Parental feeding practices and associations with child weight status. Swedish validation of the Child Feeding Questionnaire finds parents of 4-year-olds less restrictive

The Child Feeding Questionnaire (CFQ) assesses parental feeding attitudes, beliefs and practices concerned with child feeding and obesity proneness. The questionnaire has been developed in the U.S., and validation studies in other countries are limited. The aim of this study was to examine the psychometric properties of the CFQ in Sweden and the associations between parenting practices and children's weight status. Based on records from the Swedish population register, all mothers of 4-year-olds (n = 3007) from the third largest city in Sweden, Malmö, were contacted by mail. Those who returned the CFQ together with a background questionnaire (n = 876) received the CFQ again to enable test-retest evaluation; 564 mothers completed the CFQ twice. We used confirmatory factor analysis to test whether the original 7-factor model was supported. Good fit (CFI = 0.94, TLI = 0.95, RMSEA = 0.04, SRMR = 0.05) was obtained after minor modifications such as dropping 2 items on restriction and adding 3 error covariances. The internal reliability and the 2-week test-retest reliability were good. The scores on restriction were the lowest ever reported. When the influence of parenting practices on child BMI (dependent variable) was examined in a structural equation model (SEM), child BMI had a positive association with restriction and a negative association with pressure to eat. Restriction was positively influenced by concern about child weight. The second SEM treated parenting practices as dependent variables. Parental foreign origin and child BMI had direct effects on restriction, while pressure to eat was also influenced by parental education. While the results of the study support the usefulness of the CFQ in Sweden, carefully designed cross-cultural comparisons are needed to explain why the levels of restrictive feeding in Swedish families are the lowest reported.