Background

Reduced lung function is associated with increased mortality, but it is unclear how different spirometric variables are related to cause-specific death.

Objectives

To investigate the association of FEV₁ and FVC with cause-specific mortality in a large general population cohort.

Methods

The study population consisted of 26,091 men aged 30?46 years from the Pneumoconiosis Survey of Western Norway conducted in 1988?1990 with follow-up on date and cause of death for 26 years. Cox proportional hazard models were used to estimate the association between baseline FEV₁ and FVC z-scores (calculated according to GLI-2012 equations) and cause-specific mortality (European 2012 shortlist classification (EU)), after adjustment for age, body mass index (BMI), smoking habits and education.

Results

In total, 2,462 (9%) subjects died; 32% from cancer, 25% cardiovascular deaths, 14% respiratory deaths and 11% non-natural deaths. FEV₁ had stronger associations with respiratory mortality (EU 8 and 2.1.8) compared to FVC (HR for one unit z-score decrease 1.43 (95%CI 1.28, 1.60) and 1.20 (1.06, 1.36)). FEV₁ predicted non-cancer respiratory deaths (EU 8) (HR 2.26 (95%CI 1.83, 2.77)) stronger than lung cancer deaths (EU 2.1.8) (1.22 (1.07, 1.39)). FVC and FEV₁ had comparable associations with cardiovascular (EU 7) / diabetes (EU 4.1) / hematological cancer (EU 2.1.19?21) deaths, while HR for suicide deaths (EU 17.2) was higher for FEV₁ than FVC (HR 1.32 (95%CI 1.08, 1.62) and 1.22 (0.98, 1.52)).

Conclusions

Lung function predicts numerous mortality causes, both pulmonary and extra-pulmonary. Multiple mechanisms are likely to explain these associations.