Early Engagement,Lasting Impact: Predictors of Scholarly Activity in Orthopaedic Sports Medicine Fellowship Applicants

Methods

A retrospective study of orthopaedic sports medicine fellowship applicants to Cleveland Clinic was conducted from 2011 to 2013. Publication counts were determined using 2 separate PubMed queries for each applicant. The first query captured publications indexed before the date of the fellowship application to reflect research output at the time of application. The second query captured all publications indexed through the most recent follow-up at the time of data collection to reflect cumulative early career research productivity. The term “early career” in this study refers to practice after the fellowship. Applicants with no identifiable research publications were assigned a total publication count of zero. Applicants who had reapplied were only counted once to avoid duplication. ²²

To determine the current board certification status of each applicant, we used the ABOS (American Board of Orthopaedic Surgery) “Verify Surgeon” tool. The sex and type of medical degree for each applicant were obtained from the original applicant files. Current membership in the American Academy of Orthopaedic Surgeons (AAOS) was determined using the organization's “Find an Orthopaedist” tool. All collected data were entered into the REDCap software platform for secure data management and analysis. ¹⁶

Continuous variables were summarized using the median and interquartile range (IQR), while categorical variables were summarized using counts and percentages. To analyze the association between publications at the time of application and current publications (ie, at the time of follow-up), we employed negative binomial regression because of the count nature of current publication data and to address overdispersion. This regression model is appropriate for overdispersed count data in which the conditional variance exceeds the conditional mean. It is considered a generalization of Poisson regression, as it has the same mean structure but includes an additional parameter to model overdispersion. Incidence rate ratios (IRRs) and their respective 95% confidence intervals (CIs) were estimated. Analysis was conducted in R, utilizing a 2-sided test with a significance level of 0.05.

Results

A total of 305 orthopaedic sports medicine fellowship applicants were included in our analysis. Most applicants held a Doctor of Medicine (MD) degree (89.8%), were United States (US) medical school graduates (83.6%), and applied across 3 cycles: 2011 (34.4%), 2012 (33.4%), and 2013 (32.1%). At the time of application, the median number of publications was 1.00 (IQR, 0.00-2.00). By the time of follow-up, the median increased to 4.00 (IQR, 2.00-12.00) (Table 1).

OPEN IN VIEWER

Table 1

Characteristics of Fellowship Applicants ^a

	Value (n = 305)
Sex
Female	16 (5.2)
Male	289 (94.8)
Degree type
MD	274 (89.8)
DO	31 (10.2)
Medical school location
US medical graduate	255 (83.6)
International medical graduate	41 (13.4)
US medical graduate–Canada	9 (3.0)
ABOS certification	219 (71.8)
AAOS membership	152 (49.8)
Application year
2011	105 (34.4)
2012	102 (33.4)
2013	98 (32.1)
No. of publications at time of application (n = 284)	1.00 (0.00-2.00)
No. of publications at follow-up (n = 284)	4.00 (2.00-12.00)

a

Data are presented as median (interquartile range) or n (%). AAOS, American Academy of Orthopaedic Surgeons; ABOS, American Board of Orthopaedic Surgery; DO, Doctor of Osteopathic Medicine; MD, Doctor of Medicine; US, United States.

At the time of application, 42.5% of applicants had no publications. This rate dropped to 8.4% at the time of follow-up (Figure 1). Over half of applicants (53.3%) had <5 publications at follow-up, and 73.2% had <10 publications. These trends demonstrated a marked increase in career publications after a fellowship.

OPEN IN VIEWER

Figure 1.

Bar chart showing the percentage of orthopaedic sports medicine fellowship applicants with varying numbers of publications at the time of application (blue bar) versus follow-up.

Gender-based comparisons revealed persistent disparities. At the time of application, 62.5% of female applicants had no publications compared to only 25.0% of male applicants. At follow-up, 25.0% of women still had no publications compared to 7.4% of men. Similarly, 68% of women had <5 current publications versus 52% of men. Only 12.5% of women had >10 publications compared to 28.0% of men (Figure 2). The mean publication output further highlighted these differences, with women averaging 5.75 current publications versus 12.13 for men (Figure 3).

OPEN IN VIEWER

Figure 2.

Bar chart comparing the publication distribution between male and female orthopaedic sports medicine fellowship applicants at the time of application and follow-up.

OPEN IN VIEWER

Figure 3.

Bar chart displaying mean current publication counts stratified by sex among orthopaedic sports medicine fellowship applicants.

