Objectives Randomized Controlled Trials are complex undertakings, involving many different abilities, skills, knowledges and notions, some of which can only be acquired by direct experience. Objective of this teaching experiment was to assess short-term learning efficacy of a simulated RCT on knowledge of methodological issues involved in clinical research.

Method The final exam of our 400-hours course on Methods in Biomedical Research, was a 2-days simulation of a randomized placebo-controlled trial with 30 ‘patients’ in 5 ‘clinical centres’ (Africa, America, Asia, Europe, Oceania), in three editions (2015–2017, final sample size = 90). The study question was on efficacy of caffeine on cardiovascular endpoints, and safety. Participants were randomized to standard vs decaffeinated Italian espresso in a blinded fashion, and were guided to actively perform:

Day 1, morning: study design,CRF design, trial registration

Day 1, afternoon: implementation (including obtaining consent, randomization, etc)

Day 2, morning: statistical analysis

Day 2, afternoon: study report (CONSORT)

Knowledge was tested pre- and post-intervention by means of a standardized questionnaire, based on the CONSORT checklist, scoring 0–30.

Participants were also blind to coffee manufacturer, and were allowed the following concomitant medications: sugar, milk, biscuits, or chocolates.

The trial was not funded by a coffee maker.

Results Trial participants improved their median knowledge of RTCs methods from 15 points (IQR 10–24) to 28 points (IQR 24-30) (p<0.001). Participants were able to identify and discuss a large number of theoretical and practical considerations in RTC ideation, design, conduct, interpretation and publication, and reached meaningful insights into generalizability of results, protocol violations, missing data, standardization of measurement instruments, concomitant medications, blinding, GCP, study monitoring, and others.

All trial participants enjoyed the simulation, and most (there were a few protocol violations) happily sipped their cup of coffee, with no adverse event. None was lost to follow up.

Conclusions Short-term learning goals were achieved in a rapid and entertaining way. Long-term utility of the teaching exercise will be measured by means of the number of future publications by the participants in high impact journals. The effects of coffee on cardiovascular outcomes will need a much larger sample size to be ascertained, but the effect on students’ mood was striking, with no safety concerns.