Quality of life measurement in asthma.

In chronic disease, such as asthma, cure is not possible and death, though of great importance, is rare. The majority of adult asthmatics lead lives that are characterized neither by good health nor by severe impairment. To assess disease severity, monitor progression, and quantify treatment effects, this impairment of health must be measured. How is this to be done? Asthma is a disease that is characterized, in part, by variations in airways obstruction. Measurements of airways function, such as peak flow and forced expiratory volume in one second (FEV1) were the first parameters to be used for assessing disease severity. These have been invaluable tools with which to show that treatment for asthma improved airways function. It has proved a little more difficult to translate such measurements into an understanding of whether the improvement was clinically worthwhile, however. Furthermore, measurements of airways function do not fully reflect all the disease processes that occur in asthma, such as overinflation and airways plugging. Evidence from clinical trials suggests that airflow measurements may not always follow other markers of disease activity. For example, in very mild asthmatics treated with budesonide over one year, asthma symptoms and bronchodilator use improved in steroid treated patients, but not in those receiving placebo. These improvements occurred in the absence of a change in FEV1 [1]. In another study, that compared budesonide and terbutaline, at the end of a one year treatment period, the initial improvement in peak flow on starting steroids had been sustained, but asthma symptoms were returning towards pretreatment levels [2]. Clearly, in mild-moderate asthmatics, who constitute the bulk of patients with this disease, standard physiological measurements do not tell the whole story. What additional measures are required to supplement airflow measurements? Two have already been mentioned, bronchodilator use and asthma symptoms. One frequently cited treatment objective is the avoidance of acute exacerbations, but these are difficult to define. Other measures may include the use of health resources, such as visits to the physician or hospital admissions, but these may depend on a range of factors, including local medical practice. In most circumstances, these different variables are used as surrogate markers for the patient's state of health. There is little need for the use of such surrogates, however. Health can be measured directly using specifically designed questionnaires. These quantify the effects of disease on the patient's daily life and well-being in a formal and standardized manner. Unfortunately, the term "quality of life" has become attached to these instruments, which has led to certain misunderstandings about them. This has been compounded by the widespread use of this phrase, by politicians and journalists, to convey sweeping concepts about life without the inconvenience of providing specific details that may be open to scrutiny. The science of measuring health is very rigorous, but unfortunately the jargon term with all its nebulous associations has stuck very firmly. A critical part of the validation of a health questionnaire is the demonstration that its scores are related to disease severity. Despite the availability of such evidence, one criticism of these measures is that it may be possible for patients with asthma "to die with a good quality of life". This is unlikely. An examination of disease-specific quality of life questionnaires reveals that they look rather like clinical check lists albeit far more comprehensive than the histories that most of us take in routine practice. It might be reasonable, therefore, to rephrase the anonymous quote and express a concern about patients "who die despite appearing to have wellcontrolled asthma as assessed by a comprehensive clinical history". Whilst such patients do appear to exist (although they have rarely been assessed using an comprehensive disease-specific questionnaire), their existence does not constitute a valid argument against either clinical history-taking or health measurement. The concerns just discussed may arise partly through a failure to distinguish between the processes, requirements and purposes of a clinical trial involving populations of patients and those required in the treatment of an individual patient. At present, health-related quality of life questionnaires are used largely in clinical trials to quantify average changes in health. There are few routine therapies for asthma for which there is clear evidence, at a population level, of a dissociation between disease in the airways and the patients' symptoms and health. Indeed, only by carrying out studies which measure both airways function and quality of life will it be possible to test whether this does occur. The possible existence of patients who have bad asthma, yet apparently normal health and no disturbance to daily life, constitutes an argument for the development of tools with which to test this hypothesis, not an argument against them. The process of validating any new method of measurement in medicine is long and multifaceted. How many of us can detail the history of the development and EDITORIAL