Skip to main content

Table 4 Multivariate quantile regression and ordinary least squares (OLS) coefficients for 2 year BMI-for-age Z-score 1

From: Effects of early feeding on growth velocity and overweight/obesity in a cohort of HIV unexposed South African infants and children

 

OLS mean

10th percentile

20th percentile

30th percentile

40th percentile

50th percentile

60th percentile

70th percentile

80th percentile

90th percentile

Not breastfed at 12 weeks = Yes [Ref: No]

0.32 (0.04, 0.61)

0.04 (−0.36, 0.44)

0.08 (−0.35,0.52)

0.31 (−0.10,0.72)

0.29 (−0.02,0.60)

0.37 (0.07,0.66)

0.34 (0.05,0.63)

0.46 (0.18, 0.74)

0.49 (0.16, 0.83)

0.68 (0.41, 0.94)

R-squared/pseudo R-squared

0.2182

0.1204

0.1238

0.1299

0.1251

0.1165

0.1147

0.1200

0.1216

0.1640

  1. 1Values are ordinary least square (OLS) and quantile regression beta-coefficients (adjusted for birth weight, weight velocity between 12–24 weeks, 12 week BMI-for-age Z-score, maternal education, study site and intervention arm) with respective 95% confidence intervals in brackets. 466 observations were assessed in the models. With the exception of study arm, only variables that had significant association with BMI-for-age Z-score in the bivariate analysis were included in the final model. R-squared/pseudo R-squared.