Key points are not available for this paper at this time.
United States medical school seniors who matched into neurosurgery residency programs in 2024 reported an average of 37 research products (abstracts, presentations, and publications) on their residency applications.1 Equally shocking is the number of such products among US seniors who sought a neurosurgery match but were unsuccessful: 31. This “arms race” for more and more research products by medical students year over year does not seem to be losing steam.2 While the most extreme numbers of research products are seen among applicants in very competitive fields, this trend is seen across specialties including primary care–focused fields such as family medicine, internal medicine, and pediatrics (Figure).2,3The increasing volume of research by residency applicants may have escalated since the United States Medical Licensing Examination (USMLE) Step 1 transitioned to pass/fail. Data from the National Resident Matching Program (NRMP) in 2024 (the first students to experience the pass/fail USMLE Step 1) show increased growth of research product metrics among matched applicants to highly competitive specialties (Figure).3 A 2025 study of orthopedic surgery applicants predicts exponential growth of research products over the coming decade.4 Both survey-based research and student-authored opinion pieces articulate concern that research productivity is an increasingly important criterion in residency selection.5-7 In our own conversations with medical students regarding research opportunities, there is greater urgency to identify research opportunities earlier in medical school. Absent a numeric Step 1 score, the number of research products remains one metric by which applications might be sorted. With the need to distinguish themselves from their peers, students are driven to meet a perceived, ever-rising “minimum bar.”8As directors of medical school research programs, we want students to participate in research. In fact, all 3 of our institutions require such participation in the belief that all students can benefit from the associated development of critical thinking and self-directed learning skills. We also embrace student engagement in research to support the physician-scientist workforce, which is critical to the nation. However, we share deep concerns that the current state of affairs is counterproductive to achieving these goals. In our experience, students reaching to meet or exceed the most recent mean number of research products may hesitate to select research opportunities that are challenging or time-consuming in favor of research product volume. While an extra research year during medical school can be a key formative experience, some students feel pressured to add a research year primarily to “buff” their application, which requires funds and potentially introduces inequity based on financial resources.7,9One previous suggestion to address this arms race was to limit reported products to a total of 3 to 5.2 This was not well-received by applicants, who opined that (1) research productivity is one of few objective metrics by which to differentiate themselves in the residency application process; (2) additional research products could still be shared on a supplemental curriculum vitae; and (3) reporting limits may paradoxically increase the amount of poorly conducted student work as the limit would be perceived as a minimum.10,11 An alternative approach, recently published in Neurosurgery, suggests a novel “arms race control score” with a “publication value unit” that adjusts for the type of product and the presumed contribution of the student.12 We do not endorse this particular score, but such outside-the-box thinking is needed. Here, we propose several actionable reforms to favor meaningful research engagement over the sheer quantity of abstracts, presentations, and publications.Changes to the Electronic Residency Application Service (ERAS) application could help students highlight their scholarly accomplishments while concurrently tamping down the urge to add just one more presentation or case report to the total. These changes would also provide clearer data to program directors (PDs) who want to recruit applicants motivated to conduct high-quality research and in turn allow the NRMP to provide more useful data to future students.First, we urge ERAS to separate entries by applicants (and display to PDs) of abstracts/presentations from full-length publications. Currently, without tedious manual review, there is no way for PDs to differentiate a meeting abstract published in a journal supplement from a research manuscript. Presenting an abstract of research results at a professional meeting is an important contribution for future physicians early in their training. However, distinguishing these types of products is important; recent detailed analysis of neurosurgery applicants reveals that the number of indexed publications is much lower than the mean number of research products reported by the NRMP.13 Next, we suggest that each abstract/presentation be listed only once, regardless of the number of conferences at which it was presented (with each instance of presentation listed under the title). We suggest that applicants indicate each product as a research report, case report, meta-analysis, review, commentary, or other type of work. Without these changes, students are incentivized to create products in high numbers rather than engage deeply in meaningful science. Last, we suggest that ERAS separate research outputs arising from scholarship during medical school from work done prior to matriculation. This prevents differential opportunities prior to medical school from following students into their next phase of training.The NRMP publishes the mean of the composite number of abstracts, presentations, and publications among US seniors, by specialty,1 and students make assumptions about their competitiveness based on this single number. They cannot, however, draw reasonable conclusions about the data when only a mean is presented. Without measures of variability including standard deviation, quartiles, and range, the mean fails to convey the degree of data dispersion and is highly susceptible to the influence of extreme values (eg, from MD/PhD students or those who completed a research year).14 The NRMP does attempt to provide more granular data by displaying a research product histogram, but the largest “bin” is “5 or more,”1 rendering these graphics unhelpful for specialties in which nearly all applicants fall into this category.The NRMP could substantively help students by adding the range and standard deviation to the mean number of research products and adjusting histograms to account for high output. These would be meaningful changes, particularly when coupled with our recommended modifications to the ERAS application from which the data is sourced.Graduate medical education (GME) most often takes place in academic settings, supervised by academic physicians who value research and other scholarship. While this may naturally draw PDs to applicants with research experience, we suspect that applicants with smaller numbers of full-length manuscripts from robust basic science, clinical, or health systems research are just as prepared for residency training (and future research success) as those with larger numbers of abstract presentations and case reports. We encourage professional organizations, including the Accreditation Council for Graduate Medical Education, specialty-based residency PD groups, the Association of American Medical Colleges, and others to spearhead conversations among PDs across specialties to better understand the actual value placed on research products of various types in resident selection, particularly in surgical subspecialties.15 Is a greater number of products, regardless of type, simply a proxy for being a hard worker—someone resilient enough to succeed in a demanding training program? PDs have an opportunity to articulate clearly what they value in an applicant profile, and their professional organizations should support transparent communication with applicants through webinars, websites such as the FREIDA American Medical Association Residency Database, or other reliable and accessible venues. Without more clarity from PDs, digitally engaged students will access anecdotal information and advice from their peers through social media, which may be of dubious accuracy.16-18Medical student research should be a process of inquiry and learning, not a competition measured in counts of research products. The emphasis must shift from the number of lines on an application to substantive contributions that reflect learning of scientific methods that prepare students for a career in medicine. We call for increased discussion among medical school and GME educators, through their national organizations, to find solutions that appropriately value engagement in research and protect trainees from pressures that do not serve them or their chosen profession.
Wolfson et al. (Wed,) studied this question.