It’s a well-established fact that clinical trials are challenging. They are expensive and suffer from multiple difficulties such as patient recruitment and retention, unexpected changes to protocol and trial design issues.
We explore some of the major pitfalls that beset sponsors and research professionals when designing clinical trials, and also take a look at how those challenges can be overcome.
1. Not Formulating the Right Research Question
One of the first steps for a successful clinical trial is coming up with the right research question. Yet the process is often difficult, says Dr. Wilson Fandino at St. Thomas’ Hospital in London. Getting the question wrong can lead to issues later on in the trial.
For instance, methodological difficulties can arise during the operational components of the study if the question is not properly chosen. So researchers need to identify a clinical problem to be solved and set about finding the right question to achieve this.
The temptation, Fandino argues, is to formulate several research questions — but this is unwise as it could require a different trial design and sample sizes would need to be much larger. The best practice is to focus on a primary research question instead.
Once the researchers have evaluated whether a study is needed to answer the research question and have reviewed the clinical literature, they should narrow the question. Fandino gives the example of an effective a narrow question: “among children younger than 1 year of age undergoing elective minor procedures, to what extent the insertion times are different, comparing the Supreme laryngeal mask airway (LMA) to Proseal LMA, when placed after reaching a BIS index <60?”
The question is effective because it defines the target population, intervention, alternative treatment or procedure to be compared, a primary outcome and a time frame. This is also known as the PICOT approach — population, intervention, comparator, outcome, and time frame. Although effective, few use this method.
Reviewing 313 articles published in four anesthesia journals, researchers published by the Canadian Journal of Anaesthesia found that 96 percent did not use the PICOT approach. But they assert that PICOT is a helpful way of defining and stating a research question
Indeed, good research questions are as important as the methodology designed to test them, argue J. Andre Knottnerus and Peter Tugwell in an editorial for the Journal of Clinical Epidemiology. “Original ideas, creativity, clinical and scientific perceptiveness, extensive subject matter expertise and well-prepared questions and hypotheses are the starting point for scientific progress; methodology is the craft to harvest the resulting new knowledge. Both ingredients need to be present for research to succeed and make a difference.”
2. Not Engaging Patients Early On
Patient recruitment and retention are perennial problems in clinical research. Part of the problem is that sponsors and researchers do not engage patients early enough in the trial design process to know if the design fits the patients’ requirements. Early engagement, therefore, will improve the research design, says Matt Cooper, Ph.D., business development and marketing director for the National Institute for Health Research’s Clinical Research Network.
Patients have vital insight into their conditions that could help researchers shape trial design for the better. As Cooper writes, “they are the only people who can truly tell us if we are achieving the right balance between the ‘ask’ of the research and the burden of the disease.” It is this kind of thinking that shaped the CRN’s collaboration with Pfizer to create a more patient-led design initiative.
For instance, CRN organized meetings with patients and carers at the Alder Hey Children’s Hospital NHS Foundation Trust in Liverpool in the UK. Researchers designing the study found interesting results that shaped the creation of legal and compliance documents, says Sophie Evett, feasibility lead for Pfizer UK.
But eligibility criteria have also become an issue in research. The team at Clinical Trial University says eligibility criteria for phase III trials has increased to around 50, up from 31 between 2002-2012. This makes it even more challenging to find patients.
The result of too stringent eligibility criteria is often that trials end up studying patients who are not representative of the broader population that will ultimately use the treatments in the future, warns Nancy Magnier Boxx, director of quality assurance at the Atlantic Research Group.
Patient insight is crucial but some sponsors remain reluctant to include it in trial designs.
Sara Ray, senior director of research at healthcare social network Inspire, says the fear is that sponsors will encounter new knowledge that threatens the trial. While Ray acknowledges that is a risk, she says it’s riskier to ignore the patient voice. A better approach is to include patients as early as possible in the design phase.
3. Getting Sample Sizes Wrong
Population size is important. The right number of participants allows researchers to be precise in the estimates or detect differences if any between treatments exist, explain Vysaul Nyirongo, Mavuto Mukaka and Linda Kalilani in their article in the Malawi Medical Journal.
Too small a population and the results may not be statistically significant. But recruiting too many is costly, drains resources and is unnecessary to obtain statistically significant results. The researchers say the following factors need to be considered to get the right sample size:
- The sampling technique — random vs. cluster, for example.
- The variability in the population.
- The accuracy of the estimate or difference between populations required.
- The statistical model or test to be used for analysis.
When a study aims to determine the efficacy of different treatments or the severity of risk factors, selecting the right sample size is essential. Allan Hackshaw, professor of epidemiology and medical statistics at CRUK Cancer Trials Centre, says “the size of the study depends on the magnitude of the expected effect size, which is usually quantified by a relative risk, odds ratio, absolute risk difference, hazard ratio, or difference between two means or medians.”
When the true-effect size is small, the study needs to be large. Hackshaw explains that the size is needed to distinguish between a real effect and a random variation.
4. Trial Conduct Does Not Match Trial Design
The way a trial is designed and the way it is conducted are often not aligned. This is a challenge for clinical trials, according to Joanna Kelly, Barry Hounsome, Gill Lambert and Caroline Murphy at Trials. An important means of overcoming this is to select the right trial team.
The best type of team works towards a common goal under an insightful leader. The trial investigator that sources the right team will lay the foundations of success for a clinical trial, the authors argue. Key members of the team include the lead investigator, co-applicant statistician and staff, and senior specialist support staff. The lead investigator and co-applicant statistician will need to develop a strong bond based on trust. They will be reliant on each other’s expertise to make the trial succeed.
The lead investigator will also need to draw on the expertise of others such as academic colleagues, sponsors, pharmacists and statistics help desks to make the right decisions. Getting the team composition wrong can be disastrous for effective clinical research and cause the design and operations to be out of sync.
5. Making Trial Design Too Complex
The more complex the trial design, whether adaptive, basket or umbrella, the more complex the protocol. So researchers are well served by removing any unnecessary complexity. Yet they can still make a trial complex when they try to obtain desirable but not necessary data. It’s the difference between what they want and want they need, explains Susan Wiskow, senior clinical project manager at Regulatory and Clinical Research Institute.
Wiskow says to ask what is absolutely necessary to show the safety and efficacy of a treatment. What data is needed to ensure patients benefit and satisfy payers? Trial design must be steered by the pursuit of relevant and actionable data. Doing so will make trials more streamlined and cost-effective.
While this is not an exhaustive list of the common pitfalls in clinical trial design, they do point to some major concerns for sponsors and research teams. Each challenge details how best to avoid or resolve it. The common solution is careful planning before action, engaging patients and other key stakeholders and working together to develop a relationship based on trust.