Editor's note: This study was updated on Nov 1 with comments from Christopher Stark, MD.
Two new large observational studies of children have found a link between antibiotic use in the first 2 years of life and risk of obesity, with evidence suggesting the association strengthens with increased exposure to antibiotics. The two studies, however, differed significantly in the observed association.
In a study published yesterday in BMJ Gut, retrospective analysis of more than 333,000 children in the US Military Health System (MHS) database found that antibiotics consumed in the first 2 years of life were associated with a 26% heightened risk of obesity by the age of 3 and that exposure to multiple antibiotics increased the risk. The researchers also found that antacids were associated with risk of obesity, although to a lesser extent.
A study today in Pediatrics using data on more than 362,000 children from a nationwide clinical database, meanwhile, found that antibiotic use before 24 months of age was associated with 5% higher odds of overweight or obesity at 5 years of age, with a more modest dose response. The authors of the study concluded that the clinical significance of the observed weight gain in children who received antibiotics was likely negligible.
"We found these associations, but they're small, and they're likely small enough that it shouldn’t really come into the decisions that doctors and parents are making with regards to antibiotics," lead author Jason Block, MD, an associate professor of population medicine at the Harvard Pilgrim Health Care Institute and Harvard Medical School, told CIDRAP News.
The studies add to the growing body of research into connections between antibiotic use in infancy and obesity risk, which is based on the idea that antibiotics alter the gut microbiota in children in ways that could possibly lead to weight gain. But as with these two studies, previous studies have produced mixed results; most have found a positive association, but with notable differences in the strength of the observed effect.
Early antibiotic exposure and weight outcome
In the Pediatrics study, the researchers used electronic health record data on children from 35 institutions participating in the National Patient-Centered Clinical Research Network. Several of the previous studies have relied on data from single institutions, and Block and his colleagues felt a large, diverse cohort, including children with and without complex chronic conditions, would provide a better assessment of the effect of early life antibiotic use on children's weight gain. The data covered the years 2009 to 2016.
"Because of its size, we were able to look at some of the more precise questions that other studies couldn't look at as much," Block said. That included looking at the effect of specific classes of antibiotics, dose response, and how chronic conditions might affect the relationship.
"That's what this adds; the ability to explore this question across a large population and to examine it in several different ways," he added.
The primary outcome of the study was association between any antibiotic use before 24 months and body mass index (BMI) z score—a measure of relative weight adjusted for child age and sex—at 60 months, with secondary assessments of antibiotic spectrum and age-period exposures.
Of the 362,550 children eligible for the study, more than half (58%) received at least one antibiotic before 24 months, 16% received more than four antibiotics, and 35% received at least one broad-spectrum antibiotic. Overall, 28% of the children were overweight or obese at 5 years of age.
In adjusted models for children without a complex chronic condition at 5 years, Block and his colleagues estimated a higher mean BMI z score by 0.04 (95% confidence interval [CI], 0.03 to 0.05) and higher odds of overweight or obesity (odds ratio [OR], 1.05; 95% CI, 1.03 to 1.07) associated with receiving any antibiotics (compared with no antibiotics) before 24 months.
They also found that the OR for overweight and obesity were higher for broad-spectrum antibiotics than for narrow-spectrum antibiotics (1.07 vs. 1.02), and that the OR increased with the number of antibiotic exposures (1.01, 1.06, 1.07, and 1.10 for 1, 2, 3, and more than 4 antibiotics, respectively, compared with no antibiotics). "The more antibiotics you used in the first 2 years of life, the higher the odds ratio was," Block said.
Although there is a theory that exposure to antibiotics earlier in infancy—especially in the first 6 months of life when the gut microbiome is still developing—may have a bigger impact on the composition of the microbiome and a resulting effect on weight gain, the results showed that the timing of antibiotics did not significantly affect the association.
On the basis of the results, Block and his colleagues calculated that among 5-year-olds of average height, exposure to more than four courses of antibiotics (compared to none) resulted in a gain of less than a quarter of a pound. While this may add to the overall concerns about antibiotic overuse, he said, it's not enough to affect the decision to prescribe antibiotics.
