(P17-008-23) Ultra-Processed Food Intake Among Adolescents Diverges From Maternal Ultra-Processed Food Intake As Offspring Enter Adulthood in Cebu, Philippines

Objectives: The objectives of this study were to examine the extent to which ultra-processed food (UPF) intake in adolescence is predicted by maternal UPF intake and whether offspring UPF intake diverges from maternal UPF intake as offspring enter adulthood and establish diet patterns.

Methods: Our sample included offspring-mother pairs enrolled in the Cebu Longitudinal Health and Nutrition Survey birth cohort between 1994 (n=2,028) and 2005 (n=1,671). Foods and beverages from 24-hour recalls were classified via the NOVA system. The outcome and exposure were offspring and maternal UPF intake (% kcals/day), respectively. We used a two-part model combining multilevel logistic regression predicting any UPF intake and multilevel gamma regression predicting amount of UPF intake among UPF-consuming offspring, adjusted for offspring sex, maternal education, maternal household income, and offspring household urbanicity. We assessed modification by offspring age and sex. Main effect and interaction terms were considered significant at p< 0.05 and 0.10, respectively. In a sensitivity analysis, we excluded offspring-mother pairs living apart.

Results: From 1994 to 2005, the proportion consuming UPFs increased from 74% to 90% among offspring and from 72% to 86% among mothers and median UPF intake rose from 7.3% to 10.6% of daily kcals among offspring and from 2.7% to 6.5% among mothers. Adjusted for confounders, a 1-tertile increase in maternal UPF intake was associated with 30% higher odds of offspring UPF intake at age 11 (OR=exp(0.265)=1.30; 95% CI: 1.13, 1.50); relative to age 11, this association was similar at ages 15 and 18 but significantly weaker at age 21 (OR=exp(0.265-0.242)=1.02, p<sub>interaction=0.06). Among UPF-consuming offspring, a 1-tertile increase in maternal UPF intake was associated with a 7% increase in UPF intake across all ages (rate of increase=exp(0.065)=1.07; 95% CI: 1.04, 1.10). Excluding offspring-mother pairs living apart yielded similar results.

Conclusions: UPF intake in adolescence was positively associated with maternal UPF intake, and this association waned as offspring entered adulthood. Efforts to curb UPF intake among Filipino adolescents to prevent cardiometabolic disease should include multi-generation interventions.

Funding Sources: NHLBI predoctoral traineeship