• Clinical epidemiology

Hanging on the wall in my office are two obituaries, one of Dr Charles Fletcher who died in 19951 and the other of Dr Benjamin Burrows who died in 2002.2 Their pictures look over the desk where I do a great deal of my work and provide inspiration and, through their collective body of work, guidance. I would like to think that the work that I do continues in a very small way the work that these giants in our field started. Dr Fletcher was responsible for, among other accomplishments, defining the natural history of chronic obstructive pulmonary disease in his landmark study of British men.3–5 Dr Burrows founded the Tucson Epidemiological Study of Airway Obstructive Disease (TESAOD)6 that has added greatly to our knowledge of respiratory disease. Drs Fletcher and Burrows trained and mentored many of our current leaders in respiratory health and coauthored several publications.7 8 An ongoing legacy of Dr Burrows is the TESAOD, which continues to provide valuable and important information relevant to today.

The publication by Guerra et al in this issue of Thorax (see page 499) provides an example of Dr Burrows' ongoing legacy and the convergence of his work with that of Dr Fletcher in trying to better understand the natural history of lung disease.9 This analysis of the TESAOD data examined the baseline lung function, changes in lung function over the next 25 years of the study and the effects of lung function on mortality by year 34 of the study. The traditional view of the natural history of lung disease dating back to the time of Dr Fletcher has focused on lung function decline—the idea that, with each passing year, people lose some of their forced expiratory volume in 1 s and those who lose this more quickly develop chronic obstructive pulmonary disease (COPD).3 10 11 This construct has provided the basis for the main paradigms by which we attempt to determine COPD progression and the effectiveness of treatment.10 12 13 Ironically, though, the concept of ‘lung function decline’ is not terribly useful clinically. We know that an individual's lung function and reversibility testing can vary dramatically from day to day and even from test to test on the same day.14 15 Over the years I have made almost no clinical decisions in my patients with COPD on the basis of ‘accelerated decline in lung function’. Rather, I have categorised patients' lung function, asked about their symptoms and exacerbations, and determined interventions based on the information gathered.

In the paper by Guerra et al, not one analysis attempts to determine lung function decline.9 Rather, the authors classify subjects as obstructed or ‘restricted’ at baseline and follow-up visits and use these more clinically relevant categories to predict outcomes. This approach raises several issues, especially with regard to the ‘restrictive spirometric’ pattern. From the time of Hutchinson, clinicians have known that a low lung capacity resulted in a shorter life—this represents the origin of the term ‘vital vapacity’.16 In recent years the links between restrictive spirometric patterns and diabetes,17 heart failure,18 lung cancer,19 asthma20 and obesity21 have been explored. In a large study from the Mayo Clinic22 there is even evidence that this pattern can represent a subtype of COPD in some cases, with the thought process being that air trapping resulting in a smaller forced vital capacity may be one of the initial manifestations of COPD. Of course, there are a host of other problems that can result in this pattern, ranging from muscle weakness to poor effort. Furthermore, the presence of a restrictive pattern, regardless of the aetiology, predicts worse outcomes such as deaths and hospitalisations.23 This was confirmed in the analysis by Guerra et al of the TESAOD cohort.9

Guerra et al, however, took their analysis one step further by looking at longitudinal changes in lung function.9 To accomplish this, rather than looking at lung function decline they examined what category the person was in at the time of the testing (obstructive or restrictive) and further classified them into one of six categories based on this pattern over time. The bottom line was that the presence of a restrictive pattern in just one of the evaluations predicted an increased risk in mortality that was similar in magnitude to that seen with ‘recurrent obstruction’ or COPD.9

The hallmark of the work of both Dr Fletcher and Dr Burrows was that they challenged the norms of their day and were willing to create new paradigms to advance medical knowledge and result in better interventions for our patients. Is it time for a new paradigm for chronic lower respiratory diseases (to include the various subtypes of asthma, COPD and restrictive lung diseases)? I believe it is! Moving away from Dr Fletcher's paradigm of ‘lung function decline’ to one that includes transitions between spirometric categories, radiographic data, cormorbid disease and, eventually, genomic, proteonomic and metabalomic information, is a change that I'm confident both Dr Fletcher and Dr Burrows would have happily embraced. Of course, this task will not be an easy one, but the tasks that Drs Fletcher and Burrows met head on and conquered in their day were, in many ways, more challenging than what we face today. Their biggest challenge was that they did not have the body of work of these giants in our field to rely on. As we prepare for the future, I encourage all researchers to study the work of these and other historic leaders in our field and, as they did, to keep an open mind, to challenge the status quo and to keep the ultimate goal of improving the lives of our patients central to our work.