Testing health status (“quality of life”) questionnaires for asthma and COPD

Abstract

Two papers in this issue of the Journal address the problem of testing versions of health status (or "quality of life") questionnaires after translation from their original English [1,2]. Tests of the validity of health questionnaires fall into four main areas: internal consistency; discriminative properties, i.e. the ability to distinguish different levels of disease between patients; evaluative properties, i.e. the ability to detect clinically meaningful changes in health; and predictive validity, i.e. the ability to predict future changes in health or healthcare needs. It is clear from this list that the process of validation is long and complex. Indeed, it is never-ending. Each new study in which a health questionnaire is used contributes to the body of knowledge concerning its performance, and thereby to its validity.A further term used in this general context is "reliability". This is a measure of the likelihood that an instrument will give the same reading when used to measure the same thing in a different setting. This is usually tested in terms of repeatability, which is dependent on the assumption that the patients' health will not have changed between assessments. The usual approach is to recruit stable patients and make the assessments within 1-2 weeks. Some workers confine their measurements to those patients who report no overall change in their symptoms or health, as assessed by a global measure of change. At first sight this seems reasonable, but it presupposes that the patients' recall is accurate and that their estimates of change are reliable. Other tests of reliability are less frequently reported, but a recent study has examined broader aspects of the reliability of a disease-specific questionnaire, and found that it behaved similarly in a face-to-face interview compared with subsequent computer-assisted telephone interviews [3].Tests of internal consistency are usually made at the time of questionnaire development, and most questionnaire developers use methods that either directly or indirectly examine issues of internal consistency. Such tests were developed to address the problem that arose when attempts were made to develop questionnaires designed to test abstract concepts, such as anxiety or depression, for which there were no adequate reference instruments at the time, apart from clinical judgement. More recently, greater attention has been paid to tests of discriminant and evaluative properties than to internal consistency. Most widely used questionnaires have quite good internal consistency, as again reported in the two papers in this issue of the Journal [1, 2]. There have been occasional reports of low consistency [4], but such occurrences appear to be rare.The simplest way to test the discriminative properties of a new questionnaire is to compare it with another established health status instrument in the same population of patients. Unfortunately, this does not prove that the new questionnaire is measuring health, merely that the two questionnaires appear to be addressing similar...

Independent validation of the Sleep Apnoea Quality of Life Index

“…We concur with Jones' commentary that "the purpose of translation is not to produce a literal conversion of the questionnaire but to convey the spirit of the questionnaire items into different languages and cultures". 29 There is more and more evidence, including this study, that careful translation and back translation of quality of life questionnaires can produce non-English language versions that appear to behave in a very similar manner to their originals.…”

Construct Validity of the Abbreviated Mental Test in Older Medical Inpatients

“…Among components of validity, the ability of a test to discriminate clinically meaningful changes in the explored outcome are usually explored in detail, whereas the assessment of internal consistency is commonly reserved to tests exploring abstract concepts such as anxiety [27]. From this perspective, the proven high concordance between the AMT and well-validated measures of cognitive impairment could per se validate the AMT [3][4][5].…”

Recommendations for epidemiological studies on COPD

“…JONES [69] previously showed that FEV1 correlated only weakly with quality of life, measured using the St George's Respiratory Questionnaire. Quantitative morphometric analyses post mortem and surgical lung specimens, as well as radiographic studies, have shown that FEV1 is insensitive to anatomic disease severity in COPD [70].…”