Table 2.

Table 2a. Selected variables from fixed effect-only logistic regression model of current smoker status. (Equation 1). Full results in Table S1. OR (95% CI uncorrected for multiple comparisons) Holm-adjusted p-values^a
		OR for current smoker status
e^2α_C	strong clean indoor air coverage	0.661	(0.656, 0.665)	0.010
e^α_P	centered price per pack of cigarettes (¢)	0.995	(0.995, 0.995)	0.038
e^α_{P_b}	change in price’s effect at $3.28/pack	1.006	(1.006, 1.006)	0.058
e^α_E	centered educational attainment^b	0.839	(0.839, 0.839)	<0.001
e^α_E²	educational attainment²	0.976	(0.976, 0.976)	<0.001
e^α_E³	educational attainment³	1.001	(1.001, 1.001)	0.042
e^α_I	centered household income ($10,000s)	0.842	(0.842, 0.842)	<0.001
e^α_I²	household income² effect at $70,000	1.028	(1.028, 1.028)	<0.001
e^2α_B	black	0.945	(0.928, 0.963)	ns
e^2α_N	Native American^c	1.370	(1.268, 1.479)	ns
e^2α_A	Asian/Pacific Islander	0.504	(0.492, 0.516)	0.030
e^2α_L	Latino	0.473	(0.472, 0.474)	<0.001
e^2α_CB	clean indoor air law * black	1.705	(1.680, 1.731)	0.033
e^2α_CN	clean indoor air law * Native American	0.711	(0.658, 0.768)	ns
e^2α_CA	clean indoor air * Asian/Pacific Islander	0.817	(0.801, 0.833)	ns
Table 2b. Selected variables from fixed effect-only linear regression model of cigarette consumption among current smokers (Equation 2). Full results in Table S3. exponentiated coefficients (95% CI uncorrected for multiple comparisons) Holm-adjusted p-values^a
		multiplier for consumption cigarettes/day
e^2β_C	strong clean indoor air coverage	0.826	(0.826, 0.827)	0.001
e^β_P	centered price per pack of cigarettes (¢)	0.997	(0.997, 0.997)	0.015
e^β_{P_b}	change in price’s effect at $3.20/pack	1.003	(1.003, 1.003)	0.031
e^β_{E_b}	threshold ed. attainment at 12^th grade^b	0.961	(0.961, 0.961)	<0.001
e^β_I	centered household income ($10,000s)	0.992	(0.992, 0.992)	0.015
e^2β_B	black	0.877	(0.875, 0.878)	0.156
e^2β_N	Native American^c	1.006	(1.000, 1.011)	0.960
e^2β_A	Asian/Pacific Islander	0.697	(0.683, 0.712)	0.310
e^2β_L	Latino	0.606	(0.605, 0.606)	<0.001

Holm adjustment is a sequential procedure with a stopping criterion; all parameter estimates with p-values larger than that of the first rejected p-value are rejected; such rejection is indicated by “ns”.

See Table 1 for coding.

Includes American Indians, Aleutians, and Eskimo