. 2017 Feb;139(2):e20162509. doi: 10.1542/peds.2016-2509

TABLE 3.

The Association Between Maternal Smoking and Offspring ADHD Diagnoses Compared With the Associations With Three Negative Controls, Mutually Adjusted for Each Other

	Crude^a					Adjusted^b
	Hazard Ratio	P	95% CI	χ² Value	P Value χ² Test	Hazard Ratio	P	95% CI	χ² Value	P Value χ² Test
Maternal smoking	1.93	<.001	1.70–2.19	7.47	.006	1.39	<.001	1.22–1.59	2.64	.10
Paternal smoking	1.48	<.001	1.34–1.64	7.47	.006	1.19	.001	1.07–1.33	2.64	.10

Maternal smoking	2.16	<.001	1.90–2.45	17.47	<.001	1.44	<.001	1.25–1.65	2.89	.09
Grandmother’s smoking	1.45	<.001	1.30–1.62	17.47	<.001	1.22	<.001	1.09–1.36	2.89	.09

Maternal smoking Maternal smoking in previous pregnancies	1.65	<.001	1.37–2.00	0.03	.87	1.29	.008	1.07–1.56	0.06	.81
Maternal smoking Maternal smoking in previous pregnancies	1.70	<.001	1.43–2.02	0.03	.87	1.34	.001	1.14–1.59	0.06	.81

CI, confidence interval.

Maternal smoking and the negative controls mutually adjusted for each other, but not adjusted for covariates.

Maternal smoking and the negative controls mutually adjusted for each other and for the following covariates: maternal and paternal age, maternal and paternal education, maternal and paternal ADHD symptoms, maternal (prepregnancy) and paternal BMI, maternal alcohol consumption during pregnancy, parity, child’s birth year, and geographical region. χ² values were obtained from testing 3 null hypotheses that the association between maternal smoking and offspring ADHD diagnoses was equal to each of the 3 other associations (Wald tests). Degrees of freedom = 1 in all tests.