Video or paper? This is the best-supported answer in physiotherapy
Published on
The patient nods in the treatment room. He does the exercise once in front of you, it looks fine, he goes home with a printout. Two weeks later it turns out he has been doing it wrong the whole time.
That is not carelessness. That is what happens when you put movement on paper.
What goes wrong with a printout
In a randomised study in Physical Therapy, participants learned an upper-limb exercise program in three ways: demonstrated live, on video, or from a printed handout. Accuracy of performance was then measured, immediately and after a delay.
The handout group made more than twice as many errors as the other two groups (Reo & Mercer, 2004).
That is the heart of it. A printout carries all the information: the starting position, the number of repetitions, the muscle involved. What it cannot carry is the only thing that matters in an exercise, which is how the movement runs between the start and the end.
Video reaches the same level as you in the room
The most interesting part of that study is not that paper loses. It is that video does not lose to you.
Between the live instruction group and the video group no difference could be shown, neither immediately after instruction nor on delayed retention.
A later trial tested that more sharply. 58 participants learned shoulder exercises, half through a single instruction session with a physical therapist, half through a video that showed and corrected the usual mistakes. On the performance scale: 13.66 against 13.46. No meaningful difference, at level of evidence 1b (Berkoff et al., 2016).
For teaching correct technique, then, a well-made video holds its own against a consultation.
Why photographs do not close the gap
An obvious middle ground is a printout with photographs. That helps, but less than you would hope.
Participants learned simple and complex exercises from video or from still illustrations. The video group was more motivated to train at home and more confident they were doing it correctly. Every participant preferred video (Weeks et al., 2002).
Note what was measured there: not only performance, but motivation and confidence. That is exactly the factor that keeps coming out on top in the adherence literature. Self-efficacy is one of the strongest predictors of whether someone keeps up a home program.
A photograph shows you where to end up. A video shows you how to get there, and that is what makes people feel capable of it.
What video does not fix
The honest part belongs here, because video is not a switch that makes the problem go away.
Accuracy drifts over the months. In 76 caregivers of children with neonatal brachial plexus palsy, accuracy of the home exercises was tracked, with an instruction DVD and an up-front demonstration. Over twelve months that accuracy largely held, but at the shoulder it fell from 98.9 to 88.3 percent within roughly three months (Rasmussen et al., 2013). Video does not replace the check at the next appointment.
Longer is not better. A meta-analysis of home-based video exercise programs in people over 65 pooled 26 studies with 1,292 participants. Retention came out at 91.1 percent and attendance at 85.0 percent. But from that same analysis: as sessions got longer, attendance dropped sharply. And programs without live contact with a coach scored lower than sessions where that contact was present (Rihova et al., 2024).
That second point is worth repeating. Video lowers the threshold, but it does not replace you.
What a usable exercise video does
From what has actually been measured, a short checklist falls out:
- Show the movement, not the end position. That is the one thing a photograph cannot do.
- Show the mistake, not only the correct version. Berkoff explicitly tested a video that showed and corrected common errors. That is what a consultation does too.
- Keep it short. Longer sessions lowered attendance in the meta-analysis.
- Do not let it replace the consultation. Accuracy drifts; schedule a real check.
The video has to be yours
There is one thing that recurs in all three trials and that drops out of the summaries too easily: what is being measured is not "video" but someone demonstrating the exercise the way they explain it. Reo & Mercer filmed the instructor. Berkoff filmed the corrections a physiotherapist gives in the treatment room. That is not stock footage, it is a consultation on tape.
Add Rihova's finding to that: programs without live contact scored lower. A video in which the patient hears you, with your cues and your pace, is the closest you get to that contact at home.
Which is why Kinmo is built around your content, not ours. Every exercise has a photo slot and a video slot. The library you get from us is the base that lets you start straight away. The video slot is yours, and filling it is deliberately made as short a job as possible:
- Film from your phone. At your desk, you scan a QR code and your phone takes over. Already on a phone or tablet, and the camera opens directly. The phone never logs in anywhere.
- Upload a file you already have, up to 200 MB and three minutes.
- Paste a link to a video that already lives somewhere.
Film it once, reuse it with every patient who has the same complaint. Start with the ten exercises you prescribe most often; that is worth more than a library of a thousand exercises of which you use twenty.
What the literature has not examined is whether your own recording beats a professionally produced video of the same exercise. That comparison has never been made. What has been measured is that the physiotherapist on camera worked as well as the physiotherapist in the room.
Conclusion
On adherence the evidence is thin and the trials contradict each other. On the form of instruction it is not. Three separate trials, fourteen years apart, and the same result each time: moving images produce more accurate performance than a printout, and in doing so they reach the level of instruction given in the treatment room.
What they do not do is make the follow-up check unnecessary.
Every study in this piece is summarised on our research page, alongside the rest of the literature on home exercise programs. You can go straight to Reo & Mercer, Weeks and Berkoff.
Sources
- Reo, J.A. & Mercer, V.S. (2004). Effects of live, videotaped, or written instruction on learning an upper-extremity exercise program. Physical Therapy, 84(7), 622-633. PMID 15225081
- Weeks, D.L., Brubaker, J., Byrt, J., Davis, M., Hamann, L. & Reagan, J. (2002). Videotape instruction versus illustrations for influencing quality of performance, motivation, and confidence to perform simple and complex exercises in healthy subjects. Physiotherapy Theory and Practice, 18(2). DOI 10.1080/09593980290058454
- Berkoff, D.J. et al. (2016). Corrected error video versus a physical therapist instructed home exercise program: accuracy of performing therapeutic shoulder exercises. International Journal of Sports Physical Therapy, 11(5), 757-764. PMID 27757288
- Rasmussen, L., Justice, D., Chang, K.W., Nelson, V.S. & Yang, L.J. (2013). Home exercise DVD promotes exercise accuracy by caregivers of children and adolescents with brachial plexus palsy. PM&R, 5(11), 924-930. PMID 23770351
- Rihova, M. et al. (2024). Adherence and Retention Rates to Home-Based Video Exercise Programs in Older Adults: Systematic Review and Meta-Analysis. Telemedicine and e-Health, 30(11), 2649-2661. PMID 39072676