Bias – Is something that affects the reliability of a study, and could lead to the study results incorrectly favouring one treatment over another. Or, bias could lead to the study results only applying to a small sub-set of people with a condition.
A well-designed study will aim to reduce bias as much a possible so that the results of the study are reliable, reproducible and apply to the wider population of people with the condition.
There are many ways bias can be introduced into a study, for example:
- A researcher choosing when and who to invite to take part in a study, depending on how well they think they will respond to the intervention.
- A researcher choosing which group a participant is assigned to, based on which group they think the person might do better in.
- A researcher or participant knowing what group they are assigned to, and that knowledge affecting their answers to health questionnaires (either consciously or subconsciously).
Comparator – A comparator is something the intervention is compared with to find out how well the intervention works. A comparator might be the current recommended treatment for a condition, or it could be no treatment at all if no recommended treatment currently exists.
Feasibility study – A feasibility study is a small study to find out if it is possible and practical to test a treatment in a much larger study in the future. A feasibility study helps to find out if people are willing to take part and enough high quality data can be collected to answer the research question, before committing to the expense of running a large study.
Intervention – In a research study, an intervention is the treatment that is being tested. This could be a:
- new or existing medicine
- medical device
- type of surgery
- individual or group counselling
- exercise programme
- lifestyle change guide (e.g. weight management)
Interview – A discussion between a participant (or, a healthcare worker or other professional) and a researcher to obtain feedback on a specific health condition or their experience of taking part in or working on a study. Interviews are often audio-recorded so as the researcher can analyse the discussion using accurate notes.
Outcome – An outcome is something that is measured or recorded in all groups to decide if the intervention is better, worse, or the same as the comparator. An outcome should be something that can be measured robustly and is important to people with the condition.
Some examples of outcome measures are:
- the number of days spent recovering in hospital
- the time taken to be re-admitted to hospital
- how well a participant is able to perform a specific task
- how a participant feels about their health and wellbeing
Participant – A participant is someone who is taking part in a research study.
Randomisation – The process of assigning participants to a group (e.g. the intervention group or the standard care group) in a ‘randomised controlled trial’.
Most often, participants are assigned to a group using a computer programme so there is no chance of anyone influencing the group choice. However, sometimes participants might be randomised by a researcher randomly selecting from a pack of envelopes with the name of a group sealed inside. This method may be used in situations where access to a computer is not possible.
Randomised control trial (RCT) – A study in which participants are randomly assigned to one of two (or more) groups to fairly test whether an intervention (this could be a new or an existing treatment) is better than another treatment or no treatment.
Putting participants into groups at random helps create groups of people with similar characteristics, (e.g. age, sex, length of time they have been unwell) so as the intervention can be tested without being affected by ‘bias’.