Introduction: Anthropometric data are key to evaluating infant health. This study assessed the validity of parent-reported infant weight and length, and their reliability to categorise children by BMI z-score, as compared to clinical measurements.
Methods: From a cohort of 4,262 infants, parent-reported and clinically measured anthropometric data were obtained and compared at three months and one year of age.
Results: Parent-reported and clinically measured data generally correlated well. Mean differences at three months and at one year, respectively, were 0.08 kg (95% confidence interval (CI): 0.07-0.09 kg) and 0.10 kg (95% CI: 0.08-0.12 kg) for weight, 0.8 cm (95% CI: 0.8-0.9 cm) and 1.0 cm (95% CI: 0.9-1.1 cm) for length and -0.16 kg/m2 (95% CI: -0.20--0.12 kg/m2) and -0.22 kg/m2 (95% CI: -0.27--0.18 kg/m2) for BMI. Effect sizes were negligible to small. Bland-Altman plots showed clinically insignificant bias, but 95% limits of agreement were wide enough to be significant. Comparing categorisation of BMI z-score showed only fair agreement.
Conclusion: Parents' reports of measured infant weight and length are reliable at a population level in a setting with routine preventive care. Parent-reported data should not be used for assessment of individual infants, particularly not if a health condition is suspected. BMI calculated from parent-reported anthropometrics is not reliable.
Funding: None.
Trial registration: This study was registered with www.
Clinicaltrials: gov, registration number NCT01694108.
Articles published in the DMJ are “open access”. This means that the articles are distributed under the terms of the Creative Commons Attribution Non-commercial License, which permits any non-commercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.