For many individuals receiving inpatient treatment for eating disorders, mealtimes and the period that follows can be particularly challenging. Patients may engage in ritualised behaviours such as counting, checking, delaying, or attempting to compensate for food intake (Levinson et al., 2018; read Sam and Caitlin’s Mental Elf blog to learn more). Although rest is commonly recommended after meals to prevent compensatory behaviours, enforced inactivity may intensify anxiety, intrusive thoughts, and established eating disorder routines (Griffiths et al., 2016).
Previous research suggests that relaxation techniques, guided imagery, and structured activities may improve patients’ emotional experiences after meals (Griffiths et al., 2016; Shapiro et al., 2008). However, these studies did not establish whether improvements in emotional wellbeing were accompanied by reductions in eating disorder behaviours. This raises the question of whether alternative approaches to post-meal rest could help patients manage this challenging period.
A recent study by Stedal and colleagues (2026) examined whether carefully supervised, mild post-meal movement was associated with fewer ritualised eating behaviours among inpatients with eating disorders. Using ecological momentary assessment (EMA), where people report their experiences and behaviours in real time, the researchers compared periods of standard post-meal rest with periods when structured movement was introduced.
Methods
This two-week study involved adult female inpatients at two specialist eating disorder services in Norway and used a counterbalanced within-participant design. Each participant completed one week of the standard 30-minute post-meal rest period and one week that included a supervised 10-minute session of gentle movement, breathing, and body-awareness exercises before the remainder of the rest period. The order of the two conditions varied across participants to reduce potential sequencing effects.
Participants received six EMA prompts daily via a smartphone application, four of which were scheduled around mealtimes. These short survey questions asked about recent behaviours and feelings. After reporting whether they had eaten, participants indicated whether they had engaged in ritualised eating behaviours, choosing from ‘made an effort to reduce intake’, ‘eating and/or preparation rituals’ or ‘other’. Researchers analysed the repeated observations using a binomial generalised linear mixed-effects model.
Results
The sample included 33 adult female inpatients with a mean age of 27.5 years. Most participants were diagnosed with anorexia nervosa (n = 23), while four had atypical anorexia nervosa and six had bulimia nervosa. The mean baseline body mass index (BMI) was 18.4 kg/m², ranging from 12.4 to 24.1 kg/m². Fourteen participants had comorbid mental health disorders, such as depression, generalised anxiety disorder, or post-traumatic stress disorder (PTSD).
Over the two-week period, participants completed 1,503 EMA prompts, yielding a compliance rate of 81.3%, which is considered high for studies requiring repeated real-world reporting. The authors assessed ritualised behaviours in 1,192 meal-related prompts. Participants reported these behaviours in 30.8% of meal-related assessments during the rest-only week and 24.2% during the week with supervised post-meal activity, an absolute reduction of approximately 6.6 percentage points. Eating and food-preparation rituals were the most common (reported in 22% of meal-related prompts), followed by attempts to reduce food intake (14.4%). Less common behaviours, grouped under an ‘other’ category (2% of meal-related prompts), included pacing, food manipulation, and delaying eating.
A similar pattern was found after accounting for the repeated observations within participants, with the estimated probability of ritualised behaviour decreasing from approximately 26% in the rest-only condition to 16% with post-meal activity. The odds of reporting ritualised behaviour were approximately 50% lower during weeks when post-meal activity was available (OR = 0.50; 95% CI [0.34 to 0.72]), which was statistically significant (p = .0002).
Ritualised eating behaviours remained prevalent and varied substantially among participants, despite being reported less frequently during the activity week. The effect remained consistent after accounting for meal type and treatment duration, suggesting that the primary finding was not attributable solely to differences among meals or increased familiarity with treatment.
Conclusions
This small study provides preliminary evidence that a brief, structured period of mild post-meal activity may be associated with reduced ritualised eating behaviours among adult female inpatients with eating disorders. The authors suggested that medically safe, supervised, structured movement could help disrupt habitual behaviours, although its effects on post-meal distress were not examined.
