Digi-physical obesity programme shows comparable 26-week outcomes in Abu Dhabi and historical Stockholm cohort

A 26-week non-inferiority intervention study found comparable adjusted changes in body mass index z-score among children with obesity who received a digi-physical treatment programme in Abu Dhabi and children in an external historical cohort treated with the same approach in Stockholm. The analysis included 35 Abu Dhabi participants and 77 Stockholm participants with eligible follow-up data. It was not a randomized comparison with standard care.

The Abu Dhabi intervention took place at the paediatric endocrine clinic at Sheikh Shakhbout Medical City. Sixty children aged 6 to 15.9 years with obesity were recruited consecutively between June and December 2023 and included after eligibility screening. Of these, 35, or 58.3%, had eligible follow-up data at 26 weeks. The Stockholm comparison came from a previously published pragmatic trial at an obesity clinic in Stockholm; 107 patients had been included in that trial, and 77, or 72.0%, contributed data at the corresponding six-month assessment.

The intervention, called Evira, combined a digitless home body scale with a mobile application, a web-based clinic interface and a family-centred working method. Home measurements were transmitted to an application on a parent’s phone and were accessible to healthcare staff. Families and clinicians jointly set individualized goals, while the programme encouraged families to develop their own strategies rather than prescribing specific dietary or physical-activity routines. Clinicians could respond through digital communication when targets were not met or measurements were missed.

The primary outcome was the change in BMI z-score from baseline to 26 weeks. Analyses adjusted for sex, age and degree of obesity at treatment initiation. Non-inferiority was evaluated against a prespecified margin of 0.10 BMI z-score units: the criterion was met if the upper bound of the 95% confidence interval for the between-cohort difference did not exceed that margin. The study was registered as NCT06665893.

Adjusted mean BMI z-score changes were −0.20 units in Abu Dhabi, with a 95% confidence interval of −0.28 to −0.12, and −0.20 units in Stockholm, with a 95% confidence interval of −0.26 to −0.14. The difference between cohorts was not statistically significant (p=0.88). The unadjusted mean difference was 0.0004 units, with a 95% confidence interval of −0.004 to 0.005. Because the upper confidence limit of 0.005 was below the prespecified 0.10-unit margin, the researchers confirmed non-inferiority under their defined criterion.

Among participants with analysed outcomes, 45.7% of the Abu Dhabi group achieved a reduction of at least 0.20 BMI z-score units at 26 weeks, compared with 36.4% in Stockholm (p=0.35). Obesity remission occurred in 14.3% and 23.4% of the respective cohorts (p=0.27). Within the Abu Dhabi cohort, participants referred specifically for obesity had outcomes comparable to the Stockholm group. The seven participants referred primarily for other endocrine concerns showed more variable outcomes, with no clinically significant BMI z-score reduction or remission in that subgroup. The authors said this exploratory subgroup analysis illustrated possible influences of referral reason, motivation and engagement and was not a direct comparison of intervention efficacy.

Non-retention at 26 weeks was numerically higher in Abu Dhabi than in Stockholm: 41.7% versus 28.0% (p=0.07). The corresponding odds ratio from univariate logistic regression was 1.83, with a 95% confidence interval of 0.94 to 3.57. Only observed data were analysed, so treatment effects may have been overestimated if participants lost to follow-up had less favourable outcomes than those who remained. No adverse events were reported in the electronic case report form during follow-up.

Home weighing averaged approximately 4.9 measurements per patient per week in Abu Dhabi and 5.0 in Stockholm during the first month (p=0.18), falling to 2.9 and 3.0 measurements, respectively, by the sixth month (p=0.54). Over the full 26 weeks, clinics sent an average of 31.2 messages per patient in Abu Dhabi and 29.9 in Stockholm (p=0.50), while patients sent averages of 15.5 and 16.7 messages, respectively (p=0.49). The study reported home-measurement frequency at months 1 and 6 and aggregate message counts over 26 weeks, but did not establish whether message engagement declined over time.

The findings support the feasibility and potential effectiveness of the digi-physical programme in the studied Abu Dhabi population, but they do not provide a contemporaneous test against standard care. Important limitations include the non-randomized historical comparator, small sample, 26-week follow-up, substantial non-retention and reliance on observed data. Eligibility requirements for local-language communication and access to a smartphone and email may also have introduced selection bias. Both study settings had developed digital infrastructure, limiting generalizability to populations with different access or healthcare systems. The evidence is therefore promising but bounded rather than definitive proof of comparative effectiveness.

Subject of Research: Non-inferiority evaluation of a digi-physical paediatric obesity treatment programme in Abu Dhabi against a historical Stockholm cohort

Article Title: Effectiveness of a digi-physical tool and working method for paediatric obesity treatment in Abu Dhabi: a non-inferiority intervention study using an external historical comparator

Article References: Lindberg, L., Hassan Beck, R., Danielsson, P., Nielsen, T., Abdalla, A., Ekstedt, P., Chaturvedi, D., Al Qahtani, N., Marcus, C., & Deeb, A. (2026). Effectiveness of a digi-physical tool and working method for paediatric obesity treatment in Abu Dhabi: a non-inferiority intervention study using an external historical comparator. BMC Pediatrics, 26(1), Article 868. https://doi.org/10.1186/s12887-026-07623-7

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07623-7

Keywords: paediatric obesity, BMI z-score, non-inferiority study, digital health, digi-physical treatment, Abu Dhabi, Stockholm, treatment adherence, historical comparator, Effectiveness, digi-physical, tool