Engagement technology is participant-facing, so it needs the same scrutiny.

The engagement layer is where regulatory scope is most often missed

Sponsors assess the eCOA platform and the wearable. They rarely assess the app that reminds a participant to take a dose, the portal that returns a result to them, or the algorithm that flags a participant as at risk of dropping out. Each of those can cross into medical device territory depending on what it does and what a clinician or participant is expected to do with the output. A reminder is not a device. A reminder that adjusts itself based on adherence data and prompts a dose change is a different conversation.

We scope engagement design and regulatory assessment together, so retention work does not create regulatory exposure that surfaces at an inspection two years later. Each component’s qualification decision is recorded with its reasoning, so your own quality function can review and challenge it.

What we deliver

• Participant-facing technology assessment. Every app, portal, chatbot, reminder engine and returned-results feature run through the seven perimeters — with the qualification decision recorded, not assumed.
• Patient-reported outcome instrument handling. Migration of validated instruments to electronic format with the evidence trail regulators expect for measurement equivalence, plus licensing, translation and linguistic validation management.
• Education and communication materials. Written at the reading level the population actually has, in the languages the study actually runs in, and version-controlled as study documents rather than marketing collateral.
• Remote monitoring programme design. What is measured, what triggers an alert, who receives it, and what the site is obliged to do about it — defined before the first participant is enrolled, because an unanswered alert is a safety finding.
• Data protection by design. Lawful basis, minimisation, retention and cross-border transfer worked out for participant-facing data flows, including the ones that run through a vendor's consumer cloud.
• Decentralised and hybrid support models. Home nursing, telehealth visits and direct-to-participant supply, with the accountability chain documented across each handoff.
• Inclusion and accessibility. Design that works for participants with low digital literacy, limited connectivity, visual or motor impairment, or no smartphone of their own — assessed as a human-factors question, which is perimeter four.

Designing the human layer on top of the technology

Digital tools are one part of a participant's pathway of care. The segmentation model, communication scripts, coaches, call agents and escalation paths are the rest — and in a regulated trial they are study records, not customer service.

We design and run this layer together with sponsors and sites: who contacts the participant, through which channel, with what script, at what cadence, and what happens when adherence drops or an adverse event is reported. When those rules adapt based on adherence data, they are part of the protocol — and they need the same inventory, assessment, register and trigger discipline as any other participant-facing system on the page.

Participant segmentation and journey mapping

Before anyone is contacted, we map the journey — for participants and for the site and support staff who serve them:

• Stakeholder and trigger mapping. Every point where a participant can disengage or where staff must intervene: enrolment, first app open, first missed entry, device return, withdrawal. Each trigger is named, owned and linked to a perimeter.
• Adherence profiling. Not marketing personas — operational profiles built from what the study actually needs: digital skills, connectivity, device access, accessibility needs, health literacy and language. A participant who cannot use the app is a protocol deviation waiting to happen.
• Default communication personas. For each profile, a defined message, channel, frequency and tone — plus the escalation path when the script no longer fits (non-compliance, safety signal, request to withdraw).
• The adaptive journey document. A version-controlled map of which profile gets which pathway, what changes when adherence crosses a threshold, and who approves a rule change. This is the document an inspector asks for when the coaching model "felt different" in month four.

Adaptive communication and coaching systems

Adaptive engagement is not a wellness feature. When a reminder changes timing from adherence data, or a health coach contacts a participant because a risk score crossed a line, the system is intervening in the study — and the intervention must be qualified, validated and change-controlled like any other regulated function.

We help sponsors and vendors design adaptive rules that are explicit rather than emergent:

• Rule definition. What signal triggers a change — missed diary, declining step count, alert from a connected device — and what the system or coach is allowed to do next.
• Channel and intensity ladders. Push notification before SMS before call before site visit; documented so the participant experience is consistent and the data remain attributable.
• Coach and agent scripts. Version-controlled language for motivation, technical support and clinical escalation — with AE and protocol deviation routes that do not depend on individual judgement.
• Trigger register for adaptive logic. Model updates, rule changes and script revisions are triggers. The validated state of the coaching system is held the same way as the eCOA build.

A fixed reminder schedules a dose. An adaptive reminder changes behaviour from trial data. A coach who acts on that data may be support — or decision support. We assess which before participants are enrolled, not after an inspection question.