Table 2 Multivariable linear regression model to assess the association between inhalant groups and sleep latency by PSQI.

From: Dual use of e-cigarettes with conventional tobacco is associated with increased sleep latency in cross-sectional Study

 

Mean difference

95% CI

Wald Chi-squared

p-value

Age

0.04

(−0.07, 0.15)

0.42

0.52

Gender

0.023

Male vs female

−2.89

(−5.38, −0.40)

5.16

 

Race

6.10

0.047

Asian vs. White

−3.83

(−6.93, −0.73)

  

Other vs. White

−0.14

(−4.08, 3.81)

  

Ethnicity

  

4.78

0.092

Hispanic vs. non-Hispanic

−4.35

(−8.62, −0.08)

  

Unknown vs. non-Hispanic

−4.67

(−13.24, 3.90)

  

Presence of cough

Yes vs no

0.16

(−2.37, 2.68)

0.01

0.90

Inhalant group

  

8.24

0.041

Conventional vs non-smoker

0.43

(−3.82, 4.67)

  

 E-cig vs non-smoker

−0.61

(−5.28, 4.05)

  

 Dual vs non-smoker

4.08

(1.12, 7.05)

  

Pairwise group comparisons

Mean difference

95% CI

Raw p-value

Adjusted p-value

Dual vs nonsmoker

4.08

(1.12, 7.05)

0.007

0.042

Dual vs e-cig

4.70

(−0.39, 9.78)

0.07

0.198

Dual vs conventional

3.66

(−0.69, 8.01)

0.099

0.198

E-cig vs nonsmoker

−0.61

(−5.28, 4.06)

0.797

0.844

E-cig vs conventional

−1.04

(−6.96, 4.88)

0.731

0.844

Conventional vs nonsmoker

0.43

(−3.83, 4.68)

0.844

0.844

  1. This table shows the multivariable linear regression model with sleep latency as the outcome, inhalant groups, age, gender, race, ethnicity, and presence of cough as the predictors. The bottom table further shows the pairwise comparisons among inhalant groups with adjusted p-values using the method of Benjamini & Hochberg. Sleep latency was determined through the following open-ended question included within the Pittsburgh Sleep Quality Index (PSQI), “During the past month, how long (in minutes) has it taken you to fall asleep each night?”.
Back to article page