Exercise Therapy Comes Home: Video-Guided Workouts Show Promise for Anorexia Recovery

For decades, one of the most stubborn dilemmas in treating anorexia nervosa has been what to do about exercise. In many patients, physical activity is not a healthy habit but a symptom, driven by compulsive urges to burn calories and reshape the body. Yet adapted physical activity programs, in which qualified instructors guide patients through carefully structured movement, have repeatedly shown that exercise can be relearned as something restorative rather than punishing. The catch has always been access: these programs have been developed almost exclusively in hospital settings, leaving the majority of outpatients with no realistic way to benefit. A new pilot study published in the Journal of Eating Disorders suggests that the solution may be as simple and as radical as bringing the therapy into the living room.

The study, known as APAMAdom, was conducted by a team of French researchers led by Béatrice Genoux of Grenoble Alpes University Hospital and Cécile Bétry, with collaborators including Sébastien Guillaume of Montpellier University Hospital. It tested whether a home-based adapted physical activity program delivered entirely by videoconference could be safe and acceptable for adult women with anorexia nervosa being treated as outpatients. The trial was registered on ClinicalTrials.gov as NCT05803707 and approved by a French ethics committee, the CPP Sud-Ouest et Outre-Mer II in Toulouse, and all participants gave written informed consent before enrollment.

The design was deliberately pragmatic. Adult women aged 18 to 60 with anorexia nervosa and a body mass index below 18.5 kilograms per square meter were recruited through two specialized eating disorder centers attached to French university hospitals. The program itself consisted of ten sessions in total, of which eight were weekly 45-minute adapted physical activity sessions led by qualified instructors over video. Researchers tracked a broad set of outcomes before and after the program, including body weight, body composition, physical activity levels, exercise dependence, quality of life, and self-esteem. Because this was a pilot study, the results were reported descriptively, with pre-to-post comparisons carried out using a paired Wilcoxon signed-rank test, a non-parametric method suited to small samples.

Thirty-six women ultimately enrolled, and their profiles underscore how severe the illness can be in this population. The median age was 33 years, with an interquartile range of 26 to 47. The median body mass index was just 15.6 kilograms per square meter, well below the diagnostic threshold for anorexia, and the median duration of illness was a striking ten years, with some participants having lived with the disorder for up to seventeen years. These were not people at the margins of the illness; they were women in its deep chronic phase, precisely the group for whom outpatient options are scarcest.

The safety findings are the study’s most reassuring contribution. Throughout the program, body weight, body composition, physical activity levels, and exercise dependence all remained stable. Only one participant was withdrawn from the study, and that was due to weight loss exceeding five percent of her baseline body weight, a pre-specified safety threshold. No other adverse pattern emerged. For a therapy involving deliberate physical movement in patients with dangerously low body weight, that stability matters enormously, because the central clinical fear has always been that exercise interventions could inadvertently fuel weight loss or reinforce compulsive activity.

Satisfaction, meanwhile, ran high. Among the 22 participants who responded to the satisfaction assessment, 77.3 percent said they wished to continue the program. That figure is notable given the population involved: patients with anorexia nervosa often avoid or drop out of interventions that touch on the body, and the therapeutic relationship around exercise is fraught. The fact that most women wanted more sessions suggests that the videoconference format, far from feeling impersonal, may have created a comfortable middle ground between clinical supervision and domestic privacy.

The participation data, however, tell a more complicated and arguably more valuable story. Of 66 eligible participants, only 36.4 percent completed the first session, a measure the researchers call the initial uptake rate. Among the 36 women who enrolled, just five, or 13.9 percent, completed all ten sessions. The retention rate for the end-of-program visit was 63.9 percent. These numbers are sobering, but they are exactly the kind of real-world estimates that pilot studies exist to produce. A future randomized controlled trial can now be sized and structured with realistic expectations about how many patients will start such a program and how many will see it through, rather than with the optimistic assumptions that have undermined many behavioral interventions.

Beneath the headline numbers, there were hints of genuine psychological benefit. The researchers observed medium effect sizes suggesting improvement in quality of life, with a correlation coefficient of r = 0.40, and in self-esteem, with r = 0.36. In a study of this size, such differences are too small to draw firm conclusions, and the authors are careful to say so. But the direction of the effects is consistent with what inpatient adapted physical activity programs have reported, and it gives the proposed primary outcome for the future trial, quality of life, a plausible empirical footing. If a supervised movement program can lift self-esteem in women whose relationship with their bodies is profoundly damaged, without shifting weight in either direction, that would be a meaningful therapeutic gain in its own right.

The technological dimension of the study deserves attention in its own right. Telehealth expanded explosively during the COVID-19 pandemic, but mental health applications for eating disorders have lagged, partly because clinicians worried that remote formats would weaken the monitoring that keeps vulnerable patients safe. APAMAdom offers a counterexample: a structured, video-supervised intervention with explicit safety thresholds, delivered by instructors employed through a home healthcare provider, that maintained weight stability in a severely ill cohort. The funding structure is also worth noting. The study was financed by an Innovadom grant from the Agiradom association, a home healthcare provider whose instructors collected the data, but the funder had no role in the conceptualization, design, analysis, decision to publish, or preparation of the manuscript, and the authors disclose only modest honoraria and hospitality fees.

What emerges from APAMAdom is less a triumph than a blueprint, and that is precisely its value. The study demonstrates that a home-based, video-supervised adapted physical activity program is feasible, safe, and wanted by the patients who try it, while honestly documenting the attrition that any larger trial must plan around. For the millions of people worldwide living with anorexia nervosa, most of whom will never set foot in a specialized inpatient unit, the prospect of receiving structured movement therapy at home, through a screen, with a qualified instructor watching over them, represents a genuine widening of the treatment map. The randomized controlled trial that the researchers now propose, with quality of life as its primary endpoint, will determine whether that promise holds. But the first, most dangerous question, whether exercise therapy can be delivered safely outside the hospital walls, has now been answered with a cautious yes.

Subject of Research: Home-based video-supervised adapted physical activity for adult women with anorexia nervosa

Article Title: A home-based video-supervised adapted physical activity program in adult women with anorexia nervosa: a pilot study

Article References: Genoux, B., Guillaume, S., Buis, C., Bakouline, M. C., Bosson, J.-L., & Bétry, C. (2026). A home-based video-supervised adapted physical activity program in adult women with anorexia nervosa: a pilot study. Journal of Eating Disorders. https://doi.org/10.1186/s40337-026-01771-y

Image Credits: AI Generated

DOI: 10.1186/s40337-026-01771-y

Keywords: anorexia nervosa, adapted physical activity, telehealth, videoconferencing, eating disorders, pilot study, quality of life, self-esteem, exercise dependence, outpatients, body composition, mental health