Articles
Childhood predictors of lung function trajectories and future COPD risk: a prospective cohort study from the first to the sixth decade of life

https://doi.org/10.1016/S2213-2600(18)30100-0Get rights and content

Summary

Background

Lifetime lung function is related to quality of life and longevity. Over the lifespan, individuals follow different lung function trajectories. Identification of these trajectories, their determinants, and outcomes is important, but no study has done this beyond the fourth decade.

Methods

We used six waves of the Tasmanian Longitudinal Health Study (TAHS) to model lung function trajectories measured at 7, 13, 18, 45, 50, and 53 years. We analysed pre-bronchodilator FEV1 z-scores at the six timepoints using group-based trajectory modelling to identify distinct subgroups of individuals whose measurements followed a similar pattern over time. We related the trajectories identified to childhood factors and risk of chronic obstructive pulmonary disease (COPD) using logistic regression, and estimated population-attributable fractions of COPD.

Findings

Of the 8583 participants in the original cohort, 2438 had at least two waves of lung function data at age 7 years and 53 years and comprised the study population. We identified six trajectories: early below average, accelerated decline (97 [4%] participants); persistently low (136 [6%] participants); early low, accelerated growth, normal decline (196 [8%] participants); persistently high (293 [12%] participants); below average (772 [32%] participants); and average (944 [39%] participants). The three trajectories early below average, accelerated decline; persistently low; and below average had increased risk of COPD at age 53 years compared with the average group (early below average, accelerated decline: odds ratio 35·0, 95% CI 19·5–64·0; persistently low: 9·5, 4·5–20·6; and below average: 3·7, 1·9–6·9). Early-life predictors of the three trajectories included childhood asthma, bronchitis, pneumonia, allergic rhinitis, eczema, parental asthma, and maternal smoking. Personal smoking and active adult asthma increased the impact of maternal smoking and childhood asthma, respectively, on the early below average, accelerated decline trajectory.

Interpretation

We identified six potential FEV1 trajectories, two of which were novel. Three trajectories contributed 75% of COPD burden and were associated with modifiable early-life exposures whose impact was aggravated by adult factors. We postulate that reducing maternal smoking, encouraging immunisation, and avoiding personal smoking, especially in those with smoking parents or low childhood lung function, might minimise COPD risk. Clinicians and patients with asthma should be made aware of the potential long-term implications of non-optimal asthma control for lung function trajectory throughout life, and the role and benefit of optimal asthma control on improving lung function should be investigated in future intervention trials.

Funding

National Health and Medical Research Council of Australia; European Union's Horizon 2020; The University of Melbourne; Clifford Craig Medical Research Trust of Tasmania; The Victorian, Queensland & Tasmanian Asthma Foundations; The Royal Hobart Hospital; Helen MacPherson Smith Trust; and GlaxoSmithKline.

Introduction

Lifetime lung function is related to quality of life and longevity. Recent studies have highlighted that low lung function, especially low FEV1 in early adulthood, is associated with incidence of respiratory, cardiovascular, and metabolic abnormalities and all-cause mortality.1, 2 Over the lifespan, lung function progresses through phases of growth and decline3, 4 and individuals have unique lifetime lung function trajectories based on the timing and duration of these phases. Trajectories might have different risk factors and different consequences for chronic lung disease risk, particularly chronic obstructive pulmonary disease (COPD), which is expected to be the third largest cause of death globally by 2030.5 Insights into how lung function trajectories develop over the lifespan are important for lung disease prediction, prevention, and management.

Repeated lung function measurements from childhood into late adulthood are needed to identify lifetime lung function trajectories, but data are sparse. Among the few studies with longitudinal lung function measurements, only two6, 7 have published lung function trajectories based on more than two timepoints, with one study being restricted to participants with childhood asthma.7 However, neither study was able to capture the decline phase, vital for lung health outcomes in later life, because they had data only to the fourth decade. Although evidence suggests that both maximally attained FEV1 and its decline are associated with the COPD development,8 the relationship between the full expression of lung function—encompassing an individual's trajectory capturing both growth and decline—and COPD has never been reported. Furthermore, understanding of how childhood and adult factors interact to determine membership of healthy as well as adverse lifetime trajectories is crucial.

