Do reminders, apps and gamification improve home-exercise adherence? What the evidence shows

Every vendor in this space, including us, would like the answer to be a simple yes: add an app, a reminder or a game to the home program and patients will do more of their exercises. The research does not say that. It says something more useful and less tidy. Digital support sometimes helps adherence, often for a while, and sometimes not at all. The one randomized trial we found in hand therapy itself was a null result.

This post walks through that evidence, including the parts that are inconvenient for anyone selling software to clinics. If you are weighing up a gamified or app-based tool for your patients, we think you should know all of it.

The short answer

• In hand therapy specifically, the best-known trial of an app with exercise videos and push reminders found no improvement in adherence or self-efficacy.
• Across musculoskeletal care, reviews disagree. Some find a benefit at mid-term follow-up or in most included trials; one found no significant effect at all. Long-term benefit is the least certain part.
• Games have been tested in thumb rehabilitation in one small trial. Adherence was better; the primary outcome was not.
• What looks most robust is not a technology but a design principle: interventions that combine many behavior-change techniques do better than those that use only a few.
• Camera-based rep counting with playful visual feedback, the kind of tool Handara is, has not been tested in a hand therapy trial that we could find.

The hand-specific trial: an app that did not move adherence

The most directly relevant study comes from Sweden. Svingen and colleagues (Clinical Rehabilitation, 2021) ran a multi-center randomized controlled trial across four hand surgery departments with 101 patients after flexor tendon repair. The intervention group got a smartphone app with exercise videos, push-notification reminders, an exercise diary with a calendar view, written information about the surgery and a question-and-answer section. The control group got standard rehabilitation.

The primary outcome was adherence, measured with the Sport Injury Rehabilitation Adherence Scale at two and six weeks. There was no significant difference between the groups. Self-efficacy did not differ either. The authors' conclusion is plain: the app did not improve adherence or self-efficacy compared with standard rehabilitation, and more research is needed.

Two details are worth holding on to, because they cut both ways.

1. The app group did report a better exercise frequency at six weeks on one self-reported measure. That is a secondary outcome, and self-report in this population may overestimate adherence, as the authors themselves note.
2. The authors list among the limitations that the app had no personal feedback features. It reminded patients and showed them what to do, but it did not tell them how they were doing.

That second point matters for anyone designing or buying a tool. "An app" is not one thing. Reminders plus videos is a specific bundle, and in this trial that bundle was not enough.

The wider picture is mixed

Digital rehabilitation in musculoskeletal care

Three reviews look at digital support for exercise adherence more broadly, and they do not line up neatly.

• Zhang and colleagues (JOSPT, 2022) pooled 11 randomized trials with 1,144 participants with musculoskeletal conditions. Digital rehabilitation improved exercise adherence at intermediate-term follow-up, but at short-term follow-up it was no better than non-digital rehabilitation, and at long-term follow-up there was no clinically important difference.
• Bunting and colleagues (Physiotherapy, 2021) looked at chronic musculoskeletal pain. Only five trials met their criteria and two could be pooled. They found no statistically significant difference in adherence and concluded there was no evidence that digital interventions enhance adherence in that group, while calling for better research.
• Lang and colleagues (Archives of Physiotherapy, 2022) reviewed 10 randomized trials with 1,117 participants across clinical populations. Seven favored the digital intervention and three found equivalent adherence. The three that found no difference were the ones that reported longer-term outcomes. Their conclusion: digital additions can likely increase adherence in the short term, with longer-term effects less certain.

Read together, the pattern is a benefit that may appear early or in the middle of a program and then fade. That fits what many therapists see in practice: novelty helps for a while. The reviews do not establish why the effect fades, so read that explanation as a reasonable guess rather than a finding.

Games in thumb rehabilitation

Gamification has been tested in the hand, though barely. Özer and colleagues (Hand Therapy, 2025) randomized 31 patients with thumb involvement after injury or carpal tunnel syndrome to routine rehabilitation or to mobile games prescribed as home exercise. Both groups improved. There was no difference in the primary outcome. The game group did better on adherence to home exercises and on one joint-movement measure. The authors call the approach feasible and say more research is needed to establish clinical relevance.

