Patterns and Predictors of Cross‐Applications Among Otolaryngology Residency Applicants Authors: Angela Renne, Maria Armache, Francis Deng, Jenny X. Chen Published: 2025 ABSTRACT ABSTRACTObjectiveTo evaluate the applicant characteristics associated with cross‐applying to otolaryngology‐head and neck surgery (OHNS) and at least one additional specialty.MethodsWe analyzed all Association of American Medical Colleges Electronic Residency Application Service applications submitted to OHNS residency programs between 2020 and 2023. Cross‐applicants applied to OHNS and at least one other specialty. Chi‐square tests and t‐tests compared the characteristics of OHNS‐only applicants and cross‐applicants. Multivariable logistic regression identified independent predictors of cross‐application.ResultsAmong 2440 applicants to OHNS, 53.0% (N = 1293) cross‐applied, most commonly to General Surgery (N = 894, 36.6%) and Internal Medicine (N = 508, 20.8%). Cross‐applicants comprised the majority of international medical graduates (IMG, 88.4%), osteopathic applicants (DO, 86.0%), and applicants 30 years of age or older (70.6%). Compared to those who applied to only OHNS, cross‐applicants had significantly fewer research experiences (difference [Δ] = −1.50 ± 0.11 [mean ± standard error]; p < 0.001) and publication entries (Δ = −3.70 ± 0.54; p < 0.001). Independent predictors of cross‐applying included IMG (odds ratio [OR]: 7.15; 95% CI: 3.78–13.96; p < 0.001) or DO status (OR = 7.04; 95% CI: 4.37–11.63; p < 0.001), age ≥ 30 (OR = 1.76; 95% CI: 1.32–2.36; p < 0.001), being a reapplicant (OR = 5.35; 95% CI: 3.25–9.08; p < 0.001), and having fewer research experiences (OR = 0.89; 95% CI: 0.86–0.92; p < 0.001).ConclusionCross‐applying to other specialties is common for OHNS applicants. These findings illustrate the complexity of the residency application process and the importance of tailored advising for applicants.Level of Evidence4. Introduction The residency application process can be extremely competitive, particularly within small specialties such as otolaryngology–head and neck surgery (OHNS) [1, 2, 3]. With a limited number of residency positions amidst a pool of highly qualified candidates, medical students often adopt strategic approaches to enhance their chances of matching [4, 5, 6, 7]. One such strategy is cross‐application, also known as parallel application, in which applicants apply to additional specialties outside their preferred field [8, 9]. While motivations for cross‐applications vary, reasons for cross‐applying include concerns about competitiveness, match uncertainty, and strong geographic preferences [5, 10, 11]. Cross‐application remains an understudied phenomenon in OHNS [7]. To our knowledge, no prior research has examined the prevalence, characteristics, or predictors of cross‐application among OHNS applicants. A deeper understanding of these trends may offer critical insights for prospective applicants, medical school advisors, and residency programs navigating the uncertainties of the match process. In this study, we evaluated cross‐application behaviors among OHNS applicants from 2020 to 2023. Our objectives were to (1) quantify the prevalence of cross‐application, (2) identify demographic and academic characteristics associated with this behavior, and (3) determine independent predictors of cross‐application. These findings contribute to a deeper understanding of how contemporary medical students use this strategy to adapt to the competitive landscape of residency applications. Methods Study Design and Data Collection We conducted a retrospective analysis of all OHNS residency applications submitted through the Association of American Medical Colleges (AAMC) Electronic Residency Application Service (ERAS) from 2020 to 2023 along with integrated data from GME Track to assess subsequent match results. Deidentified, individual‐level demographic and application data were obtained via the AAMC ERAS Applicant Dataset for the years 2019 through 2023. Each row corresponds to a single application, with a unique identifier assigned to each applicant to allow tracking of additional applications across successive years. Data from 2019 were used exclusively to identify reapplicants for the 2020 cycle and were not otherwise included in the final analysis. This study constitutes non‐human subjects research and was deemed exempt by the Johns Hopkins School of Medicine Institutional Review Board. Key Variables and Classifications Applicants who applied to OHNS as well as at least one additional specialty