Background

Accurate and less resource-intensive approaches for collecting heights and weights data are needed to assess weight status of children in population-based studies. This study compared the accuracy of children’s weight status classification using parent-measured heights and weights and parent-estimated heights and weights against the gold standard of professional measurements.

Methods

The analytical sample of 226 children, 3-18 years of age, was drawn from the New Jersey Child Health Study. The sample included predominantly minority children from low-income households. Data on children’s height and weight collected using three types of measurements were compared in the analysis: parent-measured (Method A) using standard instructions, parent-estimated (Method B), and professionally measured (gold standard). Accuracy of overweight/obesity classification for Methods A and B were compared using sensitivity and specificity analysis.

Results

Compared to the gold standard, the sensitivity and specificity based on parent-measured values (Method A) were 77.7% and 83.6%, and those based on parent-estimated values (Method B) were 78.1% and 78.4%, respectively. Comparison of sensitivity and specificity across two methods revealed that Method A did not significantly improve the accuracy of children’s weight status classification over Method B. No differences were observed in sub-group analysis by child and parent demographic variables.

Conclusions

Parent-measured and parent-estimated heights and weights were equally accurate in identifying 4 of 5 children who were overweight/obese. To improve upon this classification, other alternatives to parent-estimates of child weight status are needed.