Birth Cohort Studies: Their Next Coming of Age

The fetus was traditionally considered to have a privileged place in utero, where it was protected from harmful environmental exposures. The paradigm that we are all born in a state of good health was challenged by research in the 1980s and 1990s that demonstrated links between reduced birth weight and increased risk for noncommunicable diseases (NCDs) in later life. The paradigm that antenatal exposures can have lifelong implications for health and well-being is now well accepted (1).

Inspired by the concept of “fetal origins of disease” and later concepts such as developmental plasticity and predictive adaptive responses, recruitment began in the 1980s to birth cohort studies designed to give insight into the premorbid physiological mechanisms linking the antenatal environment to later NCDs. To obtain an early return from their investment in time and effort, investigators explored NCD outcomes in childhood (not adulthood), and the focus of their cohorts was asthma, which was reaching an “epidemic” prevalence at that time.

The TCRS (Tucson Children’s Respiratory Study) was at the vanguard of these birth cohort studies and recruited 1,246 infants between 1980 and 1984. Infant lung function was measured during tidal breathing using recently described methodologies, and the outcomes included FRC, V̇max at FRC (V̇maxFRC), and the time to peak expiration:total expiration (tptef/te) ratio. Testing was completed in 376 infants, including 180 who were assessed before 6 months of age. Generous doses of chloral hydrate were used to induce sleep, and investigators doing the time-consuming testing might reasonably have wondered whether the measurements they were taking would add up to much.

They need not have worried. The first publication reported an association between reduced V̇maxFRC and wheeze among the 120 participants who were symptom free when tested at a mean age of 2 months (2). A further publication, which has been cited more than 4,000 times, found that the association between reduced V̇maxFRC and wheeze was only transient (3). Other cohorts of similar design found reduced V̇max FRC (4, 5) and tptef/te (6) and other abnormalities of pulmonary physiology (7) were associated with persistent wheeze or asthma, with reduced lung function persisting in some individuals from infancy through to later life (8, 9). In a paper published in this issue of the Journal, the TCRS team (pp. 1646–1655) confirm a persisting relationship between both reduced V̇maxFRC and tptef/te and asthma (10).

The latest paper from the TCRS team uses questionnaire data collected at 12 times between the ages of 6 and 32 years and applies longitudinal analyses, including a survival analysis, to the relationship between infant lung function and active asthma in later life (10). Individually, reduced V̇maxFRC and tptef/te were associated with increased risk for asthma, and new onset asthma continued to occur into the third decade. In combination, reduced V̇maxFRC and tptef/te before 8 weeks of age had a threefold increased risk for asthma by the age of 36 years. Childhood asthma has many associated risk factors with odds ratios between 1.5 and 2.0, and a ratio of 3.0 suggests that early-life lung function is an important trait relative to the many described genetic and environmental factors.

In a second tantalizing set of results in their paper, the TCRS team present a series of comparisons between infant lung function and high-resolution computed tomography in 38 individuals aged 26 years (10). Among this subset of individuals, reduced tptef/te was associated with reduced airway wall thickness and airway caliber. These proof-of-concept results might indicate that the structural airway changes associated with asthma are already in place by the first few months of life. Excitement for these findings needs to be tempered by the apparent inconsistencies with data from the TCRS and elsewhere. If tptef/te is truly related to structural lung changes, there should be an association with spirometry; V̇maxFRC, but not tptef/te, was associated with spirometry at 26 years of age. If structural changes are present in early life, they should be present on airway biopsies, but no such evidence was seen in a Finnish cohort of wheezy infants with reversible airway obstruction (11). Although replication of these findings is clearly needed, there is a challenge in tracking down the few thousand 26-year-olds on the planet for whom infant lung function data are available.

The take-home message from this article (10) is that processes that determine respiratory physiology and asthma risk into the fourth decade of life are present in early infancy and are likely active in utero. The usual limitations of birth cohort studies apply to TCRS and include recruitment bias, bias in follow-up (a relatively high proportion [36%] of participants had active asthma by 36 yr of age), and loss to follow-up. An additional limitation is that because the environment and lifestyles have changed since the early 1980s, these results may not be generalizable to individuals conceived in the early 2020s.

Where do we go from here? Can we screen for asthma with infant lung function testing? Unfortunately, we remain an impossibly long way from considering screening for asthma by infant lung function; testing lacks sensitivity and specificity, and we know that reduced infant lung function resolves in many individuals (12). Can we improve infant lung function when it is found to be reduced? Diet may help by preventing excessive weight gain in infancy, and vitamin C treatment during pregnancy may possibly soften the blow to infants’ lungs from exposure to maternal smoking (13). What we collectively can do is to carry on advocating a healthy lifestyle and “lived environment” for everyone, especially pregnant mothers.

Extended follow-up of cohorts has demonstrated that in many individuals, children do not grow out of asthma; symptoms merely go into a temporary remission. Childhood asthma is a risk factor for chronic obstructive pulmonary disease (14). Taking a step back from the early origins of asthma/chronic obstructive pulmonary disease, what is clear from birth cohorts recruited in the United Kingdom in the 1950s (where less antenatal data are available) is that there are early origins of many other common NCDs in the older population, including ischemic heart disease, chronic renal failure, and even dementia (15). The next generation of researchers is now slowly taking over the reins of birth cohorts, such as TCRS, from the colleagues who founded them. Over the coming years, as this new generation of researchers ventures forth, the birth cohorts recruited in the 1980s will enter a new age in which they will continue to yield insight into respiratory NCDs and also start shining light on pathways to some nonrespiratory NCDs.