Papers

Cost effectiveness analysis of inhaled anticholinergics for acute childhood and adolescent asthma

BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7223.1470 (Published 04 December 1999) Cite this as: BMJ 1999;319:1470
  1. Joanne Lord, lecturer in health economics (j.lord{at}sghms.ac.uk)a,
  2. Francine M Ducharme, associate professorb,
  3. Ronald J Stamp, honorary senior research fellowc,
  4. Peter Littlejohns, professora,
  5. Rachel Churchill, senior research fellowa
  1. Health Care Evaluation Unit, Department of Public Health Sciences, St George's Hospital Medical School, London SW17 0RE
  2. Departments of Pediatrics and of Epidemiology and Biostatistics, Montreal Children's Hospital, McGill University, 2300 Tupper Street, Montreal, Quebec, Canada H3H 1P3
  3. Wessex Institute for Health Research and Development, Biomedical Services Building (mailpoint 727), University of Southampton, Southampton SO16 7PX
  1. Correspondence to: J Lord
  • Accepted 24 August 1999

A recent systematic review by the Cochrane Airways Group showed that adding multiple doses of anticholinergics to β2 agonists is safe and effective in improving lung function and avoiding hospital admission for school aged children and adolescents attending casualty departments with severe acute asthma.1 The estimated reduction in the risk of admission was 9.4% (0.4% to 18.4%). This intervention presumably improves bronchodilatation until systemic corticosteroids take effect. Evidence of cost effectiveness, however, is lacking. To clarify whether scarce healthresources should be spent on this intervention we conducted an economic evaluation.

Methods and results

We used various assumptions to estimate the financial implications of treatment (see table on the BMJ's website). The costs of drug administration were not included, as anticholinergics are always given with β2 agonists and involvelittle additional manipulation. The cost of nebulisers, other drugs, and the casualty attendance were also excluded. No consideration was given to possible changes in length of stay in casualty. The effect of changing the various assumptions was tested by simple, oneway, sensitivity analysis, and by multivariate probabilistic sensitivity analysis.2 The latter is a simulation approach that enables estimation of uncertainty ranges containing 95% of replicated results.3

We estimated that treatment would cost about £8 (uncertainty range £1 to £47) per admission avoided (table). This implies a net saving of £80 (£0 to £157) per severe case treated. Varying the risk reduction within 95% confidence limits varied the mean net saving from £3 to £157 per severe case treated. More precision is expected when the Cochrane review is updated. Varying the cost of hospital admissions within the interquartile range forEnglish providers (£620 to £907) varied the mean savings from £58 to £85 per severe case treated. Changes to the dose and unit cost of ipratropium had very little effect on the results.

Results of baseline analysis, with uncertainty ranges estimated by probabilistic sensitivity analysis

View this table:

Further assumptions were used to extrapolate the findings to a national level. About 7200 children aged 5-15 years are admitted from casualty with a diagnosis of asthma each year (hospital episode statistics 1988 to 1996). About 40% of children in this age group attending casualty with asthma are admitted.4 We assumed that 50% of people with asthma attending casualty have severe asthma.5 The rate of uptake of the review recommendations was assumed to be 5% a year; therefore an additional 5% of eligible patients would be treated in the first year, 10% in the second year, and so on. If doctors treated patients with mild or moderate asthma, this would add to treatment costs with no evidence of clinical benefit. We assumed that for every five patients with severe asthma who were treated, one patient withmild or moderate asthma would be treated. Costs were discounted at an annual rate of 6%.

Net savings were estimated to be £437 800 (−£3700 to £1 078 100) over five yearsin England. An increase in the number of patients for whom treatment is indicated, or in the proportion of eligible patients who are treated, leads to a proportionate increase insavings. For example, if all eligible patients were to be treated, after five years the estimated savings would be quadrupled. A reduction in the annual discount rate from 6% to 3% leads to an increase of £47 600 in expected savings. Increasing the ratio of inappropriate to appropriate treatment from 20% to 100% leads to a small decrease (£2800) in expected savings.

Comment

The addition of multiple dose anticholinergics to inhaled β2 agonistsfor children and adolescents attending casualty with severe acute asthma would result insavings in health service resources. This finding is robust to changes in modelling assumptions, although some uncertainties remain The personal value of the health effects and avoided hospital admissions provide additional benefits that have not been quantified inthis analysis.

Acknowledgments

We thank Paul Jones and Steve Milan of the Cochrane Airways Group; Janine Bestall for her help in literature searching; Richard Atkinson for providing data on hospital episodestatistics; and Debbie Latouche and Martyn Partridge for providing data from the UK National Asthma Task Force audit. A referee gave very helpful comments on an earlier draft of the report.

Contributors: RJS had the original idea for the project. JL conducted the computer modelling and drafted the report. FMD gave specific advice on the systematic review and the modelling assumptions. All contributors participated in the design of the project and commented on drafts of the report. JL will act as guarantor.

Footnotes

  • Funding The health care evaluation unit is supported by the research and development offices of the South East and London regional offices of the NHS Executive.

  • Competing interests None declared.

  • website extra A table with baseline data is on the BMJ's website www.bmj.com

References

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