SAN ANTONIO -- Infants whose families received ongoing text message-based health behavior counseling in addition to clinic-based counseling showed healthier weight trajectories than those whose families only received clinic-based counseling, according to a pragmatic, randomized trial.
Babies assigned to the digital intervention, which was started soon after birth, had a lower average weight-for-length trajectory at 24 months than those whose families received the clinic-only intervention, with a reduction of 0.33 kg/m (95% CI -0.57 to -0.09), reported Eliana Perrin, MD, MPH, of Johns Hopkins University School of Medicine and School of Nursing in Baltimore.
Adding the digital intervention reduced incidence of obesity at 24 months by a relative 44% compared with clinic-only care (7.4% vs 12.7%; adjusted RR 0.56, 95% CI 0.36-0.88), with a number needed to treat of 19.
Importantly, however, while the intervention lowered weight trajectories at the upper end of weight-related z-scores, it did not lower already healthy ones, Perrin said in presenting findings of the Greenlight Plus trial at the ObesityWeek annual meeting. The findings were published simultaneously in .
"Early intervention strategies can prevent obesity with long-term implications for cardiometabolic health and health equity," Perrin concluded.
Mona Sharifi, MD, MPH, of Yale School of Medicine in New Haven, Connecticut, who was not involved in the trial, called the results "incredibly exciting" given how few preventive interventions have been found effective for children at that age, and because this one in particular has good scalability.
Perrin acknowledged the many prior failures in pediatric obesity prevention. Her group's prior had tried to prevent obesity with a health literacy-informed pediatric primary care-based intervention that clinicians provided at well-child visits. Although that early-life intervention reduced children's BMI z-score through 18 months, it was not sustained at 24 months. The researchers attributed that fall-through to not starting the intervention early enough and the fact that well-child visits drop off in children's second year.
Grace Murray, MS, RDN, LDN, a registered dietitian at Indiana University's Riley Children's Hospital in Indianapolis, expressed surprise that the effect size in Greenlight Plus was as high as it was. She said she sees opportunities for expansion with the program and praised the use of text messages as a supplement to the original intervention.
"People are difficult to get ahold of in person, so texting helps us get information out to people in a way that's familiar and comfortable to them," Murray told 51˶.
Since mobile phones are ubiquitous, digital intervention may also help reduce health disparities, Perrin noted.
Greenlight Plus included 900 English- or Spanish-speaking families at 17 pediatric primary care clinics from six U.S. medical centers from October 2019 through January 2022. Children were enrolled in their first 21 days of life if they were born at 34 weeks' gestation or later, had a birthweight greater than the third percentile of the World Health Organization's growth curves, and had no conditions affecting growth. Follow-up continued through January 2024.
One group of 451 families was randomized to receive clinic-based health behavior counseling from pediatric providers at the clinic during well-child visits, including eight booklets to promote healthy behaviors and tangible tools, such as sippy cups and snack containers. The second group of 449 families received the same in-clinic intervention as well as individually tailored, responsive text messages to continue supporting health behavior goals along with use of a web-based dashboard.
The booklets for both groups and the text messages were available in English or Spanish and designed with health literacy in mind. They covered breastfeeding, satiety cues, active time, sleep time, screen time, and healthy eating. The text messages used behavior change techniques, such as goal-setting, self-monitoring, and tailored feedback to support developmentally appropriate health behaviors for healthy growth. Parents self-rated their progress toward goals five times in every 2-week cycle and received "immediate adaptive feedback, tips, and encouragement based on goal progress," Perrin said.
The web-based dashboard included each cycle's goal -- such as not giving a baby any juice or other sugary drinks -- and the ability to track the child's weight, length, and percentile with an explanatory graph. It also provided tips for the goal and an explanation of its importance with illustrations and health-literacy informed text.
Among the 86.3% of children with primary outcome data at 24 months, 15.9% were non-Hispanic Black, 45% were Hispanic, 20.6% were non-Hispanic white, and 18.3% identified as other or multiple races and ethnicities. About a third of parents in both groups chose Spanish as their preferred language, and two thirds of both groups had public insurance. Rates of food insecurity were similar in the clinic-only (16.5%) and clinic-digital (15%) groups, and limited health literacy was an obstacle for more than half of the participants in both groups (57.5% and 57.3%, respectively).
The outcomes data was adjusted for the child's birthweight and biological sex as well as the parents' race, ethnicity, health literacy, preferred language, education level, household income, and household food insecurity.
Among secondary outcomes, adjusted mean difference of weight-for-length z-score (-0.19, 95% CI -0.37 to -0.02) and BMI z-score (-0.19, 95% CI -0.36 to -0.01) were also lower at 24 months in the clinic plus digital intervention group.
"The digital intervention may have also reduced health disparities as it worked especially well among populations at highest risk," Perrin said. The researchers identified trends toward greater effect among Hispanic and particularly non-Hispanic Black and non-Hispanic multi-racial participants. Families with food insecurity who received the digital intervention also had a significantly lower weight-for-length trajectory than food-insecure families in the clinic-only group at 24 months (P=0.02).
Sharifi praised the trial's examination of outcome differences between groups based on preferred language, health literacy, and food insecurity, "because a lot of studies don't even take the care to look at issues related to health equity," she told 51˶. That the program was designed with health literacy as a central focus in populations at higher risk for obesity and achieved a stronger effect in those very populations was "amazing and really fitting," she said.
Disclosures
The research was funded by the Patient-Centered Outcomes Research Institute, and study data collection and management were supported by the National Institutes of Health.
Perrin had no disclosures; one coauthor was a paid advisor to the AI-driven platform Medeloop.ai. Murray and Sharifi had no disclosures.
Primary Source
JAMA
Heerman WJ, et al "A digital health behavior intervention to prevent childhood obesity: The Greenlight Plus randomized clinical trial" JAMA 2024; DOI: 10.1001/jama.2024.22362.