K N Parkinson1, R F Drewett, A R Jones, A J Adamson. 1. Institute of Health and Society, Human Nutrition Research Centre, Newcastle University, Newcastle upon Tyne, UK. kathryn.parkinson@ncl.ac.uk
Abstract
BACKGROUND: Mothers' responses to questionnaire items assessing their child's weight status typically do not correspond to conventional clinical classifications based on body mass index (BMI). From this observation health professionals infer that mothers do not recognize overweight in their child. But the questions used have generally confounded factual judgements with values, so it is not clear whether the mothers are making factual errors, or differ from professionals in their values. METHODS: Cross-sectional study of population-based birth cohort at 6-8 years and their mothers (n = 540). An objective BMI matching task was used to determine the accuracy of mothers' recognition of their child's weight. Mothers matched their child to sex- and age-specific images of children of known BMI ranging from very thin to obese, and chose a descriptor of their child's weight of the kind used in previous research. RESULTS: Mothers tended to underestimate their child's BMI on the matching task. Matching errors significantly predicted mothers' description of their child's weight; those who overestimated their child's BMI on the matching task were more likely to say their child was overweight, while those who underestimated it were less likely to, independently of their child's actual BMI. CONCLUSIONS: Educational programmes aimed at parents of young primary school children need to address separately the factual and the evaluative components of their assessment of child weight.
BACKGROUND: Mothers' responses to questionnaire items assessing their child's weight status typically do not correspond to conventional clinical classifications based on body mass index (BMI). From this observation health professionals infer that mothers do not recognize overweight in their child. But the questions used have generally confounded factual judgements with values, so it is not clear whether the mothers are making factual errors, or differ from professionals in their values. METHODS: Cross-sectional study of population-based birth cohort at 6-8 years and their mothers (n = 540). An objective BMI matching task was used to determine the accuracy of mothers' recognition of their child's weight. Mothers matched their child to sex- and age-specific images of children of known BMI ranging from very thin to obese, and chose a descriptor of their child's weight of the kind used in previous research. RESULTS: Mothers tended to underestimate their child's BMI on the matching task. Matching errors significantly predicted mothers' description of their child's weight; those who overestimated their child's BMI on the matching task were more likely to say their child was overweight, while those who underestimated it were less likely to, independently of their child's actual BMI. CONCLUSIONS: Educational programmes aimed at parents of young primary school children need to address separately the factual and the evaluative components of their assessment of child weight.