A research year adds unpublished work, not publications

Design

Retrospective review of residency applications from a single programme, Wilcoxon rank-sum comparison

Population

196 otolaryngology-head and neck surgery residency applicants in the 2025 cycle; 55 with a dedicated research year, 90 without

Primary outcome

Number and type of research products listed in the application

Effect

Total products median 32 (IQR 21-37) vs 24 (15.75-37); difference driven by unpublished work (5 vs 3) and oral presentations (6 vs 4); published articles (5.5 vs 7) and first-author manuscripts (median 1) did not differ

Research output among otolaryngology residency applicants has climbed steeply, and with it the number of students taking a dedicated year out to produce it. This review of 196 applicants to a single programme in the 2025 cycle asks what that year actually yields, categorising every research product listed in the application by type, authorship, publication status and presentation.

The median applicant listed 26 research products (IQR 17-37) - a figure worth pausing on by itself. Applicants who had taken a research year (n = 55) listed more in total than those who had not (n = 90): a median of 32 (IQR 21-37) against 24 (IQR 15.75-37). But the excess sat in unpublished work (median 5 against 3) and oral presentations (6 against 4). Published articles did not differ significantly (5.5 against 7), and first-author manuscripts had a median of 1 in both groups.

The authors' reading is the right one: volume-based metrics are capturing something other than scholarly development, and the arms race they drive raises the expected count without a matching gain in research experience. For anyone advising a student, that is the point to convey. A year spent accumulating abstracts and middle authorships to move a median is a year spent on the metric rather than on the work, and it does not appear to produce more first-author papers than not taking one.

• Advise students that a research year did not increase published or first-author output in this cohort
• The median applicant listed 26 research products - understand what the count is actually measuring before using it
• Programme selection that weights volume rewards abstracts and middle authorship
• One first-author manuscript was the median in both groups
• Single programme, one cycle - the numbers are local, the pattern is not

Why it matters

Students are spending a year of training on a metric that does not appear to measure what selection committees think it does.

Don't overread it

Retrospective, one programme, one application cycle - it describes an association, not what a research year causes.

The statistics, in plain English

Medians with wide interquartile ranges are doing a lot of work here. The research-year group's 32 (IQR 21-37) and the comparison group's 24 (IQR 15.75-37) overlap almost completely at the upper end, meaning the most productive applicants in both groups look alike. The breakdown is more informative than the total: the categories that differed - unpublished work and oral presentations - are the ones with the lowest barrier to accumulating, while the categories that did not differ are the ones requiring a completed piece of work. This is a single programme in a single cycle, so the absolute numbers describe who applied there, not the national applicant pool.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.