Why the future of evidence generation must be designed around how specialty practices actually work.
By Tom Sanders
"Innovation only matters when it becomes a part of care."
Clinical research and clinical care have always shared the same ambition: to improve the lives of patients. Yet for much of modern healthcare, they have operated in parallel rather than together. One has been optimized to answer scientific questions through carefully controlled protocols and structured data collection. The other has been optimized to care for people whose lives rarely unfold according to a protocol.
For years, we've accepted the tension between those two worlds as inevitable. Research required additional infrastructure. Clinical care required flexibility. If a specialty practice wanted to participate in research, it was generally expected to take on more work, more documentation, more coordination, more staff, more administrative oversight. That was simply the cost of participation.
For a long time, that assumption was understandable. Today, it deserves another look.
Specialty care has changed dramatically over the past decade. Practices are caring for increasingly complex patients while managing workforce shortages, growing documentation requirements, expanding referral networks, and mounting pressure to improve measurable outcomes. Every additional responsibility competes with the same finite resource: the attention of clinicians and care teams.
At precisely the moment healthcare needs more real-world evidence, many of the systems designed to generate that evidence continue to assume providers have capacity they no longer possess. That isn't a staffing problem. It's a design problem.
The difference between scientific rigor and operational friction
None of this diminishes the importance of rigorous research. Carefully designed studies remain the foundation of medical progress, and the discipline required to generate trustworthy evidence should never be compromised. But scientific rigor and operational complexity are not the same thing.
Over time, research has accumulated layers of administrative activity that extend well beyond the scientific questions being asked. Additional workflows are introduced to support documentation, coordination, patient identification, communication, and follow-up. Individually, each step is reasonable. Together, they can become difficult to absorb within the rhythm of a busy specialty practice.
This distinction matters because the greatest barrier to broader research participation may not be a lack of interest. It may be the accumulation of friction.
Healthcare professionals rarely resist research because they question its value. Most understand that today's standards of care exist because previous generations of clinicians participated in studies, contributed data, and helped answer difficult questions. What they resist is the growing distance between how research is organized and how care is actually delivered.
Clinical reality is not a controlled environment
Anyone who has spent time inside a specialty practice understands that clinical care follows a different logic than a research protocol. A physician's day is shaped by unexpected referrals, procedures that take longer than anticipated, incomplete medical records, anxious patients, urgent phone calls, insurance requirements, and countless decisions that cannot be scheduled in advance. Success depends less on controlling variability than on responding thoughtfully to it.
That reality is neither a flaw nor an inconvenience. It is simply the nature of caring for people. Yet many research models continue to assume that clinical care can pause long enough to accommodate additional administrative work. When participation depends on adding new processes rather than integrating with existing ones, research begins to compete with the very care it hopes to improve.
The result is often subtle. Practices participate in fewer studies. Opportunities are deferred. Patients who might have contributed to meaningful evidence are never identified. Valuable clinical insight remains trapped inside everyday care because the effort required to capture it exceeds the time available to do so.
The problem isn't that clinicians don't believe in research. It's that healthcare has become too complex to ask them to do everything twice.
Designing around reality instead of ideal conditions
Perhaps we've been asking the wrong question. For years, we've asked how specialty practices can participate more effectively in research. A more useful question might be how research can participate more effectively in specialty care.
That shift in perspective changes almost everything.
Rather than asking providers to build new workflows around studies, we can begin by understanding the workflows that already exist. Rather than introducing new administrative layers wherever possible, we can identify where information naturally flows, where patients are already moving through the healthcare system, and where meaningful evidence can be gathered without disrupting clinical care.
The objective isn't to make research easier. The objective is to make research more compatible with reality.
History suggests that the most successful systems rarely succeed by asking people to fundamentally change their behavior. They succeed by reducing unnecessary effort. Good design doesn't eliminate complexity; it organizes complexity in ways that feel almost invisible to the people using it. Healthcare deserves the same philosophy.
The future of evidence generation
The need for real-world evidence will only continue to grow. As diagnostics become more sophisticated, therapies more personalized, and care pathways more interconnected, understanding what happens outside highly controlled research environments becomes increasingly important.
Questions about diagnostic follow-up, patient adherence, implementation, workflow variation, and long-term outcomes cannot be answered through laboratory science alone. They require observing healthcare as it actually functions, across thousands of everyday interactions that collectively shape patient outcomes.
Generating that evidence will require a different kind of infrastructure. One that respects the realities of clinical practice instead of asking practices to temporarily become something else.
At Aton Health, that belief has shaped our thinking from the beginning. Rather than designing systems around idealized research environments, we've focused on understanding how specialty practices actually work and asking how evidence generation can fit naturally within those existing patterns of care.
The distinction may seem subtle, but we believe it is fundamental. Healthcare doesn't need research that asks clinicians to stop being clinicians. It needs research that learns from the work they are already doing.
If the next generation of evidence is going to reflect the reality of patient care, it must first respect the reality of the people providing it.
A Question Worth Asking: If participating in research depends on adding more administrative work to an already overburdened specialty practice, are we creating evidence that reflects healthcare as it exists or healthcare as we wish it existed?
About the Author:
Tom Sanders is CEO and co-founder of Aton Health. Through this ongoing essay series, he explores the systems, operational realities, and clinical pathways that determine how healthcare innovation becomes better patient care.