"Ghost" publications among applicants to a general surgery residency program.

OBJECTIVE: To determine the incidence of potentially fraudulent (or "ghost") publications in applications to a general surgery residency program.
METHODS: Electronic Residency Application Services applications submitted in 2005 to the general surgery residency program were reviewed in an IRB-approved study. No identifiers were collected. Publications were checked against Medline, PubMed, ISI Web of Science, and Google. Nonverifiable publications were then submitted to the medical librarian for verification. Ghost publications were defined as journals, books, or meetings that cannot be verified; verified journals without the listed publication; or verified publications without an applicant author. Data analyses were performed using univariate and multivariate regression analysis for nonparametric data. A p value < 0.05 was considered significant.
RESULTS: Four hundred ninety-three applications were received. Thirty-one percent (150 of 493) of applicants listed a total of 596 publications, including 30 abstracts, 359 journal articles, and 207 chapters. Thirty-three percent (196 of 596) of the publications could not be verified: 7 abstracts, 177 journal articles, and 12 chapters. The distribution of ghost publications was skewed toward the journals subgroup (p < 0.001). Positive predictors of ghost publications were age and foreign medical school. The sole negative predictor was enrollment in a top-10 US research medical school.
CONCLUSION: A disturbingly substantial fraction of publications listed on Electronic Residency Application Services applications cannot be verified. This might indicate a need for greater mentorship and oversight for medical school applicants. It is unknown whether this behavior predicts lack of integrity in other professional settings.