Table 2.

Adjusted differences (95% confidence intervals) in systolic blood pressure (mmHg) associated with an interquartile range increase (IQR) in lead concentrations in the National Health and Nutrition Examination Survey (NHANES).

	Log blood lead	Predicted patella lead		Predicted tibia lead
		Full model	Reduced model	Full model	Reduced model
All participants^a	0.91 (0.57, 1.25)	1.45 (0.96, 1.93)	1.98 (1.16, 2.80)	2.09 (1.55, 2.63)	2.21 (1.69, 2.74)
By gender^b
Men	0.74 (0.31, 1.19)	1.74 (1.15, 2.34)	1.86 (1.21, 2.51)	2.27 (1.63, 2.91)	2.35 (1.70, 2.99)
Women	0.46 (−0.07, 0.99)	1.44 (0.60, 2.29)	2.21 (1.31, 3.11)	2.51 (1.52, 3.51)	2.54 (1.63, 3.45)
By age
≥50 years	1.52 (0.72, 2.37)	2.15 (1.28, 3.03)	2.32 (1.36, 3.27)	2.53 (1.57, 3.50)	2.37 (1.45, 3.28)
<50 years	0.44 (0.09, 0.78)	0.60 (−0.01, 1.21)	1.18 (0.51, 1.85)	1.38 (0.71, 2.05)	1.62 (0.93, 2.31)

Note: Interquartile range (IQR) for bone lead is 16.2 μg/g for predicted patella lead from the full model, 16.0 μg/g for predicted patella lead from the reduced model, 10.9 μg/g for predicted tibia lead from the full model, and 10.8 μg/g for predicted tibia lead from the reduced model. Interquartile range for log-transformed blood lead is 0.89.

Models for all participants were adjusted for age, gender, race/ethnicity, education, NHANES survey cycles, smoking status, cumulative cigarette smoking (pack-year), body mass index, and alcohol consumption.

Stratified models by gender have the same covariates included in the model for all participants except gender.