The results highlight the importance of further research on the effects of carefully supervised movement that considers the patient’s preferences as a possible element of inpatient care. Importantly, they do not indicate that it is advisable to promote unregulated exercise or to remove medical precautions.
Strengths and limitations
Strengths
A major strength of the study is its within-person design, as each participant experienced both treatment conditions. This approach enabled comparisons within the same individuals rather than between separate groups, thereby minimising the influence of stable personal differences such as diagnosis, personality, symptom history, or typical coping strategies. Additionally, the counterbalanced order reduced the likelihood that all participants received the intervention at the same stage of treatment.
Another advantage is the use of EMA. Since the thoughts and behaviours associated with eating disorders can vary from one meal to the next, it may be more informative to ask participants about their most recent experiences than to ask them to recall an entire week or their whole treatment admission; this would reduce recall bias and support more accurate information about moment-to-moment experiences (Schaefer et al., 2020). The high number of recorded observations through repeated assessments in this study (81.3% completion rate and 1,192 meal-related observations) strengthened the within-person analyses, although the sample itself remained small.
The involvement of three individuals with lived experience helped refine the smartphone protocol and movement videos. This collaboration ensured that patient perspectives informed the study’s development and showed how lived experience involvement can influence both research methods and clinical procedures. This collaboration helped ensure that patient perspectives informed the intervention’s development and demonstrated how lived experience involvement can influence both research methods and clinical procedures.
Limitations
This study has several important limitations. First, the small sample size included only adult female inpatients in Norway, predominantly with anorexia nervosa. This limits the generalisability of findings to broader populations, such as adolescents, men, non-binary individuals, outpatients, or those with other eating disorders. Participants also required medical clearance to take part, limiting what the findings can tell us about patients for whom physical activity may be unsafe.
Second, because the study was not blinded, both participants and staff were aware of the movement programme. This awareness, along with additional factors such as changing location, staff support, and structured attention, may have influenced outcomes. Therefore, it is not possible to separate the effect of movement from these other factors, such as expectation, social interaction, or environmental change.
Third, the study relied on self-reported data about the presence or absence of rituals, which does not capture the intensity or duration of these behaviours, or their impact on recovery. Repeated smartphone prompts could also have altered participants’ awareness or behaviour, potentially biasing responses. Moreover, the study did not examine the intervention’s effect on anxiety, nutritional intake, weight restoration, or long-term outcomes. As a result, the reduction in ritualised behaviours should be interpreted as evidence of an association, as opposed to causal evidence that movement was responsible for the observed outcomes.
Finally, as each condition lasted only one week, it remains unclear if the observed effects would persist over time.
Implications for practice
The current evidence does not justify a general shift from post-meal rest to exercise. However, it challenges the assumption that complete inactivity is always optimal for managing post-meal discomfort. For some patients, carefully supervised, low-impact movement may offer an alternative to remaining seated during the post-meal period, although further research is needed to understand its effects on anxiety and distress.
Clinical services should not interpret this intervention as unrestricted exercise. The authors recommend that movement be introduced within medically safe, structured limits, and under supervision, in line with wider evidence on supervised exercise in anorexia nervosa (Miñano-Garrido et al., 2025; Ng et al., 2013). Drawing on wider evidence, activities should also be tailored to the individual’s nutritional status, cardiovascular stability, injury risk, medication use, history of compulsive exercise, and stage of recovery (Brunet et al., 2021). The main aim is to support emotional regulation and help patients build tolerance to post-meal discomfort.
The study further suggests that collaboration (including through patient choice and personal experience) was central to developing and implementing the intervention. The intervention received input from former inpatients and involved both staff and patients, with the aim of fostering an inclusive environment.
