Paired or Pooled Analyses in Continuing Medical Education, Which One Is Better?

Journal of European CME (05/27/23) Robles, Jessica H.; Harb, Kathleen J.; Nisly, Sarah A.

An assessment of live continuing medical education (CME) sessions for advanced practice providers sought to determine whether paired or pooled analyses yielded better results. The data covered eight sessions conducted across six different dates. The process involved asking participants identical questions for each topic before the session, immediately afterwards, and on a rolling four- to six-week follow-up poll. The sessions provided 72 total questions, with each asking nine questions divided by knowledge, competence and performance. The researchers included responses from 55 questions in the analysis. The pre versus post-content questions returned 323–956 matched response pairs, while the pre versus follow-up questions yielded 41–236 pairs. Pooled analysis yielded 1260–2285 responses in the pre versus post content framework, while 777–1431 responses resulted from the pre versus follow-up pooled analysis. All of the individual questions produced a statistically significant improvement in correct responses in both analytical schemes. Thirty-three questions (59%) indicated a statistically significant improvement in correct responses in the paired pre versus follow-up analysis. Thirty-five questions (62%) in the pooled pre versus follow-up analysis yielded similar results. The researchers note that this preliminary research "questions the need for paired data in CME data and sparks opportunities for more conversations and data insights in this space." They believe additional research should consider whether meaningful insights can be construed from other factors or finer points within the data.

Read More