The landscape of clinical research is undergoing transformative shifts. With advances in technology come improved capabilities to work with patients and their data, not just in a clinical setting but well after implementation. There’s a lot of potential here: faster and more comprehensive studies, a greater understanding of specific interventions, and improved outcomes for patients. However, researchers also need to understand the parameters and ethics around these tools before adding them to their workflow.

While these advancements aren’t new, they’re more ubiquitous and useful than ever before, notes Jamie M. Robertson, PhD, MPH, program director of the Foundations of Clinical Research and the Global Clinical Scholars Research Training programs at Harvard Medical School and senior research scientist at Brigham and Women’s Hospital.

“I think all these trends have been around for a long time,” says Robertson, “but now we’re finally at a point where the tools are more broadly available and more functional for use in application.” She highlights three trends that have the biggest potential to change clinical research in the coming years: artificial intelligence (AI), real-world evidence studies, and patient-reported outcomes. 

The Sweeping Influence of AI

Robertson and her colleagues are currently undertaking research to understand the benefits of artificial intelligence on clinical research, as the scope is not yet known. Technology experts are constantly coming up with better, more effective large language learning models, and Robertson says the potential is enormous. 

“AI will have the power to help with several aspects of clinical research,” she explains. “There are companies opening up every day promising to help with clinical research workflows, clinical practice workflows, and even clinical education workflows—how we can train the future workforce and replace some of the high-cost aspects of our work.”

Researchers should still exercise a healthy amount of caution, she warns. Beyond the ethical complexities of having generative AI produce research analysis and summaries, these tools can hallucinate information, over-generalize, and even provide incorrect instructions. 

“I think that AI can help speed up many of the processes that are tedious and challenging,” she says. “It can help us come up with code to do data analysis, and with potential scenarios. But it also underscores the need for people who are interacting with AI to be knowledgeable about the right and wrong within their work—to use their creativity and personal knowledge to problem-solve around the information it gives us.”

The Growing Value of Real-World Evidence Studies

When patients interact with interventions in real life, not just in laboratory conditions, it provides a deeper and fuller understanding of that intervention, says Robertson. “We get lots of good efficacy data when we do very controlled trials, but these real-world evidence studies are becoming an important part of how we do clinical research. We can see how an intervention plays out over months and years, with different groups of people than were originally studied and people interacting with variables outside that controlled setting.”  

There’s already a wealth of data via electronic medical systems, which store patient data from medical visits, as well as health insurance claims. Additionally, pharmaceutical companies now provide patient data in regard to their products, such as adverse events and prescription pickups. The Food and Drug Administration (FDA) has started allowing researchers to submit anonymized real-world evidence to complement clinical trials, which allows for a broader, more representative study of populations beyond the recruitment for a particular study. 

“For example, it might be easy to recruit large groups of white women in the United States to be part of your clinical trial studying breast cancer, but it’s unlikely that you’re going to be able to recruit enough men who have breast cancer: a small portion, but still an important part of that population. Similarly, our clinical trials might have some Black, Asian, or other individuals of color, but it might not be wholly representative of the population.” 

Supplementing clinical trials with real-world data is an important piece of the research. In other words, says Robertson, “How does this actually function when you don’t have people who are reporting into an investigator every week and hoping to be a good trial participant?”

Delving Deeper into Patient-Reported Outcomes

Asking patients to measure their own health status is not a new metric. “We’ve been asking about pain for a long time,” notes Robertson. “But for some, a certain level might be more agreeable for some people than for others, and may depend on different factors, such as what you do for work, what your lifestyle looks like, and so on.” 

Thus, the innovation around patient-reported outcomes is the increased effort and opportunity to understand how people measure their own health status, as well as the ability to tailor treatment based on that data. 

“What do patients actually want? It’s great that we’re able to shrink your tumor or control your blood pressure. But if that means that you’re unable to play with your children or walk your dog or write at your computer for long periods of time, it might be less exciting for you. That might not be a tradeoff that you’re willing to make,” Robertson explains.

Listening to patients when they describe their condition and talking about the impacts of either the condition or the treatment on their life is more important than ever. “In a chronic condition, for example, we’re trying to help lower pain—without giving you a bunch of side effects that’ll mean you still can’t function,” Robertson says. 

Designing Patient-Centric Clinical Trials

If there’s a single throughline to these innovations, it’s the increased centrality of patients in this work. Researchers should keep this in mind as they move forward with their trials, Robertson explains.

“People’s priorities have changed. I think a lot of people are no longer willing to go to the doctor and just accept a lower quality of care. Furthermore, the focus on quality and patient satisfaction is also helping to drive some of this. It’s no longer decisions being made from the top down,” she says.