Skip to main content
. 2018 Oct 22;99(6):1606–1612. doi: 10.4269/ajtmh.18-0143

Table 4.

Association between water quality (safe vs. unsafe)† and growth outcomes (LAZ, WAZ, and WLZ) at birth, 6 months, 9 months, and the time of the L:M test in unadjusted and adjusted linear regression models‡

Outcome Unadjusted linear regression model Adjusted linear regression model
Growth at birth
LAZ 0.65 (0.06, 1.24)* 0.57 (0.10, 1.04)*
WAZ 0.18 (−0.06, 0.43) 0.15 (−0.12, 0.42)
WLZ −0.38 (−1.04, 0.28) −0.38 (−1.02, 0.27)
Growth at 6 months
LAZ 0.40 (−0.27, 1.08) 0.16 (−0.24, 0.56)
WAZ 0.35 (0.18, 0.52)* 0.23 (0.06, 0.41)*
WLZ −0.02 (−0.64, 0.60) 0.13 (−0.32, 0.58)
Growth at 9 months
LAZ 0.25 (−0.40, 0.89) 0.10 (−0.28, 0.48)
WAZ 0.35 (0.11, 0.60)* 0.23 (−0.03, 0.49)
WLZ 0.18 (−0.33, 0.69) 0.16 (−0.31, 0.63)
Growth at L:M test (12–16 months)
LAZ 0.39 (0.13, 0.65)* 0.29 (0.00, 0.58)*
WAZ 0.29 (0.16, 0.43)* 0.20 (0.05, 0.34)*
WLZ 0.08 (−0.24, 0.40) 0.14 (−0.17, 0.44)

L:M = lactulose:mannitol; LAZ = length-for-age Z-score; WAZ = weight-for-age Z-score; WLZ = weight-for-length Z-score. Cells present β coefficient and 95% confidence interval, * P-value < 0.05.

† Safe water is defined as the lack of the presence of Escherichia coli contamination according to the results of a compartment bag test. Unsafe water is defined as any E. coli contamination detected.

‡ Unadjusted and adjusted regression models adjusted for subcounty clustering. Adjusted regression models control for gender of child, gender of household head, mother’s height, caregiver education level, family size, and asset score. Adjusted regression model for 6 months, 9 months, and the L:M test time-point controls for LAZ, WAZ, and WLZ, respectively, at birth.