Table 4

Prediction of airflow limitation (LLN) by multivariate regression analyses

Crude OR (95% CI)Model 1 OR (95% CI)Model 2 OR (95% CI)
HIV, yes versus no1.00 (0.81 to 1.22)0.97 (0.77 to 1.21)0.86 (0.66 to 1.11)
Age, per decade*0.89 (0.85 to 0.94)0.97 (0.92 to 1.03)1.03 (0.95 to 1.11)
Sex, male versus female*1.03 (0.89 to 1.19)0.81 (0.69 to 0.94)0.89 (0.72 to 1.12)
Ancestry
 ScandinavianRefRefRef
 Other European0.87 (0.69 to 1.10)0.92 (0.72 to 1.15)0.92 (0.66 to 1.27)
 Middle East/Indian subcontinent1.07 (0.77 to 1.46)1.15 (0.81 to 1.58)
 Other0.95 (0.60 to 1.43)1.07 (0.65 to 1.69)0.88 (0.41 to 1.65)
Pack-years of smoking, per 10 pack-years1.32 (1.29 to 1.37)1.35 (1.31 to 1.39)1.33 (1.27 to 1.39)
Passive smoking during childhood, per 5 years1.12 (1.06 to 1.17)1.04 (0.98 to 1.10)
Prematurity, yes versus no1.29 (0.98 to 1.68)1.43 (1.07 to 1.89)
Educational level
 No educationRefRef
 Short (<3 years)0.77 (0.54 to 1.10)0.95 (0.65 to 1.41)
 Vocational0.81 (0.62 to 1.07)0.94 (0.70 to 1.28)
 Middle length0.60 (0.46 to 0.80)0.79 (0.58 to 1.09)
 University degree0.54 (0.40 to 0.73)0.88 (0.63 to 1.24)
Pneumonia/acute bronchitis within the past 10 years, n (%)
 NoneRefRef
 1–5 times1.10 (0.97 to 1.25)1.41 (1.17 to 1.70)
 6–10 times1.64 (1.35 to 1.98)3.01 (1.75 to 5.02)
 >10 times3.68 (2.31 to 5.69)4.68 (2.29 to 9.21)
  • * Age and sex were matching variables and effect sizes cannot be interpreted. Model 1 was adjusted for age, sex, ancestry, cumulative smoking (pack-years) and HIV status. Model 2 was adjusted for the same variables, as well as preterm delivery, passive smoking during childhood, educational status and pneumonia within the past 10 years.

  • Pneumonia/bronchitis was defined as an episode requiring sick leave and/or consultation by medical doctor in the past  10   years .

  • LLN, lower limit of normal.