CME examinations: one year of experience

Medical professionals require ongoing education to maintain competency. Prior research has shown that structured CME programs help physicians stay updated with clinical advancements. However, no prior work had resolved how to evaluate the effectiveness of these programs using statistical methods. The Journal of the American Academy of Dermatology introduced a CME credit system in 1979. This system allowed physicians to earn credit by completing examinations on published articles. The program's success depended on consistent participation and measurable outcomes. The Academy office developed a computer-based analysis to assess the exams. This approach aimed to improve the quality and reliability of the CME program. The first year of the program provided a dataset for evaluating its performance. This period offered insights into how the exams functioned and how physicians responded.

The goal of this study was to evaluate the first year of the CME examination program. The researchers wanted to understand how the exams were performed and how the statistical tools applied to them functioned. They aimed to assess the reliability and difficulty of the exams. The study focused on the period from July 1979 to June 1980. During this time, the program had an average monthly participation of 748 physicians. The researchers sought to determine whether the exams met educational standards. They also wanted to identify any patterns in performance or item difficulty. The findings would help refine the program for future use.

The researchers used a computer-based analysis program to evaluate the CME exams. The program included six statistical components. First, they performed an item analysis to assess individual questions. Second, they calculated difficulty and discrimination indices for each question. Third, they generated descriptive statistics to summarize the data. Fourth, they determined the reliability coefficient of the exams. Fifth, they analyzed the distribution of raw scores among participants. Sixth, they examined the percentile distribution of scores. These methods provided a comprehensive view of the program's performance. The data was collected from July 1979 to June 1980.

The first year of the CME program saw an average of 748 participants per month. The item analysis revealed variations in question difficulty and discrimination. Some questions had higher discrimination indices, indicating better performance. The difficulty indices showed that certain questions were more challenging than others. Descriptive statistics highlighted the overall score trends among participants. The reliability coefficient suggested the exams were consistent in measuring knowledge. The raw score distribution showed a range of performance levels. The percentile distribution indicated how participants ranked relative to each other.

The CME examination program demonstrated consistent participation and measurable outcomes. The statistical tools used provided valuable insights into the program's effectiveness. The item analysis helped identify high- and low-performing questions. The difficulty and discrimination indices supported the program's educational value. The reliability coefficient confirmed the exams' consistency. The score distributions showed a wide range of participant performance. These findings supported the continuation of the program. The researchers proposed further analysis to improve future exams.