Characterizing early child growth patterns of height-for-age in an urban slum cohort of Bangladesh with functional principal component analysis

Table 4 Correlations between HAZ measurements and functional principal component scores: FPC1 score and adj-FPC2 score^a

HAZ or HAZ Difference	Boys		Girls
HAZ or HAZ Difference	FPC1 Score ^b	Adj-FPC2 Score^c	FPC1 Score^b	Adj-FPC2 Score^c
HAZ at enrollment	−0.849	0.292	−0.971	−0.136^d
HAZ at 6 months	−0.940	0.100^e	−0.982	−0.179^d
HAZ at 12 months	−0.996	−0.304	−1.000	−0.350
HAZ at 18 months	−0.980	−0.427	−0.992	−0.459
HAZ at 24 months	−0.979	−0.425	−0.990	−0.467
Change of HAZ %tiles from birth to age 24 months	−0.156^d	−0.919	−0.047^e	−0.794
Change of HAZ %tiles between age 6 and 18 months	−0.062^e	−0.905	−0.028^e	−0.787

Abbreviations: HAZ height-for-age z-score, adj-FPC2 adjusted FPC2 score
^aReported as Pearson correlation coefficient
^bFPC1 score was highly correlated with HAZs over times (p-value < 0.0001), but not correlated with the change of HAZ percentiles (%tile) from birth to age 24 months and that between 6 and 18 months in both boys and girls
^cAdj-FPC2 was highly correlated with the change of HAZ percentiles (%tile) from birth to age 24 months and that between 6 and 18 months (p-value < 0.0001), but only weakly or not correlated with HAZs over times in both boys and girls
^{d, e}All the correlations were significantly different from zero with p-value < 0.0001, except those indicated as ⁴ 0.008 < p-value < 0.05 and ⁵ p-value > 0.1

ISSN: 1471-2431