were classified as cross‐applicants. We examined several applicant characteristics to understand factors associated with cross‐application. Demographic variables included self‐reported age, sex (male/female), and race and/or ethnicity. Applicants who did not report all these demographic data were also excluded (N = 163). Age was categorized as ≤ 26, 27–29, and ≥ 30 years to represent younger, average, and older applicants at the time of application, respectively. Applicants self‐identified their race and ethnicity from the standardized options (American Indian or Alaska Native, Asian, Black or African American, Hispanic/Latino/Spanish origin, Native Hawaiian or Other Pacific Islander, White, or Other). For analytic clarity, responses were recoded into four mutually exclusive categories: non‐Hispanic Asian (hereafter, Asian), non‐Hispanic Black (hereafter, Black), Hispanic or Latino (hereafter, Hispanic), and non‐Hispanic White (hereafter, White). Applicants who selected other or multiple race and ethnicity options were excluded because their small numbers precluded reliable, discrete intergroup comparisons (N = 148). Variables describing applicant educational background included medical school type (public MD‐granting school student or graduate, private MD‐granting school student or graduate, international medical graduate [IMG], or osteopathic [DO] school student or graduate), membership in the honor societies Alpha Omega Alpha (AOA) and Gold Humanism Honor Society (GHHS), and geographical region of the applicant's medical school as classified by the AAMC (Central, Northeast, South, or West). Applicants from schools with late AOA elections in senior year were grouped with those reporting no AOA chapter at their institution, as these memberships would not have been available to programs at the time of application review. Additionally, we analyzed other applicant factors, including the total number of specialties applied to alongside OHNS, number of work and volunteer experiences, research experiences, publication entries (which includes oral/poster presentations, peer‐reviewed and non‐peer reviewed publications, abstracts, book chapters, or other articles), as well as peer‐reviewed publications. For each application year, unique applicant IDs were used to distinguish between first‐time applicants and reapplicants. Applicants whose ID appeared for the first time in a year were classified as first‐time applicants. If the same applicant ID reappeared in a subsequent cycle, that second (or later) application was classified as a reapplication. For analytic purposes and clarity, applications from applicants who submitted more than twice were excluded from the analysis (N = 15), defining reapplicants strictly as those applying for the second time. The primary outcome variable was whether an applicant cross‐applied to a different specialty in addition to OHNS. We subsequently assessed OHNS match rates among applicants with available match data, comparing those who applied exclusively to OHNS with those who cross‐applied to other specialties. Statistical Analysis Descriptive statistics were used to quantify the frequency of cross‐application and identify the most commonly cross‐applied specialties. Linear regression assessed the trend of cross‐applying across 2020 to 2023. To compare categorical variables between OHNS‐only and cross‐applicants, we employed chi‐square tests. Independent t‐tests were used to assess differences in continuous variables, such as research experiences, publication entries, and number of peer‐reviewed publications. To identify independent predictors of cross‐application, we constructed a multivariable logistic regression model using bidirectional stepwise regression with the lowest Akaike Information Criterion (AIC). All relevant predictors were included, including demographic characteristics (age, sex, race/ethnicity, and medical school region), academic metrics (medical school type, Alpha Omega Alpha [AOA] and Gold Humanism Honor Society [GHHS] membership), and application characteristics (work experiences, volunteer activities, research involvements, publication entries, peer‐reviewed publications, and reapplicant status). The total number of applications submitted was also retained in the model to control for overall application volume and better isolate the effects of the other predictor variables. Statistical significance was set at p < 0.05, and analyses were conducted using R (Version 4.4.0, R Foundation for Statistical Computing, Vienna, Austria). Results Cohort