residency and Beyond: A Single Institutional Analysis

Abstract

Background/Objectives: Research is widely recognized as an essential element of medical education across all levels. This study explores the association between publishing before, during, and after general surgery residency, shedding light on research productivity trends in medical education. By examining a diverse cohort of general surgery residents, the research aims to uncover factors influencing research output at various training levels. Methods: This study analyzed graduates’ Curriculum Vitae from a single institution’s general surgery residency program from 2017–2024. Linear regression models assessed whether variables encompassing total research output (manuscripts, abstracts, posters, and podium presentations) could predict graduates’ research outcomes in the next stage of training. Results: Out of the 38 eligible graduates, 20 submitted their Curriculum Vitae and were included in the study, resulting in a response rate of 52.6%. Total research activity before residency was not found to be predictive of total research activity in residency (t(18) = 1.020, p = 0.322) but was found to be related to total research activity in fellowship (t(10) = 5.399, p < 0.005). Total research activity during residency was not found to be related to fellowship research (t(10) = 0.834, p = 0.424). No sequential correlation was found between undergraduate, medical school, residency, or fellowship publications. Additionally, there was no correlation between first authorship and publication total in the subsequent phase of medical education. Conclusions: Research productivity during medical school does not have predictive value for future research during residency but is found to be predictive of future works in fellowship. These findings suggest that the research emphasis should be strategically managed throughout medical education to sustain long-term academic engagement.

1. Introduction

Research has long been acknowledged as a vital element of medical education [1]. Scientific research plays an integral role in advancing knowledge and enhancing patient outcomes. Scholarly research during undergraduate medical education positively impacts the student and eventual surgeon’s likelihood to transition to fellowship and become an academic surgeon [2,3]. In 2008, over 90% of general surgery program directors stated that they consider both basic and clinical research “almost always or all the time” when evaluating candidates [4] and that research experiences serve as a frequent discussion point during residency interviews and the rank process [5]. In fact, previous studies have shown that medical students matching into general surgery residency programs have a significantly greater number of research accomplishments when compared to unmatched applicants [6]. In the 2022 Match, U.S. M.D. seniors reported an average of 4.0 research experiences [7]. The number of research experiences reported on the Electronic Residency Application Service (ERAS) has a positive correlation with successfully matching across all surgical subspecialties and the eventual pursuit of fellowship [8,9]. Recent updates to medical student education, including a pass/fail reporting system for the United States Medical Licensing Examination (USMLE) Step 1 and the implementation of dedicated research rotations at many medical education programs [10], have seemingly heightened the importance of tangible research for the prospective surgery resident.

While it is generally accepted that publishing as a medical student and/or resident helps demonstrate professional proficiency, contrasting views continue when comparing publication metrics to surgical or clinical skills [11,12]. Previous manuscripts have demonstrated a correlation between medical student or resident publications to subsequent career publications in fields other than general surgery (neurology [13], radiology [14], and urology [8]). Several systematic reviews have investigated the association of previous research performance [15], years of training [16], and gender [16] with research productivity; however, there is a scarcity of studies that comprehensively follow individuals from medical school through to fellowship within the field of general surgery. This longitudinal perspective is crucial to understanding the continuity and evolution of research productivity across different stages of medical education and training.

The present study aims to identify whether an association exists between publishing before, during, and after general surgery residency. Secondly, our study aims to evaluate the trend of research productivity among medical students, residents, fellows, and attending physicians over the course of their medical education and early careers. By examining the research productivity of a large cohort of general surgery residents with unique backgrounds and varying academic experiences, we aim to provide a comprehensive understanding of the elements that foster research output among surgical trainees at all levels of their medical education. Such insights will contribute to the existing literature and facilitate the development of educational initiatives to optimize and understand the current state of research training within general surgery residency programs.

