Table 4.

Effect of calcium supplementation^a by baseline blood lead level.

Baseline blood lead level	No. (calcium/placebo)	β-Coefficient	%Δb	p-Value
Among all women with follow-up (intent-to-treat analysis)
< 5 μg/dL	349 (183/166)	−0.07	−7%	0.08
≥ 5 μg/dL	208 (100/108)	−0.19	−17%	0.003
Among those women with compliance ≥ 50%c (as-treated analysis, among high compliers)
< 5 μg/dL	292 (162/130)	−0.15	−14%	0.01
≥5 μg/dL	183 (87/96)	−0.19	−17%	0.004
Among those women with compliance < 50%c (as-treated analysis, among low compliers)
< 5 μg/dL	57 (21/36)	0.29	34%	0.02
≥5 μg/dL	25 (13/12)	−0.18	−17%	0.49

^aAdjusting for baseline blood lead level (log-transformed), maternal age, dietary calcium intake at baseline, daily energy intake at baseline, treatment group, and trimester of pregnancy.

^bPercent change; log-transformed outcome variable, thus estimate of treatment effect: 1 − e^−β.

^cWe assessed compliance by pill count at each visit and analyzed as proportion of expected pills used between baseline (first trimester) and end of follow-up (8 months’ gestation).