Abstract
BACKGROUND: While the household context is important for lifestyle behavior interventions, few studies have examined parent-child associations for diet and physical activity (PA) changes over time in a rapidly urbanizing country. We aimed to investigate changes in diet, screen time, and PA behaviors over time in children and their parents living in the same household, and examine the parent-child association for these behaviors. METHODS: We studied dietary, screen time, and PA behaviors in 5,201 parent-child pairs (children aged 7-17y) using longitudinal data from the China Health and Nutrition Survey (1991, 1993, 1997, 2000, 2004, 2006, and 2009). We collected three-day 24-h recall diet data to generate percentages of energy from animal-source foods, away-from-home eating, and snacking from 1991-2009, which are known urbanization-related behaviors. We used a seven-day PA recall to collect screen time (hours/week) and leisure-time sports participation (yes/no) since 2004. We examined the changes in children's and parents' behaviors over time using random-effects negative binomial regression for diet and screen time, and random-effects logistic regression for leisure-time sports. We then regressed each of the behaviors of offspring on each of their parents' same behaviors to examine the parent-child association, using the same set of models. RESULTS: We observed increases in energy from animal-source foods, eating away-from-home, and snacking, as well as screen time and leisure-time sports in parents and children over time, with different rates of change between children and their parents for some behaviors. We found positive parent-child associations for diet, screen time, and PA. When parental intakes increased by 10 % energy from each dietary behavior, children's increase in intakes ranged from 0.44 to 1.59 % total energy for animal-source foods, 0.17 % to 0.45 % for away-from-home eating, and 2.13 % to 7.21 % for snacking. Children were also more likely to participate in leisure-time sports if their parents participated in leisure-time sports. CONCLUSION: Our findings support household-based health behavior interventions targeting both children and their parents. However, generation-specific intervention strategies may be needed for children and adults, especially for dietary behaviors, which changed differentially in children versus parents in this rapidly modernizing population.