How to evaluate treatment outcomes in childhood obesity

In the Swedish one-year study, 107 children and adolescents received digitally supported treatment at an outpatient clinic in Stockholm. Results were compared with 321 age- and sex-matched patients receiving standard care across other clinics, identified via the Swedish childhood obesity treatment register (BORIS).

The three-year study followed these same cohorts longitudinally, providing data on how outcomes were maintained over an extended period.

| Outcome | With Evira | Standard care | | - | - | - | | Mean change in BMI z-score at one year | −0.30 | −0.15 | | Adjusted mean change in BMI z-score at three years | −0.29 | −0.12 | | Proportion achieving obesity remission during three-year follow-up | 31.8% | 18.7% |

Results from Hagman et al., published in the *International Journal of Obesity* 2022 and 2025. Patients were not randomised; the comparison was conducted using matched registry controls.

BMI z-score, also termed BMI SDS, standardises BMI relative to a reference population for the child’s age and sex. This allows clinicians to track relative weight trajectories as the child grows.

Obesity remission is defined here as a child reducing their BMI below the diagnostic cut-off for obesity. This metric provides a categorical perspective distinct from mean changes in cohort BMI z-score. It does not automatically imply the child has achieved a normal weight or that the need for clinical follow-up has ended.

When comparing clinical outcomes, metrics and follow-up intervals must be strictly aligned. A six-month change and a three-year outcome answer different clinical questions. Furthermore, verify how studies account for missing data and loss to follow-up. Completer-only analyses often paint a different picture from intention-to-treat analyses that account for the entire baseline cohort.

Children in Evira’s one-year study were aged 4.1–17.4 years at baseline. The digitally supported group demonstrated superior outcomes compared with controls across both younger (<12 years) and older (≥12 years) cohorts. This supports clinical utility across a broad paediatric age range, though family support strategies must still be tailored to the child's developmental stage. Read the one-year study.

When conducting a local appraisal, consider factors beyond age. How long have patients been in treatment? What is the distribution of obesity severity and medical complexity? Which families struggle to attend regular hospital appointments?

For tertiary paediatric centres, it is important to identify which segments of your patient cohort match the secondary outpatient population evaluated in the study. Patients referred for highly specialised care may present with different clinical complexities. This influences both realistic outcome expectations and how clinical pathways should be structured.

The evaluated treatment model comprised home measurements, visual progress tracking, individualised goals, and asynchronous messaging with the clinical team, integrated alongside scheduled clinic visits. The reported outcomes reflect the totality of this combined clinical package. Read the study design.

For clinics planning to adopt Evira, practical clinical governance forms a key part of the assessment. Sufficient time and defined responsibility must be allocated to monitor patients, provide feedback to families, and adapt care plans based on incoming data.

A clinical demonstration following an end-to-end patient journey is often more informative than a simple software feature tour. Observing how a team detects a trend, communicates with the family, and re-evaluates the clinical plan clarifies how the tool functions in real-world practice.

Define at the outset which patient cohorts you will monitor and at what intervals you will audit service delivery. Track relative weight trajectories alongside retention rates and loss to follow-up.

Equally, evaluate implementation metrics. Are families engaging with onboarding? Are they receiving timely clinical feedback? Are cross-cover arrangements functioning effectively? Tracking these parameters helps explain clinical outcomes and highlights where service pathways require refinement.

When comparing outcomes against historical clinic data, ensure patient cohorts, baseline characteristics, and follow-up durations are comparable. Variations in clinical staffing or concurrent service reconfigurations must also be accounted for.