Designing for Delegated Digital Health Access : An Interaction Design Study Supporting Informal Caregivers within eHealth Infrastructures

Sammanfattning

Across Europe, the forty to fifty million people who provide informal care to relatives increasingly do so through digital health systems that were not designed to accommodate them. In most countries, no formal mechanism exists to make that role legitimate. Informal caregivers share credentials, use patients' devices, and attend appointments in person just to access information that should be reachable digitally. This is not an individual failure, nor a technical one. It is a structural condition that existing systems were not designed to see. This project investigated how interaction with digital health systems can be designed to support the gradual and negotiated delegation of access between patients and their informal caregivers, within eHealth infrastructures where formal proxy mechanisms are absent or inadequate. It was conducted through Research through Design across Spain, Italy, and France, three countries chosen because each represents a distinct stage of the same unresolved problem. Twenty-three participants took part across three phases: contextual inquiry, co-design workshops with patient and caregiver dyads (pairs in which both members of a caregiving relationship participated), and think-aloud evaluation of a medium-fidelity prototype. Three findings shaped the design. Informal caregivers have built sophisticated parallel documentation systems, handwritten notebooks, shared folders, phone notes, because the official system provides no equivalent. An expert participant who is simultaneously the information architect of the Andalusian Health System and an informal caregiver confirmed that revocation of delegated access is technically trivial and was never designed as a user-facing function not because it could not be, but because no one imagined it would be needed. And most consistently across every complete dyad: the research session was the first occasion on which participants had discussed the delegation of digital health access directly. Across relationships of five, ten, and fifteen years, the conversation had not happened. No system, and no caregiving relationship, had created the conditions for it to occur. These findings produced a design response covering three interaction flows: configuring delegation, modifying and revoking it, and acting as a delegate while contributing structured observations to the clinical record. Twenty-five design principles were synthesised from the fieldwork, fifteen confirmed across participant profiles, delegation types, and national contexts. The think-aloud evaluation corrected the design in three specific ways: a new screen that resolved a critical perspective-shift failure, the removal of a severity classification that placed clinical judgment on the informal caregiver, and the replacement of pre-selected permission defaults with named starting configurations. In Research through Design, these corrections are not incidental; they are the mechanism through which the prototype produced knowledge that workshops alone could not generate. The project contributes on two levels. At the level of design knowledge, the twenty-five principles constitute a transferable framework for anyone designing delegated health access. At the level of theoretical contribution, the prototype operationalises Nissenbaum's concept of contextual integrity through mechanisms that existing systems do not provide: a relational identification field that encodes the norm of the specific caregiving relationship rather than a generic proxy category; a granular permission structure that allows the patient to calibrate access to match the actual norms of that relationship; and an attributed observation channel that preserves the distinction between patient-generated and caregiver-generated clinical data. Two principles document where interface design reaches its boundary: the policy threshold that currently excludes the entire sample from formal delegation rights, and the structural absence of contingency networks that no delegation interface can substitute. The barriers are not technical. They are design decisions that were made without the informal caregiver in the room, and that this project has attempted to make visible and designable.