Table 4.

Effect of Cut-off level for respiratory index on the weighted longitudinal association between Wave 2 tobacco use and selected constructs and improvement in functionally important respiratory symptoms at Wave 3, Population Assessment of Tobacco and Health (PATH) Study^a

Risk factor at baseline	Improvement of respiratory symptoms over time
	Model 1: Respiratory index cutoff ≥ 3b				Model 2: Respiratory index cutoff ≥ 2b
	Symptomatic Wave 2→Asymptomatic Wave 3				Symptomatic Wave 2→Asymptomatic Wave 3
	(Unweighted N = 1582) → (29% became asymptomatic)				(Unweighted N = 2339) → (21% became asymptomatic)
	Unweighted N	Improve %	Adjusted RRc	95% CI	Unweighted N	Improve %	Adjusted RRc	95% CI
Wave 2 past month tobacco used
Never	250	36%	Ref	Ref	389	25%	Ref	Ref
Former	160	28%	0.89	[0.61, 1.30]	255	26%	1.19e	[0.83, 1.71]
Exclusive use categories
Cigarette	559	26%	0.86	[0.63, 1.18]	786	13%	0.57	[0.40, 0.82]
E-cigarette	23	47%†	1.64e	[1.04, 2.58]	43	32%	1.59e	[0.89, 2.85]
Cigars	25	29%†	0.87	[0.35, 2.20]	47	36%	1.43e	[0.87, 2.37]
Smokeless tobacco	29	21%†	0.76	[0.36, 1.59]	40	23%	1.32e	[0.71, 2.47]
Hookah	14	43%†	1.28	[0.56, 2.95]	16	32%	1.60	[0.54, 4.71]
Multiple use categories
Cigarette + e-cigarette	152	28%	0.95	[0.64, 1.42]	225	15%	0.72	[0.47, 1.11]
Combustible only	180	23%	0.78	[0.50, 1.21]	267	17%	0.76	[0.52, 1.11]
Otherf combustible + noncombustible	185	25%	0.91	[0.61, 1.34]	264	20%	0.97e	[0.67, 1.39]
Other smoke-related exposures
Cigarette pack years (per each additional 5 pack years)	1582	N/A	0.89	[0.84, 0.94]	2339	N/A	0.92	[0.87, 0.98]
Second-hand smoke exposure (per each additional 5 hrs/week)	1582	N/A	0.98	[0.96, 1.01]	2339	N/A	0.96	[0.94, 0.99]
Past-month marijuana useg
No	1128	30%	Ref	Ref	1682	22%	Ref	Ref
Yes	454	25%	0.86	[0.67, 1.11]	657	17%	0.87	[0.65, 1.17]

N = 16 295 adult respondents without chronic obstructive pulmonary disease or other nonasthma respiratory disease and with PATH Study longitudinal (all-waves) weights and complete data on all variables.

In Model 1, symptom improvement is defined as moving from a symptom score of ≥3 to <3. In Model 2, symptom improvement is defined as moving from a symptom score of ≥2 to <2.

All risk ratios (RR) adjust for the variables in the table, age, sex, race/ethnicity, education, income, BMI, CHF, heart attack, diabetes, cancer, regular use of beta blockers, angiotensin receptor blockers, or ace inhibitors, and living in an urban area.

Data are not presented for exclusive pipe (N = 44), and dual e-cigarette + smokeless (N = 20) users due to small sample size. Never tobacco user category includes former experimental (e.g., lifetime use of <100 cigarettes) users; former established user category includes all established users (e.g., lifetime use of more than 100 cigarettes) who did not use a tobacco product in the past 30 days.

These exclusive product risk ratios are significantly different from exclusive cigarette use.

Other than exclusive dual cigarette and e-cigarette users.

Marijuana use variable does not distinguish between combustible and noncombustible use.

Estimate should be interpreted with caution because it has low statistical precision. It is based on a denominator sample size of less than 50, or the coefficient of variation of the estimate or its complement is larger than 30%.

Supplementary Table 6 reports standard errors for all the weighted estimates presented in this table and adjusted RRs for all the variables in the model.