July 22, 2020
Clinical trial management software in 2020 and beyond
The pandemic did bring about a powerful positive phenomenon—an unprecedented level of cross-nation collaboration in search of suitable vaccines and treatments. Here’s the figure showcasing the number of COVID-19 trials registered worldwide as of July 13, 2020:
And yet, it has posed a serious challenge—the need to streamline and standardize diverse clinical trial practices to let all participants reap the benefits. This requires well-balanced clinical trial management software (CTMS), which may call for combining healthcare software development with business intelligence technologies. This is not to mention the need to cope with the common clinical trial challenges, which are here to stay. We’ll explore these matters in detail.
Clinical trials have never been easy to implement. They are time- and effort-intensive with results that are not always predictable. This year, the traditional set of challenges has been further augmented by the pandemic. So what are they now?
The studies vary greatly in terms of the number of participants and duration. However, the costs are stably high, starting from $10 million and climbing up to hundreds of millions. According to the JAMA Internal Medicine study, in 2018, the costs of a trial ranged from a modest $2.1 million for a trial involving just 4 patients to test a drug for a rare hereditary disorder to $346.8 million for a new cardiovascular drug trial, the number of patients unknown (presumably high). The research points out that a trial amounted to $41,117 per patient on average.
In 2019, about 40% of studies failed to onboard the target number of participants, and the trend is still on. But recruitment is not the only problem—for the study to pass all the phases successfully, patients should stay in for a relatively long time. Unsurprisingly, at least one-fifth of participants stably drop out along the way:
It doesn’t end there. Researchers are very likely to lose another 10% of patients, as the average clinical study dropout rate is 30%. Why do people choose to leave? The reasons vary from the lack of improvements and condition deterioration to inconvenience with accessing the trial site, with the latter also driving costs up.
We’ve been accustomed to believing in the pharmaceutical axiom: successful drug development doesn’t happen overnight. Even in 2018, it took 2.5 to 4 years to work out a new drug:
However, in the pandemic times this duration is not a good fit, as even 2.5-year-long cycles mean many lost lives. And here comes the first pandemic-induced challenge—the need for lightning-speed drug development.
Pharma companies and regulatory bodies do their best to adapt—the FDA has already issued dozens of simplified approval guidelines for COVID-19 vaccines, while major pharmaceutical companies have made an effort to drastically reduce the 10-year-long vaccine development cycle to just 12 months. Nevertheless, the efforts haven’t been very fruitful so far, and experts are not too positive about the progress either:
If the goal was to optimize the likelihood of figuring out the best treatment options, the system is off course. Too often studies are too small to answer questions, lack real control groups, and put too much emphasis on a few potential treatments.
Head of Clinical Policy and Strategy at Verily Life Sciences and Google Health, Ex-FDA Commissioner
Yet, other studies steadily bring insights. This is the case with the UK-based RECOVERY trial. Though the study has been on for three months only, the researchers have managed to cross out two popular COVID-19 therapy drugs as ineffective and prove the previously unknown benefit of another medicine.
So why some trials go south while others prove efficient? Here’s another expert’s opinion:
It’s a huge amount of wasted effort and wasted energy when actually a bit of coordination and collaboration could go a long way and answer a few questions.
Professor of Medicine at Oxford University, Lead Researcher in RECOVERY
Collaboration and coordination are what CTMS can help with. Let’s see how it works in the modern environment.
Clinical trial management software is a complex system that uses robotic process automation (RPA) to speed up or facilitate some processes, provide quality analytics, including predictive one, and allow for secure data storage and management. Clinical trial management is one of the most promising fields for automation in healthcare. Smart automation can help clinicians and researchers solve trial challenges, and patient recruitment is number one.
Surprisingly, patient recruitment for clinical trials is often a manual process. Researchers have to explore available electronic health records to detect suitable patients. Of course, studies differ in scale and the needed number of participants, but this is still time-consuming and error-prone: oftentimes, manual screening results in missed or erroneously assigned patients, which may be even more harmful. Besides, strict recruitment deadlines are still there. Obviously, in times of a pandemic, this mode of recruitment can’t address the urgency. Luckily, there are other viable alternatives.
To facilitate manual patient selection, researchers usually choose between two strategies: outsourcing and automation. Clinical researchers may delegate patient selection to qualified agencies. However, among obvious benefits such as rapid search with tried-and-true methods, this strategy has its cons—extra cost and the inability to establish a good rapport with patients due to a lack of pre-trial interactions. This may lead to dropouts in the course of a trial.
Automation with the help of machine learning in healthcare assists researchers to overcome the recruitment challenge above. The system may select patients, but researchers still need to contact them at some point before the trial. An AI-driven CTMS was introduced at the emergency department of the Cincinnati Children’s Hospital Medical Center. It screened the EHRs, selected those patients who met the trial requirements, and displayed the results on a clear-cut dashboard, thus leveraging healthcare BI to facilitate reporting and informed decision-making. In 2018, the researchers evaluated the system and found out that the screening time dropped by 34% compared to manual recruitment, which helped meet the recruitment timeline and even save time for other activities.
BI is not the only ML-run solution that can find its way into CTMS. Clinical researchers started to make use of predictive modeling in healthcare, which can also help them with patient onboarding, just in another way—a strategic one. Applying predictive modeling at the very beginning, researchers can precisely estimate the time needed to enroll participants from certain locations. Modeling also helps measure the onboarding progress as the trial goes on, which allows managing the process efficiently and keeping costs stable.
