The ABC of RCTs
Nearly 70 years ago, the first ever modern randomised controlled trial (RCT) was conducted in the UK 1. The 6-month long investigation took place in hospitals in London and Middlesex, and tested the efficacy of treatments for pulmonary tuberculosis – streptomycin versus bed rest 2. By the end of the trial, 7% of patients who received streptomycin had died versus 27% of control patients, a statistically significant effect that could be entirely accounted for by the difference in treatments.
Since this time, RCTs have typically been used to evaluate medical interventions, however they’re now increasingly employed by behavioural scientists to determine the success of behavioural nudges. In their simplest form, an RCT involves evaluating interventions by comparing them to a control. The randomisation component ensures that any bias is minimised, making RCTs one of the most rigorous experimental designs on offer 3. Considering their growing use, here is some ‘Hunting Dynasty-approved’ advice on how to design and run RCTs in business.
Step 1: Designing a trial
- Always begin your RCT by forming a clear set of objectives you want to achieve. This helps to establish the exact behaviour and sample of individuals you’re interested in targeting.
- Next, develop your intervention(s). Of course, this is the critical aspect of designing any RCT – you want to make sure you create strategies that truly effect behaviour.
○ How? Root your interventions in pre-existing evidence and/or recommended criteria such as APEASE 4. These criteria are useful as they consider factors like affordability, acceptability and practicability, ensuring real-world viability of your intervention
- For really robust findings, you may also want to conduct a power analysis. This provides an estimate of the number of participants needed for you to feel confident in your results.
Step 2: Running a trial
- Wherever possible, you should keep conditions across your trial constant, apart from the intervention itself – in order to properly isolate its effect.
○ How? Make sure intervention and control groups are treated in the same way and keep the method of data collection the same
- Given that your RCT will be conducted alongside daily business tasks and activities, you may need to be economical with time without sacrificing the validity of the trial.
○ How? Use a data set that is already routinely collected and observe changes that result from the intervention; Use a computer programme to perform random allocation (this also has the benefit of reducing bias in your trial); Establish a core project team who are responsible for the running of the trial
- Last but not least, ethics are an essential consideration for any research trial, (regardless of whether it’s medically- or business-focused). You’ll probably need to seek ethical approval from internal stakeholders or participants, and ensure you can store the data confidentially.
So, there you have it: our introduction to designing and running RCT’s in business. Although their origins are medical, RCTs are gradually helping to solve modern behavioural challenges that are experienced in a variety of contexts. Please get in touch with us if you’d like any more advice.
1 Bothwell, L. E., & Podolsky, S. H. (2016). The emergence of the randomized, controlled trial. New England Journal of Medicine, 375(6), 501-504.
2 Streptomycin in Tuberculosis Trials Committee (1948). Streptomycin treatment of pulmonary tuberculosis. British Medical Journal, 2(4582), 769-782.
3 Concato, J., Shah, N., & Horwitz, R. I. (2000). Randomized, controlled trials, observational studies, and the hierarchy of research designs. New England Journal of Medicine, 342(25), 1887-1892.
4 Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6(42), 1-11.
HUMAN BEHAVIOUR – AND HOW
TO CHANGE IT
For more speak with Davina (Client Services Director) or Oliver (Founding Director)
+44 (0)843 289 2901