An introduction to clinical trial designs
Clinical trials are an essential part of the Australian and global health system. They provide medical practitioners with innovative new tools to prevent, identify and treat a wide range of diseases and conditions. This improves the health and well-being of people everywhere.
To be carried out successfully, clinical trials must be planned and carried out with considerable care. As early as possible in the process, it’s important to decide not only what you will be testing and who will be involved but how you will do it – or the trial design.
Let’s review some of the most common trial design elements, from control groups to randomisation, and consider how they may impact your research.
Clinical trial designs: A quick summary
- Each clinical trial should be designed based on the trial goals and participants.
- There are various trial designs to select from, including controlled, randomised, blind and crossover designs.
- You should consider your participants, including how many there are and who they are.
Read on to learn about clinical trial designs in more detail.
What is a clinical trial?
A clinical trial is a form of research that uses human volunteers, both patients and non-patients, to test new medical interventions. These interventions may be drugs, devices, surgical techniques or operations, diagnostic tests, or lifestyle programs.
Clinical trials seek to provide better treatment and care options than those already available. They help develop medical advancements that improve the length and quality of people’s lives and even help prevent deaths.
However, by involving living people, they are required to follow a large number of ethical, legal and practical guidelines. They also involve human researchers and must account for the impact of subjective bias.
The different clinical trial designs
Clinical trials are designed to be as safe and easy as possible for their participants and efficient operations for researchers. It is important that the results are accurate and trustworthy.
Below are some of the most common designs and design factors for clinical trials in Australia.
A controlled trial
A controlled clinical trial is a trial that uses a control group. Trials use multiple groups, including a control group, to minimise doubt over whether the treatment or other factors are driving results.
An uncontrolled trial, or a trial without a control group, is more commonly used in earlier trial phases when the focus is on safety, side effects and ideal dosage. Control groups are then introduced in Phases 2 and 3.
Participants will be allocated to one of two or more groups, also known as arms. For instance, if there are two groups, they are:
The test group or arm
This group is given the intervention being studied, also known as the experimental intervention.
The control group or arm
This second group is given the currently available standard of care for whichever disease or condition is being analysed. Alternatively, they may be given a placebo.
There may be multiple test groups or multiple test combinations. For instance, the test group above, or a second test group, might be given the new intervention in addition to the current standard treatment.
The researchers will closely monitor and assess results for both or all groups. If a test group is showing significantly better results than the control group, researchers may conclude that the new intervention, and not chance or natural recovery, is driving this improvement.
If the test group isn’t exhibiting serious side effects, the researchers may decide to end the trial at this point. They may alter the design of the trial or move onto a new phase so that more participants can receive and potentially benefit from the new treatment.
A randomised trial
If a clinical trial is randomised, or involves randomisation, this means that participants are randomly allocated to testing groups. This way, researchers can remove their personal bias or influence as a potential contributing factor to the test results.
Phase 1 and Phase 2 clinical trials tend to be non-randomised, meaning participants are in a single test group. This is also a common design for other forms of clinical research, like observational studies.
Randomisation is introduced with Phase 2 trials and is used in most Phase 3 trials. These phases test the safety and effectiveness of new interventions using hundreds or even thousands of participants. They are all randomly allocated into two or more groups, including a control group.
Computers and digital technology make this process possible. At the start of the trial, all participants receive a code number. These numbers are randomly allocated by computer to the different test groups.
Ultimately, researchers want to be confident that any differences in results between the two or more groups are due to the new intervention – not their influence as the organisers, or any other factors.
Clinical trials that combine randomisation and control groups are commonly known as randomised controlled trials (RCTs). This is seen as a highly effective trial design and is used widely in Australia.
A blind trial
A blind trial, also known as a masked trial, is also designed to minimise the impact of personal bias on test results. The participants and/or the researchers do not know who is receiving the new intervention and who is receiving a different treatment or a placebo.
In a single-blind trial, the researchers and staff know who is receiving the new intervention, but the participants do not. In a double-blind trial, neither the researchers nor the participants know who is receiving the control and experimental treatment. The researchers only find out once the trial has ended and the results are being analysed.
This blinded approach is only used when there is no discernible difference for the participants and/or researchers between the control and experimental treatments. It helps ensure that neither group is influencing outcomes with their knowledge or opinions of events.
For instance, sometimes issues arise when participants feel invested in the success of the experimental intervention. If they know they are receiving this new treatment, they may report more positive results than what they are actually experiencing, as a way of supporting the research efforts or pleasing the researchers or doctors involved.
When trials aren’t blinded, they are sometimes referred to as open-label studies. This approach is common when researchers want to compare treatments or understand the long-term use of a treatment by the intended patients.
A crossover trial
In crossover trials, the people in each trial group receive one treatment for a time and then have the chance to ‘cross over’ to the other treatment. This approach gives researchers more data on the effectiveness of the experimental intervention compared to the standard of care.
If participants register positive results with the first treatment, they won’t swap with the other group. Similarly, if the new intervention isn’t working as intended with the first group to try it, the second group won’t cross over.
