Background Reduced ventilatory function is an established predictor of all-cause mortality in the general population. We sought to verify this among lifelong non-smokers and investigate the association with circulatory disease and cancer deaths using mortality follow-up from UK national population surveys.

Methods In UK Biobank, among 149,348 white never-smokers aged 40–69 years at entry, there were 1,357 deaths over a mean 4.6 years of follow-up. In the Health Surveys for England (HSE) 1995, 1996, 2001 and Scottish Health Surveys (SHeS) 1998 and 2003 combined there were 503 deaths among 6,604 white never-smokers aged 40–69 at entry, followed for a mean 14.3 years. Spirometry reference equations (Global Lung Initiative 2012) were used to derive age-sex-height- adjusted standard deviation (z) scores for forced expiratory volume in the first second (FEV1) and forced vital capacity (FVC). The associations of z-scores for FEV1 and FVC with deaths from all causes, circulatory disease and cancers were examined in Stata using proportional hazards models adjusted for age, sex, standing height, socio-economic status, region and survey.

Results In the HSE and SHeS combined dataset, decreasing z-scores for FEV1 and FVC were each associated to a similar degree with increased all-cause mortality (hazard ratios per SD decrement 1.17, 95% CI 1.09–1.25 for zFEV1 and 1.19, 1.10–1.27 for zFVC). This was replicated in UK Biobank (HRs per SD 1.24, 1.18–1.30 and 1.27, 1.20–1.33, respectively). In HSE-SHeS, zFEV1 and zFVC were also associated to similar degrees with mortality from circulatory diseases (HRs per SD 1.21, 1.05–1.38 for zFEV1 and 1.22, 1.06–1.40 for zFVC). These associations were stronger in Biobank (1.48, 1.33–1.65 and 1.52, 1.36–1.70, respectively). For cancer mortality, corresponding hazard ratios were more consistent: 1.10, 1.00–1.22 for zFEV1 and 1.12, 1.01–1.24 for zFVC in HSE-SHeS and 1.11, 1.05–1.18 and 1.13, 1.06–1.20, respectively, in Biobank. Among 102,950 white never-smokers in Biobank with spirograms most closely approximating internationally recommended acceptability and reproducibility criteria, of whom 881 died, z-scores for the FEV1/FVC ratio were significantly associated with all-cause mortality when modelled jointly with zFEV1 but z-scores for FEV1/FVC were not associated with survival independent of zFVC.

Conclusion FEV1 and FVC each predict all-cause mortality, circulatory disease mortality and cancer mortality among white lifelong non-smokers in the UK. In those who are able to complete spirometry to stringent international standards, FVC is a more influential predictor than FEV1. However, in settings where end-blow quality is difficult to assess, FEV1 is the more generally applicable predictor.