Quantifying the literature of computer-aided instruction in medical education

Purpose. To characterize the literature about computer-aided instruction (CAI) as it relates to medical education. Method. A descriptive study using the Medline and ERIC databases, reviewing articles pertaining to CAI for medical professionals. The literature was evaluated with respect to type of article, journal, language, and year of publication. Results. The search yielded 2,840 citations, 92% of which were in English. The number of journals with at least one citation was 747. Less than 5% of the 5,147 authors had three or more articles published in the CAI literature. Of the citations with abstracts, 60% were demonstrations of a CAI application, 11% were media-comparative studies, and 13% were analyses of the CAI field. While the pace of article publication increased markedly over time, the percentages of article types did not change significantly over time. Less than 10% of CAI articles appeared in core medical journals. Conclusion. Medical CAI is an increasingly popular topic of research and publication. However, these studies appear in journals with smaller circulations, are predominantly demonstration articles, and are generally written by authors with two or fewer publications: Evaluation articles remain less common. A series of analytic articles has appeared offering substantive suggestions for better research design. These suggestions appear to have gone unheeded. CAI investigators need be more aware of the gaps in the existing body of CAI publications, and the inherent difficulties of this type of research, if this literature is to move beyond this early stage of development.