Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Measuring quality of prescribing for asthma


Which factors can be used as an indication of the overall quality of prescribing is one of those seemingly simple questions for which there are only complicated answers. For a start prescribed drugs are seldom just used for a single complaint. If they are used for many, that confuses things mightily. Then there's the population covered. GP practices in particular are subject to enormous differences in demography, wealth or deprivation, ethnicity, and the burden of disease. So when we see a paper that indicates that certain factors may relate to prescribing quality, even if they are preliminary [1], it deserves examination.


The setting for the study was north Staffordshire. Two single-handed practices were chosen, one with a low ratio of corticosteroid to bronchodilator use (C:B ratio 0.24), and the other with a high ratio (C:B ratio 1.53). The practices were located within the same deprived area. These ratios had been calculated from an official database of all prescriptions issued by the practices between December 1993 and February 1994.

Patients from the practices who were prescribed drugs for asthma between March 1994 and August 1995 were found from a search of the practice databases. The patients were sent a validated questionnaire about the severity of their asthma symptoms during the previous month. They were also asked about smoking and occupation.

Patient notes were also examined to discover co-morbidity, and the strength of the diagnosis of asthma. Age and sex were also extracted, and a residential measure of deprivation calculated from the 1991 census for districts where patients lived.


There were 366 patients prescribed asthma drugs in the two practices, about 7% of the list size in each practice. There was no difference between the age of patients, but there were differences in smoking, social class, deprivation score and diagnosis of comorbid conditions. About 80% of patients returned the questionnaire.

The symptom severity score was significantly higher in the low corticosteroid to bronchodilator practice (Figure 1). The difference remained after adjustments for age, sex, diagnosis, smoking and deprivation status.

Figure 1: A higher ratio of corticosteroid to bronchodilator prescribing was associated with less severe asthma symptoms


This paper is a really interesting read. It emphasises that these are preliminary results only, and from only two practices. But there are some interesting methods being used here, and the discussion puts them into context. Congratulations are in order for the sponsorship from the local hospital in Stoke on Trent.

Measuring the quality of prescribing has been a frequently-asked question of Bandolier . So far there seems little good information, but perhaps we are not looking in the right place. We'd love to know if someone has some answers.


  1. M Shelley et al. Is the quality of asthma prescribing, as measured by the general practice ratio of corticosteroid to bronchodilator, associated with asthma morbidity? Journal of Clinical Epidemiology 2000 53: 1217-1221.
previous or next story in this issue