NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women's and Children's Health (UK). Diarrhoea and Vomiting Caused by Gastroenteritis: Diagnosis, Assessment and Management in Children Younger than 5 Years. London: RCOG Press; 2009 Apr. (NICE Clinical Guidelines, No. 84.)

Cover of Diarrhoea and Vomiting Caused by Gastroenteritis

Diarrhoea and Vomiting Caused by Gastroenteritis: Diagnosis, Assessment and Management in Children Younger than 5 Years.

Show details

Appendix BHealth economics of ondansetron

Introduction

Children presenting with acute gastroenteritis often have high levels of vomiting. There is currently no widely accepted method of treatment for the cessation of vomiting in such patients in the UK. Vomiting is not only distressing to both patients and their families but can also interfere with the oral rehydration process and can limit the success of oral rehydration therapy (ORT). Children who are unable to tolerate rehydration fluids orally are then given intravenous fluid therapy (IVT), which in turn has additional significant resource implications. It is thought that the use of anti-emetics may be effective in the cessation of vomiting and may in turn help with the successful delivery of ORT, thereby reducing the need to treat with IVT. Consequently, this would have cost saving implication for the NHS through fewer admissions for IVT. Ondansetron is recognised as an effective anti-emetic and is used frequently in post-operative patients and also in oncology. The GDG felt that an economic analysis would help guide recommendations regarding its use.

Economic analysis

A literature review identified several articles investigating the effectiveness of ondansetron in children with vomiting due to gastroenteritis. The evidence was for both oral and IV uses of ondansetron but the evidence for the efficacy of IV ondansetron was limited and thus the economic analysis only considers oral administration. None of the articles retrieved included any data regarding costs or cost-effectiveness.

A simple economic model was constructed to assess the cost-effectiveness of ondansetron. The model probabilities are taken from a meta-analysis that pooled the results of three RCTs160,163,164 comparing the effects of administration of oral ondansetron against placebo in children with vomiting. Model probabilities are presented in Table B.1. The three main outcomes considered are:

Table B.1. Model probabilities.

Table B.1

Model probabilities.

  • cessation of vomiting
  • need for IVT
  • hospitalisation.

Model costs are taken from the decision analytical model for the cost-effectiveness of IVT versus ORT for children with dehydration (Appendix A). All cost inputs are presented in Table B.2. The cost for ondansetron was obtained from the British National Formulary for Children (BNFC 2007).209 It is assumed that only one dose of ondansetron (at a strength of 4 mg) is given to the vomiting patient.

Table B.2. Costs used in the model.

Table B.2

Costs used in the model.

The effectiveness of ondansetron is measured by an increase in the cessation of vomiting. The need for further IVT and hospitalisation are both measures of resource use subsequent to treatment. To the extent that ondansetron leads to lower hospitalisation and need for IVT, there will be savings to at least partially offset some of the costs of treatment.

The analysis calculates the difference in effect between placebo and ondansetron for the three outcomes from the pooled meta-analysis. Any savings in potential downstream costs is then calculated and this is attributed to the difference in effect achieved by taking a dose of ondansetron. This downstream saving is calculated by multiplying the difference in effect by the costs of treatment. It is also important to note that this saving also needs to be offset against the cost of treatment, i.e. the cost of ondansetron. Therefore:

Net savings=downstream savings-treatment cost of ondansetron

Results

The results of the analysis are listed in Table B.3. Note that the outcome of ‘cessation of vomiting’ was only reported in two of the trials.163,164 Net savings are derived by multiplying the difference in effect by the costs of treatment.

Table B.3. Net savings with ondansetron using the pooled meta-analysis.

Table B.3

Net savings with ondansetron using the pooled meta-analysis.

Once the treatment cost is taken into account, this gives a net saving of approximately £52 with ondansetron compared with placebo. Given that ondansetron also leads to a benefit in terms of increased cessation of vomiting, this implies that the treatment is dominant, providing additional clinical benefit at reduced cost.

Sensitivity analysis

It is important in economic evaluation to take into account uncertainty. Therefore, a probabilistic sensitivity analysis (PSA) was undertaken to reflect the fact that the meta-analysis only provides an estimate of the ‘true’ effect size. To carry out the PSA, the model was run for 10 000 simulations with the values of the model probabilities sampled from a probability distribution, as indicated in Table B.1. The cost values were accepted as fixed as these are based on published cost data. The results of the probabilistic simulations are shown in Figure B.1.

Figure B.1. Probabilistic sensitivity analysis of the net costs and effects of ondansetron (n = 10 000).

Figure B.1

Probabilistic sensitivity analysis of the net costs and effects of ondansetron (n = 10 000).

