Family support delivered through ordinary pediatric practices helped more children reach a weight-development goal after one year in the STARKIDS trial. Six months later, the advantage over a single structured counseling session was no longer clear. That contrast makes the study useful: it asks not only whether accessible care can help, but what happens when regular contact ends.
The trial report, published October 14, 2026 in The Lancet Regional Health – Europe, evaluated a German program combining face-to-face sessions with online resources. It was a family lifestyle intervention, not a medication trial or a test of an app used alone. Its findings offer a reason to examine how ongoing support is organized, without promising lasting weight loss from a short course.
What did the one-year result actually measure?
The researchers defined success as a reduction of at least 0.2 in BMI z-score. This expresses BMI relative to an age- and sex-specific reference, rather than kilograms lost. In growing children, it is a different measure from the percentage body-weight loss often reported in adult drug trials.
At 12 months, the analysis that accounted for missing measurements estimated that 33% of STARKIDS participants and 22% of the comparison group reached that threshold. The reported difference was 11 percentage points, with a 95% confidence interval of 3 to 20 points. Among children with measurements at that visit, the corresponding counts were 83 of 251 and 37 of 170. These are response proportions from the trial, not 33% and 22% reductions in body weight.
At 18 months, the study did not establish a difference between groups in reaching the same threshold. This does not mean every child lost all earlier progress, or prove that the two approaches are equivalent. It means the evidence for an additional STARKIDS benefit was no longer convincing at that follow-up. Our guide to reading weight-loss trials explains why the measured outcome and time point must travel with a headline result.
More than an online program—and more than no treatment
Practices, rather than individual families, were randomized. Of 89 assigned practices, 71 enrolled families. The trial enrolled 611 families with a child aged 3–17 who had overweight or obesity; 559 completed baseline assessments and started their assigned care. The main analysis included 343 in STARKIDS and 216 in the comparison group.
During the first year, STARKIDS offered five core sessions, an optional additional session and an online resource library. The published intervention description specifies 90-minute appointments every three months with trained medical assistants. Topics included eating, enjoyable activity, media habits, sleep, family life and managing setbacks. The digital material supported work between appointments. This describes the research program, not a schedule families should prescribe for themselves.
The comparison group received one structured 30-minute counseling appointment and written recommendations, alongside usual care. It was therefore an active comparison, not a group denied support. Neither the whole program’s results nor its digital component can be treated as a comparison against doing nothing. The distinction also matters when interpreting randomized evidence gathered in everyday care.
The later support step reached few eligible families
The original protocol planned a second step for families whose children had not made sufficient progress after the first year. Local public health services would offer one or two counseling sessions over the next six months, helping families connect with further support. Online access continued.
In practice, just 24 of 162 eligible families—15%—attended that step, according to the results report. That is an important delivery problem. An offered service cannot provide much support if most eligible families do not use it.
Low attendance is a plausible clue to the loss of the between-group advantage, but it does not prove the cause. The study did not randomly assign families to different maintenance schedules. The authors propose keeping later support closer to the familiar practice and tapering contact more gradually. Whether that redesign produces more durable benefits still needs testing.
Weight response did not translate into a clear wellbeing gain
Adolescents’ self-reported quality of life did not differ significantly between groups at either follow-up. Parents’ ratings of their children’s quality of life showed a small difference favoring the comparison group at 18 months, which the authors did not regard as clinically meaningful. This was a parent report about the child, not a measure of the parent’s own wellbeing. The trial’s quality-of-life and safety results include no treatment-related serious adverse events; that does not establish the absence of every possible harm.
About a quarter of the 559 participants who started assigned care lacked measurements at each follow-up. The main analysis used multiple imputation, estimating missing values under assumptions about why they were missing. A separate analysis of complete cases produced consistent findings, but cannot remove every concern about missing data. Practices were unblinded, and the study took place in one German state with German-language eligibility requirements.
What should families take from this?
The practical question is how support continues after initial appointments. Families discussing a program can ask who provides follow-up, what happens if attendance is difficult, and how wellbeing is assessed alongside growth. Our questions for a dietitian can help prepare a conversation; a child’s needs require pediatric assessment rather than an adult weight-loss target.
The trial registration records completion in December 2024, so this is newly published research, not a newly finished trial. The full paper, protocol and registry were inspected for this report; supplementary files and the separate economic evaluation were not. The evidence supports a one-year benefit from accessible family care while leaving the best way to maintain that advantage unresolved.