Description The analytic dataset included 2440 applicants who applied to OHNS residency programs from 2020 to 2023 after excluding missing demographic variables and reapplicants who applied more than twice. Most applicants identified as male (62.3%, N = 1520), while 37.7% (N = 920) identified as female. The largest group of applicants was aged 27–29 years (54.3%, N = 1324), followed by those aged ≥ 30 years (27.7%, N = 677), and those ≤ 26 years (18.0%, N = 439). Applicants were predominantly White (57.3%, N = 1397), followed by Asian (26.4%, N = 643), Hispanic (10.1%, N = 247), and Black (6.3%, N = 153) applicants. Among these applicants, 46.5% (N = 1135) attended public MD programs, 34.4% (N = 839) private MD programs, 9.9% (N = 242) DO programs, and 9.2% (N = 224) were IMGs. Recent Cross‐Application Trends Among the 2440 OHNS applicants, 1293 (53.0%) cross‐applied to at least one other specialty (Table 1). By year, 55.4% (N = 353) cross‐applied in 2020, 48.1% (N = 272) in 2021, 50.3% (N = 292) in 2022, and 57.2% (N = 376) in 2023. There was no statistically significant trend in cross‐application rates over this four‐year period (β = 0.008, p = 0.77). The cross‐applicants applied to a median of 2 specialties (interquartile range: 1–4 specialties). The most frequently cross‐applied specialties were general surgery (N = 894, 36.6%), internal medicine (N = 508, 20.8%), and anesthesiology (N = 295, 12.1%). TABLE 1Cross‐applications among otolaryngology residency applicants (2020–2023, N = 2440).SpecialtyCountGeneral surgery894Internal medicine508Anesthesiology295Interventional radiology (integrated)219Dermatology200Vascular surgery (integrated)197Emergency medicine192Orthopedic surgery192Diagnostic radiology185Family medicine179Radiation oncology151Neurological surgery110Pediatrics98Obstetrics and gynecology89Physical medicine and rehabilitation70Plastic surgery (integrated)68Neurology67Pathology57Psychiatry55Child neurology22Urology20Thoracic surgery (integrated)18Public health and general preventive medicine4Other385Applied to otolaryngology1 other specialty5652 other specialties2143 other specialties1244 other specialties895+ other specialties301 Note: This table summarizes the number of OHNS applications to other specialties. The top 20 cross‐applied specialties are listed in order of frequency. The “Other” category includes pooled counts of less frequently selected specialties. Cross‐applications among otolaryngology residency applicants (2020–2023, N = 2440). Note: This table summarizes the number of OHNS applications to other specialties. The top 20 cross‐applied specialties are listed in order of frequency. The “Other” category includes pooled counts of less frequently selected specialties. Characteristics of Cross‐Applicants Personal and academic characteristics differed between cross‐applicants and OHNS‐only applicants (Table 2). Sex distribution differed between groups (p < 0.001), with the majority of male applicants (57.0%) cross‐applying. Age was a strong differentiating factor (p < 0.001): 70.6% of applicants aged ≥ 30 were cross‐applicants, while only 36.7% of those aged ≤ 26 cross‐applied. Racial and ethnic distribution differed significantly between groups (p = 0.003): 60.8% of Black and 61.9% of Hispanic applicants cross‐applied, compared to 50.9% and 51.5% of Asian and White applicants, respectively. Among domestic applicants, geographic region also played a role (p < 0.001): 54.5% of applicants from Southern medical schools cross‐applied, while 44.7% of Northeast applicants cross‐applied. TABLE 2Comparison of applicant characteristics between single‐application (OHNS‐only) and cross‐application, 2020–2023.CategoryCross application, N = 1293 (count, %)OHNS‐only application, N = 1147 (count, %) p * Sex< 0.001Female427 (46.4%)493 (53.6%)Male866 (57.0%)654 (43.0%)Age< 0.001≤ 26161 (36.7%)278 (63.3%)27–29654 (49.4%)670 (50.6%)30+478 (70.6%)199 (29.4%)Race0.003Asian327 (50.9%)316 (49.1%)Black93 (60.8%)60 (39.2%)Hispanic153 (61.9%)94 (38.1%)White720 (51.5%)677 (48.5%)Region0.003Central270 (48.3%)289 (51.7%)Northeast289 (44.7%)358 (55.3%)South416 (54.5%)348 (45.5%)West120 (48.8%)126 (51.2%)Medical school type< 0.001DO208 (86.0%)34 (14.0%)International (IMG)198 (88.4%)26 (11.6%)MD‐Private343 (40.9%)496 (59.1%)MD‐Public544 (47.9%)591 (52.1%)GHHS< 0.001Member124 (43.8%)159 (56.2%)No chapter258 (61.6%)161 (38.4%)Not member911 (52.4%)827 (47.6%)AOA a < 0.001Late elect85 (49.4%)87 (50.6%)Member177 (29.3%)428 (70.7%)No chapter271 (59.2%)187 (40.8%)Not member760 (63.1%)445 (36.9%)ReapplicantYes237 (87.5%)34 (12.5%)< 0.001No1069 (48.7%)1125 (51.3%)* p values derived from Chi‐square tests. a No chapter and late elect were collapsed into a single category as residency programs would not have data for late elect applicants. Comparison of applicant characteristics between single‐application (OHNS‐only) and cross‐application, 2020–2023. p values derived from Chi‐square tests. No chapter and late elect were collapsed into a single category as residency programs would not have data for late elect applicants. In terms of academic characteristics, medical school type varied significantly across groups (p < 0.001); 88.4% of IMGs and 86.0% of DOs applied to another specialty in addition to OHNS, respectively, compared to only 40.9% and 47.9% of applicants from private or public MD schools, respectively (Table 2). AOA status was strongly associated with cross‐application behavior as well (p < 0.001, Table 2). Reapplication status was also a strong distinguishing factor (p < 0.001); 87.5% of reapplicants cross‐applied. Cross‐applicants had fewer research experiences (mean difference [Δ] = −1.50, standard error [SE] = −0.11; p < 0.001), ERAS publication entries (Δ = −3.70, SE = 0.54; p < 0.001), peer‐reviewed publications (Δ = −0.83, SE = 0.12; p < 0.001), and volunteer experiences (Δ = −1.18, SE = 0.13; p < 0.001) compared to OHNS‐only applicants (Figure 1). However, cross‐applicants demonstrated significantly more work experience (Δ = +0.54, SE = 0.08; p < 0.001). Differences in application characteristics between OHNS‐only applicants and cross‐applicants. Positive values indicate characteristics greater among cross‐applicants; negative values indicate characteristics greater among OHNS‐only applicants. Statistical comparisons were performed using independent t‐tests. Predictors of Cross‐Application The final logistic regression model predicting cross‐application status was derived using bi‐directional stepwise selection with the best‐fitting model including the following variables: age group, medical school type, AOA membership, reapplicant status, number of work experiences, volunteer experiences, research experiences, and number of publication entries (Table 3). Medical school type was a strong predictor of cross‐application. Compared to applicants from public MD programs, IMG (OR: 7.15; 95% CI: 3.78–13.96; p < 0.001) and DO applicants (OR: 7.04; 95% CI: 4.37–11.63; p < 0.001) had significantly higher odds of cross‐applying. In contrast, MD applicants from private institutions showed no significant difference in cross‐application behavior (OR: 0.94; 95% CI: 0.74–1.20; p = 0.63). Applicants who were AOA members had significantly lower odds of cross‐applying compared to non‐members (OR: 0.55; 95% CI: 0.42–0.72; p < 0.001), while those from schools with effectively no AOA chapter (including late‐elect institutions) did not differ significantly from non‐members (OR: 0.89; 95% CI: 0.67–1.18; p = 0.42). TABLE 3Multivariable logistic regression identifying independent predictors of cross‐application among otolaryngology (OHNS) applicants. Odds ratios (ORs) reflect the likelihood of cross‐application relative to the reference category for each variable – defined in the Methods section under “Key variables and classifications”.VariableOdds ratio95% confidence interval p Age group0–260.840.62–1.120.2427–29 (reference)———30+1.761.32–2.36< 0.001Medical school typeDO7.044.37–11.63< 0.001International7.153.78–13.96< 0.001MD‐private0.940.74–1.200.63MD‐public (reference)———AOA membershipMember0.550.42–0.72< 0.001No chapter0.890.67–1.180.42Non‐member (reference)———Reapplicant5.353.25–9.08< 0.001Number of work experiences1.051.00–1.100.062Number of volunteer experiences0.970.94–1.000.037Research experiences0.890.86–0.92< 0.001Publication entries0.980.97–0.99< 0.001 Note: The model adjusts for total number of applications submitted. Multivariable logistic regression identifying independent predictors of cross‐application among otolaryngology (OHNS) applicants. Odds ratios (ORs) reflect the likelihood of cross‐application relative to the reference category for each variable – defined in the Methods section under “Key variables and classifications”. Note: The model adjusts for total number of applications submitted. Age group was another important factor. Applicants aged ≥ 30 years were significantly more likely to cross‐apply compared to those aged 27–29 (OR: 1.76; 95% CI: 1.32–2.36; p < 0.001), while applicants aged ≤ 26 years did not differ significantly from those aged 27–29 (OR: 0.84; 95% CI: 0.62–1.12; p = 0.24). Several academic and application behaviors also played a role in cross‐application behavior. Reapplicants had substantially higher odds of cross‐applying than first‐time applicants (OR: 5.35; 95% CI: 3.25–9.08; p < 0.001). Greater research involvement was associated with reduced likelihood of cross‐application; each additional research experience decreased the odds by approximately 11% (OR: 0.89; 95% CI: 0.86–0.92; p < 0.001). Similarly, a higher number of publication entries was also associated with lower odds of cross‐application (OR = 0.98, 95% CI: 0.97–0.99; p < 0.001). Volunteer experiences were also associated with lower odds of cross‐applying (OR: 0.97; 95% CI: 0.94–1.00; p = 0.037), while the number of work experiences was not a statistically significant predictor (OR: 1.05; 95% CI: 1.00–1.10; p = 0.062). Match Results Of 2440 total applicants, final match data were available for 2251 (92.3%). Match data were missing for 53 OHNS‐only applicants (4.6%) and 136 cross‐applicants (10.5%). Among applicants who applied only to OHNS with final match data (n = 1094), 1058 (96.7%) successfully matched in OHNS and 36 (3.3%) did not. Among cross‐applicants (n = 1157), 260 (22.5%) matched in OHNS and 897 (77.5%) did not. Of those who cross‐applied, 495 (42.8%) ultimately matched into general surgery and 166 (14.3%) into internal medicine (Table S1). Discussion Our study demonstrates that over half of OHNS residency applicants between 2020 and 2023 cross‐applied to at least one additional specialty. Among cross‐applicants, general surgery emerged as the most frequently chosen alternative specialty, followed by internal medicine and anesthesiology. Several applicant characteristics were found to be predictive of cross‐application behavior: medical school type, reapplicant status, applicant age, research experiences, and number of ERAS publication entries. Medical school type was the most significant predictor: IMG and DO applicants had substantially higher odds of cross‐applying compared to U.S. allopathic applicants. This likely reflects a strategic response to lower match rates historically reported among IMGs and DO applicants in OHNS, suggesting a protective behavior to mitigate the risk of going unmatched [12]. Reapplicant status also emerged as one of the strongest predictors of cross‐application behavior, with reapplicants having more than five times the odds of cross‐applying compared to first‐time applicants. This suggests that individuals re‐entering the match may pursue broader strategies to diversify their applications to improve match outcomes. Older applicants (≥ 30 years) also demonstrated a greater likelihood of cross‐applying. This may reflect the added personal or financial responsibilities often faced by non‐traditional applicants, who may be less willing to risk going unmatched [13]. Similarly, applicants with fewer research experiences, publication entries, or honors such as AOA membership were more likely to cross‐apply. These findings align with known predictors of match success in competitive fields in which research involvement and academic recognition are important factors for matching [4, 7, 10, 14, 15]. Notably, race/ethnicity and sex did not independently predict cross‐application behavior. These results have important implications for how we advise students applying to OHNS. On one hand, there are several drawbacks to cross‐applying [5, 6, 16]. Doing so substantially increases application volume and higher associated financial burdens for applicants [2, 16, 17]. Additionally, ongoing stigma surrounds applicants who pursue more than one specialty, particularly in competitive fields like OHNS, where applicants may be expected to project commitment to the specialty [6, 14]. Our analysis of match outcomes further demonstrates that applicants who cross‐applied were substantially less likely to match in OHNS compared with those who applied exclusively to OHNS (96.7% vs. 22.5%). This may potentially reflect underlying differences in applicant competitiveness. On the other hand, our findings suggest that cross‐application is a common practice, undertaken by 53% of OHNS applicants, and may offer some applicants a pragmatic way to mitigate the risk of going unmatched when applying to competitive specialties like OHNS [5, 6, 18]. However, cross‐applicants in our cohort generally had fewer predictors of OHNS match success, which likely contributes to their substantially lower match rate. At the same time, more than one in five (22.5%) cross‐applicants still matched into