Randomised control trials (RCTs) can be one of the most powerful tools to test whether something is making a positive difference. On 20 June 2013 we are hosting the launch of Randomise Me, a new website developed with Ben Goldacre, which will enable anyone to set up and run their own trial.
Randomised control trials (RCTs) can be one of the most powerful tools to test whether something is making a positive difference. On 20 June 2013 we are hosting the launch of Randomise Me [this site is no longer active], a new website developed with Ben Goldacre, which will enable anyone to set up and run their own trial.
What questions would you like to answer? For instance, ever wondered whether coffee gives you heart palpitations? Whether the reading app used in your classroom really does improve children's attainment? Whether your new marketing campaign is increasing volunteer recruitment? If you want to know answers, then run a trial on Randomise Me and find out.
Randomised Control Trials may sound complex, but they simply involve taking a group, such as a group of patients, children, schools, or others, splitting them into groups at random, and then giving one intervention to one group, and another intervention to the other. The differences between each group are then observed to see if one intervention has achieved its supposed outcome. They are commonly used in medicine, but are much less common in other areas, such as children's services, social care or education. Randomise Me is going to help remedy this.
Randomise Me is a free to use, online trials generator we have developed with author, Guardian journalist, and self proclaimed "nerd cheerleader", Ben Goldacre. Trials can be set up and run online, or people can use the site to structure the trial, and then download a dataset to complete. In both instances, Randomise Me will help to analyse and understand the data.
We hope that Randomise Me will be the RCT equivalent of Survey Gizmo or Survey Monkey, helping strip out the perceived complexities of the method and increase their use by a range of different individuals and organisations. Our ambition is that Randomise Me will enable anyone to set up and turn their own trial, helping to robustly test different issues and interventions in an agile and low-cost way. This is a great leap forward for anyone wanting to know the impacts of something in their lives.
Yet, RCTs are a contentious subject for some. One of these views is that it is "unethical" to "withhold" an intervention from the control group. We greatly disagree. Any method can be done badly or unethically, but this isn't the case here. Instead, this view presupposes that the intervention is beneficial - when testing that assumption is the whole point of the experiment. Secondly, often we are testing a new intervention: the 'control group' may simply continue with current provision. And thirdly, many programmes are not funded sufficiently to treat everyone: access may already be restricted to a select group. This rejection of random assignment techniques because of misunderstandings is a huge missed opportunity for improving our understanding of what is working.
We recognise that RCTs are not the answer for every project. Indeed, I frequently find myself saying "don't put method before question" when discussing evaluation strategies, yet I firmly believe, that trials should and could be used much more than they currently are.
The launch of Randomise Me is therefore a very exciting leap forward. Making it easier for trials to be set up and run will help to normalise their use in different fields, whilst greatly increasing our understanding of what is working - and what isn't. And the fact that this can be done in an easy to use, agile, low cost way, is even more of a bonus.
Randomise Me is being built by Better Data, a non-profit health IT company run by Ben Goldacre and Carl Reynolds. Website details will be available on 20 June 2013.
If you would like more details on Randomise Me contact Ruth Puttick at [email protected]