Skip to content
UXE Award

Healthcare UX design awards: designing for people who are not at their best

What we look for in healthcare and care products: clarity under stress, clinical safety, accessibility as a baseline, and interfaces that work for carers, clinicians and patients at once.

The UXE editors5 min read

Health products are used by people who are frightened, tired, in pain, or working a twelve-hour shift. That single fact should change every design decision, and in the entries that win healthcare UX design awards, you can see that it did.

1. Who is actually holding the phone?

Almost no health product has one user. There's the patient, the family member managing appointments on their behalf, and the clinician with ninety seconds between patients. The strongest entries name which of these the product is for, and design for that person specifically rather than averaging the three.

Designing for three audiences

Averaged

  • One interface for patients, carers and staff
  • Clinical language softened until it's ambiguous
  • Permissions ignored — everyone sees everything
  • Designed for the calmest possible moment

Specific

  • Distinct modes, or distinct products
  • Plain language for patients, precise terms for clinicians
  • Proxy access designed, not improvised
  • Designed for the worst realistic moment

2. Clarity when the reader is not at full capacity

Comprehension drops under stress, medication and fatigue. This is a design constraint, not an excuse for condescension. What it means in practice:

  • One instruction per screen where the stakes are high. Dosage, preparation for a procedure, what to do about a symptom.
  • Absolute dates and times, never relative. "Tuesday 6 October, 9:15am" — not "in 12 days".
  • Numbers written the way people say them, with units attached every time.
  • Critical information in text, not only in an icon, a colour or an image.
  • A reading level aimed at a tired person, tested with real readers rather than a formula.

3. Safety is a UX property

In clinical tools, an interaction design mistake is a patient-safety event. Editors with healthcare experience look for the specific patterns that prevent them.

Four safety patterns editors check
  1. 01

    Confirm identity

    Who is this record about? Visible at all times, not only on entry.

  2. 02

    Constrain input

    Units fixed, ranges validated, look-alike drug names separated.

  3. 03

    Warn proportionately

    Alerts reserved for real risk. Alert fatigue is a design failure.

  4. 04

    Record the action

    Who did what, when — visible in the interface, not only in a log.

Alert fatigue deserves particular attention: a product that warns about everything has trained its users to dismiss warnings, which is worse than not warning at all. Entries that can show how they reduced alerts while keeping the important ones score highly.

4. Accessibility is the baseline, not the differentiator

The population using a health product includes, by definition, more people with disabilities, temporary impairments and low digital confidence than the general population. An inaccessible health product excludes exactly the people it exists for.

Editors will test with a screen reader, at the largest text size, with a keyboard only, and with reduced motion. They will also look for the things a checklist misses: whether a form can be completed with shaky hands, whether a timeout gives enough time to someone reading slowly, whether an emergency action is reachable without signing in.

A large share of health interactions are carried out by someone other than the patient — a parent, an adult child, a partner. Most products handle this with a shared password, which is a security failure and a dignity failure at once.

Designing proxy access properly (clear role, visible identity, revocable, with the patient able to see who has access) is unusual enough that it materially lifts a score. If you've built it, say so prominently.

6. Offline, low-bandwidth and old devices

Hospital wifi is bad. Community and home-care staff work in buildings with no signal. Patients on low incomes use older phones for longer. A health product that assumes a good connection and a recent device has designed for the wrong person.

What to send

  • A test account with realistic clinical or patient data, and a way in that doesn't need a sales call.
  • The highest-stakes screen in the product: medication, results, or an appointment change.
  • Your accessibility audit — including the failures and the fix plan.
  • Evidence of testing with the actual population, not colleagues. Even small-n moderated testing counts, and disclosed limitations are fine.
  • One measured outcome: missed appointments, completion of a pre-procedure form, time to find a result.

Health entries usually go to Service Experience or Product Experience with a Health subcategory, most often alongside Inclusive & Ethical Design. Enter a project.

Results, and the hardest design question in health

Showing test results directly to patients is now common, and it is the single hardest content-design problem in the field: the same words have to be truthful, comprehensible and not needlessly frightening, read by someone who may be alone.

  • State the result plainly, then what it means, then what happens next — in that order.
  • Say whether anyone has reviewed it yet, and when they will.
  • Give the reference range in words as well as numbers.
  • Never let "abnormal" stand alone as a label with no interpretation.
  • Make the route to a human visible on the same screen.

Designing for shift work

Clinical tools are used at 3am by someone on their eleventh hour. That has design consequences that are rarely considered: high contrast that survives a dimmed ward, targets that work with gloves, no reliance on fine colour distinctions, and never a destructive action adjacent to a frequent one.

Four clinical design priorities
  • Confirm identity persistently4Whose record is this? Visible at all times, not only on entry.
  • Separate look-alike actions4Discharge and delete should not be adjacent, or similar in colour.
  • Preserve work on interruption3Clinical work is interrupted constantly. Drafts must survive it.
  • Reduce alerts3Fewer, better alerts beat more. Alert fatigue is a safety issue.

Questions we get asked

  • Can we enter a clinical system that contains patient data?

    Yes, with a demonstration environment containing synthetic data. Never send us real patient data — an entry containing it would be withdrawn.

  • Do we need a formal accessibility audit?

    Not required, but for health products it is the strongest single piece of evidence you can attach, and Inclusive & Ethical Design requires an accessibility statement.

  • How is patient-facing work judged against clinician-facing work?

    By the same criteria, with different expectations of density and vocabulary. A clinical tool is allowed to be dense; it is not allowed to be ambiguous.

  • What evidence matters most in a health entry?

    Testing with the actual population — patients, carers or clinicians — even at small sample sizes, with the limitations stated. Internal review is not the same thing and editors can tell.

Assessed work in health & care

Chosen by score from the work the editors have assessed, not by hand.

  • healthcare
  • accessibility
  • industry

Written by

The UXE editors

The team that selects the work, uses it, scores the seven criteria and writes every verdict on this site.