OHNS, indicating that the strategy can be beneficial for select individuals. Given these nuances, individualized mentoring is essential to help applicants realistically assess their competitiveness by considering predictors of OHNS match success and determine whether cross‐applying or focusing on an alternative specialty is most appropriate. Medical school advisors and residency leadership should support applicants in developing informed, personalized strategies that balance competitiveness with long‐term career goals [5]. An additional consideration is the potential impact of cross‐application on the interview and match processes of other specialties. Cross‐applicants to OHNS may occupy interview slots or training positions in other fields like general surgery or internal medicine that might otherwise go to applicants applying solely within those specialties. However, the impact is likely limited, as cross‐applicants comprise a small fraction of the total applicant pools in these larger specialties [8, 12]. Moreover, although OHNS applicants are typically competitive, cross‐applicants may not be as strong and may not necessarily displace stronger candidates in other fields. Still, these behaviors may have downstream implications particularly if cross‐applicants later transition into OHNS residency after initially matching into a different specialty. This study has a few limitations. First, the dataset used is limited to applicants who participated in ERAS from 2020 to 2023, therefore excluding those who applied to ophthalmology, which uses a separate application system. This exclusion may lead to a small underestimation of true cross‐application rates within OHNS. Additionally, while we examined a broad range of demographic and academic factors, unmeasured variables such as couples matching or geographic preference could have influenced the decision to cross‐apply but were not captured. Rank list preferences were not available in our dataset; however, NRMP data suggest that most cross‐applicants who rank both OHNS and other specialties list OHNS as their top choice [12]. Further, the ERAS data encompasses the entire application cycle, including those who later entered the Supplemental Offer and Acceptance Program (SOAP). Since SOAP applicants may apply more broadly due to limited availability of positions in their field, this could inflate cross‐application rates. However, SOAP applicants to OHNS represent a very small minority of the overall OHNS applicant pool and likely had a minimal effect on our estimation of cross‐application behavior. Lastly, while we attempted to accurately capture reapplicant status, our data to assess reapplicants only reflect applications submitted between 2019 and 2023. As a result, individuals who initially applied prior to 2019 and reapplied during the 2020–2023 study period may have been misclassified as first‐time applicants. However, year‐by‐year analysis demonstrated a consistent proportion of reapplicants across all four application cycles from 2020 to 2023, suggesting that any underestimation of reapplicants is likely small and unlikely to significantly affect our findings. Further research is warranted to explore the motivations, decision‐making processes, and match outcomes of cross‐applicants. Qualitative studies could provide insight into applicant experiences, including the role of mentorship, geographic preferences, and perceived stigma. Additionally, examining the long‐term career trajectories and satisfaction of cross‐applicants could help determine whether this strategy impacts professional development and specialty retention. Conclusion Cross‐applying is a common strategy among otolaryngology applicants. Further research is needed to explore the underlying motivations for cross‐application and how these trends might evolve in future residency cycles. Funding The authors have nothing to report. Disclosure The authors have nothing to report. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Table S1: Matched specialty distribution among applicants with cross‐specialty and otolaryngology‐only (OHNS‐only) applications. Values represent the number (N) and percentage (%) of applicants who matched into each specialty, based on available match data. Of 2440 total applicants, final match data were available for 2251 (92.3%), including 1094 OHNS‐only and 1157 cross‐specialty applicants. Match data were missing for 53 OHNS‐only applicants (4.6%) and 136 cross‐specialty applicants (10.5%). Percentages are calculated within each respective application group.