Major contributions by and the future scope of cohort studies to advance respiratory and sleep medicine

Abstract

Prospective cohort studies are valuable in establishing the aetiology and prognosis of diseases. These studies can provide the best evidence on potential interventions that can either be trialled in randomized controlled trials or inform prevention strategies when it is not feasible or ethical to conduct trials. Alternatively, retrospective cohort studies can be conducted either when the cohort has already been recruited or is identified retrospectively. This approach has been particularly helpful in occupational health, which can draw upon workplace records for data. This commentary briefly outlines the contributions made by prospective cohort studies to the fields of respiratory and sleep medicine. While there are a number of population-based and clinical cohort studies in respiratory health, few span early childhood to adulthood, with three (Tasmanian Longitudinal Health Study (TAHS), Melbourne Epidemiology Study of Childhood Asthma (MESCA) and National Child Development Study (NCDS)) reaching well into the sixth decade of life. Together, these studies have changed multiple paradigms in asthma, chronic obstructive pulmonary disease (COPD) and lung function trajectories. They have shown that there are multiple early-life longitudinal wheeze phenotypes, with some affecting lung growth, and that lung function deficits in COPD are partly established in early life. Not only is the aetiology of COPD related to multiple risk factors, but also early life and adult risk factors interact with each other to increase the risk of COPD. While a majority of childhood asthma resolves by adulthood, asthma that persists from childhood is the major contributor to severe asthma burden in adults. Severity of asthma, female sex, parental history of asthma and atopic disorders (particularly eczema and allergic rhinitis) are risk factors for the persistence of childhood asthma into adult life. As studies that started in childhood have matured in the past 5 years, establishing how lifetime lung function trajectories deviate from the normal trajectory has received increasing interest. Six population-based forced expiratory volume in 1 s (FEV1) trajectories from the TAHS have now been described (Fig. 1), and 75% of COPD cases that develop by age 53 years arises from three trajectories that demonstrated lung function deficits from childhood. It has also been shown that half of those with COPD at 62 years of age had normal FEV1 before 40 years of age and a rapid decline in FEV1 thereafter, while the other half had low FEV1 in early adulthood and a subsequently normal decline. The discrepancy between these two studies in the contribution made by early-life lung function deficits to the overall COPD burden (75% vs 50%) may reflect the relative increase made by accelerated lung function decline to COPD developed by 62 years of age. Earlylife risk factors that increase the risk of COPD, as well as low lung function trajectories, include early life asthma and bronchitis, and early life exposures to infection and smoking. There is increasing evidence that early-life risk factors interact with adult asthma, smoking and occupational exposures to increase the risk of COPD in a multiplicative fashion. Given the current evidence on the interactions between lifetime risk factors, these risk profiles could be used to identify those who are at risk of developing COPD. These risk profiles could also be used to diagnose COPD in combination with symptoms, lung function measurements and smoking history, approaches that are commonly used in clinical practice. To use such complex information practically in busy clinical settings, the development of risk prediction tools would be essential. Longitudinal cohort studies in sleep medicine have also played a key role in establishing the high prevalence of disorders, key aetiological risk factors and associated co-morbidity risks. The Wisconsin Sleep cohort was established over 30 years ago and performed gold-standard laboratory polysomnography at 4-year intervals in 1500 state employees. It established the high prevalence of sleep-disordered breathing in the middle-aged population—affecting 24% of men and 9% of women. It has also confirmed the aetiological role played by weight gain and obesity in the development of obstructive sleep apnoea (OSA). In addition to causing a symptom burden, including fatigue and excessive daytime sleepiness, OSA has been strongly associated with depression, hypertension, cardiovascular disease, cerebrovascular disease and increased mortality. The Wisconsin Sleep cohort was the first to demonstrate that the incidence of depression and hypertension increased over long-term follow-up in subjects with OSA. These data formed the basis for subsequent intervention trials showing that continuous positive airway pressure (CPAP) could lead to improvements in blood pressure. The association between OSA, cardioand cerebrovascular disease and mortality was further strengthened by data from the Sleep Heart Health Study (SHHS), a sub-study of the well-known Framingham cohort. The SHHS has shown that severe OSA (where the apnoea–hypopnoea Index (AHI) is ≥30 events/h) is associated with an increased risk of incident stroke and all-cause mortality, adjusted for known confounders. Although the association between OSA and mortality in the SHHS was only significant for males <70 years of