Table 4.

Adjusted mean differences in term birth weight for a +10-percentage point increase in the probability that private well arsenic concentrations exceed the respective threshold, overall and stratified by county-level rates of private well use.

Term Birth Weight (grams) β (95% CI)
	Overall	0–25% private well use	>25–50% private well use	>50–75% private well use	>75% private well use
No. of Births	3,305,090	2,810,765	393,685	91,114	9,526
No. of Counties	3,105	1,522	1,091	415	76
Boosted regression tree: 1 μg/L
≤1 μg/L	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)
>1 μg/L	−3.12 (−4.05, −2.18)*	4.19 (−5.33, −3.05)*	−1.78 (−3.50, −0.06)*	−3.63 (−6.99, −0.27)*	1.49 (−7.65, 10.60)
Boosted regression tree: 5 μg/L
≤5 μg/L	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)
>5 μg/L	−2.73 (−4.19, −1.26)*	−4.86 (−6.81, −2.91)*	−0.61 (−3.06, 1.85)	−2.51 (−6.69, 1.68)	3.52 (−16.30, 23.30)
Boosted regression tree: 10 μg/L
≤10 μg/L	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)
>10 μg/L	−3.85 (−6.07, −1.64)*	−6.86 (−10.10, −3.65)*	−1.45 (−4.80, 1.91)	−2.87 (−9.01, 3.28)	20.10 (−19.90, 60.00)
Random forest classification: 5 and 10 μg/L
≤5 μg/L	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)	0.00 (reference)
>5 to ≤10 μg/L	−4.39 (−7.91, −0.86)*	−6.91 (−11.40, −2.46)*	−1.67 (−7.57, 4.24)	1.54 (−11.70, 14.70)	−16.10 (−51.30, 19.20)
>10 μg/L	−1.54 (−4.16, 1.06)	−3.38 (−7.00, 0.25)	−1.33 (−5.55, 2.89)	−2.58 (−9.85, 4.69)	32.00 (1.30, 62.80)*

Note: CI, confidence interval. Asterisks denote statistical significance. Estimates are adjusted for maternal age, race/ethnicity, marital status, educational attainment, smoking during pregnancy, pre-pregnancy BMI, rurality/urbanicity, and the average annual concentration of particulate matter with aerodynamic diameter less than 2.5 μm in the county with nested random intercepts for maternal county and state of residence.