You demonstrate four exercises, the patient nods, and on Thursday they do two of them, one of them wrong. This is not a motivation problem. It is a memory problem, and the research on it is old, consistent, and mostly ignored.
How much is lost
Roy Kessels' 2003 review in the Journal of the Royal Society of Medicine summarised decades of studies on patients' memory for medical information. Between 40 and 80 percent of what a clinician says is forgotten immediately. Of what is remembered, almost half is remembered wrong. The more information presented, the smaller the proportion recalled correctly. A four-exercise programme with sets, reps, frequency, and cues for each is somewhere between 20 and 40 separate facts delivered at the end of a session, often while the patient is putting their shoes on.
What the evidence says helps
- Write it down. Kessels found written information is remembered better than spoken and supports adherence, with the caveat that it fails patients with low literacy or a different first language.
- Use pictures. The most striking finding in the review: in one study about 14 percent of spoken instructions were recalled correctly, versus more than 80 percent when the same instructions were given as simple pictographs.
- Be specific. "Rest for two days" is remembered; "rest for a while" is not. "Three sets of ten, morning and evening" beats "a few times a day".
- Say what the categories are before you fill them. Telling the patient "I am going to give you the diagnosis, then the plan, then what to do at home" improved recall in Kessels' sources. Just presenting things in a logical order did not.
- Put the instructions first and say they matter. Items presented first and flagged as important are remembered best.
The physiotherapy trial that tested it
A 1998 trial in Manual Therapy randomised 96 patients with acute or sub-acute low back pain to receive their home exercises either verbally alone or verbally plus a written, illustrated handout. Mean compliance was 38 percent in the verbal-only group and 77 percent with the handout. The trial is small, unblinded for patients and therapists, and scored 6 out of 10 on the PEDro scale, so treat the exact figures loosely. The direction, a doubling of adherence from a sheet of paper, has not been contradicted since.
A format that works
- One exercise per block: name, a picture or short clip, the dose (sets, reps or hold time, days per week), and one cue. Not three cues. One.
- Lead with the exercise that matters most. If they only do one, make it that one.
- Give the pain rule once, in plain words, at the top: what level is acceptable during, and when to stop or message.
- Say what to do when it goes wrong: fewer reps, shorter hold, skip and message. A patient with a fallback keeps going; one without one stops.
- Fewer exercises. Four well-remembered exercises outperform seven half-remembered ones.
How KineTrue fits
KineTrue assigns each exercise with its own name, description, repetitions or hold duration, and difficulty level, so the patient opens the app and sees the dose rather than reconstructing it from memory. The camera-guided session then shows the movement as they do it, which does the work of the pictograph in real time. For the patient-facing explanation of what a programme is for, see what is a home exercise program, and why did my therapist prescribe one?
Written and visual instructions support the explanation you give in the room; they do not replace it. Patients with low literacy, cognitive impairment, or a different first language may need a carer involved or a slower, repeated demonstration rather than more paper.