Approaches to restrictive feeding: Associations with child weight and eating behavior.

Background: Identifying differences in how mothers communicate restriction of their children's eating may be important to understanding the effects of restriction on children's intake and weight status.

Objectives: To characterize mothers' restrictive statements by affect and directness, and examine cross-sectional associations between restrictive statement types and children's body mass index and eating behaviors.

Methods: Mother-child dyads (N = 223, mean child age 5.9 years) participated in a structured eating task. A coding scheme reliably characterized mothers' restrictive statements. Mothers completed measures of child enjoyment of food, food responsiveness, and satiety responsiveness, and child anthropometrics were measured. Poisson regression was used to test associations between type of restrictive statements and child BMI z-score (BMIz) and eating behaviors, adjusting for covariates.

Results: Higher child BMIz was associated with mothers' more frequent use of negative direct restrictive statements, but not other types of statements. This association was stronger among girls (RR (95% CI) = 2.28 (1.45-3.59)) than boys (RR (95% CI) = 1.49 (1.05-2.10)). Among girls, but not boys, higher enjoyment of food and lower satiety responsiveness were associated with more frequent positive direct restrictive statements (RR (95% CI) = 1.63 (1.20-2.21) and RR (95% CI) = 1.94 (1.29-2.92), respectively). For both sexes, mothers' use of positive indirect restrictive statements was more frequent among children with higher enjoyment of food (RR (95% CI) = 1.38 (1.11-1.72).

Conclusions: The statements mothers use to restrict their children's eating vary in affect and directness. Child characteristics, such as sex, BMI, and the presence of specific eating behaviors, are associated with differing approaches to restriction by mothers.