Measuring disease (2nd edition) A review of disease-specific quality of life measurement scales Ann Bowling Open University Press 395 pages£25 paperback£65 hardback This is a welcome second edition of the 1995 original and supplements Measuring Health (2nd edition, Open University Press, 1997) by the same author.

The author says her aim is '- to introduce the key literature on the psychometric properties of measures of disease-specific quality of life, including the symptoms scales often used alongside them'.

Methodological issues are covered in sufficient depth to give the less experienced reader an understanding of the psychometric properties of scales. These include validity, sensitivity and reliability.

Validity considerations include the extent to which the scale measures the underlying concept of interest and has relevance within the particular social and clinical setting of its use. Scale sensitivity is discussed, emphasising that its importance will vary depending on the purpose for which the scale is being used.

Factor structure and the differences in scaling methods are explained. This is important as the unwary might make inappropriate inferences by not appreciating, for example, the somewhat arbitrary nature of ordinal scaling.

Guidance follows about when to use generic, domainspecific or disease-specific quality-of-life assessment scales in particular circumstances. The reader is also referred to fuller discussions of all these issues.

The remaining chapters are organised around particular conditions - cancers, psychiatric conditions, and psychological morbidity, respiratory, neurological, rheumatological and cardiovascular conditions, and a final chapter with a collection of other diseases or conditions.

Each starts by considering important domains to measure in that condition. For cancers, for example, this might include known side-effects of particular therapies or changed body image, whereas in neurological conditions, aspects such as gait speed and cognitive impairment may be needed.

Some 50 specific scales are discussed.

Clinical teams looking for an outcome measure for routine service use would find enough information to make a preliminary choice. So, too, would anyone setting out to evaluate new service developments.

This book will encourage less experienced researchers to consider key questions before rushing to use a well-known and validated scale that might not suit them. The more experienced will find the extensive referencing helpful.

Dr Finbarr Martin Consultant physician and senior lecturer, Guy's and St Thomas' trust.