16.8% of children on GLP-1 drugs developed a nutritional deficiency within a year

Of 2,031 children aged 10 to 17 who started a GLP-1 prescription between 2017 and 2022, 16.8% had a nutritional deficiency diagnosed within the following year. Vitamin D accounted for most of it, at 12.4%. The cohort came from a national insurance claims database covering more than 100 million patients, and every child in it had no deficiency diagnosis on record before treatment began.

The analysis was led from Ann & Robert H. Lurie Children’s Hospital of Chicago and appeared in the journal Childhood Obesity in September 2026. Justin Ryder, PhD, the hospital’s vice chair of research, is the name attached to the clinical message: nutritional support has to be part of care once a GLP-1 drug is started. Ryder is quoted in Pharmaceutical Journal saying that, as pediatric use of these drugs becomes more widespread, clinicians need to understand the risks during periods of rapid growth, and the study is built to put a first number on one of them. The numbers behind that message are narrower than a quick read of the percentage suggests.

Diagnosis codes behind the 16.8%

The figure is a share of diagnoses, not of laboratory results. According to the release summary of the Lurie work, “nearly 16.8%” of the young patients carried a diagnosed nutritional deficiency within one year of beginning treatment. A Pharmaceutical Journal report on the paper adds the earlier checkpoint: 10.2% of the children had a deficiency diagnosis within 180 days, so more than a third of the one-year total had already appeared by the six-month mark.

The group was mostly female (61.5%), with a mean age of 15. About two-thirds (67.2%) had type 2 diabetes and 62.6% had an obesity diagnosis, per the paper’s cohort breakdown as reported by Drug Topics. Those overlapping diagnoses matter, because children with obesity or diabetes can carry low vitamin D before any prescription is written. The cohort excluded anyone with a prior deficiency diagnosis, but an undiagnosed low level would not show up in claims.

Liraglutide, dulaglutide and semaglutide

The prescriptions skewed toward the older drugs. Liraglutide, sold as Victoza for diabetes and Saxenda for weight, made up 78.6% of the prescriptions, dulaglutide (Trulicity) 10.4% and semaglutide (Ozempic and Wegovy) 9.1%. That mix reflects the 2017-2022 window, and it means the 16.8% describes mostly liraglutide patients rather than the weekly semaglutide injections that dominate adult prescribing today.

Beyond vitamin D, the individual counts were small. Drug Topics lists nutritional anemia at 1.55%, iron-deficiency anemia at 1.44% and muscle loss at 0.8%. Vitamin D therefore carried the result: a 12.4% rate that, in the authors’ framing, raises a bone-health concern in an age group that is still building skeletal mass, which is why Ryder singled out calcium and iron alongside it.

Counseling that arrived late

The second finding is about follow-up. Only 5% of the children received nutritional counseling within 30 days of starting treatment, and fewer than 25% had it within six months, according to the Lurie Children’s release. Drug Topics gives the 180-day share as 23.3% and the average wait for a first nutrition visit as 149 days after the first dose.

One result runs against intuition. Children who did get early counseling had a higher deficiency rate, 23.2% against 14.8%. The authors cautioned that this most likely reflects referrals made after a deficiency had already been found, not counseling causing it. Ryder’s own line, quoted in coverage of the paper, is that “nutrients such as vitamin D, iron and calcium are of particular concern during adolescence, when deficiencies may have lasting implications for skeletal health and overall development.”

Claims codes and the missing comparison

The design has limits the authors themselves state. A Contemporary Pediatrics summary lists them: the data rest on diagnosis codes rather than laboratory confirmation, baseline diet and supplement use could not be accounted for, and the effect of the drug cannot be separated from that of the underlying obesity or diabetes. No group of similar children who were not given GLP-1 drugs is reported, so the study cannot say how many of the 16.8% would have been diagnosed anyway.

It may also undercount. NutritionInsight reports that the authors consider claims data likely to miss subclinical cases, since a child is only counted if a clinician recorded the deficiency. The two biases pull in opposite directions, and the authors stop short of weighing one against the other.

What the cohort does establish is the size of the follow-up gap. Of 2,031 adolescents handed a GLP-1 prescription between 2017 and 2022, only 5% were counseled on nutrition inside the first month, and the 16.8% diagnosis rate was measured against that near-empty backdrop.

The database ends in 2022, so children started on newer products since then are not represented.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview