How recent is recent? Retrospective analysis of suspiciously timeless citations.

OBJECTIVE: To quantify the time lag between biomedical articles and the studies they describe as "recent," a term widely used to imply timeliness despite rarely reflecting the actual age of the cited evidence. DESIGN: Retrospective analysis of suspiciously timeless citations based on a structured PubMed search of 20 predefined "recent" expressions. SAMPLE: 1000 English language, full text biomedical articles in which a "recent" expression is directly linked to a citation. MAIN OUTCOME MEASURE: Time lag in years between citing articles and their referenced "recent" studies. RESULTS: The age of the cited "recent" studies varied widely. The citation lag ranged from 0 to 37 years (mean 5.53 years, median 4 years, interquartile range 2-7). The most frequent lag was one year (n=159, 15.9%), and 177 references (17.7%) were at least 10 years old. Citation patterns varied across medical specialties: critical care, infectious diseases, genetics, immunology, and radiology showed shorter median lags (around two years), while nephrology, veterinary medicine, and dentistry displayed substantially longer lags (ranging from 8.5 to 14 years). Among expressions, "recent approach," "recent discovery," and "recent study" were linked to older references, whereas "recent publication" and "recent article" had much fresher citations. The citation lag was similar across world regions and gradually decreased over time, with the most recent publications showing the shortest lags. Journals with high impact factors (≥12) cited more up-to-date work. CONCLUSIONS: This playful analysis suggests that "recent" is applied with striking elasticity across biomedical literature. While some authors cite genuinely recent work, others stretch the definition to decades. Readers and reviewers should take "recent" claims with a grain of chronological salt.