Research in context

Evidence before this study

Although accelerated lung function decline has been shown to be associated with development of chronic obstructive pulmonary disease (COPD), the relationship between the full expression of lung function—encompassing an individual's lifetime trajectory capturing both growth and decline—and COPD has never been reported. We searched for articles in PubMed from inception up to July 28, 2017 using the search terms “lung function”, “growth”, “decline”, “pattern*”, and “trajector*”. From 266 identified papers, only two classified lung function trajectories on the basis of more than two repeated lung function measurements. However, one was restricted to participants with childhood asthma and both were unable to capture the lung function decline phase due to duration of the follow-up (maximum age at last measurement being 32 years). Additionally, understanding how child and adult factors interact to determine membership of healthy as well as adverse lifetime trajectories is crucial to inform lifetime preventive strategies and promote lung health. Such evidence does not currently exist.

Added value of this study

Our findings of associations between distinct lung function trajectories and risk of COPD provide novel insights into the role of lifetime lung function trajectories in the course of COPD, and show the potential for interventions that promote healthy lung function and reduce COPD risk. Our study is the first to model lung function trajectories from childhood to the sixth decade of life in a general population. We identified six distinct trajectories. Three trajectories exhibiting lower lung function in childhood with subsequent normal or accelerated decline had increased risk of COPD and collectively accounted for most COPD cases. Most importantly, moderate–severe COPD cases arose only from these three trajectories. Early life factors including allergic diseases, lung infections, parental asthma, and maternal smoking predicted three unfavourable lung function trajectories. Personal smoking amplified the effect of maternal smoking on belonging to the worst lung function trajectory.

Implications of all the available evidence

Reduction of maternal smoke exposure and personal smoking and encouragement of immunisation are identified as public health targets to prevent adverse lung function trajectories and reduce future COPD burden. Clinicians and patients with asthma should be made aware of the potential long-term implications of non-optimal asthma control throughout life, and this should be investigated in future intervention trials.

Despite inter-individual variation in lung function trajectories within a general population, distinct subpopulations might follow similar FEV1 trajectories. The Tasmanian Longitudinal Health Study (TAHS),9 a population-based cohort study with multiple assessments of lung function, provided an opportunity to investigate lung function trajectories into the sixth decade. We aimed to (1) characterise population FEV1 trajectories from the first to the sixth decade, (2) investigate their childhood predictors and interaction with adult asthma and personal smoking, and (3) relate trajectories to COPD risk.

Section snippets

Study design and data collection

Six waves of TAHS9 were used. TAHS began in 1968 when 8583 Tasmanian children born in 1961 and attending school in Tasmania were enrolled. At age 7 (baseline), the children underwent a clinical examination including pre-bronchodilator spirometry and their parents completed a questionnaire. Follow-up assessments were done at 13, 18, 45, and 50 years of age with pre-bronchodilator spirometry measured. In the most recent follow-up in 2015, when the participants were 53 years old, all those from

Results

Of 8583 participants from the original cohort, pre-bronchodilator and post-bronchodilator spirometry were done at 53 years in 2689 participants. 2438 participants had at least two waves of lung function data at age 7 years and 53 years and comprised the study population (appendix). The participants included had similar baseline characteristics to those not included except for more female participants, more participants with eczema, and fewer smoking parents among those included (appendix).

Discussion

Our study is unique in that, to our knowledge, it is the first to characterise lung function trajectories in a large general population sample from early childhood to the sixth decade. This analysis is of considerable importance because understanding lifetime lung function is crucial for population-based interventions to promote healthy trajectories and prevent unhealthy ones. We identified six distinct FEV1 trajectories including two novel trajectories: early below average, accelerated decline

References (39)

  • MC Matheson et al.

    Cohort profile: the Tasmanian Longitudinal Health Study (TAHS)

    Int J Epidemiol

    (2017)
  • MR Miller et al.

    Standardisation of spirometry

    Eur Respir J

    (2005)
  • PH Quanjer et al.

    Multi-ethnic reference values for spirometry for the 3–95-yr age range: the global lung function 2012 equations

    Eur Respir J

    (2012)
  • GL Hall et al.

    The Global Lung Initiative 2012 reference values reflect contemporary Australasian spirometry

    Respirology

    (2012)
  • DS Nagin

    Analyzing developmental trajectories: a semiparametric, group-based approach

    Psychol Methods

    (1999)
  • A Heather et al.

    Latent class growth modelling: a tutorial

    Tutor Quant Methods Psychol

    (2009)
  • BL Jones et al.

    A note on a Stata plugin for estimating group-based trajectory models

    Sociol Methods Res

    (2013)
  • EJ Williamson et al.

    Introduction to causal diagrams for confounder selection

    Respirology

    (2014)
  • D Firth

    Bias reduction of maximum likelihood estimates

    Biometrika

    (1993)
  • Cited by (368)

    View all citing articles on Scopus

    Contributed equally; joint senior authors

    View full text