Regression analysis using a negative binomial model identified several statistically significant predictors of the current publication count (Table 2). Publication count at the time of application was strongly associated with higher future productivity (IRR, 1.23 [95% CI, 1.18-1.30]; P < .001). Applicants with a Doctor of Osteopathic Medicine (DO) degree had significantly fewer current publications than their MD counterparts (IRR, 0.52 [95% CI, 0.32-0.89]; P = .014). Conversely, international medical graduates had higher research productivity than US medical graduates (IRR, 1.65 [95% CI, 1.10-2.54]; P = .019). Membership in the AAOS was also positively associated with publication count (IRR, 1.54 [95% CI, 1.15-2.04]; P = .003). Sex was not significantly associated in the multivariable model (IRR, 1.23 [95% CI, 0.60-2.31]; P = .538). The model explained 47.8% of the variance (Nagelkerke R² = 0.478).

OPEN IN VIEWER

Table 2

Predictors of Current Publication Productivity ^a

	IRR (95% CI)	P Value
Degree type (DO vs MD)	0.52 (0.32-0.89)	.014
Medical school location (international vs US medical graduate)	1.65 (1.10-2.54)	.019
Sex (male vs female)	1.23 (0.60-2.31)	.538
AAOS membership (yes vs no)	1.54 (1.15-2.04)	.003
Previous publication	1.23 (1.18-1.30)	<.001

a

AAOS, American Academy of Orthopaedic Surgeons; DO, Doctor of Osteopathic Medicine; IRR, incidence rate ratio; MD, Doctor of Medicine; US, United States.

Discussion

The principal finding of this study was that there was a strong correlation between publication count at the time of the sports medicine fellowship application and future early career research output; specifically, each additional publication at the time of application increased future research output by 23%. Our findings suggest that early engagement in research, even before the start of fellowship training, can have a profound impact on one's trajectory as an academic surgeon. By fostering a culture of inquiry and providing opportunities for scholarly activity during medical school and residency, training programs can help cultivate the next generation of orthopaedic surgeon-scientists. Additionally, our results highlight the importance of involvement in professional societies such as the AAOS, which may provide networking opportunities, mentorship, and resources to support ongoing research endeavors. We believe that current requirements for scholarly activity are flexible and may allow programs to tailor research expectations to their institutional strengths. However, this variability may result in unequal research exposure across training programs. Rather than increasing formal requirements, a more effective approach may involve standardizing access to mentorship, protected research time, and institutional research support early in training. Such efforts could help reduce disparities in research productivity while preserving program-level autonomy.

Our results are consistent with prior studies demonstrating the importance of research experience during residency and a fellowship for future academic success. We found that for each additional publication at the time of application, there was a 23% increase in the rate of current publications. Acevedo et al ¹ found that orthopaedic surgeons who were more productive in research during their training tended to continue publishing at higher rates as attendings. They observed that overall, surgeons produced more research after training than during, but there was a strong correlation between publishing during and after residency. This aligns with the findings of Carr et al, ⁷ who showed that orthopaedic residents with a higher total number of publications, especially first-author publications, were more likely to pursue academic careers than those with lower research output. Cvetanovich et al ¹⁰ analyzed the research productivity of sports medicine fellowship faculty and found that faculty employed at academic centers had significantly higher publication metrics compared to those not at academic centers. Rompala et al ²¹ focused specifically on shoulder and elbow surgeons and determined that research output before and during training, particularly during a fellowship, was predictive of academic productivity as an attending. Similarly, Clark et al ⁹ surveyed sports medicine fellowship directors and discovered that most programs require fellows to complete research projects and that the presence of dedicated research assistants was associated with higher average publication rates for fellows. DeFroda et al ¹¹ looked at research trends among residents applying for sports medicine fellowships, noting an increase in publication quantity and quality in recent years, likely driven by the competitive nature of the subspecialty. Collectively, these studies underscore the value of early and sustained engagement in research throughout surgical training. By providing opportunities for scholarly activity and cultivating an interest in academic pursuits, residency and fellowship programs can help shape the next generation of surgeon-scientists who will advance the field of orthopaedics. Aspiring surgeons should recognize the importance of research and actively seek out mentorship and experiences to develop their skills in this domain, as it may have a profound impact on their future career trajectory.

