Teaching Tips: randomisation for trials
Chris Del Mar describes a group exercise that enables students to appreciate how trials work, and how they can go wrong.Key Concepts addressed:
- 2-1 Comparisons are needed to identify treatment effects
- 2-5 People should not know which treatment they get
The educational solution
Students are handed an envelope as they enter the lecture theatre for a session introducing randomised controlled trials. They are asked not to open it until instructed.
The lecture proceeds. At the point when randomisation is discussed, the students are invited to supply some information about themselves: how many are female; live inter-state, have ever smoked etc. The data are written up on a white board in a Table.
Then I ask if anyone has already opened their envelope (almost invariably 1-2% have), and these data are also tabulated, also checking their state and past smoking status. I also ask if anyone has been able to work out what the envelopes contain: explaining that we put in two types of confectionery (popular brands of either a boiled, or minty, sweet). Some know, and I record this into the Table. And I ask how they were able to check (usually they have held the envelope to the light, or simply palpated its contents and guessed), recording this also. I ask them to check if they were right by fully opening their envelopes, recording how many guessed correctly (usually most). I invite everyone to open their envelope and consume their contents. Finally, I ask how many have swapped theirs with a neighbour because of preferences (usually surprisingly many).
The following discussion
After the commotion has settled down, I open a discussion, and ask the students to help me draw up a CONSORT statement of the outcomes of this RCT. The breaking of blinding is recorded, together with the attrition to the ‘other treatment’, and we use the previously collected data about the students to test if the randomisation was successful (the Baseline Table of so many reported RCTs).
I test knowledge and understand by means of multiple choice questions in the end of year exams to check there is some persistent understanding. I also revisit the teaching session the following year when we start to critically appraise randomised controlled trials, pausing to talk about randomisation. This enables a spiral of learning re-enforcement.
Where the idea came from
Sandy Pirozzo was a lecturer in the Masters of Public Health program at the University of Queensland who was extremely original in her teaching ideas, and she devised these techniques.
Chris Del Mar [email protected] Oct 2017