• Richard R Ranney, John C Gunsolley, Lois S Miller, and Morton Wood.
• Baltimore College of Dental Surgery, Dental School, University of Maryland, 21201, USA. rranney@dental.umaryland.edu
• J Am Dent Assoc. 2004 Aug 1;135(8):1146-53.

BackgroundLicensure examinations in dentistry have become an increasing concern, owing to ethical issues in the use of patients, difficulties in seeing relationships between outcomes of licensure examinations and performance in educational programs, and questions on the reliability of "one-shot" clinical examinations. Using data from a nine-year period, the authors compared the results of clinical licensing tests and the academic class ranks of the candidates.MethodsThe authors studied data for 835 dental school graduates of one school from 1994 through 2002. They compared the dental graduates' results from the North East Regional Board, or NERB, of Dental Examiners examinations with their class ranks. The authors used analysis of variance to analyze the differences among passing, failing and "no data" groups, kappa statistic and logistic regression for variation, and receiver operating characteristic, or ROC, curves for diagnostic utility.ResultsThe class rank of graduates who passed and failed NERB's restorative section of the examination did not differ. Differences for other sections of the examination were statistically significant but small. The variation in restorative and manikin exercises over time was highly significant. No consistency existed between these tests, and their ROC curves indicated no utility for diagnosing class rank.ConclusionsThe authors' analysis of nine years' data called into question the reliability and validity of initial licensure examinations based on certain of the one-time tests used by NERB. Future study should determine if the results generalize to other schools and clinical testing agencies.Practice ImplicationsIf the results of this study can be generalized to all U.S. licensure examinations, basing licensing decisions on clinical licensure examination alone risks licensure decisions of low validity. Use of patients in examinations of questionable validity may be unethical because they may have been subjected to risk of irreversible damage without contribution to a valid decision-making process by the licensing authority.

18 keywords...

hide…