Scientists find genetic link between BMI and sleep duration in children

May 11, 2021

Childhood weight issues and sleep are likely connected. (Unsplash/Neonbrand)

For years, observational studies of children have documented an association between weight and sleep habits. New research has begun to clarify this association, showing for the first time that children genetically predisposed to have higher body mass indices also had shorter average sleep times.

In a paper published April 26 in Sleep Medicine, a group of epidemiologists and psychologists examined whether sleep duration and obesity, which are both heritable traits, share common genetic pathways, known as genetic etiology. It is the first study to establish a relationship between genetic obesity risk and objective sleep duration in children. 

Victoria Garfield, the study's first author and a research fellow at University College London, told The Academic Times that she wanted to understand why BMI and sleep duration are so frequently linked in epidemiological studies. In previous research, she found no evidence of a causal relationship between these traits and little evidence of shared genetic underpinnings in adult data. Thus, she wanted to go beyond classical observational associations in a pediatric data set to understand whether there may be such a relationship in children, even if one was not observed in later life.

"We hypothesized that essentially, it could be that genetic variants that predispose to a higher BMI may also predispose to children experiencing shorter sleep. And that's what we found," Garfield said. "We found that higher BMI genetic risk scores were associated with shorter total sleep time." Genetic risk scores estimate the cumulative contribution of genetic factors to a specific trait of interest in a person, highlighting potentially shared genetic sources of the two traits, according to the paper. 

Few prior studies have examined the relationship between a genetic risk score for obesity and self-reported sleep duration in adults, and none of them have found any causal associations. However, this does not rule out potential underlying genetic influences linking BMI and sleep duration in children, the authors said, perhaps via previously understudied biological pathways.

"For example, evidence suggests that short sleep can lead to alterations in the 'hunger hormones,' leptin (which promotes satiety) and ghrelin, and that these changes lie on the causal pathway between sleep duration and obesity," the authors reported in the paper.

The researchers analyzed data from the Trondheim Early Secure Study, an ongoing longitudinal childhood development study in Norway. A total of 452 6-year-old Norwegian children comprised the sample. Garfield and her colleagues analyzed data that included the children's BMI, genotypes and objectively measured total sleep times.

Body height and weight of the children were measured using digital scales, and their DNA was extracted and stored from a saliva sample. The children wore accelerometers on their hips for seven consecutive days to measure their sleeping time. Parents and guardians also completed a questionnaire about their child's health and activity during the study period. 

The genetic risk of obesity for each child was inferred using a 32-single nucleotide polymorphism, or SNP, weighted genetic risk score of the child's BMI. An SNP is a common type of genetic variation involving changes to one nucleotide in a person's DNA.

"The analysis was quite straightforward," said Garfield. "Because what we were doing was quite novel, we didn't do anything too complex and we decided to run some classical correlations and linear regressions to see whether there was any association simply between the BMI genetic risk score in children and objectively measured sleep duration from waist accelerometry at age 6." But while the authors found a strong association using linear regression analyses, which attempt to assess the relationship between variables using linear equations, they did not find the same strong association when calculating the Pearson's correlation, which measures the strength of a linear association between two variables.

"We found that higher genetic risk of obesity was associated with 5 minutes less sleep in objectively measured [total sleep time] independently of age, sex and BMI, in this sample of 6-year-old children," the authors said in the study. "These results suggest that shared genetic [etiology] may exist between BMI and total sleep time in early life." 

The results also highlight one potential genetic pathway that may explain this association. As the researchers stated in the paper, the 32-SNP BMI genetic risk score that they produced included an intron — a nucleotide sequence within a gene that does not code for any proteins and is removed by RNA splicing — in two genes associated with appetite and obesity. These genes are both highly expressed in the hypothalamus, one of the key brain regions involved in sleep.

"This is only one example of a relevant pathway, and future research could perform downstream analysis to explore whether any robustly associated BMI genetic variants are on the causal pathway for sleep duration," the authors said. 

They also acknowledged the limitations of their sample, as the children were all of European descent, 84.5% of them were within the normal weight range for their age and more than 95% of them were white. The findings need to be replicated using larger, more diverse samples of children, potentially involving different age groups and other sleep phenotypes.

"What we're trying to get people to understand is that these two phenotypes [BMI and sleep duration] are really complex, but we have found something interesting. We found that if you have a higher BMI, it could be that you also have problems with shorter sleep and that this could be due to genetic predisposition to higher BMI," Garfield said.

The study, "Shared genetic architecture underlying sleep and weight in children," published April 26 in Sleep Medicine, was authored by Victoria Garfield and Clare H. Llewellyn, University College London; and Lars Wichstrøm and Silje Steinsbekk, Norwegian University of Science and Technology.

We use cookies to improve your experience on our site and to show you relevant advertising.