One trial in adolescents gets quoted everywhere. It is a good trial and this site has its own page on it. This is what the rest of the evidence looks like when you pool all of it.
What was pooled
Eighteen randomized controlled trials — eleven in obesity, six in type 2 diabetes, one in prediabetes — covering 1,402 participants, of whom 838 received a GLP-1 and 564 placebo. [1] Mean age was 13.7, with a range from six to seventeen. Fifty-nine per cent were girls.
The pooled results: HbA1c fell 0.44 percentage points, 95% CI −0.68 to −0.21. Fasting glucose fell 9.92 mg/dL, 95% CI −16.20 to −3.64. Body weight fell 3.02 kg, 95% CI −4.98 to −1.06. BMI fell 1.45 points, 95% CI −2.40 to −0.49, with BMI standard deviation score down 0.20 and BMI percentile down 7.24. Systolic blood pressure fell 2.73 mm Hg, 95% CI −4.04 to −1.43. Gastrointestinal adverse events rose.
Half a year
The number that should stop you is the duration. Median treatment across the entire randomized evidence base in children was 0.51 years, with an interquartile range of 0.25 to 1.00.
A thirteen-year-old starting one of these drugs is not making a six-month decision. Obesity in adolescence is a chronic condition and the drugs work while taken. So the relevant exposure is years, possibly decades, and there is essentially no randomized evidence at that horizon in this age group. Not bad evidence. Absent evidence.
The safety signal people ask about
Suicidal ideation and behaviors were not significantly different between drug and placebo in this pool. That is genuinely reassuring and it is bounded by the same half-year median, in trials that were not designed or powered to detect it. The adult version of the question has its own longer story.
Gastrointestinal effects did rise, which the authors flag as warranting attention in long-term management. In a population that has to keep eating enough to finish growing, that is a more consequential side effect than it is in adults.
Why this page exists on a site that cannot sell to them
Nothing on this roster ships to anybody under eighteen, and no seller here offers a pediatric service. This page is not a buying guide.
It is here because the adolescent evidence base is the clearest available example of something this market does constantly: one strong trial becomes the whole story, and the thinness behind it stops being visible. Grown-up buyers meet the same pattern in much weaker form all the time. Parents being asked to decide deserve the full picture, and nobody in this market is in the business of providing it.