Adaptive trial design
Adaptive clinical trials are flexible and efficient because they use ongoing results to make changes during the trial based on pre-set rules. This approach saves time and money, often needs fewer participants, and can be used in all phases of clinical research. For example, if early data shows that a particular treatment isn't working, the trial can be adjusted to stop testing that treatment and focus on more promising ones. Despite their advantages, these designs haven't been widely adopted due to a lack of understanding and some misuse of the term in the clinical community.
The key feature of adaptive trials is the ability to make planned changes, such as adjusting sample sizes, stopping ineffective treatments early, or changing how patients are allocated while keeping the trial's integrity intact. This flexibility leads to more efficient trials, helping to stop ineffective treatments sooner, needing fewer participants, and reaching conclusions faster. However, challenges like getting funding, ethical approvals, and clear communication need to be addressed to fully benefit from adaptive trials.
Types of comparison
Any controlled trial design with two or more groups will require an alternative treatment to compare to the new intervention.
The two most common types of comparison are the standard treatment of care and a placebo.
A standard intervention
Clinical trials often compare a new and experimental intervention with a standard intervention. This is a common and widely used drug, treatment, tool or other intervention that is accepted as the most effective option available for a particular disease or condition.
The standard intervention gives researchers a useful starting point for contextualising the results of their tests. If the new intervention produces better results, the trial may progress to further stages.
A placebo
Researchers also regularly compare a new intervention with a placebo.
A placebo is a harmless “dummy” substance or treatment. Often a sugar pill or saline injection, it is made to be indiscernible from the new intervention being tested, with the same look, taste or feel. However, it doesn’t have any active or inert therapeutic ingredients or beneficial effects.
Like the standard intervention, the use of placebos helps contextualise test results. If the new intervention produces better results than the placebo, researchers have an indicator of success that may move the trial forward.
Potential clinical trial participants are always informed ahead of time if a placebo will be used. Yet, it is common for placebos to be used in blinded trials, where participants do not know if they are receiving the placebo or new intervention. This helps to minimise the ‘placebo effect’.
According to studies on this effect, when people believe they are receiving an effective treatment and their symptoms or pain will be reduced, they are more likely to see improvements. Even reassurance from GPs or medical professionals can help people feel better. For researchers, this is an unhelpful effect as it can mask the accuracy of the intervention.
Other important factors to consider in your clinical trial
Two other important considerations for a clinical trial design is the size of the test group, or groups, and the eligibility of participants.
The number of participants
Researchers want to be confident that the success of new interventions is not due to chance. In addition to using design factors like control groups and randomisation, they also carefully control the number of participants.
The size of test groups is determined using mathematical calculations, with the goal of achieving statistically significant results. If ten or fewer people are tested, any improvements identified may not be seen as meaningful. But if hundreds or thousands test an intervention, and the majority see benefits, researchers have a clearer indication of its effectiveness.
Who the participants are
Another important part of designing a clinical trial is understanding who is taking part, and why. Researchers will follow a plan, known as a protocol, that describes best practices for the trial, including who is eligible for participation and the methods and procedures researchers will use.
It is common for the health status of participants, or other factors like age and gender, to influence these design factors. In many trials, the participants are patients with a disease or condition that a new intervention is designed to treat. The researchers will avoid giving them a placebo if this lack of effective treatment might put them at risk.
When and how do investigator-led research teams confirm their trial design?
Investigator-led trials, also known as investigator-initiated trials, are usually undertaken by health institutions or within an academic setting. They are funded through grants or sponsorship from government agencies and departments, research groups, foundations or charities.
Conversely, industry-led trials are initiated and funded by commercial entities like pharmaceutical, biotechnical or medical device companies.
Both investigator-led and commercially sponsored trials will involve multiple detailed stages for planning, initiation and completion. They may test similar interventions and involve similar phases and designs.
However, investigator-led trials are more likely to address more niche research questions or to test treatments for less common conditions. They also typically have less funding compared to commercially sponsored trials, which can affect how long they run for and how many people are involved.
If you are undertaking an investigator-led clinical trial, here are the two key points at which you may decide on your trial design.
Trial planning
At this early stage, you will devise both a trial plan and protocol. Your trial plan will be a top-level summary with a wide range of information, including your proposed design and methodology. It will also include overall objectives and procedures, resource allocation and budget, data and risk management, and timelines and milestones. Building on this summary, your trial protocol will be a comprehensive document that outlines the above information and other factors, including participant eligibility and ethics and regulatory considerations.
Trial initiation
This is the second stage of your trial before you officially begin and complete your research. You may have an opportunity to carry out a pilot study or trial. Like a beta test, this pilot will give you a chance to refine various aspects of your design and procedures before the full-scale trial begins.
Frequently asked questions
A number of regulations apply to clinical trials in Australia, including national and international ethics guidelines, codes of conduct, and laws. This regulatory framework is designed to ensure research integrity and to protect the interests and well-being of trial participants. All clinical trials in Australia must also be approved by an ethics committee before they can proceed. This is usually granted by one of approximately 200 Human Research Ethic Committees (HRECs) in organisations around the country.
A useful starting point may be the Australian Government’s Australian Clinical Trials website, and in particular, the page for researchers titled ‘Starting a clinical trial’. This page includes several factors that researchers should consider when designing a clinical trial, including questions about the proposed research question and whether the trial can provide a clear answer. This page also includes useful questions and resources regarding participant and staff planning and legal requirements.
