Translation of the letter WE SENT YOU

This is how it feels to receive a letter you can't understand.

We sent a letter in a different language (in this case, Turkish) because this is how NHS communications arrive for millions of patients in England.

Send more information to your inbox

This is some text inside of a div block.
This is some text inside of a div block.
Why we did it

Two separate populations. The same outcome.

Neither group is refusing care. Both are recorded the same way in your recall report.

Population one: language
5.1m

people in England and Wales whose main language is not English. More than a million of them speak English poorly or not at all.

Source: Census 2021, Office for National Statistics.
Population two: literacy
1 in 6

adults in England has very poor literacy, reading at or below the level expected of an eleven-year-old. The NHS aims for a reading age of nine to eleven.

Sources: National Literacy Trust; NHS England accessible information guidance.

These are different problems and they need different fixes. A translated letter does nothing for the second group; plain English does nothing for the first. Both end up in the same column of your report.

What it actually takes

Reaching everyone is not a better first message. It is refusing to stop at the first message.

Four things have to happen. None of them are clever. All of them are work that practices do not have hours for.

01

Language and format the patient can use

Preferred language taken from the record, plain-English wording at a reading age of nine, large print and SMS where a letter will not land.

02

Repeated and varied contact

A scheduled sequence over weeks, worded differently each time and sent at different times of day, not one text on a Tuesday morning.

03

Escalation when a channel fails

An undelivered text escalates to a letter; an unopened letter escalates to a call. Failure of one route triggers the next automatically.

04

Reporting who is still unreached

A named list of patients no channel has touched, not a response rate. You can act on a list. You cannot act on a percentage.

What the channel data actually shows

Non-English-speaking patients convert lower on every channel. No single channel fixes it, which is why the sequence matters more than the send.

NHS App
21.9%
25.2%
Accessible SMS
12.9%
17.0%
NHS Email
9.3%
12.8%
Best send time
2pm
9am
Non-English-speakingEnglish-speaking
Source: Appt Recall production analytics across 150+ practices. A programme that sends everyone at 9am has chosen the optimal time for one group and a poor one for the other.

The same patient, two recall processes

Mrs Y, 62, due for bowel screening. Main language Turkish. Nothing about her changes between these two rows.

Contacted once
Day 0 · Letter
English, reading age 15
Recorded as a non-responder
Contacted four times, four ways
Day 1 · NHS App
Bilingual, Turkish & English
Day 7 · SMS
Turkish, large print
Day 18 · Letter
Reworded, afternoon send
Day 30 · Call
Interpreter line
Booked on the fourth contact
What it produced

Uptake up, and the gap narrower at the same time

+18pp

increase in cervical screening uptake, with income-related inequality in uptake eliminated in participating practices rather than widened.

Source: Imperial College London evaluation, NHS Cancer Programme.
64.6%

of non-English-speaking patients booked their NHS Health Check, against 54.1% of English-speaking patients. The gap runs the other way when the message is built for the patient.

Source: Appt Recall production analytics, last 12 months.
Selected for the NHS Innovation Accelerator · 150+ GP practices across 7 NHS regions, on EMIS and TPP
Document Icon
Case studies

Appt Health is on a mission to reach every patient

See how practices are reaching patients that one letter in English never could.

See Appt Health working in your practice

A 20-minute call. We’ll show you exactly what Appt Health can do for a practice like yours.

Book a 20-minute call

Someone from the Appt Health team will reply the same working day.