Skip to main content
. Author manuscript; available in PMC: 2009 Feb 17.
Published in final edited form as: Am J Clin Nutr. 2007 Aug;86(2):488–495. doi: 10.1093/ajcn/86.2.488

TABLE 4.

Association between tertile (T) of the cumulative average of the Western pattern and newly diagnosed chronic obstructive pulmonary disease (COPD) according to BMI (3 classes)1

Intake
Western pattern T1 T2 T3 P for trend2
Lean subjects (BMI ≤ 20)
 Cases (n) 16 54 51
 Person-years 10 758 10 168 10 278
  Age-adjusted RR (95% CI) 1.00 3.48 (2.00, 6.07) 4.87 (2.83, 8.40) <0.001
  Multivariate RR (95% CI)3 1.00 2.42 (1.38, 4.25) 2.70 (1.55, 4.70) <0.001
  Multivariate RR (95% CI)4 1.00 2.19 (1.19, 4.04) 2.76 (1.37, 5.56) 0.006
Normal-weight subjects (BMI: 20–25)
 Cases (n) 100 120 117
 Person-years 79 489 78 945 74 534
  Age-adjusted RR (95% CI) 1.00 1.34 (1.03, 1.75) 1.63 (1.24, 2.12) <0.001
  Multivariate RR (95% CI)3 1.00 1.17 (0.90, 1.53) 1.24 (0.95, 1.62) 0.12
  Multivariate RR (95% CI)4 1.00 1.11 (0.82, 1.49) 1.17 (0.74, 1.55) 0.72
Overweight and obese subjects (BMI >25)
 Cases (n) 75 78 91
 Person-years 79 403 81 555 84 928
  Age-adjusted RR (95% CI) 1.00 1.37 (0.99, 1.91) 1.70 (1.24, 2.34) 0.001
  Multivariate RR (95% CI)3 1.00 1.29 (0.92, 1.79) 1.46 (1.06, 2.01) 0.02
  Multivariate RR4 (95% CI)4 1.00 1.30 (0.91, 1.88) 1.50 (0.95, 2.35) 0.08
1

RR, relative risk. The reference category is based on the lowest category of intake. The Western pattern × BMI interaction was significant, P = 0.03.

2

Based on each intake category and modeled as continuous variables in a Cox proportional hazard model.

3

Cox proportional hazard models adjusted for age, smoking status, pack-years, pack-years2, and exposure to secondhand tobacco smoke.

4

Cox proportional hazard models adjusted for age, smoking status, pack-years, pack-years2, exposure to secondhand tobacco smoke, menopausal status, race-ethnicity, spouse’s educational attainment, physician visits, US region, physical activity, multivitamin use, and energy intake.