PURPOSE: There is a dearth of data regarding journal-based continuing medical education (CME) programs. Deutsches Arzteblatt has been publishing CME articles since 2004. Articles are accompanied by a test on the article's content as well as a readers' evaluation questionnaire. The goal was to find out whether a journal-based CME program can be successfully tailored to a multidisciplinary audience. SUMMARY: This study is an analysis of 706,995 participations (76,486 participants) between September 2004 and August 2007. Main dependent variables were the percentages of correctly answered test questions. Independent variables were subgroups defined by demographic and job-related (e.g., specialty) variables, and characteristics emerging from the participants' evaluations of the articles. On average, 9.55 (SD = 0.82) out of 10 test questions were answered correctly. Significant differences were found between the subgroups, but these were small. Significant intersubgroup differences for difficult questions were larger yet still small in absolute terms. Reader satisfaction was high. CONCLUSIONS: Our results indicate that these articles are suitable for a wide range of physicians. The high percentage of correct answers reflects the understanding in Germany that CME questions should not mimic examinations but rather check whether readers have read the article attentively.
PURPOSE: There is a dearth of data regarding journal-based continuing medical education (CME) programs. Deutsches Arzteblatt has been publishing CME articles since 2004. Articles are accompanied by a test on the article's content as well as a readers' evaluation questionnaire. The goal was to find out whether a journal-based CME program can be successfully tailored to a multidisciplinary audience. SUMMARY: This study is an analysis of 706,995 participations (76,486 participants) between September 2004 and August 2007. Main dependent variables were the percentages of correctly answered test questions. Independent variables were subgroups defined by demographic and job-related (e.g., specialty) variables, and characteristics emerging from the participants' evaluations of the articles. On average, 9.55 (SD = 0.82) out of 10 test questions were answered correctly. Significant differences were found between the subgroups, but these were small. Significant intersubgroup differences for difficult questions were larger yet still small in absolute terms. Reader satisfaction was high. CONCLUSIONS: Our results indicate that these articles are suitable for a wide range of physicians. The high percentage of correct answers reflects the understanding in Germany that CME questions should not mimic examinations but rather check whether readers have read the article attentively.