On 6 October 2016, the IMI GetReal project hosted a webinar “Identifying and addressing the operational challenges of pragmatic trials”. The webinar showcased GetReal’s Work Package 3 (WP3), providing guidance on the optimisation and feasibility of pragmatic trials. The webinar also presented PragMagic, a decision support tool being developed by WP3 to help researchers design pragmatic trials.
Rick Grobbee, Professor of Clinical Epidemiology at the University Medical Center Utrecht, started the webinar by discussing the uncertainties surrounding a new drug at the time of launch despite data from Randomised Clinical Trials (RCT) needed to get regulatory approval.
Patients included in phase III clinical trials may not be representative of the population to which the drug is eventually prescribed, and individual patient characteristics can modify the response to drugs, changing how a drug works under usual care conditions compared to RCTs. This is where Real World Evidence (RWE) can help, as RWE reflects the entire spectrum of patient responses.
However, how can we use RWE to assess safety and effectiveness of the drug as it happens in usual care to usual patients?
Rick continued the webinar by describing how observational studies aim to be closer to the real-world by studying treatment strategies while explanatory trials (ETs) study the effects of a medicine in a more isolated and standardised setting. He introduced the concept of Pragmatic trials (PT) where – like in traditional explanatory trials (ETs) – randomisation into treatment groups is applied to assure that there is no prognostic incomparability between patient groups. He emphasized how he regards randomization as an important design element as “in real-life, drug use is not random, patients with more severe disease will use drugs more often, patients cannot tolerate a drug and will not use it; as a consequence there will be bias in everything we measure.”
Apart from the randomization, a PT aims to be conducted in a similar manner as an observational study, trying to investigate treatment outcomes as realistically as possible in daily practice.
To illustrate the concept, the webinar highlighted two ongoing pragmatic trials: The Salford Lung Study is investigating the use of a new inhaler for COPD against the standard of care. An important feature of the study is the linkage to electronic health records (EHRs) for outcome measurements. The SCOT trial is a post-launch pragmatic trial featuring a broad inclusion via GPs, aiming at wide generalizability and using EHRs as a source for outcome information.
Professor Grobbee concluded his presentation by saying, “there is a true need for more pragmatic trials in the world. RWE is more needed than ever to guide decisions of policy makers, payers, physicians and patients. But the design implications are not always straightforward.”
In the second part of the webinar, Dr. Mira Zuidgeest, assistant professor at the University Medical Centre Utrecht, introduced the decision support tool PragMagic, which is being developed by WP3. PragMagic is developed to help study teams designing a pragmatic trial to understand the complex interplay between design features which make a trial more pragmatic, their possible operational challenges and the impact on factors like generalisability, operational feasibility, ethics, validity, precision, and acceptability by stakeholders.
For example, the recent Retropro and eLung trials were conducted in a primary care setting to reflect daily practice and increase generalizability of the trial results. 59% of GP sites showed an interest in participating, however, due to operational challenges such as required GCP training, less than 4% of the sites actually recruited patients. The participating GPs differed from those not participating and patient recruitment was hampered, negatively affecting both feasibility of the trial and generalizability of the trial results.
According to Dr. Zuidgeest, PragMagic aims to help trial designers and decision-makers in the design and planning of pragmatic trials, giving them insights to the consequences of design choices and allowing them to visualize the interplay between the various design options, operational challenges and implications. While PragMagic will not be a decision-making tool, it could help balancing the intended generalizability of pragmatic trials with a valid and feasible trial design.
The first version of PragMagic will be available at the beginning of 2017 and can be piloted during the GetReal closing event in Brussels on the 24th of November.
Iris Goetz, MD MSc Research Advisor Epidemiology at Eli Lilly joined the extensive moderated and open Q&A session that concluded the Webinar. Numerous questions regarding the practicalities of pragmatic trials where discussed, including the selection of usual care and the legal frameworks of conducting PTs.
You can view the entire webinar here
Download the Presentations here
 Van Staa et al. Health Technol Assess 2014;18:1-146