Research methods

Researchers plan and carry out studies in a way that ensures results are accurate and not caused by chance. This means they have to follow strict guidelines. They also need to make sure their own – or the participants’ – ideas or beliefs about the research don’t unfairly influence the results. This section describes some methods used to make sure clinical trials are fair and reliable.

Learn more about:

• Randomised controlled trials
• How randomisation works
• Non-randomised trials
• Standard treatment and placebos
• Crossover studies
• Blinded studies

Listen to our podcasts on Coping with a Cancer Diagnosis and Making Treatment Decisions

Randomised controlled trials

Randomised controlled trials (RCT) are the best way to test if a new treatment is effective. This is because they help prevent bias. Bias occurs when the results of a trial are influenced by human choice, expectations or other factors not related to the treatment being tested.

Most Phase 3 trials and some Phase 2 trials are randomised. This means participants in the study are randomly assigned into two or more groups, or arms, and the results of the different groups are compared.

Test or experimental group (or arm) – This group is given the experimental intervention. Sometimes, the experimental treatment is given in addition to the current standard treatment.

Control group (or arm) – This group receives the current standard treatment for the disease.

When randomly allocated groups are compared with each other, it is possible to reliably work out which treatment is better. This is because researchers can be certain that the results are related to the treatment, and not to any other factors (see also Blinded studies).

How randomisation works

Non-randomised trials

In a single group (arm) trial, all participants receive the same experimental treatment. This method may be used for Phase 1 and Phase 2 trials, or where the cancer being treated is rare and it is hard to conduct a randomised trial.

Standard treatment and placebos

Standard treatment

This is the current most effective treatment or care given to people for their disease or condition. For example, the standard treatment for newly diagnosed early breast cancer is surgery to remove the breast tumour, often followed by radiation therapy, hormone therapy and/or chemotherapy.

In some cases, active surveillance or watchful waiting is the best currently available standard of care. For example, active surveillance is recommended for some early thyroid cancers that aren’t causing any symptoms and are considered to be low risk.

Placebo

This is an inactive or mock treatment made to look, taste or feel like the treatment being tested, but that doesn’t have any active (therapeutic) ingredients (if a medicine) or beneficial effect (if another type of treatment). Examples of placebos are sugar pills and saline injections.

A placebo is used to compare treatments to see whether the patients’ outcome is because of the actual treatment or because of other factors associated with being in the study. If the people given the experimental treatment show more improvement than those given, this provides stronger evidence that it’s the experimental treatment that is responsible. Participants will be told if a study uses a placebo, but will not be told which treatment they are receiving.

In cancer treatment trials, placebos may be used:

• together with the standard treatment, for example, one group receives the existing standard therapy plus the experimental treatment, and the other group receives the standard treatment plus a placebo
• on their own, when there is no existing standard treatment to compare against an experimental treatment.

Crossover studies

In these studies, participants in each trial arm receive their assigned treatment for a period of time before swapping to the other treatment. This enables all participants to experience all treatments, and helps to confirm which is the most effective. Crossover studies are often used when researchers feel it would be difficult to recruit participants willing to risk not receiving a promising new treatment.

Blinded studies

In a blinded study, participants don’t know which arm of a study they’re in. Some randomised controlled trials are called double-blind studies as neither the participant or trial team members know who is receiving the experimental or control treatment. In a double-blind trial, even the lead researchers only discover who is in each arm of the study at the end of the trial when the results are being analysed.

Blinding is used only when participants can’t tell the difference between the two types of treatment. It is not used when the control and experimental treatment are noticeably different – for example, it would be hard to disguise surgery and massage from the participant.

The aim of blinding is to reduce bias in the reporting of benefits and side effects. If you don’t know which treatment you’re having, the results are less likely to be influenced by your or your doctor’s thoughts. For example, if you or your doctor knew you were having the experimental treatment, you might report that you’re feeling better than you actually are because you believe you are receiving an effective treatment. If necessary for safety reasons, your doctor can find out what treatment you’re having by contacting those running the study.