Correspondence to: Paul W. Jones, MD, PhD, Respiratory Medicine, Cardiac and Vascular Science, St. George’s, University of London, London SW17 0RE, UK; e-mail: firstname.lastname@example.org
The article by Polley et al1 in this issue of CHEST (see page 295) illustrates the use of symptom questionnaires to quantify cough and gain a better understanding about its relationship to the underlying diseases. Its findings are predicated on the validity of the instruments that were used.
Laboratory measurements and biomarkers of disease have turned medicine from a profession that was solely skilled based to one that is founded on solid science. However, the ultimate outcome of treatment is to improve clinical outcomes. These are things that are identified by the patients themselves: symptoms, well-being, and ability to perform daily tasks and function as healthy people do.2 Laboratory measurements of disease severity are only markers of outcome. Indeed many may be surrogate markers (ie, markers of other markers), for example diffusion capacity of the lung for carbon monoxide as a marker for CT lung densitometry as a marker of emphysema.
An observer cannot estimate symptoms. They are subjective and experienced only by the patient. The term subjective has a pejorative connotation in medicine. It suggests that the way the patient feels and thinks will influence his or her estimation. Most damningly, it implies unreliability, lack of rigor, not verifiable. So, the patient’s estimate of his or her symptoms is not reliable. If he or she cannot be trusted to make the measurement, no one can. Therefore, symptoms are therefore not scientifically quantifiable.
Let us step back from this nihilism and think through the issues. It is true that mood state and personality can influence how we judge things. A healthy adult, having been awake all night with a fretful child, judges the world differently in the morning than after a good night’s sleep. Previous experience of a symptom may influence current estimates.3 But, despite all of those factors, in clinical practice we all ask our patients to quantify symptoms and make management decisions based on those apparently unscientific and unreliable assessments.
The science of converting people’s answers to questions into valid numbers developed in education, and moved into medicine about 40 years ago. However, the first clinical quantification scale in pulmonary medicine is older than that. The British Medical Research Council Dyspnoea Scale was developed half a century ago by experienced clinicians, without any modern psychometric tools. It only got two things wrong: first it is not actually a dyspnea scale because it measures disability due to breathlessness, and second it asks about two different activity imitations in one question. These things aside, it has stood the test of time although it was only properly tested some 40 years after it was developed.4
The science of developing reliable and responsive questionnaires is now sophisticated and rigorous and continues to develop. Most recently, mathematical modeling techniques have been applied to questionnaire development and testing to ensure that the instrument has reliable measurement properties and behaves like a ruler: the same distance between two points means the same difference in severity, regardless of where they lie along the scale. For a more extensive explanation of this important methodology, see the online supplement to Meguro et al.5 This method of questionnaire development to some extent supersedes those used in the development of those questionnaires but does not render them invalid.
The term validity is often associated with a range of jargon terms in this context to describe different aspects of testing whether an instrument truly and reliably measures what it purports to do. Many of the methods were developed because the construct being measured was abstract (eg, “intelligence” or “mood”). There were no external references against which to compare the new instrument. The process of validating instruments is continuous, as illustrated in this article, which provides important new validation data about two existing cough questionnaires but also tells us more about cough and its interrelationships.
The two cough questionnaires correlated well with each other (convergent validity). It is also reassuring to see that the severity of cough in coughers was the same across disease groups, but the health of the patients measured with a generic instrument was different (divergent validity). It was equally reassuring to see that cough did not correlate with FEV1; why should it? Having got these validation steps out of the way, it is then interesting to see that some components of cough are more important in some conditions than others, and that the influence of cough on overall quality of life may differ between diseases. However the numbers of patients are small, and these observations deserve further testing.
Symptom measurement by well-designed questionnaires has taught us a lot and can change perceptions. Health status measurement has shown us just how important are COPD exacerbations6–7 True, no questionnaire is perfect and most can be improved,5 but let us not forget that FEV1 is affected by at least two entirely separate mechanisms, but that has not hindered its adoption as a measurement of immense value.
Dr. Jones is Professor of Respiratory Medicine, Cardiac and Vascular Science, St. George’s, University of London, London SW17 0RE, UK.
Become a CHEST member and receive a FREE subscription as a benefit of membership.
Individuals can purchase this article on ScienceDirect.
Individuals can purchase a subscription to the journal.
Individuals can purchase a subscription to the journal or buy individual articles.
Learn more about membership or Purchase a Full Subscription.
Institutional access is now available through ScienceDirect and can be purchased at myelsevier.com.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited:
Customize your page view by dragging & repositioning the boxes below.
Enter your username and email address. We'll send you a reminder to the email address on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.