Measuring What Matters: Realigning Clinical Endpoints with Patient Reality

What does it actually mean for a treatment to work?

In clinical research, the answer is usually straightforward: a statistically significant change in a biomarker, a reduction in symptom scales, a slowing of disease progression, or an improvement in overall survival. These endpoints are the bedrock of regulatory approvals and scientific validation. Yet, a fundamental question often gets lost in the data: are we measuring what actually matters to the patient?

A trial can meet its primary endpoint, a laboratory value can normalize, and a clinical score can improve—all while the person receiving the therapy notices zero meaningful difference in their day-to-day life.

Through my patient advocacy podcast, On One Condition, I regularly talk with individuals navigating life with chronic health conditions. Two of those conversations, one with Parkinson’s advocate Cathy Molohan and another with hypoparathyroidism advocate Katie Gillick, brought this disconnect into sharp focus.

The gap between clinical markers and lived experience

In our discussion, Cathy referred to Parkinson’s as a "snowflake disease." Two people can share the exact same diagnosis yet experience entirely different combinations of symptoms. While one person struggles primarily with motor tremors, another might be overwhelmed by stiffness, depression, severe fatigue, or sleep disruption. Because Parkinson’s lacks a simple, definitive biomarker, trials heavily rely on standardized rating scales. But if an intervention improves a secondary motor score without touching the debilitating sleep issues keeping a patient from working, has the trial truly captured functional success?

Katie’s experience with hypoparathyroidism highlights a different side of the same coin. During a Patient-Focused Drug Development (PFDD) meeting with the FDA, she and fellow advocates described how drastically the condition varies depending on age, onset, and individual circumstances. A pediatric patient dealing with congenital hypoparathyroidism faces vastly different daily realities than an adult managing post-surgical complications.

More importantly, Katie pointed out a discrepancy that plagues many chronic condition trials: patients can have lab results that fall within a 'normal' clinical range while still suffering from severe fatigue, brain fog, muscle spasming, tingling, or numbness.

A normalized biomarker proves a biological mechanism; it does not guarantee a restored life. This doesn't make the biomarker invalid, but it underscores its limits. If our trial designs stop at biological markers, we risk declaring victory on paper while leaving the patient's daily burden untouched.

Technology as a bridge: Capturing real-world function

To bridge the gap between clinical metrics and daily reality, we need to look beyond the intermittent snapshots taken during site visits. A patient visiting a clinic every few weeks provides only a brief, highly artificial window into their health. Disease progression and treatment response happen in the spaces between those visits.

Digital health technologies offer a path forward, if we deploy them to answer the right questions:

• Electronic patient-reported outcomes (ePROs): By inviting patients to report symptoms, functional abilities, and quality of life directly, ePROs bring structured patient perspectives into the trial dataset alongside physiological data.
• Wearables and remote monitoring: Continuous passive data collection (such as gait, tremor severity, physical activity, sleep architecture) captures how a treatment performs in the patient's actual living environment rather than in a controlled clinical setting.

However, technology is only as valuable as the intent behind it. Are we using wearables and ePROs simply to collect more data points against rigid, sponsor-defined protocol endpoints? Or are we using them to measure changes relative to the patient’s own baseline and functional priorities?

If a wearable detects a slight change in gait, that metric becomes meaningful only when paired with patient context, such as an ePRO confirming that the participant can now walk to the grocery store without stopping to rest. Bringing physiological context, digital telemetry, and patient reporting together allows us to evaluate a drug's true impact on daily life.

The digital burden paradox

As clinical trials adopt more digital tools, a new challenge has emerged: the temptation to measure everything simply because we can.

More questionnaires. More smart-watch notifications. More active tasks. More data streams.

Every additional assessment asks time, energy, and cognitive effort from someone who is already managing an illness. I recently had an opportunity to talk to the founder of a clinical site, who shared a story that has stayed with me: one of her patients withdrew from a study entirely due to the daily burden of using the required study technologies. The ePRO questionnaires had become a chore, as well as a daily reminder that they were not gaining any benefits from participating in the trial.

It is very unfortunate that tools intended to make trials more "patient-centric" can end up doing the exact opposite. When technology creates friction, participants disengage, skip assessments, or drop out, resulting not only in a poor patient experience but in missing data and compromised study integrity.

We need conscious deployment of technology. Before adding a device or ePRO module to a protocol, trial designers should ask:

1. Does this assessment answer a question that matters to the patient's quality of life or functional capabilities?
2. Does the clinical value of this data point justify the daily friction placed on the participant?
3. If we cannot explain how this data will directly inform our understanding of meaningful treatment response, why are we asking the patient to collect it?

Co-designing what we measure

Shifting the focus from collecting more data to measuring what matters requires involving patients long before a protocol is finalized.

Patients can tell researchers which symptoms are most disruptive, what a meaningful functional improvement looks like, and what data collection methods fit realistically into their routines. Clinical scientists and technology vendors can then translate those priorities into validated, practical measures.

• Cathy's experience demonstrates how patient insights in Parkinson's can guide engineers to target continuous wearable metrics—like gait stability or tremor duration—that directly reflect independence.
• Katie's advocacy work shows how community input can help regulators and sponsors re-evaluate trial endpoints, ensuring that symptom relief is prioritized alongside lab values.

The goal is not to create a completely bespoke endpoint for every individual participant, nor is it to replace scientific rigor with subjective opinion. The goal is to define what matters, using patient insights to decide which measurements are worth making, and then leverage technology to capture those measurements, as unobtrusively as possible.

Realizing the opportunity

Clinical research has unprecedented access to patient data, but data volume is not a substitute for clinical relevance.

At the intersection of patient advocacy and trial technology lies an opportunity to transform trial design. It is not merely about making clinical trials more digital; it is about making the evidence we generate more meaningful.

Patients do not experience a primary endpoint. They experience whether a treatment allows them to work, care for their families, sleep through the night, or live with less pain. If we truly want to evaluate whether a therapy works, our endpoints must start measuring what matters to the people living with the disease.

This article is inspired by conversations from the On One Condition podcast, where patients, caregivers and advocates share their lived experiences to help create a more human approach to healthcare and clinical research.

Sylvain Berthelot, host, On One Condition podcast