2. Materials and Methods

2.1. Participants and Data Collection

This study was reviewed and approved by the Institutional Review Board of Tulane University School of Medicine (protocol number: 2022-1320, approval date: 16 August 2022). Informed consent was obtained from all participants prior to their inclusion in the study, ensuring that they were fully aware of the study’s purpose and procedures. Additionally, all data were anonymized to maintain participant confidentiality and protect their privacy. The present study consisted of recent graduates and the current cohort of fifth-year residents from a single general surgery residency program over a seven-year period (2017–2024). A request for Curriculum Vitae (CV) was sent in accordance with the Association of Program Directors in Surgery (APDS) protocol for surveys. CV was used as the gold standard for the list of total research activity (consisting of either first or co-authored publications, oral and poster presentations, and published abstracts) as well as basic demographics (such as date of birth, gender, race, country of origin, career timelines (such as undergraduate institution, medical institution, fellowship institution), and secondary degree or gap years). It was assumed as such that graduates accurately updated their CVs in anticipation of fellowship or subsequent career interviews. A single research project was eligible to be included as an abstract, podium presentation, and publication in total research experience counts.

2.2. Publication Categorization

General surgery residency programs in the United States conclude annually at the end of June. For the purposes of this study, research output was classified based on the academic year, defined as the period from 1 July to 30 June of the following year. This classification method was chosen to best reflect the cyclical nature of academic activities in general surgery residency programs. Graduation dates were obtained from the individual’s CV, accounting for any time away from residency. Research output was assigned to specific years as indicated in the CV and categorized according to distinct training periods. Research conducted during gap years—defined as periods outside formal medical school, residency, or fellowship programs—was documented and analyzed separately.

2.3. Statistical Analysis

Spearman correlation tests (S) were used to examine the correlations between research output and graduation years, as well as the correlations between research productivity measures. Linear regression was performed to determine predictors of total publications and research output. The statistical analysis and figure production were conducted using RStudio v3.6.0. 2023/2024 graduates were excluded from both correlations and regressions pertaining to fellowship. The p-value is lower than the reporting threshold of the software; it is formatted as ‘p < 0.005’.

3. Results

Out of the 38 eligible graduates, 20 submitted their CVs and were included in the study, which equates to a response rate of 52.6%.

3.1. Total Research Activity

The total research activity before residency (undergraduate and medical school) was not found to be predictive of residency total research activity (t(18) = 1.020, p = 0.322) but was found to be predictive of fellowship total research activity (t(10) = 5.399, p < 0.005). Total research activity during residency was not found to be related to fellowship research (t(10) = 0.834, p = 0.424). Moreover, the total research activity count before residency was not found to be predictive of total research activity during residency (t(18) = 0.211, p = 0.835) but found to be predictive of total research activity during fellowship (t(10) = 5.399, p < 0.005). Additionally, publications during residency were found to be unrelated to fellowship total research activity (t(0) = 0.391, p = 0.704).

3.2. Poster and Oral Presentations

Poster and oral presentations before residency were not found to be predictive of residency total research activity (t(18) = 1.379, p = 0.38531), as well as not predictive of fellowship total research activity (t(18) = 0.890, p = 0.385). Poster and oral presentations during residency were found to be unrelated to fellowship total research activity (t(7) = 0.596, p = 0.564). When separated, the number of poster presentations in residency (t(7) = −1.940, p = 0.0936) was not found to be related to fellowship publications, but the number of podium presentations (t(7) = 2.729, p = 0.0294) was.

These findings are summarized in

Table 1. Predictors of total research activity.

		t-Value	p-Value	Degrees of Freedom
Total research activity before residency predictive of total research activity in…	Residency	1.020	0.322	18
	Fellowship	5.399	<0.005	10
Total research during residency is predictive of total research activity in fellowship		0.834	0.424	10
Publications before residency is predictive of total research activity in…	Residency	0.211	0.838	18
	Fellowship	5.399	<0.005	10
Publications during residency predictive of total research activity in fellowship		0.391	0.705	10
Poster and oral presentations before residency predictive of total research activity in…	Residency	1.379	0.38531	18
	Fellowship	0.890	0.3385	18
Poster and oral presentations during residency predictive of fellowship research		0.596	0.564	7

Gold highlight indicates significance.

. Interestingly, a positive correlation was observed between the total amount of basic science research a resident participated in and the total clinical research (S = 262.97, p = 0.042).

