“Train up a child in the way he should go: and when he is old, he will not depart from it.” Proverbs 22:6

I have three children, two of whom are adults, and the third of whom is now taller than me. These children, despite sharing similar genes, are all very different in their personalities, demeanours, likes and dislikes. I am convinced that almost all of the traits they now have as adults, or near adults, are pretty much what they emerged from the womb with. Other parents I have discussed this observation with have similar stories—it may be that most of what affects us in adulthood has been established before we start grade school.

Almost all of the epidemiological studies of chronic obstructive pulmonary disease (COPD) have focused on risk factors in adulthood such as cigarette smoking, occupational exposures, exposures to biomass smoke and respiratory infections.1 2 3 4 COPD is clearly a disease that manifests itself in adulthood with a peak onset of incidence in the fifth to sixth decades of life.5 The paper in the current issue of Thorax by Svanes et al6 explores a different possibility—that COPD is a disease of childhood that becomes manifest in adults (see page 14). If this observation is correct, it may require us to reconsider some of our long-held assumptions on COPD.

According to the analysis by Svanes et al,6 combining paediatric risk factors (such as a personal history of asthma, a parental history of asthma or early respiratory infections) with early exposure to tobacco smoke resulted in a risk that rivalled or exceeded that seen from the traditional COPD risk factor of cigarette smoking. The strongest risk factor was a diagnosis of “asthma” as a child, which was associated with a 10-fold higher risk of stage 2 COPD in men and a 4-fold higher risk of stage 2 COPD in women.

The possibility that COPD starts in childhood is certainly provocative and has been the focus of several recent reviews or editorials.7 8 The argument that this is “hype” looks at 50 years worth of epidemiological and clinical work demonstrating that cigarette smoking is—at least in the developed world—the primary risk factor for COPD and that, in the absence of smoking, this disease would be greatly diminished. The supporters of this viewpoint might also say that focusing on anything other than tobacco smoking is a poor use of resources for a disease that is largely preventable. There has been sentiment in some circles to actually include cigarette smoking in the definition of COPD, saying that disease occurring in never smokers is, by definition, something other than COPD.9

The alternative argument that childhood origins of COPD provide “hope for better understanding” is based on more recent data pointing to a substantial burden of COPD among never smokers and evidence based on cohorts of people followed since childhood. While cigarette smoking remains an undeniable risk factor, it is clear that not all smokers have the same risk of developing COPD and its complications.

Several possibilities may explain why we are now seeing evidence supporting the childhood origins of COPD. The first may be somewhat definitional. The increased willingness to accept impaired lung function as “confirmatory” of COPD is a recent change. While the precise definition may be controversial and well beyond the scope of this editorial, the idea that the presence of “abnormal lung function” is sufficient to diagnose COPD dramatically altered our epidemiological impression of disease burden, including the concept of “undiagnosed disease”.10 11 12 Another possibility is the increasing recognition that both asthma13 and COPD14 have multiple phenotypes. Some asthma phenotypes in children do look like COPD in that they seem to be related to “irritant” exposure and show some evidence of impaired lung function.15 The improved treatment and survival of premature infants and their resulting chronic lung disease of prematurity is thought to be a potential risk factor for chronic pulmonary disease in adulthood.16 While this is an extreme example, it seems entirely feasible that less severe respiratory insults, especially when combined together, might result in respiratory diseases, including COPD, in adulthood.

So does early “training” of the respiratory system, paralleling the words of the proverb above, result in a tendency towards respiratory health or respiratory disease in adulthood? This certainly makes sense, and some recent data support the finding that measures of lung function at a mean age of 2 months track into young adulthood.17 Does this change our approach towards known risk factors such as cigarette smoking or occupational dust exposures? Of course not. While I believe adults entering adulthood with poor lung function are probably at an increased risk for developing COPD, there is no reason to believe that young adults with normal or even supernormal lung function are protected from COPD development. My hope is that a better understanding of the early life factors that increase the risk of disease might lead to better early detection and intervention efforts.