Clinical Trials

Feelings about being allocated (randomised) to a treatment group

In some trials (especially Phase 1 and Phase 2 trials) everyone taking part will get the new treatment or intervention. (See ‘Non-randomised trial designs and other studies’).
 
However, many trials compare a new treatment with the standard or usual treatment by setting up two groups of people. One group of people, the experimental group, is given the new treatment. The other group, called the control group, is given the standard treatment. If no standard treatment exists, the control group may not be given any specific treatment or may be given a placebo. Some trials compare more than two groups, perhaps comparing a new treatment with two or more well-established treatments. (Sometimes you may hear the experimental group called the ‘trial group’ or ‘intervention group’. This can be confusing, as all the groups, including the control group, are part of the trial, and people in the control group may also be given an intervention, in the form of the standard treatment).
To make sure that each group contains a similar mix of people, many trials are ‘randomised’. This means that people are allocated at random to one of the groups in the trial, often by using a computer programme. When people are randomised they have an equal chance of being in either trial group. Random allocation helps ensure we are comparing two very similar groups of patients, so if one group does better than another, it is very likely to be because the treatments being compared have different effects, and not because of differences between the people in the groups. The helpful UK Clinical Research Collaboration booklet on Understanding Clinical Trials explains these and other terms (see Resources).
 
Many of the people we talked to had been invited to take part in a randomised clinical trial or RCT. Two linked issues crop up in discussing people’s feelings' firstly, how well they understood generally what randomisation is and why it matters, and secondly how they felt about it in their own case. (See also ‘Information and questions’).
 
A lot of research evidence suggests that staff may find it hard to explain randomisation and that the idea confuses or worries some patients. They may feel that terms such as ‘the toss of a coin’ or ‘pulling it out of a hat’ sound too trivial for what is to them a very serious issue - even though patients we talked to used these terms themselves. One difficulty is that often people would prefer to be in one group rather than another.
Wendy was not unusual in having quite a strong preference to be in the group receiving the new treatment – in her case to get the extra drug, which she hoped would improve her chances of beating cancer. Others had a preference to be in the control group. Ronny was in a trial comparing three options (active monitoring, radical surgery or radiotherapy) for prostate cancer. He agreed to be randomised, but thought he might have dropped out of the trial if he had been allocated to the radiotherapy group. (See also ‘Thinking about withdrawing from a trial’).
Some people’s preferences were so strong that they decided they could not take part at all, especially if they knew they could get the treatment they wanted outside the trial (see also ‘Deciding not to take part’ and ‘Feelings about being in a placebo-controlled trial’). Randomised trials are done when we don’t know which treatment is best, in other words when the relative merits and disadvantages of different treatments are uncertain. It is important to realise that in about half of trials, the new treatment will turn out to be better, but in the other half it will turn out to be worse. If, in spite of the treatment uncertainties that the trial has been designed to address, people would strongly prefer to have one of the treatments being compared, they should not volunteer for the trial.
Of course trials are done because there is a possibility that a new treatment will be better than the standard or control treatment, and it may have already been shown to work for other conditions or groups of patients. However, the trial may find the new treatment is no better. Amanda felt more needed to be done to explain to people that trials are run where we genuinely do not know which treatment is best. This might help reassure people who have a preference and are worried about being randomly allocated to a group.
In some ways, random allocation can help people by removing the need to make a difficult decision. As Ronny (above) argued, letting the computer allocate him “would take the decision away from me.” However, this was based partly on a belief that the computer was calculating the best treatment for him based on his personal medical details, which is not in fact the case.
 
Alex understood that the decision was random, and was happy with this as a way to make the decision when doctors were unsure what to recommend.
Gill also commented:
“If you’re randomised into a trial where there are three different methods, or surgical techniques… as long as you’ve got a reasonable outcome you can live with you don’t mind... But you’d have to have that feeling that you didn’t quite know which one you wanted to choose.”
 
Lester agreed that people need to feel confident that the trial really is on a topic where we do not know the best treatment. But that in turn depends on whether patients trust their doctor. (See also ‘Deciding not to take part’).
Although no one can be entered into a trial without their consent, some people we talked to trusted their doctor to tell them all they needed to know and were not very interested in the details of the trial, whereas others wanted to gather much more information. There is considerable debate about how much information is ‘enough information’ to be sure consent is properly informed. (See also ‘Information and questions’). Elvan did not remember being randomised or that the trial he was in was comparing one group with another, but he was happy to trust his doctor that going into a trial was a good thing for him to do.
Very few people said they genuinely had no preference for one treatment rather than another. As Lester suggested, “Randomisation is difficult the more serious the issue is... If it’s something like about my survival or a big element of my quality of life, randomisation is a big ask.” Others agreed that randomisation was easier to consider in trials where someone was not seriously ill or the differences between treatments were small. However, Jayne was comfortable with the idea of randomisation even in the case of serious illness, perhaps partly because she was studying evidence-based medicine.
More commonly, there were several people who had a mild preference for one group but did not really mind if they ended up in another group. It is worth remembering that the control group is usually the best current treatment, if one exists, and it is what you would receive if you did not enter the trial.
A few people wondered why we could not have trials where people are allowed to choose which treatment they want and then the outcomes are compared.
There are trials called ‘patient preference trials’, in which patients with treatment preferences are allowed their desired treatment, and those without strong views are randomised as usual. However, in analysing such trials it is not clear how to allow for the fact that patients who choose a treatment may not be comparable with other groups.
Some people wondered about ways in which trials might be affected by the types of people who are recruited. Danny, for example, discovered that trial appointments might clash with summer holidays and that she might be excluded as a result. She commented, “Now, it raised the question with me… are they just trials for people that can go up there at the drop of a hat, or are retired and sick, or unemployed?” Gill wondered whether the way people were approached or the kind of information given out might make some people more likely than others to join a trial. However, in a randomised trial this will not affect the result, because the different types of people will be distributed equally between the groups.
 
Amanda and Rosamund reflected on the importance of well-designed trials with high numbers of participants. (The number needed in each trial – the ‘sample size’ - to get reliable results will depend on the condition, the treatment being tested and how big or small an effect the treatment is expected to have. See also ‘Different types of trial’).

Last reviewed June 2013
Last updated July 2011

Feedback

Please use the form below to tell us what you think of the site. We’d love to hear about how we’ve helped you, how we could improve or if you have found something that’s broken on the site.

Make a Donation to healthtalk.org





Find out more about how you can help us.

Send to a friend

Simply fill out this form and we'll send them an email