3.3. Publication Rates

There was no significant correlation between undergraduate publications and medical school publications (S = 649.81, p = 0.182). Likewise, no correlation was found between medical school publications and residency publications (S = 1041.1, p = 0.358), nor between residency publications and fellowship publications (S = 142.11, p = 0.096). When assessed together, the total amount of publications published before residency did not correlate with residency publications (S = 778.34, p = 0.434). However, the combination of undergraduate, medical school, and residency total publications did correlate with total publications in fellowship (S = 103.85, p = 0.026).

3.4. First-Authorship

There was a significant correlation between the number of first-authorship publications in medical school and the total number of medical school publications (S = 605.33, p = 0.013). However, no significant correlations were found between the number of first-authorship publications in medical school and the total number of residency publications (S = 1205, p = 0.694) or fellowship publications (S = 266.89, p = 0.8365). On the other hand, there was a significant correlation between the number of first-authorship publications in residency and the total number of residency publications (S = 194.52, p < 0.005). However, no significant correlation was found between the number of first-authorship publications in residency and the total number of publications in fellowship (S = 250.44, p = 0.700). Overall, the total number of first authorships in both medical school and residency was not found to be correlated with the total number of publications in fellowship (S = 248.92, p = 0.668).

3.5. Research Years and Fellowship Pursued

Out of the 20 residents in the present cohort, 7 (35%) took at least one research year during residency. There was no significant correlation between research years taken either in medical school or residency and the total number of publications during their medical education (S = 133.46, p = 0.074). Interestingly, the most common field pursued via fellowship in individuals taking a research year were three residents who pursued plastic surgery, followed by one each in transplant, colorectal, surgical oncology, minimally invasive, and endocrine surgery. Interestingly, the most pursued fellowship from our cohort was Trauma and Acute care, for which no resident took a research year. The full list of fellowships pursued in the cohort can be found in

Table 2. Research activity by specialty.

Fellowship Pursued	Number of Entrants
General Surgery Private Practice	2
Transplant	1
Cardiothoracic	1
Trauma and Critical Care	6
Colon and Rectal	1
Plastic Surgery	3
Surgical Oncology	1
Minimally Invasive Surgery	3
Endocrine Surgery	2

.

3.6. Overall Publication Trends

Between the 2017 and 2023 graduating classes, publication rates before residency showed a slight increase (R2 = 0.323, p = 0.050), whereas publication rates during (R2 = 0.013, p = 0.634) and after residency (R2 = 0.017, p = 0.684) stagnated. Figure 1, Figure 2 and Figure 3 highlight these trends in mean and median publication rates over time.

4. Discussion

Research holds significant importance in both general surgery residency and fellowship applications, with an evident increase in undergraduate and postgraduate research participation in recent years [17]. The current study explores the relationship between pre-residency scholarly research activities—such as publications, poster presentations, oral presentations, and abstracts—and research involvement during both residency and fellowship. Our findings indicate that general surgery residents who engaged in research prior to their residency were not more inclined to participate in research during their residency. However, these same individuals were more likely to continue research endeavors during their fellowship years.

This outcome appears consistent with the growing emphasis on research in residency programs, particularly given the recent shift to a pass/fail grading system for the Step 1 exam. Faculty, attending physicians, and current residents are increasingly emphasizing the value of research, likely influencing medical students to produce more scholarly work before entering residency. Despite this push, our data show that an uptick in pre-residency publications does not necessarily translate into increased research activity during the residency period. Interestingly, when assessing the publication count alone, no correlation was found between the likelihood of publishing during undergraduate to medical school, medical school to residency, or residency to fellowship. These results corroborate previous studies suggesting that undergraduate medical students may be increasing their research output during medical school in pursuit of research without a clear understanding of the optimal number of publications to aim for or the impact of their work [18].