Future research would benefit from larger, multi-site samples that include individuals of diverse genders, ages, diagnoses, and medical risk levels. A valuable next step would be a trial comparing gentle movement with an attention-matched non-movement activity, such as guided relaxation or a comparable group session. This design could help determine whether reduced ritualised behaviours result from movement itself or from related factors, including reduced boredom, staff interaction, distraction, breathing exercises, or a structured post-meal plan. Researchers should also evaluate anxiety, distress, intrusive thoughts, and emotion regulation alongside the presence of ritualised behaviours. Extended follow-up will be necessary to determine whether any benefits persist after discharge. While objective measures may be considered, they should be used judiciously, as excessive monitoring could inadvertently reinforce eating disorder preoccupations.
In the absence of further evidence, it is important to note that there are currently no universally accepted guidelines for safely incorporating exercise into the treatment of anorexia nervosa, and not all services have staff trained to supervise movement in severely malnourished patients. Therefore, services should assess whether they have the necessary expertise and safety procedures in place before modifying any existing practices.
Statement of interests
Ghufran Al Syed has no conflicting interests to declare.
Edited by
Dr Nina Higson-Sweeney.
Links
Primary paper
Kristin Stedal, Line Wisting, Joseph Wonderlich, Anne L. Wennersberg, Annette Skarsbakk, Ida Aarmo, Selma Ø. Lie, & Ingrid Funderud (2026). Post-meal activity and eating disorder behaviors: An ecological momentary assessment (EMA) study. International Journal of Eating Disorders, 59(10), 2153–2160. https://doi.org/10.1002/eat.70137
Other references
Brunet, J., Del Duchetto, F., & Wurz, A. (2021). Physical activity behaviors and attitudes among women with an eating disorder: A qualitative study. Journal of Eating Disorders, 9, Article 20. https://doi.org/10.1186/s40337-021-00377-w
Griffiths, E., Hawkes, N., Gilbert, S., & Serpell, L. (2016). Improving the post-meal experience of hospitalised patients with eating disorders using visuospatial, verbal and somatic activities. Journal of Eating Disorders, 4, Article 9. https://doi.org/10.1186/s40337-016-0098-y
Han, X. S., & Lloyd, C. (2025). How do anorexia patients and clinicians feel about supported mealtimes?. The Mental Elf.
Levinson, C. A., Sala, M., Fewell, L., Brosof, L. C., Fournier, L., & Lenze, E. J. (2018). Meal and snack-time eating disorder cognitions predict eating disorder behaviors and vice versa in a treatment seeking sample: A mobile technology based ecological momentary assessment study. Behaviour Research and Therapy, 105, 36–42. https://doi.org/10.1016/j.brat.2018.03.008
Miñano-Garrido, E., Catalán-Matamoros, D., Pagán Acosta, G., & Gómez-Conesa, A. (2025). The effect of anorexia nervosa inpatient physiotherapy adapted program (ANIPAP): A clinical trial. Journal of Bodywork and Movement Therapies, 42, 337–343. https://doi.org/10.1016/j.jbmt.2024.12.043
Ng, L. W. C., Ng, D. P., & Wong, W. P. (2013). Is supervised exercise training safe in patients with anorexia nervosa? A meta-analysis. Physiotherapy, 99(1), 1–11. https://doi.org/10.1016/j.physio.2012.05.006
Schaefer, L. M., Engel, S. G., & Wonderlich, S. A. (2020). Ecological momentary assessment in eating disorders research: Recent findings and promising new directions. Current Opinion in Psychiatry, 33(6), 528–533. https://doi.org/10.1097/YCO.0000000000000639
Shapiro, J. R., Pisetsky, E. M., Crenshaw, W., Spainhour, S., Hamer, R. M., Dymek-Valentine, M., & Bulik, C. M. (2008). Exploratory study to decrease postprandial anxiety: Just relax! International Journal of Eating Disorders, 41(8), 728–733. https://doi.org/10.1002/eat.20552