Shari L. Barkin
An intensive behavioral intervention for children with overweight and their parents that included skills-building sessions and phone calls did not lead to changes in BMI trajectory after 3 years, whereas a smaller parenting education intervention for parents of young infants was associated with a modest reduction in BMI z score during the same period, according to findings from two studies published in JAMA.
“Given the challenges associated with effective obesity treatment, recent focus has been on childhood obesity prevention,” Shari L. Barkin, MD, MSHS, the William K. Warren Foundation endowed chair and professor in theVanderbilt University School of Medicine, Nashville, Tennessee, and colleagues wrote in the study background. “The developmental origins of disease hypothesis suggest that early life influences can alter a person’s lifelong health trajectory, linking early obesogenic exposures and rapid weight trajectory to common chronic adult conditions, including coronary artery disease and type 2 diabetes. These problems are especially salient for families from traditionally underserved minorities, who are confronted with significant barriers, such as poverty.”
In the first study, Barkin and colleagues analyzed data from 610 parent-child pairs recruited from 54 physician offices in Nashville, Tennessee between August 2012 and May 2014 (91.4% Latino) as part of the Growing Right Into Wellness (GROW) study. Children were aged 3 to 5 years at baseline with weight ranging from high-normal to overweight, but not obese, with a parent who qualified for at least one service for underserved populations, such as Medicaid (mean child age at baseline, 4 years; 51.9% girls; 56.7% with annual household income below $25,000). Researchers randomly assigned 304 families to a 36-month multicomponent behavioral intervention, including 12 weekly skills-building sessions followed by monthly coaching telephone calls for 9 months and a 24-month sustainability phase providing clues to action, and 306 families were assigned to attend six school-readiness sessions conducted by the local library, delivered over 36 months (controls).
Primary outcome was child BMI trajectory over 36 months. Prespecified secondary outcomes included parent-reported child dietary intake and community center use.
At 36 months, mean child BMI was 17.8 kg/m² in the intervention group and 17.8 kg/m² in the control group. Researchers observed no differences in BMI trajectory over 36 months for either group (P = .39). The intervention group had a lower mean caloric intake vs. controls (1,227 kcal per day vs. 1,323 kcal per day, respectively), and children in the intervention group used the community center more than control children (56.8% vs. 44.4%, respectively).
“The prevalence of obesity observed in both the intervention and control groups was similar to the regional prevalence of obesity for Latino children (37.7%), indicating that the behavioral intervention did not alter the usual pattern of obesity in this low-income minority population,” the researchers wrote.
“A family-based, community-centered behavioral obesity prevention trial for underserved preschoolers can lead to some sustained healthy behavior change, but that was not sufficient to result in differences in early BMI growth trajectories,” Barkin told Endocrine Today. “What this means is that one size doesn’t fit all. Childhood obesity prevention in low-income, underserved populations could require a higher amount of sustained behavior change to affect BMI growth in early childhood.”
Barkin said obesity prevention must begin in early life.
“Obesity is not the result of one thing at a time,” Barkin said. “It is a complex problem affected by a dynamic interaction between biology and behavior over multiple developmental periods. When we try to distill how we prevent something by doing only one thing, we miss the mark.”
In the second intervention, Ian M. Paul, MD, MSc, professor in the department of pediatrics at Penn State College of Medicine in Hershey, Pennsylvania, and colleagues analyzed data from 279 mother-newborn pairs recruited shortly after delivery between January 2012 and March 2014 (mean age of mothers, 29 years; 86% white; 86% privately insured). Researchers randomly assigned the mothers to either a responsive parenting intervention, advising parents on how to respond to their child’s needs across four behavior states (drowsy, fussy, sleeping and alert) or a home safety intervention (controls). Research nurses conducted four home visits during infancy and annual research center visits. The responsive parenting curriculum focused on feeding, sleep, interactive play and emotion regulation. Primary outcome was BMI z score at 3 years. Secondary outcomes included prevalence of overweight and obesity at 3 years.
At age 3 years, children in the intervention group had a lower mean BMI z score vs. controls (–0.13 vs. –0.15; absolute mean difference, –0.28; P = .04). However, mean BMI percentiles between intervention children vs. control children did not differ significantly at 3 years.
“Differences between study groups were modest, not all study outcomes achieved statistical significance, and the pattern of significant results from age 2 to 3 years varied,” the researchers wrote. “However, the direction of the differences consistently favored the responsive parenting intervention group.”
Impact of age, study design
In a related editorial, Jody W. Zylke, MD, deputy editor of JAMA, and Howard Bauchner, MD, editor in chief of JAMA, noted that both interventions had good retention and adherence; however, the different findings may be the result of differing study designs.
“Single-center studies are known to overestimate intervention effects, such that the responsive parenting intervention trial may have produced a positive result that will not be replicated in other sites,” Zylke and Bauchner wrote. “Alternatively, the contrasting outcomes may be the result of differences in the content or delivery of the interventions. The skills-delivery component of the multicomponent behavioral intervention was delivered in group settings at community centers or by telephone; the responsive parenting intervention was delivered at home and research center visits. However, a likely explanation relates to the populations of children and families enrolled.”
Zylke and Bauchner added that the age of the children in both studies — preschoolers vs. young infants — was also a likely factor, noting that interventions to prevent obesity should begin as early as possible,
“By preschool age, habits of the child or family may be too ingrained to alter,” they wrote. – by Regina Schaffer
Disclosures: Paul reports he has served on advisory boards for Boehringer Ingelheim, Johnson & Johnson and Pfizer. Zylke and Bauchner report no relevant financial disclosures.