Table 4. Cox proportional hazards models for all-cause mortality
Males
(n = 5,494, Deaths = 3,535)
Females
(n = 6,771, Deaths = 3,632)
Never-smokers
(n = 2,555, Deaths = 1,181)
Data for lung function level∗
(Examination 2)
Q1 FEV1/height2
Q2 FEV1/height2
Q3 FEV1/height2
Q4 FEV1/height2
Asthma†
Chronic bronchitis†
Shortness of breath†
HR 95% CI
1.00
1.27 (1.13–1.44)
1.60 (1.42–1.81)
2.31 (2.04–2.62)
0.98 (0.82–1.17)
1.29 (1.18–1.40)
1.52 (1.36–1.70)
(n = 4,253, Deaths = 2,842)
HR 95% CI
1.00
1.39 (1.22–1.57)
1.83 (1.62–2.07)
2.65 (2.33–3.01)
0.99 (0.83–1.18)
1.32 (1.20–1.46)
1.50 (1.36–1.67)
(n = 5,426, Deaths = 2,958)
HR 95% CI
1.00
1.54 (1.20–1.98)
1.70 (1.33–2.16)
2.15 (1.68–2.76)
1.13 (0.78–1.63)
1.16 (0.94–1.44)
1.38 (1.11–1.70)
(n = 1,822, Deaths = 860)
Data for lung function decline‡
(Examination 1 to 2) 95% CI 95% CI
Q1 FEV1 slope
Q2 FEV1 slope
Q3 FEV1 slope
Q4 FEV1 slope
Asthma
Chronic bronchitis
Shortness of breath
Q1 FEV1 relative slope§
Q2 FEV1 relative slope
Q3 FEV1 relative slope
Q4 FEV1 relative slope
Asthma
Chronic bronchitis
Shortness of breath
FEV1 below LNL§
Asthma
Chronic bronchitis
Shortness of breath
FEV1 decline of ≥ 90 ml/yr
Asthma
Chronic bronchitis
Shortness of breath
∗ Models adjusted for age at examination 2.
†Asthma, chronic bronchitis, and shortness of breath represented as dichotomous variables in all models.
‡Models adjusted for baseline age, height-adjusted baseline lung function (FEV1/height2), and height. See Table 1 for slope and relative slope quartile values.
§FEV1 relative slope (slope FEV1/baseline FEV1) and LNL (Longitudinal Normal Limit).
HR
1.00
1.14
1.20
1.67
0.87
1.24
1.43
1.00
1.10
1.23
1.55
0.84
1.24
1.45
1.48
0.87
1.25
1.44
1.37
0.87
1.25
1.47
95% CI
(1.02–1.27)
(1.12–1.39)
(1.23–1.53)
(1.68–2.10)
(0.78–1.16)
(1.14–1.43)
(1.21–1.54)
(0.95–1.43)
(1.06–1.61)
(1.23–1.91)
(0.73–1.80)
(0.84–1.45)
(0.97–1.66)
(0.98–1.23)
(1.09–1.37)
(1.39–1.73)
(0.68–1.04)
(1.12–1.37)
(1.28–1.65)
(1.35–1.62)
(0.71–1.07)
(1.13–1.39)
(1.27–1.64)
(1.27–1.48)
(0.70–1.07)
(1.13–1.39)
(1.30–1.67)
HR
1.00
1.24
1.37
1.88
0.96
1.27
1.36
1.00
1.25
1.22
1.81
0.91
1.28
1.38
1.58
0.94
1.30
1.36
1.55
0.97
1.29
1.37
(1.11–1.39)
(1.09–1.37)
(1.63–2.01)
(0.74–1.10)
(1.15–1.44)
(1.22–1.55)
(1.46–1.72)
(0.77–1.15)
(1.16–1.45)
(1.21–1.54)
(1.42–1.68)
(0.80–1.18)
(1.15–1.44)
(1.22–1.55)
HR
1.00
1.17
1.31
1.53
1.15
1.10
1.27
1.00
1.18
1.13
1.59
1.11
1.09
1.26
1.42
1.11
1.11
1.29
1.37
1.14
1.10
1.29
(0.96–1.46)
(0.91–1.40)
(1.29–1.95)
(0.71–1.74)
(0.83–1.44)
(0.97–1.65)
(1.22–1.66)
(0.71–1.74)
(0.84–1.46)
(0.99–1.69)
(1.17–1.61)
(0.73–1.79)
(0.83–1.45)
(0.99–1.68)
increasing trend for all outcomes, but less statistical significance (Tables 2–4).
Among the 4 criteria, quartiles for the slope had higher
HRs, but goodness of fit (AIC) was generally better for
the relative slope. The slightly lower risk and better fit for
the relative slope are reasonable given the adjustment for
baseline lung function level. As for the 2 limits (the LNL and
the 90 ml/yr limit), the HRs were similar, but often slightly
higher for the LNL, and the goodness-of-fit was often better
for the LNL.
Investigating critical rates of lung function decline
HRs for COPD morbidity suggest a critical rate at the third
quartile of the slope with risks greater than 2.5 times those for
the reference (Figure 1a), corresponding to declines starting
at 60 ml/yr for males and 57 ml/yr for females (Table 1). For
COPD or CHD mortality, the critical rate is more likely at
the fourth quartile of the slope, with risks nearly doubled for
males and tripled for females (Figure 1b). For all-cause mor-
tality, fourth quartile risks were increased by 67% and 88%
for males and females, respectively (Figure 1c), correspond-
ing to declines starting at 118 ml/yr for males and 95 ml/yr
for females (Table 1).
