Turning Poor Phase II eCOA Compliance into a Phase III Data Surplus with 21,000+ Days of Quality Data

Situation

A global Phase III trial for a new therapy targeted at a rare dermatological condition affecting children spanned 11 countries and 35 sites and required complex data capture from multiple caregivers. The participants’ caregivers needed to track the children’s sunlight exposure, symptoms, and related events in real time throughout the day during the 40-month study. Low volume and quality of data captured during Phase II using a legacy technology meant the sponsor was looking for a new partner to capture data more reliably and reduce the burden on families.

Risk

The previous tool was very difficult to use, requiring a 20-page paper manual, and resulted in poor-quality data and very low compliance. The practical difficulty of recording data accurately while supervising an active child was a key barrier to overcome.

Diagnosis

uMotif identified the need for three interlinking diaries covering sun exposure, symptoms, and attacks and including start and stop times, since episodes could span more than one day. As the condition is triggered by exposure to sunlight, local sunrise and sunset times needed to be captured through geographic lookup. Built-in edit checks would be essential to help caregivers enter data correctly.

Actions

The design process began with the caregivers themselves. Conversations surfaced not only practical challenges but the emotional context of managing a rare condition in a child. uMotif designed potential solutions and tested them directly with families, refining the design through iterative feedback.

The resulting eCOA design used three distinct diary types — exposure, symptom, and event logs — each with its own visual language to give caregivers immediate clarity about what actions and information were needed and when.

A persistent status view showed open and completed entries at a glance. Built-in edit checks made it structurally impossible to enter conflicting start and stop times, embedding data quality at the point of capture, and location pre-saving meant sunlight data was collected automatically, without requiring manual input at each session.

Result

Diary data was required on at least 5 days each week tracking various aspects of exposures and impacts. Participants captured more data than expected through the application, logging over 1,900 diary days more than required — 21,000 diary days versus the expected 19,000.

Families gave strongly positive feedback on the new application, and both the volume and quality of data captured improved markedly compared to Phase II.

Lessons Learned

Understand what data is truly necessary before designing the tool. Then prioritize and design for the user, not just the endpoint. Build in edit checks so the technology does the hard work, not the person entering the data. Involving end users throughout the design process, to gather feedback and iterate, is essential to a solution people can actually use and want to use, especially when it is a caregiver who must do the recording.