Recommendations from quality of life scales are not simple

EDITOR—As someone who is guilty of adding to the large number of disease specific quality of life (QOL) scales, let me add a note of caution to the pleas made by Garratt et al for guidance and recommendations for the users of these scales—however understandable that plea is.1 QOL scales are not like thermometers or spirometers, where the reading is independent of the type of patient.

A QOL scale is just a shopping bag of experiences (or questions) that are put together to form a scale, rather like the retail price index. The retail price index is a shopping bag of goods for an “average” shopper, even though most people are not that average shopper. The scale value obtained from a QOL scale depends on the overlap between the items in a scale and the patient's own experience of disease. So, for example, if there is a generic QOL scale and there are many pain items but no items on sleep disturbance, then arthritis will come out worse than asthma. The same logic applies to disease specific scales. If there are questions about sport, and most patients are elderly, then there is little complaint about how health has affected their sporting life.

It is the nature of QOL assessment that the “best” scale is always best for a particular purpose, where purpose is defined in terms of disease, population, and treatment. An example will illustrate this. We developed a scale to measure QOL in patients with chronic obstructive pulmonary disease (COPD), which was subsequently used to assess improvement after rehabilitation.2 Some of the items measuring COPD severity did not change—they were indicators of severity that were encouraged by rehabilitation (for example, using pillows at night; eating regular, smaller meals). Subsequently, we developed a scale for evaluating rehabilitation by selecting items that showed change after rehabilitation.3 In other words, we selected the items from our big trolley to make a smaller basket that measured what we wanted. QOL scale selection is a matter of selecting baskets of items that match a particular purpose.

A final warning: QOL researchers always recommend their own scales. It is a personal thing—to the author, the scale is a bit like his or her child. The more “successful” scales tend to be those whose authors are better promoters. The last person you should come to for a recommendation about QOL scales is someone who is an author—including myself.