Improving assessment of child growth in a pediatric hospital setting

Abstract

Background: Accurate anthropometric measurements are essential for assessing nutritional status, monitoring child growth, and informing clinical care. We aimed to improve height measurements of hospitalized pediatrics patients through implementation of gold standard measurement techniques.

Methods: A quality improvement project implemented computerized training modules on anthropometry and standardized wooden boards for height measurements in a tertiary children's hospital. Heights were collected pre- and post-intervention on general pediatric inpatients under 5 years of age. Accuracy of height measurements was determined by analyzing the variance and by comparing to World Health Organization's defined biologically plausible height-for-age z-scores. Qualitative interviews assessed staff attitudes.

Results: Ninety-six hospital staff completed the anthropometry training. Data were available on 632 children pre- and 933 post-intervention. Training did not increase the proportion of patients measured for height (78.6% pre-intervention vs. 75.8% post-intervention, p = 0.19). Post-intervention, wooden height boards were used to measure height of 34.8% patients, while tape measures and wingspan accounted for 42.0% and 3.5% of measurements, respectively. There was no improvement in the quality of height measurements based on plausibility (approximately 3% height-for-age z-scores measurements flagged out of range pre- and post-intervention), digit preference (13.4% of digits pre- and 12.3% post-intervention requiring reclassification), or dispersion of measurements (height-for-age z-scores standard deviation 1.9 pre- and post-intervention). Staff reported that using the wooden board was too labor consuming and cumbersome.

Conclusions: Our findings suggest that efforts to improve anthropometric measurements of hospitalized children have multiple obstacles, and further investigation of less cumbersome methods of measurements may be warranted.

Keywords: Anthropometry; Child health; Growth; Pediatrics.

Fig. 1

Flowchart of included subjects

Fig. 2

Height-for-age z-score distribution for children under 5 years, pre^A and post^B intervention. HAZ: Height-for-age z-score (HAZ) The vertical bars represent measured data, the solid line represents the expected distribution of z-scores, and the dashed line represents ±2 standard deviations. Panel a depicts the pre-intervention distribution while panel b depicts the post-intervention distribution. World Health Organization standard deviation ranges for data quality assessment for HAZ 1.20–1.30 [13]

Fig. 3

Weight-for-age z-score distribution for children under 5 years, pre^A and post^B intervention. WAZ: Weight-for-age z-score (WAZ) The vertical bars represent measured data, the solid line represents the expected distribution of z-scores, and the dashed line represents ±2 standard deviations. Panel a depicts the pre-intervention distribution while panel b depicts the post-intervention distribution. World Health Organization standard deviation ranges for data quality assessment for WAZ 1.17–1.46 [13]

Fig. 4

Weight-for-height z-score distribution for children under 5 years, pre^A and post^B intervention. WHZ: Weight-for-height z-score (WHZ) The vertical bars represent measured data, the solid line represents the expected distribution of z-scores, and the dashed line represents ±2 standard deviations. Panel a depicts the pre-intervention distribution while panel b depicts the post-intervention distribution. World Health Organization standard deviation ranges for data quality assessment for WHZ 1.08–1.50 [13]