"For each individual decision that doctors and parents are making, it probably shouldn't come into the calculus, because the effect is just not enough," he said.
Antibiotic, antacid exposure
In the other study, a team of pediatricians from military medical centers in Texas and Maryland looked at data on all infants born into the MHS database from October 2006 to September 2013, focusing on exposure not only to antibiotics but also to histamine 2-receptor antagonists (H2RAs) and proton pump inhibitors (PPIs)—medications that are commonly prescribed in infancy and are also known to have microbiota-altering effects. They assessed the association between exposure to these drugs during the first 2 years of life and obesity starting at age 3.
Overall, 333,353 children met the inclusion criteria for the study; 72.4% were prescribed antibiotics, while 11.8% were prescribed H2RAs and 3.3% received PPIs. Fourteen percent developed obesity.
After adjusting for confounding factors, a prescription for antibiotics was associated with an increased risk of obesity (hazard ratio [HR], 1.26; 95% CI, 1.23 to 1.28), irrespective of the type of antibiotic. As in the Pediatrics study, the risk increased with the number of antibiotics received, but more dramatically, rising from an HR of 1.12 for one antibiotic prescription to 1.42 for more than four. H2RAs and PPIs were also associated with an increased risk of obesity, but the effect was significantly smaller (adjusted HR 1.01 and 1.02 for H2RAs and PPIs, respectively).
The authors of the study say that their findings are noteworthy because of the large size of the cohort and because antibiotic overprescribing is a known problem. National estimates show that at least a third of outpatient antibiotic prescriptions are unnecessary. They suggest the findings warrant further research into how antibiotics affect the gut microbiome, and that evidence of specific effects on gut bacteria could provide clinicians with additional motivation to reduce unnecessary prescribing.
"It would be productive if this study helps shift the mindset of healthcare providers to further appreciate that medication exposures are a provider-driven modifiable risk factor," study co-author Christopher Stark, MD, a pediatrician at William Beaumont Army Medical Center, said in an email. "The increased risk for obesity may ultimately be small, but these exposures should be judiciously limited to specific clinical scenarios."
But Stark and his co-authors also note that the study did not include information on maternal weight, smoking status, and other comorbid health problems—information that could account for some of the differences that were seen in the exposed populations. "The lack of data on maternal weight and other comorbid conditions is a significant limitation," Stark said. "We will rely on future research studies to fill these gaps and assist with determining how these factors affect long-term growth in pediatric patients."
Correlation or causality
Ultimately, though they differ in the strength of the observed effect, both studies indicate that antibiotic use in early life is positively associated with increased weight gain. On that, the findings are in line with previous studies and meta-analyses. "There seems to be some consistency that, if there is a relationship, it's a small one," Block said.
The question then, Finnish researchers Antti Saari, MD, PhD, and Ulla Sankilampi, MD, PhD, write in a commentary that accompanies the study by Block and colleagues, is whether there is a causal relationship between early life antibiotic exposure and future weight gain.
"Because obesity is a multifactorial condition wherein both genetic and multiple environmental factors play roles in its pathogenesis, a relatively weak association does not necessarily exclude causality," they write.
To date, there has been some research conducted in mice showing how antibiotic use changes microbial diversity in the gut and affects metabolism and weight gain. In addition, antibiotics are known to promote growth and weight gain in food-producing animals. But determining whether antibiotics cause weight gain in humans is difficult.
"It's really hard to get at causality explicitly," said Block. "The truth is, when you find these small relationships, it make you wonder a lot about the causal issues, because you're much more likely to have other unmeasured confounders that could explain small relationships than you would with big relationships," he said.
While Saari and Sankilampi suggest a well-designed prospective study could help answer the question, Block says there are some sophisticated statistical techniques that would enable researchers to look at the association between antibiotics and weight gain more closely and "ratchet up" the argument that the relationship is causal. "That's an important next step that should be taken," he said.
Oct 30 BMJ Gut study
Oct 31 Pediatrics study
Oct 31 Pediatrics commentary