Minimal prevalence of authorship misrepresentation among internal medicine residency applicants: Do previous estimates of "misrepresentation" represent insufficient case finding?

Background: High rates of authorship misrepresentation have been documented among medical trainees. Objective: To assess misrepresentation among internal medicine residency applicants while comparing searches used by previous authors (searches 1 and 2) to a more comprehensive strategy (search 3). Design: Review of 497 residency applications. Setting: Two university-based internal medicine residency programs. Measurements: Search 1 was limited to MEDLINE. Search 2 added Current Contents, Science Citation Index, and BIOSIS and included searching journals by hand. Search 3 added seven other databases and contacts to librarians, editors, and coauthors. Results: 224 applicants reported 634 articles; 630 (99%) were verified. The number of applicants with misrepresented citations varied depending on the search used (56 applicants [25%] in search 1 vs. 34 applicants [15%] in search 2 vs. 4 applicants [1.8%] in search 3). Conclusions: Using a comprehensive search, we found substantially less misrepresentation than had been reported. Previous studies probably overestimated the magnitude of the problem.