What are Randomised Controlled Trials?
A randomised controlled trial (RCT) is an experiment carried out under highly controlled conditions. It uses the same measurements over time, comparing two or more groups of people. Each group will undergo a different treatment or therapy, and the aim is to see whether these lead to changes. Researchers usually compare those having treatment with a control group who isn’t having treatment.
Whether people end up in one group or another is decided randomly. In some RCTs those receiving the therapy don’t know whether, or which therapy they’ve having. This is known as a single blind RCT. In some, the researchers don’t know which therapy, if any, is being given either. This is known as a double blind RCT. With comparisons of medicines this is more easy to do, as you can’t tell what is in a pill from its appearance. However, with a treatment like psychotherapy it’s not possible to keep the sort of therapy hidden. This is because psychotherapy is all about conscious interaction.
The RCT is seen to be the strongest form of scientific evidence to show cause-and-effect. It can show how effective a therapy is, and also how safe it is. Whether a condition gets worse or improves as a result of the therapy can be made clear through this type of study.
Many practical considerations need to be taken into account when designing, conducting, and analysing the results of an RCT. The numbers of people in each ‘arm’ may be quite high in order to be meaningful. As a result they tend to be quite costly.
RCTs and Cognitive Analytic Therapy
Randomised Controlled Trials (RCTs) are considered by many to provide the strongest and most convincing form of research evidence. Despite the challenges described already, a number of smaller randomised controlled trials have been completed which study CAT for groups of people with the same medical or psychiatric diagnoses.
The results of these have been encouraging, showing that CAT may be a helpful therapy for people with a number of conditions. These include those receiving a diagnosis of eating disorder and also bipolar disorder. There has been RCT evidence supporting CAT for people with complex difficulties who have attracted a diagnosis of borderline personality disorder. (Emotionally unstable personality disorder is another term for this cluster of difficulties.) There have also been encouraging results from an RCT carried out with people managing diabetes.
So why can’t you just do more research like this?
There are obstacles in the way of building up a larger bank of RCT evidence for CAT, for several reasons.
Less emphasis on diagnosis in CAT
As outlined already, therapists use CAT to work with people across many different medical and psychiatric diagnoses, and often when no diagnosis fits. Rather, repeating patterns and difficulties in relationships are a common theme. The individual’s target problems are more of a focus than conditions that can be measured using standardised questionnaires. So comparing large groups of people who have been given the same diagnoses can be more difficult. And as each target problem is unique to that person, they can’t be compared easily. Meaningful measures of change will differ from one person to another
Smaller sample numbers
A further limitation is that CAT is it not as commonly available in NHS services as some other forms of therapy. As a result it is more difficult and costly to find the numbers of people required for an RCT. Studies have to be carried out over several different services, which can add to time and cost.
CAT is not a manualised approach
CAT doesn’t follow a manual for therapy, like some therapies do. The general approach may be the same, following the same stages and phases. However therapists adapt what they do to match what a person needs. They “push where it moves” and adapt their approach according to what works best for the individual. They also draw upon a range of therapy skills to help support change in the middle phase of therapy. As a result, one CAT therapy may look quite different from another. This too makes them harder to compare in an RCT.
So how can anyone be sure that the therapy is actually CAT instead of something else?
To help assess whether the therapist is doing “satisfactory CAT”, a tool called the Competence in CAT (CCAT) was developed. The CCAT helps assess how therapists perform in ten domains of competence. These domains are based on more detailed descriptions of each of the ten areas of therapist performance.
The CCAT is used in both training and research. For research purposes, the CCAT can be used by raters to score recordings of CAT sessions. This helps researchers know whether the therapist is using CAT skillfully. Using the CCAT is a check for researchers to know if the therapy was satisfactory CAT.
More recently, a more detailed Competence Framework for Cognitive Analytic Therapy was developed. You can see more about CAT's Competemce Framework at the UCL website at this link.
Recent research trials
A study completed recently looking at patient preference in guided self help for anxiety using materials based on either CAT or CBT (cognitive behavioural therapy). This was a type of RCT called a “partially randomised patient preference trial”.
We thank Dawn Bennett, Samantha Hartley, Steve Kellett and Peter Taylor, for helpful contributions to this page.
Read more about single case experimental design
Evidence Through Large Groups – Randomised Controlled Trials by ACAT Public Engagement Team
CC BY-SA 4.0 
