Can smoking cessation be taught online? A prospective study comparing e-learning and role-playing in medical education

Abstract

Objectives: We compared the effect of different didactic formats - e - learning and role-playing - on medical students' knowledge and counselling skills in smoking cessation training.

Methods: At a German medical school, 145 third-year students were randomly allocated to attend an online course with video examples or an attendance course with role-playing. Students were trained in smoking cessation counselling according to the 5A's (ask, advise, assess, assist, arrange) for approximately 90 minutes. Practical skills were measured in an objective structured clinical examination (OSCE) and represent the primary endpoint of this prospective comparative study. Additionally, changes in theoretic knowledge were assessed by pre - and post - interventional questionnaires and a final written exam.

Results: In the OSCE, overall scores were higher in the attendance group (Mdn=70.8 % vs. 62.8 %; U=119; p=.087, n=36), but a statistical advantage was only found in one single counselling sequence ("Assist": Mdn=66.7 % vs. 51.4 %; p = .049) and the rating of the standardised patients (M=4.7 vs. 4.2 out of 5 points, t_(27.836)=2.0, p=.028). Students' results (n=130) from self-assessment and written exams suggest that both approaches are equally well suited to increase theoretical knowledge. The online course was more time efficient (90 vs. 73 minutes).

Conclusions: Seminar and web-based training seem equally well suited for transferring knowledge and skills on tobacco cessation counselling. Considering their particular strengths, these two teaching approaches could be combined.

Keywords: e-learning; medical education; objective structured clinical examination; smoking cessation.

Figure 1.

The distribution of the OSCE scores is displayed section by section as boxplots showing the median (horizontal line) and the interquartile range (IQR; lengths of the box). The vertical lines (whiskers) extend from the hinge to the largest/smallest value no further than 1.5*IQR from the hinge, values beyond these upper and lower bounds are considered outliers. Significant difference only in the section “Assist” (p = .049, n=36).