The PSA results showed that in every simulation ondansetron led to a greater cessation in vomiting than placebo, providing very strong evidence for the efficacy of the intervention. It also showed ondansetron to be cost saving in 99.96% of simulations,

Discussion

The baseline result shows that ondansetron is dominant compared with placebo, with increased cessation of vomiting and a net saving of £51.68. A PSA suggested this cost-effectiveness conclusion was not sensitive to parameter uncertainty in the model’s probabilities. This is because the magnitudes of the effect sizes for cessation of vomiting and reduced resources are so large (and the confidence intervals sufficiently narrow) that they are most unlikely to be due to chance. As the costs of hospitalisation and IVT are high relative to treatment, a real reduction in the costs associated with these events is likely to offset the initial ondansetron drug cost.

There are a number of caveats to be considered when interpreting the above results. Firstly, the PSA assumed independence in the three probabilistic model parameters. In practice, it would be expected that a reduction in hospitalisation and IVT is dependent on the efficacy of ondansetron. Therefore, a more sophisticated model would demonstrate an inverse relationship between the net costs of ondansetron and increased cessation of vomiting. Nevertheless, the importance of this should not be overstated. As the independent variable, the distribution of cessation of vomiting rates is not affected by the assumption of independence. Although a more sophisticated model may show a greater uncertainty around the mean net costs, the efficacy of the intervention would still be likely to produce some offsetting savings and the probability of a net cost saving would remain high.

The two studies that reported the outcome of ‘cessation of vomiting’ occurring ‘a few hours’ after ondansetron is taken use different time frames from each other. One study163 looked at the proportion of children who vomited while receiving ORT whereas the other164 measured the frequency of emesis during the 48 hour period after enrolment. All the studies in the meta-analysis were undertaken in the USA and treatment practice differs in England and Wales. In the NHS, a child would be admitted as an inpatient if their vomiting had not stopped within 4 hours of taking ondansetron in the emergency department, regardless of whether the child ultimately receives IVT. This inpatient admission would incur costs. Therefore, the time frame used in the US studies may not be the most relevant to UK practice, particularly with respect to subsequent resource use. It cannot be assumed that the savings imputed from US models of care would be transferable to an NHS setting.

The inclusion criteria for the studies included in the meta-analysis also varied. One study163 included children with at least one reported episode of vomiting within the 4 hours preceding triage whereas another164 enrolled patients with five episodes of vomiting in the preceding 24 hours. The authors felt that the inclusion criteria for those being given ondansetron should be restrictive in order to best identify the group of patients who would benefit from ondansetron. This is in line with the view of the GDG who also agreed that not every child should be given ondansetron but only those who are most likely to benefit, i.e. those patients who would otherwise fail ORT and go on to IVT. It is in this group of patients that savings could be made.

None of the three above-mentioned studies reported any significant adverse events or complicating side effects from the use of ondansetron and the economic analysis has not taken into these into account. Nevertheless, the BNFC209 reports several possible side effects from ondansetron, ranging from headaches to chest pain and seizures. The chance of these side effects occurring, although small, could lower the health-related quality of life improvement of ondansetron. It is therefore important to remember the importance of any potential harms that may be of clinical importance and may differ systematically between those who are treated with ondansetron and those who are not.

Changes in diarrhoea in response to treatment was not included in this model because it was unclear whether ondansetron worsened this outcome and more importantly because the GDG queried the clinical significance of diarrhoeal outcomes reported in the studies. Two trials in the meta-analysis163,164 reported a statistically significant increase in the frequency of diarrhoea as an adverse event of ondansetron. A third study160 showed that the number of children with fewer episodes of diarrhoea was less with ondansetron but the difference was not statistically significant. To more accurately determine the cost-effectiveness of ondansetron, it would be important to know the clinical significance of any increased diarrhoea and whether it led to a concomitant increase in the use of healthcare resources.

In the economic model, it was assumed patients are given a single oral dose of ondansetron in order to reduce vomiting. This is consistent with two of the studies included in the meta-analysis.163,211 However, the third study164 gave a single oral dose of ondansetron in hospital but also provided discharged patients with an additional five doses of ondansetron to be used every 8 hours for a total of 2 days. Although this approach would increase the cost of ondansetron, repeated home doses of ondansetron may also help in delivering persistent benefit and consistently reduce hospital admission. This would clearly have implications for the economic analysis.

Finally, treatment costs were restricted to the cost of ondansetron. To the extent that other costs, such as staffing, in administering oral ondansetron have been omitted, there will be a bias, albeit small, in favour of ondansetron in this analysis.

Although ORT has been proven to be a clinically effective and cost-effective treatment for children suffering from dehydration, it remains underused, especially when the child is vomiting. Clinicians are more likely to choose IVT in scenarios where vomiting is a major symptom, and therefore a safe and effective method of controlling vomiting, such as ondansetron, may increase the use of ORT. The simple model in this appendix is suggestive of potential clinical and economic benefits of ondansetron; however, more evidence, particularly with regard to diarrhoeal outcomes, was felt to be necessary by the GDG to justify its use in routine practice.

Copyright © 2009, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK63842

Views

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...