Skip to main content
. 2019 Feb 1;199(3):302–312. doi: 10.1164/rccm.201807-1345OC

Table 3.

Multinomial Logistic Regression Models for Airway Hyperresponsiveness as Measured from Age 11 to Age 26 Years*

Airway Hyperresponsiveness z-Score CC16 All Participants (n Subjects = 555; n Observations = 1,228)
No Asthma (n Subjects = 493; n Observations = 1,033)
Nonsmoker (n Subjects = 485; n Observations = 894)
adjRRR (95% CI) P Value adjRRR (95% CI) P Value adjRRR (95% CI) P Value
  Continuous CC16            
No drop   (Base outcome)
Mild   0.92 (0.77–1.10) 0.352 0.89 (0.73–1.09) 0.270 0.88 (0.70–1.09) 0.232
Moderate 0.75 (0.63–0.89) 0.001 0.76 (0.62–0.92) 0.006 0.63 (0.50–0.80) <0.001
Severe 0.66 (0.55–0.81) <0.001 0.63 (0.50–0.79) <0.001 0.58 (0.45–0.73) <0.001
  Category  
No drop   (Base outcome)
Mild  High CC16 (Reference group)
 Middle CC16 1.18 (0.80–1.73) 0.404 1.22 (0.81–1.82) 0.343 1.41 (0.91–2.18) 0.126
 Low CC16 1.19 (0.77–1.84) 0.426 1.21 (0.77–1.91) 0.405 1.38 (0.82–2.32) 0.221
Moderate  High CC16 (Reference group)
 Middle CC16 1.25 (0.84–1.86) 0.274 1.25 (0.81–1.94) 0.317 1.53 (0.94–2.48) 0.088
 Low CC16 1.86 (1.22–2.85) 0.004 1.88 (1.19–2.98) 0.007 2.62 (1.57–4.39) <0.001
Severe  High CC16 (Reference group)
 Middle CC16 1.45 (0.96–2.18) 0.078 1.80 (1.12–2.89) 0.015 1.52 (0.97–2.39) 0.067
 Low CC16 2.30 (1.46–3.61) <0.001 2.64 (1.54–4.51) <0.001 3.05 (1.80–5.18) <0.001

Definition of abbreviations: adjRRR = adjusted relative risk ratio; CC16 = club cell secretory protein 16; CI = confidence interval.

*

Subject-clustered sandwich estimators of SE were used in multinomial logistic regression models to adjust for within-subject correlation. All models were adjusted for sex, survey year, ethnicity, parental education, and maternal smoking at Year 6. Airway hyperresponsiveness severity (mild, moderate, severe) was defined at each survey based on tertiles of provocative concentration of methacholine causing a 20% fall in FEV1 (see Methods section for more information).

Circulating CC16 levels were included as z-scores in the model.