Design for Health

Why this matters

Healthcare is among the most complex systems a designer can work in. Multiple stakeholders (patient, caregiver, clinician, hospital, regulator, payer), multiple scales (body, system, society), and the cost of a mistake can be physical and real. It’s not regular product UX.

I came here through two paths: personal (I’ve lived with IBS for over 14 years) and academic (Master’s in Design for Health and Wellbeing, ESAD.CR, 2024, with a thesis on a conversational AI interface to support IBS self-management). This page consolidates what I learned designing for this context, and opens out to the posts where I go deeper.

The premise that orients everything: design in health means designing with the people who live the problem, not for them. Everything else follows.

The designer’s role in health

Peter H. Jones, in Design for Care (2013), describes five critical roles a designer plays in this context. They work as a compass:

1. Dealing with the complexity of health systems. Design breaks processes into manageable pieces and makes sure each digital solution fits existing workflows.
2. Centring the person. Treat the patient as an active participant in their own health, not a passive recipient.
3. Innovating in clinical practice and services. Collaboration between designers and clinical teams identifies where digital solutions improve processes, decisions, and outcomes.
4. Improving UX and usability. Especially in electronic health records, where friction costs hours of clinical time every day.
5. Facilitating organisational change. Designers help promote collaboration and a focus on the individual inside systems that tend to be hierarchical and cautious.

Human-Centred Design + Design Thinking

The methodology I use (and that most of this field uses) combines two things:

Human-Centred Design (HCD) is a principle. ISO 9241-210:2019 defines six pillars: design grounded in explicit understanding of users, tasks, and environments; users involved throughout; design refined by user-centred evaluation; iterative process; covers the whole experience; multidisciplinary team.

Design Thinking is a way of doing it. The five phases (IDEO): Empathise, Define, Ideate, Prototype, Test. Non-linear, iterative, with going back and forth.

In practice, HCD is the “what” and Design Thinking is the “how”. They run together. Covered in Design Thinking in healthcare with the worked example from my thesis.

Designing with patients

One of the most underrated points of this work is how much recruitment matters. For my thesis I invited 99 people with IBS to participate. Eight made it through to sessions. Ratio of 12 to 1, and even then with calendar conflicts, variable physical condition, and the need to make every session remote.

Working with participants who live with a chronic condition asks for:

• Clear inclusion criteria (age, diagnosed for long enough, minimum digital literacy)
• Ethical exclusion criteria (sensitive health state, unrelated comorbidities)
• Written informed consent, guaranteed anonymisation
• Remote sessions (going to a physical space might be impossible on a bad day)
• Flexible schedules and the ability to reschedule without friction

I go deeper in Designing with patients, not for them. The post applies to any health context, not only IBS.

Patient journeys: what you find when you map them

The thesis surfaced a number that stuck with me: people with IBS can have symptoms for five years before getting an official diagnosis. In Portugal the wait for a gastroenterology appointment can exceed 600 days. The direct cost of IBS in the UK runs at £1.2-2 billion per year.

These numbers only show up when you do the work of mapping the actual patient journey: from first symptom to GP visit, from failed tests to misdiagnosis, from diagnosis to trial-and-error with diet and medication, from crisis episodes to attempts at self-management with no support.

What the designer takes from these maps:

• Friction points (where people give up or stall)
• Support gaps (where no system serves them)
• Critical moments (a flare-up, an emotional crisis)
• Intervention opportunities (where a digital tool can make a difference)

More in Patient care journeys: when diagnosis takes 5 years.

Conversational AI in health

For my thesis I prototyped a conversational AI interface to support IBS self-management. The context demands extra care: answers have to be correct (health), clear (people who are suffering), and useful at the right moment (a crisis, not a manual).

There’s a principle from linguistics that applies directly: Grice’s Maxims (Grice, 1975). For a conversation to work, both sides must cooperate along four rules:

• Quantity: give the right amount of information, no more, no less
• Quality: be truthful, avoid saying what you don’t have evidence for
• Relation: stay on topic
• Manner: avoid obscurity and ambiguity, be brief and orderly

These maxims weren’t designed for AI, but they work as conversational design heuristics. I go deeper in Grice’s Maxims as conversational design heuristics.

For the broader AI background, see The Agentic Era and the Design for AI guide. For ethics, see Ethical Principles in AI Design.

Terms to know

In digital health there’s a set of terms that overlap but aren’t equivalent. eHealth is the umbrella. It covers anything involving electronic processes in health: records, prescriptions, telemedicine. Telemedicine is the subset that covers remote clinical services: telemonitoring, teleconsultation, tele-assistance. mHealth is the subset that lives on mobile: apps, wearables, sensors.

Full glossary with the differences table in eHealth, Telemedicine, mHealth: a glossary for designers.

Where to start this week

Three concrete steps:

1. Pick a condition that touches you. It can be personal, family, professional. Genuine motivation keeps you on the problem for the long time it takes.
2. Map the current patient journey. No solution, no product. Just what happens today, from first symptom to last appointment. You’ll see things.
3. Talk to at least three people who live with the condition. Not in a focus group, in individual long unstructured interviews. It’s the foundation for everything else.

Design in health isn’t a sprint. It’s a practice held up by evidence, ethics, and iteration. If something here stuck, write to me.

FAQ

What is Design for Health?

The design of products and services for healthcare. It covers accessibility, health literacy, medical ethics, and working with vulnerable populations. It differs from regular UX by the weight of consequence (errors can affect health) and the complexity of the systems involved.

How is it different from regular UX?

Users are patients in a vulnerable state. The process asks for longer, narrative sessions, with co-design. There’s ethical care around sensitive data, specific regulation (GDPR, MDR, FDA), and systems with multiple stakeholders (patient, clinician, hospital, regulator).

What is a patient care journey?

The sequence of events a patient goes through, from first symptom to self-management. Mapping it reveals friction points, support gaps, critical moments, and design opportunities. I go deeper in Patient care journeys.

How do you recruit patients for research?

Clear criteria (age, diagnosed long enough, digital literacy), written informed consent, remote sessions by default, flexible scheduling, and a realistic ratio (around 12 invites per actual participation). Detail in Designing with patients, not for them.

Which principles to use for conversational AI in health?

Grice’s Maxims: right quantity, verifiable quality, relation to topic, and clear manner. They work as conversational design heuristics, especially where errors cost more. More in Grice’s Maxims as conversational design heuristics.

Posts in this guide

• Patient care journeys: when diagnosis takes 5 years
• Designing with patients, not for them
• Grice’s Maxims as conversational design heuristics
• Design Thinking in healthcare: what changes
• eHealth, Telemedicine, mHealth: a glossary for designers

Other guides: Design for AI, Inclusion and Diversity.