Association of maternal characteristics and behaviours with 4-year-old children's dietary patterns.

This study examined the association of family and maternal characteristics with preschool children's dietary patterns. Trained interviewers evaluated subsample 3422 mothers and children enrolled in the population-based birth cohort Generation XXI (Porto, Portugal, 2005-2006). Maternal characteristics and behaviours (exercise, smoking habits, diet and child-feeding practices) and family characteristics were evaluated. Maternal diet was classified by a dietary score, and children's dietary patterns were identified by latent class analysis. Odds ratios (OR) and confidence intervals (95% CI) were estimated by multinomial regression models. The analysis was based on a framework with four conceptual levels: maternal socio-economic position (SEP) at 12 years, maternal socio-economic and demographic characteristics at child's delivery, family characteristics and maternal behaviours at child's 4 years. Three dietary patterns were identified in children: high in energy-dense foods (EDF); low in foods typically consumed at main meals and intermediate in snacks (Snacking); higher in healthy foods; and lower in unhealthy ones (Healthier, reference). Lower maternal SEP had an overall effect on children's diet (low vs. high SEP; EDF, OR = 1.76, 95% CI: 1.42-2.18; Snacking, OR = 1.73, 95% CI: 1.27-2.35), while maternal education was directly associated with it (≤9 vs. >12 schooling years, EDF, OR = 2.19, 95% CI: 1.70-2.81; Snacking, OR = 2.22, 95% CI: 1.82-3.55). Children whose mothers had worse dietary score were significantly more likely to follow unhealthier patterns (first vs. fourth quartile; EDF, OR = 9.94, 95% CI: 7.35-13.44, P-trend < 0.001; Snacking, OR = 4.21, 95% CI: 2.94-6.05, P-trend < 0.001). Maternal diet was the key factor associated with children's diet, above and beyond socio-economic and demographic characteristics, accounting for one-third of the determination coefficient of the fully adjusted model. At preschool age, interventions should give a particular focus on maternal diet and low SEP groups. [ABSTRACT FROM AUTHOR]