Infant feeding and the development of obesity: what does the science tell us?

Apr 11 2008

On Wednesday, April 9, 2008, a symposium at the American Society for Nutrition's annual meeting at Experimental Biology was held in which noted scientists discussed new infant feeding studies that used methodology such as randomized clinical trials (involving breastfeeding promotion) as well as sibling pairs analysis.

These studies may offer new insights into possible associations between infant feeding and health outcomes such as obesity.

The symposium, “Infant Feeding and the Development of Obesity: What Does the Science Tell Us",” sponsored by the International Formula Council (IFC)* and chaired by Linda Adair, Ph.D., professor of nutrition at the University of North Carolina at Chapel Hill, brought together international experts in the field of infant nutrition to present their recent findings.

Featured researchers included David Barker, M.D., Ph.D., professor of clinical epidemiology at the University of Southampton, UK and professor of Cardiovascular in the Department of Medicine at the Oregon Health and Science University, whose soon-to-be published study examines breastfeeding in a large group of sibling pairs that were followed into their late 60s.

“This type of study design controls for maternal factors,” according to Dr. Barker. “Differences in the long-term effects of breast and bottle feeding may reflect differences in the mothers rather than the effects of feeding itself.” Maternal factors include maternal health status, maternal care-giving, mother–child interactions or other health-related behaviors of the mother that may interfere with determining the association of infant feeding and health outcomes and the strength of any possible associations.

Other study designs such as the randomized clinical trial on breastfeeding and health outcomes in infants in Belarus recently conducted by Michael Kramer, M.D., a pediatrician and perinatal epidemiologist at McGill University in Montreal, provide evidence that research design can have a significant impact on infant feeding study results. In a randomized infant feeding clinical trial, known as the gold standard in research, infants would be randomly assigned to be breastfed or formula-fed; however, such trials are generally not feasible in infant feeding research, since most mothers determine their infant's feeding method. Dr. Kramer randomly assigned hospitals to implement breastfeeding promotion practices and standard care. He found that despite the substantial increase in prolonged and exclusive breastfeeding among mothers receiving the intervention, there were no differences between their children (n= 7,108 subjects) and the children of mothers from the control hospitals (n= 6,781 subjects), that did not implement breastfeeding promotion practices, on several measures of adiposity at 6.5 years of age.

These findings challenge the concept that breastfeeding reduces the risk of obesity in childhood, as some other studies have found. Speaking of the strength of his study design and the fact that other studies due to their design may have been prone to inherent bias, Dr. Kramer noted, “Previously reported beneficial effects on these outcomes (measurements of obesity) may be the result of uncontrolled confounding and selection bias.” The study performed by Dr. Kramer represents the largest randomized trial done to date in the area of human lactation.

Other researchers at the symposium presented the outcomes of more traditional epidemiologic studies that were based on observational designs and thus had limitations such as not being able to control subjects' behavior. Beth Mayer-Davis, Ph.D., R.D., professor of nutrition and diabetes researcher at University of North Carolina at Chapel Hill, discussed findings from her research on infant feeding and diabetes in ethnic groups in a United States population. In her observational study of less than 300 subjects, mothers of children with diabetes were asked to recall if and for how long they breastfed their infant. Dr. Mayer-Davis reported that breastfeeding appeared to reduce the risk for development of type 2 diabetes in youth, possibly mediated in part by weight status in childhood.

Nancy Butte, Ph.D., professor of pediatrics at the Children's Nutrition Research Center at Baylor College of Medicine, discussed early infant feeding practices and effects on obesity. Dr. Butte presented her work from the VIVA LA FAMILIA study, an observational study with a cohort of 1,030 Hispanic children. Dr. Butte noted that although breastfeeding appeared to have a small protective effect against childhood obesity, other genetic and environmental factors “far supersede infant feeding practices as risk factors for childhood obesity.”