We collected information on respiratory health and risk factors across multiple generations of the same family and showed that health and lifestyle (eg. smoking and being overweight) of parents (and perhaps even grandparents) may influence a child’s respiratory health. Mothers who smoke during pregnancy are more likely to have children with asthma and we provided further evidence that fathers who smoke (particularly during early adolescence) are more likely to have children with asthma (even if they stopped smoking many years before conception). We developed appropriate complex statistical methods to do these analyses.
We showed that children with asthma, persistent wheezing through childhood, low levels of physical activity, obesity and early puberty may not achieve their maximal potential lung growth. Three lung function trajectories during the first five decades of life (“early below average, accelerated decline”, “persistently low” and “below average”) are strongly associated with COPD. Early-life predictors of these trajectories include childhood asthma, bronchitis, pneumonia, hayfever, eczema, parental asthma, and maternal smoking. Children with frequent asthma plus allergies, especially if they smoke as adults, seem to be particularly at risk of COPD.
We showed that in adult life, obesity, low physical activity, poor diet (particularity diets low in anti-oxidant containing foods such as fresh fruit and vegetables), and occupational exposure to some agents contribute to lung function decline and COPD. Chronic asthma or increased airway responsiveness may lead to accelerated loss of lung function and a higher risk of COPD. Poor sleep and sleep apnoea are associated with respiratory symptoms but not related with accelerated lung function decline
Where possible, we used sophisticated analytical approaches (for example Mendelian Randomisation) to provide strong evidence that associations between some of these risk factors are causal and not an artefact of other associations within the data. These approaches showed that early puberty is detrimental to lung function growth and development of asthma. Oestrogens and androgens may influence lung function but there were some inconsistencies.
We published a systematic review of studies that examined peripheral blood DNA methylation and associations with lung function/COPD – no study included large populations followed over a long time. Using the ALEC cohort biobanks we used state of the art tests to confirm that low lung function is associated with a measurable change in DNA-methylation – most of these changes are clearly linked with smoking. We also showed that the effects of the SERPINA 1 gene on lung function are unlikely to be mediated through DNA methylation.
Although smoking in the European population is decreasing, an analysis of the ALEC cohorts provides strong evidence that decreases are not even across Europe. Adolescent females in particular seem to show little change in smoking behaviours. Our health impact analysis suggested considerable decreases in COPD prevalence in Europe if tobacco prices were increased.
Using anonymised data from the ALEC cohorts we developed Bayesian methods to impute values where data is missing so that the information can be used most effectively. We developed a risk prediction tool for COPD - our earlier published systematic review showed there was no adequate or widely applicable instrument currently available that takes account of all relevant factors. We found that the best prediction model we could build was not highly discriminatory for lifetime risk of COPD but can be used by the public to assess how changing their lifestyles (particularly smoking) could alter their personal disease risk. We used ’Value of Information’ analysis to identify economically justifiable research priorities for the use of this risk tool as a motivation for smoking cessation. It , suggested resources should be prioritised towards effectiveness trials. The risk prediction tool was demonstrated to patients’ organisations and received positive feedback regarding its potential use in their work.
Our work has been published and/or presented at scientific meetings (particularly at the European Respiratory Society International Congress, the major European conference for respiratory physicians and lung researchers)