That is an encouraging signal for adherence, from a small trial, with a null primary outcome. It is not evidence that games improve recovery.

What seems to work: many behavior-change techniques, combined

The broadest piece of evidence is an overview of 19 systematic reviews covering 205 trials by Ley and Putz (Systematic Reviews, 2024). Pooled across four meta-analyses (booster sessions, behavior-change techniques, goal setting and motivational interventions), additional adherence interventions had a small overall effect, with low certainty of evidence. The abstract notes moderate evidence for booster sessions, supervision and graded exercise.

The finding that matters most here: interventions using eight or more behavior-change techniques had a significantly higher effect (standardized mean difference 0.29) than those using fewer than eight, which showed no clear effect. Techniques such as goal setting, action planning, social support, and feedback and self-monitoring recur across the included reviews.

One caveat makes this review both more and less relevant to software. It explicitly excluded reviews that analyzed only digital apps or tools, including gamification, exergames, virtual reality and tele-rehabilitation. So it tells you what ingredients tend to work. It does not tell you whether any particular app delivers them.

What has not been tested

We looked for randomized trials of camera-based exercise tracking, automatic repetition counting, or playful visual feedback such as particle animations in hand therapy. We did not find any.

That gap includes Handara. Handara has no evidence base of its own. We have not run a trial, we have no adherence data from real patients yet, and nothing in this post should be read as evidence that our approach works. The studies above test other tools: reminder apps, video libraries, mobile games and broader behavior-change programs. Their results may or may not transfer.

We point this out because vendor blogs on this topic tend to cite the positive reviews, skip the null results and imply that their product inherits the good news. It does not work that way, for us or for anyone else.

Why the results might be mixed

The trials do not answer this directly, but the literature offers a few plausible reasons.

• "Digital" covers very different things. A push reminder, a video library, a game and a remote program with therapist contact are not the same intervention. Pooling them can blur real effects in both directions.
• Reminders address forgetting, not everything else. A Dutch study of physiotherapy practice by Sluijs, Kok and van der Zee (1993) found the main factors linked to non-compliance were the barriers patients perceive and encounter, a lack of positive feedback, and helplessness. A notification touches none of these directly.
• Feedback may be the missing piece. The Svingen app had none; Sluijs names its absence as a correlate of non-compliance; Ley and Putz list feedback and monitoring among the recurring techniques. That is a pattern worth testing, not a proven mechanism.
• Adherence is hard to measure. Most trials rely on self-report or exercise diaries, which can overestimate what patients actually did. Ley and Putz flag adherence measurement in physiotherapy as a research gap.

Practical take-aways for choosing any tool

Whatever you use, including nothing digital at all, these questions follow from the evidence above.

1. Count the behavior-change techniques. Ask what the tool actually does besides send reminders. Does it support goal setting, action planning, self-monitoring, feedback and social support? Ley and Putz suggest combinations matter more than any single feature.
2. Make feedback visible. Patients should see that they did the exercise and how it went, and you should see it too. A tool that only reminds and instructs is the bundle that failed in the flexor tendon trial.
3. Keep the program short and simple. Every extra exercise is another perceived barrier. Fewer, clearer exercises are easier to fit into a day and easier to give feedback on.
4. Plan for the fade. Several reviews suggest effects are strongest early or mid-program. Build in follow-up contact or booster sessions rather than expecting an app to carry adherence for months.
5. Ask vendors what evidence they actually have. Is there a trial of their product, in your population? Or are they citing reviews of other tools? "Research shows digital tools improve adherence" is not the same as "our tool improves adherence".
6. Check how adherence is measured. If a tool reports adherence, find out whether it is self-reported, logged automatically or observed. That decides how much you can trust the dashboard.
7. Run your own small test. If you try a tool, decide beforehand what you will look at, such as completed sessions over the first six weeks, and compare honestly with what you saw before.

The honest summary for clinicians is this: reminders and apps are not a reliable fix on their own, games show early promise in a single small hand trial, and the strongest evidence points to combining several behavior-change techniques with visible feedback. Any tool, ours included, should be judged against that, and against its own data once it has some.