Skip to main content
. 2021 Apr 7;114(1):90–100. doi: 10.1093/ajcn/nqab052

TABLE 3.

Mean differences in human milk and blood thiamine biomarkers, between treatment groups1

Unadjusted mean difference, µg/L P value2 Adjusted mean difference,3 µg/L P value2
Human milk total thiamine concentration,4 µg/L 
 1.2 mg − placebo (0 mg) 29.05 (12.17, 45.94) <0.0001 30.09 (13.30, 46.88) <0.0001
 2.4 mg − placebo (0 mg) 38.66 (19.88, 57.45) <0.0001 39.84 (21.21, 58.47) <0.0001
 2.4 mg − 1.2 mg 9.61 (−8.32, 27.54) 0.5 9.75 (−7.99, 27.49) 0.5
 10 mg − placebo (0 mg) 37.25 (18.71, 55.79) <0.0001 38.65 (20.35, 56.95) <0.0001
 10 mg − 1.2 mg 8.19 (−9.30, 25.68) 0.6 8.56 (−8.85, 25.97) 0.6
 10 mg − 2.4 mg −1.42 (−20.93, 18.10) 1.0 −1.19 (−20.20, 17.93) 1.0
 Overall <0.0001 <0.0001
Whole blood ThDP,5 nmol/L
 Mother
  1.2 mg − placebo (0 mg) 43.03 (12.02, 74.05) 0.002 43.17 (12.64, 73.70) 0.002
  2.4 mg − placebo (0 mg) 46.24 (13.17, 79.32) 0.002 46.01 (13.36, 78.66) 0.002
  2.4 mg − 1.2 mg 3.21 (−30.28, 36.71) 1 2.84 (−30.14, 35.82) 1
  10 mg − placebo (0 mg) 77.23 (44.79,109.67) <0.0001 76.78 (44.81,108.76) <0.0001
  10 mg − 1.2 mg 34.19 (1.96, 66.42) 0.03 33.61 (1.93, 65.29) 0.03
  10 mg − 2.4 mg 30.98 (−2.63, 64.59) 0.08 30.77 (−2.40, 63.95) 0.08
  Overall <0.0001 <0.0001
 Infant
  1.2 mg − placebo (0 mg) 56.60 (24.50, 88.69) <0.0001 57.01 (25.31, 88.72) <0.0001
  2.4 mg − placebo (0 mg) 42.04 (7.52, 76.55) 0.01 42.35 (8.00, 76.70) 0.009
  2.4 mg − 1.2 mg −14.56 (−50.71, 21.58) 0.7 −14.66 (−50.52, 21.20) 0.7
  10 mg − placebo (0 mg) 47.31 (16.02, 78.61) 0.0007 46.83 (15.88, 77.78) 0.0007
  10 mg − 1.2 mg −9.29 (−40.65, 22.08) 0.9 −10.18 (−41.22, 20.86) 0.8
  10 mg − 2.4 mg 5.28 (−27.80, 38.35) 1 4.48 (−28.22, 37.18) 1
  Overall <0.0001 <0.0001
ETKac5
 Mother
  1.2 mg − placebo (0 mg) −0.05 (−0.08, −0.01) 0.004 −0.05 (−0.08, −0.01) 0.004
  2.4 mg − placebo (0 mg) −0.06 (−0.09, −0.02) 0.0002 −0.06 (−0.09, −0.02) 0.0002
  2.4 mg − 1.2 mg −0.01 (−0.04, 0.02) 0.7 −0.01 (−0.04, 0.02) 0.7
  10 mg − placebo (0 mg) −0.07 (−0.11, −0.04) <0.0001 −0.07 (−0.11, −0.04) <0.0001
  10 mg − 1.2 mg −0.03 (−0.06, 0.00) 0.1 −0.03 (−0.06, 0.00) 0.1
  10 mg − 2.4 mg −0.02 (−0.05, 0.02) 0.7 −0.02 (−0.05, 0.02) 0.6
  Overall <0.0001 <0.0001
 Infant
  1.2 mg − placebo (0 mg) −0.04 (−0.08, 0.01) 0.1 −0.04 (−0.08, 0.01) 0.1
  2.4 mg − placebo (0 mg) −0.04 (−0.09, 0.01) 0.1 −0.04 (−0.09, 0.01) 0.1
  2.4 mg − 1.2 mg −0.00 (−0.04, 0.03) 1 −0.00 (−0.04, 0.03) 1
  10 mg − placebo (0 mg) −0.06 (−0.10, −0.02) 0.003 −0.06 (−0.10, −0.02) 0.003
  10 mg − 1.2 mg −0.02 (−0.06, 0.01) 0.3 −0.02 (−0.06, 0.01) 0.3
  10 mg − 2.4 mg −0.02 (−0.06, 0.02) 0.6 −0.02 (−0.06, 0.02) 0.6
  Overall 0.009 0.009
1

Data are mean differences (95% CIs) unless otherwise indicated. Results are based on analysis of 100 imputed datasets. ETKac, erythrocyte transketolase activity coefficient; ThDP, thiamine diphosphate concentrations.

2

Post hoc (Tukey) adjusted P values for multiple comparisons.

3

Adjustments: human milk concentrations, adjusted for human milk total thiamine at 2 wk and health center; ThDP, adjusted for health center (mother and infant) and maternal thiamine diphosphate concentrations (corrected for hematocrit) at 2 wk (mother only); ETKac, adjusted for health center (mother and infant) and erythrocyte transketolase activity coefficient at 2 wk (mother only).

4

Linear mixed-effects model was used to analyze the data at 4, 12, and 24 wk with a treatment by time point interaction model. Mean differences between treatment groups collapsed across all time points (4, 12, and 24 wk) are reported, as the treatment by time interaction P value was >0.05 and hence was dropped from the analysis model.

5

Linear regression was used to estimate the mean differences between treatment groups at 24 wk.