TABLE 4a.

Associations between maternal iron status at early and mid-pregnancy with systolic blood pressure in late pregnancy (multivariable linear regression).

Gestational age	Predictor	Systolic blood pressure (mmHg)
		Model 1 (n = 190)			Model 2 (n = 189)			Model 3
		B	95% CI	p	B	95% CI	p	B	95% CI	p
< 18 weeks	Hb (g/dL)	−0.46	−1.46, 0.53	0.358	−0.87	−1.87, 0.13	0.089	-	-	-
	Fer (μg/L)	−0.01	−0.02, 0.01	0.376	−0.01	−0.02, 0.01	0.273	-	-	-
	sTfR (mg/L)	0.04	−0.15, 0.22	0.709	0.08	−0.10, 0.26	0.407	-	-	-
		Model 1 (n = 189)			Model 2 (n = 188)			Model 3 (n = 188)
22 weeks	Hb (g/dL)	−1.04	−2.06, -0.02	0.046	−1.10	−2.11, -0.09	0.033	−0.75	−1.59, 0.09	0.081
	Fer (μg/L)	−0.01	−0.04, 0.02	0.660	−0.02	−0.04, 0.01	0.310	−0.01	−0.03, 0.02	0.695
	sTfR (mg/L)	−0.03	−0.33, 0.27	0.858	−0.03	−0.32, 0.28	0.831	−0.07	−0.31, 0.17	0.566

CI, confidence interval; Hb, haemoglobin; Fer, serum ferritin; sTfR, soluble transferrin receptor. Associations were assessed with multivariable linear regression, B-values and 95% confidence intervals. Model 1 is adjusted for maternal age and living standards measure (socio-economic status); model 2 is adjusted additionally for gestational age, CRP and maternal BMI at baseline, HIV status throughout pregnancy and aspirin prescription during mid-pregnancy; model 3 is adjusted additionally for blood pressure at ± 22 weeks’ gestation. p < 0.05 was considered significant and p between 0.05 and 0.08 had a tendency towards significance.