For healthcare professionals only

Education And Debate

Development of a symptom based outcome measure for asthma

BMJ 1994; 309 doi: (Published 22 October 1994) Cite this as: BMJ 1994;309:1065
  1. N Steen,
  2. A Hutchinson,
  3. E McColl,
  4. M P Eccles,
  5. J Hewison,
  6. K A Meadows,
  7. S M Blades,
  8. P Fowler
  1. Centre for Health Services Research, University of Newcastle upon Tyne, Newcastle upon Tyne Department of Public Health Medicine, University of Hull, Hull University of Leeds, Leeds East Gloucestershire NHS Trust
  1. Correspondence to: Professor Hutchinson.
  • Accepted 25 July 1994

Measuring symptom specific health outcome is complex, but the methodologies now exist to develop measures with the appropriate properties. As one element of a major programme to develop multidomain health outcome measures for chronic disease, a symptom based measure for asthma care has been developed for use in general practice and outpatient departments. This article outlines the development process, which used a framework recently described in the theoretical literature to show the constraints that scientific criteria place on the development of outcome measures and the means of overcoming such limiting factors. Although substantial effort is required to undertake a rigorous process of development, useful tools are the result. Two five item, symptom based outcome measures for adult asthma are described.

Although health outcome measurement is now becoming a priority for the NHS, substantial scientific hurdles are yet to be overcome. In the move from the assessment of health status to the evaluation of health outcome, a recent series of papers in the BMJ emphasised that not all health status measures have the scientific attributes required of an outcomes measure (for instance, evidence of responsiveness to change) and not all outcome measures are health status measures in psychometric terms (for instance, peak flow reading in the management of asthma).1,2 Other authors have suggested that health outcome indicators could be used to compare populations that are subject to a range of influences, rather than just to evaluate the effect of a specific intervention.3

Fitzpatrick and colleagues identified important scientific issues in the measurement of health status and health outcome.1,2 Measures must have the properties of validity and reliability. Validity relates to the effect of systematic error; an instrument is valid to the extent that it measures what it purports to measure.4 Reliability relates to …

View Full Text

Log in

Log in through your institution


* For online subscription