There’s another technological solution uniting ML, statistics, and data mining—predictive analytics. Researchers have started to use it to visualize and mitigate potential risks in ongoing trials and speed up approvals. The technology was used in a study by AI & big data researchers from the Massachusetts Institute of Technology (MIT). The team developed complex algorithms that took into account over 140 features of then ongoing drug development trials to predict their outcomes and the likelihood of approval. The tools reached the predictive measure of 0,78 for phase 2 approvals and 0,81 for phase 3 approvals respectively.
This approach helps reduce uncertainty and exorbitant costs, stimulating well-informed investments. What’s more, during a pandemic, it acquires yet another value—the approach may help reduce the financing of inefficient trials and boost investment in those that do yield results.
Trial participants don’t like to travel to trial sites, and it’s not a whim. In fact, about 70% of potential participants live over 2 hours away from a trial site. This inconvenience causes them to drop off early, as site visits may take much time, effort, and money. Luckily, there appeared an alternative—siteless trials, which may be the best choice in the present epidemiological situation. So how does it work?
To be efficient, siteless, or virtual, trials need a powerful cloud-based CTMS, which allows each participant, researcher, and clinician to connect, upload, consult, and review the needed data. This hub should provide for secure data storage and management, as well as drive insights for non-tech specialists. To make it all possible, siteless trials need to use a range of technologies, from mobile apps and remote patient monitoring to robust analytical solutions and telemedicine tools. The latter is to ensure seamless communication between researchers and participants. But is it a win-win strategy? Well, it depends.
First of all, this trial arrangement leaves adherence to patients, who differ in their levels of responsibility. While some patients play strictly by the rules, others take their medicines as they please and don’t bother to report a missed dose. On the trial scale, such an attitude results in incorrect or incomplete results.
There’s also another source of erroneous data looming around—unpredictable accidents. For instance, a remote monitoring device may break down or internet connection may be disrupted. As a result, valuable study data gets hampered. Besides, there are concerns about healthcare data security, as personal and biometric patient data is processed in numerous locations. That’s why additional CTMS security measures, e.g. two-factor authentication, may be needed.
In addition, according to Ken Getz, Director of the Tufts Center for the Study of Drug Development, whatever the economic benefits, patients don’t want to be “in a trial vacuum,” left all alone in the impersonal digital environment. For this very reason, researchers now opt for the hybrid trial model. It combines remote data collection with clinical site visits, which allows the team to add personal touch to researcher-participant interactions. For example, VirTrial developed a customizable platform that combines email, text, and video to streamline communication between all involved parties. Still, it looks like other vital functions can still be performed only on-site, which makes this CTMS not completely suitable for virtual trials.
Though siteless trials are still under development, they are there to remain, and the best proof is their acknowledgment by prominent regulatory bodies—the FDA and the Medicine & Healthcare products Regulatory Agency (MHRA) have already developed the guidelines for running virtual trials, which calls for further CTMS enhancements.
According to the 2020 Clinical Trial Management System Marketing Trends survey by Medgadget, CTMS development efforts have been centered on leveraging big data solutions, and this focus is likely to prevail over the forecast period.
In its turn, in The Future of Clinical Trials, Accenture predicts further digitalization of trials to the point when completely virtual trials may become mainstream. This presupposes a new step in the CTMS evolution: directing patients, their families, and clinicians to relevant trials through the use of intelligent agents. Driven by AI and ML technologies, these agents will solve the onboarding issue completely. Besides, data management processes will be simplified. The pace of AI development will soon make real-time data capture a reality. What’s more, the data will be processed by intelligent agents and connected devices, which will reduce the rate of human error.
The approach to data security and ownership will also change. All actors (participants, clinicians, and even devices and clinical systems) will manage to upload and send data across the network via decentralized applications operated by multiple users in a secure environment built on a healthcare blockchain.
The present developments enabled researchers to visualize two powerful trends of clinical research—global studies and personalized (precision) research driven by universal data standards. With global studies in place, this will mean instantaneous drug approval worldwide.
Clinical trials are a complex task ridden with many bottlenecks—ever-growing costs, complex patient enrollment, and unstable retention, often disrupted by participants’ wants and whims.
Yet, there is a way to solve these challenges at least in part by introducing powerful clinical trial management software (CTMS). Capable of intelligent automation, this system can help streamline and speed up workflows, which in clinical trials often helps upscale the overall efficiency and avoid extra costs. Even if CTMS doesn’t address all the challenges, it can still significantly reduce their impact.
Learn how healthcare automation with NLP, RPA, predictive analytics and more intelligent tools frees up medics to treat more patients, faster.
Providers can see a range of benefits from healthcare BI. Find out how to unlock them with Itransition.
Find out how machine learning personalizes patient care, alleviates pressure off doctors, and opens opportunities for innovative medical research.
Learn about the rules to follow and misconceptions to avoid while implementing predictive modeling in your organization.
The array of patient engagement technologies is growing steadily. But how to choose a solution that works? Check out our guide.
Clinical decision support software speeds up decision-making, and yet its adoption lags behind. We look into the ways to facilitate it. Learn more.