A unique approach to trial design
Liam Smeeth, Professor of Clinical Epidemiology at the London School of Hygiene & Tropical Medicine, talks to us about the Statin WISE trial and why an N-of-1 trial design will provide a statistical powerful trial without the need to recruit thousands of patients.
The research - Trial overview
Statins are one of the most widely used drugs and very effective at preventing heart attacks and strokes, but some people stop taking them because of muscle aches and pains. However, there is limited trial data on statins and their effects on muscle pain.
In order to investigate whether there is a direct link between statins and muscle pain Professor Liam Smeeth and his team, at the London School of Hygiene and Tropical medicine have set up the Statin WISE Trial. This is the first of its kind to look at the relationship between statins and muscle pain. It is also one of the first trials in the UK to use an N-of-1 trial design.
A unique approach
An N-of-1 design is where a single patient is compared with themselves, they are the main comparison.
In the case of the Statin WISE Trial, participants will take statins for a period and then a placebo for a period in a random order and in a blinded way; neither the researchers nor the participants know what they are taking when. Participants will do this for 12 months over which time they complete a symptom diary. The pain scores that the participants’ record will be looked at to see if there are any excess scores during the statin periods. This will tell us whether it is true that statins cause muscle pain or not. The results are then pooled allowing them to be more generalisable.
Why this trial design?
One of the big benefits of using an N-of-1 design is the number of participants required. If this trial were designed as a conventional RCT thousands of patients would have had to be recruited. By using the N-of-1 design, only 200 patients need to be recruited to achieve the same statistical power and precision.
Challenges of this design
This design has not been used very widely and there is limited literature on it.
The topic area has been a big challenge as many patients feel that statins cause the muscle pain and so they are not keen to go back onto the statins.
MHRA insisted that a pre-trial screening test was done which was a complicating factor especially when the main aim was to keep the trial as simply as possible.
Advice for others
If you are planning on using this trial design, it is important to have clear data about timings. You need to know when you would expect the intervention to have an effect and once the intervention is stopped, when things to go back to normal, so that you can decide on the length of the treatment periods and design the trial.
You will need a placebo to run a blinded trial, which means there are more regulations and ethical approval that will need to be gained.
the match between trial group and placebo group is perfect, since it is the same person. Everyone on the trial benefits from the intervention being tested, nobody is left untreated for a long period. This allows much greater statistical precision, and a more powerful result from a much lower level of recruitment.