Refining cost-effectiveness analyses using the net benefit approach and econometric methods: An example from a trial of anti-depressant treatment

Abstract

Background

Economic evaluation analyses can be enhanced by employing regression methods, allowing for the identification of important sub-groups and to adjust for imperfect randomisation in clinical trials or to analyse non-randomised data.

Aims

To explore the benefits of combining regression techniques and the standard Bayesian approach to refine cost-effectiveness analyses using data from randomised clinical trials.

Method

Data from a randomised trial of anti-depressant treatment were analysed and a regression model was used to explore the factors that have an impact on the net benefit (NB) statistic with the aim of using these findings to adjust the cost-effectiveness acceptability curves. Exploratory sub-samples' analyses were carried out to explore possible differences in cost-effectiveness.

Results

The analysis found that having suffered a previous similar depression is strongly correlated with a lower NB, independent of the outcome measure or follow-up point. In patients with previous similar depression, adding an selective serotonin reuptake inhibitors (SSRI) to supportive care for mild-to-moderate depression is probably cost-effective at the level used by the English National Institute for Health and Clinical Excellence to make recommendations.

Conclusions

This analysis highlights the need for incorporation of econometric methods into cost-effectiveness analyses using the NB approach.

Keywords:

• Depression
• cost-effectiveness
• anti-depressant

Acknowledgements

The THREAD study was funded by the National Institute for Health Research (Health Technology Assessment) programme, England. The opinions expressed in this paper are the authors and do not represent the views of the Department of Health for England. The other members of the THREAD study group were J. Chatwin (Primary Medical Care, Aldermoor Health Centre, University of Southampton, UK), C. Dowrick (University of Liverpool, UK), A. Tylee (Institute of Psychiatry, King's College London, UK), R. Morriss (University of Nottingham, UK), R. Peveler (Primary Medical Care, Aldermoor Health Centre, University of Southampton, UK), M. Leese (Institute of Psychiatry, King's College London, UK), T. Harris (King's College, London, UK), M. Moore (Primary Medical Care, Aldermoor Health Centre, University of Southampton, UK), R. Byng (Peninsula Medical School, Plymouth, UK), G. Brown (King's College, London, UK), S. Barthel (Institute of Psychiatry, King's College London, UK), H. Mander (Primary Medical Care, Aldermoor Health Centre, University of Southampton, UK), A. Ring (University of Liverpool, UK), V. Kelly (Institute of Psychiatry, King's College London, UK), V. Wallace (Institute of Psychiatry, King's College London, UK), M. Gabbay (University of Liverpool, UK), T. Craig (King's College, London, UK) and A. Mann (Institute of Psychiatry, King's College London, UK).

Declaration of Interest: The authors report no conflict of interest.