General surgery residency stands out from other graduate medical education programs, as a significant number of residents, approximately 36%, interrupt their training to pursue full-time research [18]. Our findings align with this trend, as 35% of the cohort in question took at least one research year during their training. It should be noted that the institution in which this study was conducted only requires one research experience prior to graduation. However, as previously mentioned, these research experiences did not predict a continuation of scholarly activity into fellowship. One possible explanation for this lack of continuity in research between residency and fellowship could be the time constraints that residents and fellows face in dedicating themselves to research activities [19]. To explore this possibility, we assessed the correlation between first authorship publications, used as a proxy to evaluate time commitment and subsequent research output. Surprisingly, we found that the number of first authorships a resident published during medical school was not a strong predictor of their total research output in the later stages of medical education. The same lack of correlation was observed for first authorship in residency and total output in fellowship. These findings raise questions about why residents and fellows who previously took the lead in research initiatives either lost interest in research or stopped pursuing it as they progressed through higher medical education.

Our study suggests that traditional markers of research engagement and productivity may not significantly influence a candidate’s trajectory through residency or fellowship. This highlights that the motivation and opportunity to engage in research can evolve due to various pressures as medical students, residents, and fellows advance in their careers. The current model may inadvertently emphasize first-author publications as a “golden ticket” for residency applications without fostering a long-term commitment to academic research. This emphasis on first authorship may lead residents to focus solely on first-authorship publications to be competitive applicants. Research productivity has been linked to continued academic pursuits and an increased likelihood of pursuing fellowship training after residency [20]. Johnson et al. suggest that research is essential for applicants to top-quartile academic research programs for residency [21]. However, this emphasis on research has also contributed to the growing problem of application inflation for the general surgery match [22].

The phenomenon of “burnout”, characterized by exhaustion due to chronic exposure to stressors, is well-documented among both practicing physicians and those in training. This may account for the stagnation in research productivity found in our cohort during residency, which typically lasts five years and is known for its rigorous workload. Ishiguro et al. found that residents who pursue scholarly work do not publish the majority of their research output during these years [23]. Our findings do not support a linear relationship between the number of papers published before residency and those published during residency or fellowship. Engaging in research prior to residency does not enhance the likelihood of publishing scholarly work after residency begins. Other variables are likely involved, and further research is needed to identify and account for these factors. Additionally, not all residency candidates have research experience, and there is no conclusive evidence that publishing as a medical student or resident inherently improves clinical skills. Interestingly, 60% of respondents in our study had only one or zero publications before starting residency. The pre-residency publication status cannot definitively predict a resident’s potential for future publications, as all individuals in our cohort exceeded the residency program’s requirement of one publication (IQR: 1–32).

Future research should aim to better understand the reasons for declining research productivity among aspiring surgeons as they progress through medical education and strive for more comprehensive and transparent statistics regarding research involvement during general surgery residency and fellowship applications. This could include disclosing hours invested and leadership roles on projects through Electronic Residency Application Service (ERAS) and considering unpublished projects for evaluation. Emphasizing research for medical students applying for general surgery residencies should be approached cautiously to avoid application inflation and to ensure a holistic assessment, considering clinical skills, teamwork, communication, and commitment to patient care. Additionally, further investigation is needed to explore factors influencing general surgery residents’ research interest, such as funding availability, mentorship, career goals, and the balance between research and clinical duties, alongside assessing institutional support and workload’s effect on research participation to foster research-friendly environments.

This study has several limitations. First, it is based on data from a single residency program, which may not reflect the emphasis on academic productivity in other programs. This limits the generalizability of the findings. Secondly, the low response rate (52.6%) introduces potential bias, making it difficult to draw definitive conclusions from the limited data set. Additionally, relying solely on publication dates may not accurately reflect the actual timeline of research activities, potentially misclassifying publications and skewing results. The accuracy of the study also depends on the reliability of the respondents’ CVs, which were not cross-referenced with online databases, introducing the possibility of over- or under-reporting. Lastly, this study is limited to a single institution, which may not fully represent general trends across different residency programs. Future studies involving multiple institutions will be necessary to confirm and expand upon these findings.

5. Conclusions

This institution-specific analysis suggests that pre-residency research productivity does not reliably predict research engagement during residency. However, it appears to be a stronger indicator of continued scholarly involvement during the fellowship. While academic programs often value prior research experience as a marker of scholarly potential, our findings suggest that research activity during residency or designated research years does not necessarily predict an academic career. This highlights the importance of evaluating early research accomplishments alongside other aspects of a candidate’s professional trajectory and refining the criteria used to assess academic potential in medical training.