The situation
A virtual care provider needs a single mobile experience across patients and clinicians. Patients range from confident smartphone users to elderly people using an app for the first time, often while they are feeling unwell.
Clinicians have minutes between appointments and cannot afford a system that adds clicks or loses visit notes. Health data rules shape every design decision.
The challenge
Wide range of users
Intake has to work for someone unwell, using assistive technology, on an older device.
Unreliable networks
Video needs to degrade gracefully and reconnect without forcing the patient to start again.
Regulated data
Protected health information must be safeguarded on the device, in transit and in logs.
The approach
- 1
Clinical and patient researchWeeks 1–3
Shadow clinicians, interview patients and turn intake questions into a shorter adaptive flow.
- 2
Accessible prototypeWeeks 4–7
Test with screen readers, large text and real patients before committing to the build.
- 3
Build the two appsWeeks 8–16
Patient and clinician apps share one codebase, with weekly builds issued to the pilot group.
- 4
Pilot and launchWeeks 17–20
Run a limited clinic pilot, tune video settings on real networks and roll out in stages.
Key decisions and trade-offs
React Native with native video modules
Why: One team can serve both platforms while keeping low-level control of the call experience.
Trade-off: Native modules must be maintained separately for each OS.
Adaptive intake instead of a long form
Why: Each patient answers only the relevant questions, which reduces drop-off.
Trade-off: Clinical teams must help maintain the branching logic.
Audio-first fallback on weak networks
Why: A clear conversation is better than a frozen picture when bandwidth falls.
Trade-off: Clinicians temporarily lose visual assessment.
How success is measured
| Measure | Baseline | Target | How it’s measured |
|---|---|---|---|
| Intake time | Measured during pilot | 35% shorter | Product analytics on completed intakes |
| Call recovery | Patient must rejoin | <2s reconnect | Client-side call telemetry |
| Accessibility | Not assessed | WCAG 2.2 AA | Automated checks and manual audit |
| Patient rating | New product | 4.7+ in store | App Store and Google Play ratings |
Key takeaways
- Test accessibility with people, not only with scanners.
- A shorter form is a clinical design problem as much as a UX one.
- Plan the failure states of a video call before the happy path.