Prescribing and hospital admissions for asthma in east London

BMJ 1996; 312 doi: (Published 24 February 1996) Cite this as: BMJ 1996;312:481
  1. Chris Griffiths, general practitioner, senior lecturera,
  2. Jeannette Naish, general practitioner, senior lecturera,
  3. Patricia Sturdy, research officera,
  4. Filomena Pereira, lecturer in medical statisticsb
  1. a City and East London General Practice Database Project, Department of General Practice and Primary Care, St Bartholomew's and the Royal London School of Medicine and Dentistry, Queen Mary and Westfield College, London E1 4NS
  2. b Department of Epidemiology and Medical Statistics, Queen Mary and Westfield College
  1. Correspondence to: Dr Griffiths.
  • Accepted 23 November 1995

Admission rates for asthma in 1991-2 were 80-100% above national averages for all age groups in east London.1 We have shown in east London general practices that a higher ratio of prophylaxis to bronchodilator prescribing occurs in training practices and those approved for health promotion band 3 and asthma surveillance.2 We have also explored the relation of appropriate prescribing for asthma to other local practice characteristics.3 We investigated excessive asthma admission rates in patients from these practices by studying the relation between asthma prescribing and admissions.

Methods and results

Data on asthma admissions by age for east London residents in 134 out of 163 practices covered two years from April 1992 and included some 1602 patients (800 in 1992-3 and 802 in 1993-4). Ninety eight per cent of admissions for asthma were acute and only 3% of all patients admitted were not allocated to a practice. Data were obtained from the integrated district and regional information system with the international classification of disease code 493. Rates per thousand patients per practice were calculated from the average number of patients admitted per year; this excluded readmissions within the same year. The denominators were the resident population of east London in each practice at June 1993 and June 1994. We also investigated admission rates in the age groups under 5, 5-64, and 65 and over.

Our asthma prescribing data have been described elsewhere.2 While only one year's prescribing data were available (April 1992 to March 1993), our experience from a parallel study in 24 local practices is that the prescribing ratios remained almost constant during our two year study period.

Table 1 presents the mean (SD) asthma admission rates for different categories of the prophylaxis to bronchodilator ratios. These ratios have been divided into four groups by the 25th and 75th percentiles and the median. Statistical significance was determined by Cuszick's test for trend. Table 1 shows that practices prescribing higher ratios of prophylaxis to bronchodilator medication had on average lower admission rates for asthma. Significant trends were observed in those aged 5-64. For young children and the elderly, however, we found neither significant trends nor differences in admission rates between categories of the prescribing ratios (prescribing items ratios: for under 5s P=0.61; for the elderly P=0.17; prescribing cost ratios: for under 5s P=0.38; for the elderly P=0.62).

Table 1

Asthma admission rates (per 1000 patients per practice) by age group and the ratio of prophylaxis to bronchodilator prescribing in east London practices

View this table:


We have shown an association between asthma prescribing and morbidity experienced by patients with asthma, as reflected in admissions to hospital. Practices with higher prescribing ratios had lower admission rates to hospital. This relation has been reported by other workers.4 The diagnostic coding for asthma is probably most secure in the middle age band: the lack of association in those aged under 5 and in the elderly may reflect the presence of other conditions such as chronic cough and chronic obstructive pulmonary disease, possibly classified under the asthma code.

Prescribing studies are often crude and may be complicated by problems such as indication and compliance5; ours is no exception. Nevertheless, that we can relate this marker to a patient outcome such as admission rate is encouraging. While this relation is not necessarily causal (higher prescribing ratios may simply be a marker for other aspects of good asthma care), it seems sensible to promote more appropriate prescribing for asthma, particularly in practices with low prophylaxis to bronchodilator ratios. Methods might include facilitating local asthma guidelines, practice based medical education, and financial incentives. Further studies should address other patient and practice factors that may be associated with asthma admission rates and investigate the relation between prescribing patterns and a wider range of patient outcomes.

We thank Jacqui Bobby for providing us with asthma admissions data and the Prescription Pricing Authority for its information, which was obtained by JN when she was medical adviser to the family health services authority. We also thank Kath Moser and Mike Chambers for their support with the general practice database.


  • Funding Wellcome Trust.

  • Conflict of interest None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract