Skip to content
Curious Context Labs

case study · Healthcare · Custom app

Patient instructions people can actually follow

How approved leaflets and discharge instructions can be rewritten in plain language and translated, with clinicians approving every version.

business
Hospitals
build
Custom app
timeline
6 weeks to first batch
read
3 min

01 the business

Children's and maternity hospital

300 leaflets, patients speaking 40 languages

Leaflets are written in clinical language and only a few are translated. Updates take months to reach every version. Families leave with instructions they struggle to follow.

02 before

The problem

Many patients struggle to understand medical instructions, especially after discharge or before a procedure. Leaflets are written in clinical language, translations are slow and costly, and updates take months.

  • Leaflets are hard to read.
  • Few translations exist, and they lag behind updates.
  • Versions drift out of date.

03 the build

What the AI does

  1. Step 1: Starts from your approved clinical leaflets and instruction templates.
  2. Step 2: Rewrites them at a reading level you choose, with short sentences and clear steps.
  3. Step 3: Translates them into the languages your patients speak most.
  4. Step 4: Keeps the clinical content identical to the approved source and marks every change for review.
  5. Step 5: Publishes approved versions to your patient portal or print templates, with version dates.

paused: waiting for a person to approve

How it was put together

  • Started from approved clinical leaflets and templates.
  • Rewrote them at a chosen reading level with clear steps.
  • Translated them into the most-spoken languages.
  • Linked every version to its source so updates flow through.

04 human in the loop

Where a person stays in charge

A clinician approves every plain-language version and every translation before use, ideally with a native speaker reviewing translations.

  • Clinicians approve every plain-language version and translation.
  • Native speakers review translations where possible.
  • Clinical content must match the approved source.

05 rollout

From kickoff to live in 6 weeks to first batch

  1. Weeks 1–2Choose the 30 most-used leaflets and target languages.
  2. Weeks 3–6Rewrite, translate and review the first batch.
  3. OngoingAdd leaflets monthly and update versions when sources change.

06 after

What changes

Patients receive instructions they can follow, in their language. Updates reach every version quickly because the source and its versions are linked.

worked example · your numbers will differ

Assume 300 leaflets, each costing 4 hours of clinician and writer time to rewrite by hand, against 1 hour to review a draft.

300 × 3 hours saved = 900 hours, before counting translation costs.

What to measure

  • Leaflets available in plain language and each language.
  • Time from source update to all versions updated.
  • Patient understanding checks.
  • Calls asking to explain instructions.

07 evidence

What the research says

  1. Only 12% of US adults had proficient health literacy; 36% had basic or below basic health literacy.

    US National Center for Education Statistics, The health literacy of America's adults: results from the 2003 National Assessment of Adult Literacy (opens in a new tab) · 2006

  2. Around 1 in 10 patients is harmed in health care, and more than half of that harm is preventable.

    World Health Organization, Patient safety fact sheet (opens in a new tab) · 2023

08 lessons

Mistakes to avoid

  • Changing clinical meaning. Simplify words, not advice.
  • Skipping translation review.
  • Unlinked versions. They drift out of date.

09 questions

Common questions

Why plain language?
The US national assessment found only 12% of adults had proficient health literacy.
Can it personalise instructions per patient?
It can fill approved templates with appointment details, but clinical content stays as approved.

keep reading

All case studies ->

Tell us how your business runs. We'll tell you where AI fits.

A free 30-minute call. You talk, we ask questions, and you leave with a few concrete ideas — or an honest answer that AI isn't the right fix yet.

Working with teams around the world.