Our results highlight persistent gender disparities in research productivity within orthopaedic sports medicine, findings that reflect broader systemic challenges affecting women throughout the specialty. Despite increasing efforts to enhance female representation, women remain significantly underrepresented in orthopaedic surgery overall, comprising only approximately 6% to 8% of practicing orthopaedic surgeons and 19% of current residents, with sports medicine ranking among the subspecialties with lower female representation. ⁸ Research by Leal et al ¹⁹ demonstrated that the prevalence of female orthopaedic sports medicine specialists increased from only 5% in 2010 to 6% in 2019, indicating a compound annual growth rate of merely 0.2%, suggesting that at current rates, achieving gender parity could take over 2 centuries. While our study found that sex was not significantly associated with publication count in the multivariable model, the raw data revealed substantial differences in mean publication output between men and women, with women averaging 5.75 publications compared to 12.13 for men. These productivity disparities may reflect multiple interconnected factors including differential access to research opportunities, mentorship availability, and institutional support systems. The underrepresentation extends beyond research metrics to encompass leadership positions in which women hold <4% of fellowship director positions and have minimal representation at division chief levels in sports medicine according to recent analyses. ¹⁹ Encouragingly, recent data from Harris et al ¹⁵ suggest that female match rates into orthopaedic surgery residency have shown improvement over the past decade, with the adjusted odds ratio for female and underrepresented minority applicants increasing significantly from 2011 to 2021, indicating that targeted diversity initiatives may be yielding positive results. However, White et al ²⁴ found that while female applicants’ match rates were reflective of their application rates overall, persistent disparities remain in subspecialty representation. The intersection of research productivity expectations with ongoing gender disparities in academic advancement suggests that while women are increasingly entering the field, they may still face barriers to achieving the publication metrics traditionally associated with career advancement in academic orthopaedic sports medicine. Furthermore, Anand et al ³ demonstrated that certain subspecialties such as hand surgery have seen significant growth in female fellowship matches, while others including spine, trauma, and sports medicine have shown no significant change in female representation over recent years. ²⁴ These findings underscore the critical need for comprehensive interventions that address not only research opportunities but also mentorship programs, leadership development, and institutional cultures that support equitable career advancement for women throughout their training and early career phases.

We also found that applicants with DO degrees had lower research productivity compared to those with MD degrees. This difference may reflect variations in research opportunities and support across medical school curricula. ² As osteopathic and allopathic residency programs continue to integrate under a single accreditation system, it will be important to ensure that all trainees have access to robust research infrastructure and mentorship. Programs such as research fellowships and dedicated research tracks may help level the playing field and provide targeted support for residents interested in pursuing academic careers. ¹² Although this study reflects applicants from 2011 to 2013, broader trends in orthopaedic training since that time, including increased female representation, integration of osteopathic graduates, and rising publication expectations for fellowship applicants, may have improved but still remain highly relevant. While applicant characteristics have evolved, disparities in access to research mentorship and infrastructure persist. The associations observed in this cohort reinforce the importance of continued advocacy and institutional efforts to provide equitable research opportunities across all training pathways.

The higher publication counts observed among international medical graduates likely reflect a strong selection effect. International medical graduates who successfully match into US orthopaedic residency and fellowship programs often demonstrate exceptional academic credentials including substantial research productivity, which may be necessary to overcome structural barriers related to training pathway differences, visa considerations, and program familiarity. Prior studies have similarly shown that international medical graduates who match into competitive surgical specialties tend to exhibit higher research output compared with their US-trained counterparts, suggesting that scholarly productivity serves as a key differentiator within this applicant population. ²⁰

Our study has several limitations that should be considered when interpreting the results. First, we focused on a single institution's fellowship applicants over a limited time period, which may limit the generalizability of our findings. Second, our follow-up publication counts may include studies initiated during late residency or fellowships that were published after the completion of training. As a result, follow-up publication totals may not exclusively reflect independent research productivity after training. Accordingly, our findings should be interpreted as predictors of early career publication trajectory rather than purely academic activity after a fellowship. Third, we relied on publicly available data sources to ascertain publication counts and other variables, which may be subject to errors or omissions. Fourth, we did not assess the quality or impact of publications, which may be an important factor in evaluating research productivity. ⁶ Finally, our regression model explained less than half of the variance in publication counts, suggesting that other unmeasured factors may play a significant role in predicting research output. Although sex was not independently associated with publication count on multivariable analysis, this should not be interpreted as an absence of gender disparity. Substantial unadjusted differences were observed, with women averaging less than half the publication output of men, likely reflecting persistent systemic factors such as disparities in mentorship, research infrastructure, and institutional support. Additionally, the small number of female applicants limits statistical power to detect sex as an independent predictor, despite meaningful differences in productivity. Future studies should explore how early research productivity relates to eventual career trajectory, including the likelihood of pursuing academic versus predominantly private practice. Additionally, evaluating the impact of formal research mentorship programs, protected research time, and institutional research infrastructure during residency and fellowship training may provide important insights into strategies that promote sustained scholarly productivity in the long term.

Conclusion

Our study demonstrated that early research engagement resulted in a trajectory toward sustained scholarly activity in orthopaedic sports medicine. Each additional publication at the time of the fellowship application increased future research output by 23%, while AAOS membership provided an additional 54% boost in productivity. Significant disparities persisted, with DO graduates producing 48% fewer publications than MD graduates and women averaging half the output of men, highlighting the need for equitable research opportunities across all training pathways.