OBJECTIVE: To estimate the cost effectiveness of representatives of three different classes of antidepressants used in major depression in the UK NHS. DESIGN, PATIENTS AND INTERVENTIONS: A decision-tree model for the treatment of major depression was constructed by interviewing UK GPs and psychiatrists (as part of a Delphi panel). An important part of the tree was that patients in primary care were treated until remission (pre-morbid state). Three classes of antidepressants (serotonin and noradrenaline reuptake inhibitors [SNRIs; venlafaxine], selective serotonin reuptake inhibitors [SSRIs; fluoxetine, paroxetine and fluvoxamine] and tricyclic antidepressants [TCAs; amitriptyline]) were compared by populating the tree with clinical success rates determined by a meta-analysis and a clinical trial. Where there were insufficient data from clinical trials a Delphi panel was used. Costs within the tree were taken from UK data sources. Six-monthly costs and cost effectiveness were then calculated. MAIN OUTCOME MEASURES AND RESULTS: Treatment costs for 6 months were pound 1285 for venlafaxine, pound 1348 for SSRIs and pound 1385 for amitriptyline. Cost effectiveness as measured by cost per symptom-free day was pound 21 for venlafaxine, pound 26 for SSRIs and pound 32 for TCAs (2001 values). Incremental cost-effectiveness analyses showed a treatment strategy of using venlafaxine and switching if necessary to an SSRI was dominant over all other strategies considered. Sensitivity testing demonstrated that the cost of an SSRI could be reduced to 4 pence daily and amitriptyline to zero before the expected 6-monthly cost of venlafaxine ceased to be the lowest. CONCLUSION: The SNRI, venlafaxine, may be a cost-effective option compared with the SSRIs and TCAs when used as a first-line drug for depression in primary care in the UK. As this is a model, cost effectiveness can be suggested but not proven.
OBJECTIVE: To estimate the cost effectiveness of representatives of three different classes of antidepressants used in major depression in the UK NHS. DESIGN, PATIENTS AND INTERVENTIONS: A decision-tree model for the treatment of major depression was constructed by interviewing UK GPs and psychiatrists (as part of a Delphi panel). An important part of the tree was that patients in primary care were treated until remission (pre-morbid state). Three classes of antidepressants (serotonin and noradrenaline reuptake inhibitors [SNRIs; venlafaxine], selective serotonin reuptake inhibitors [SSRIs; fluoxetine, paroxetine and fluvoxamine] and tricyclic antidepressants [TCAs; amitriptyline]) were compared by populating the tree with clinical success rates determined by a meta-analysis and a clinical trial. Where there were insufficient data from clinical trials a Delphi panel was used. Costs within the tree were taken from UK data sources. Six-monthly costs and cost effectiveness were then calculated. MAIN OUTCOME MEASURES AND RESULTS: Treatment costs for 6 months were pound 1285 for venlafaxine, pound 1348 for SSRIs and pound 1385 for amitriptyline. Cost effectiveness as measured by cost per symptom-free day was pound 21 for venlafaxine, pound 26 for SSRIs and pound 32 for TCAs (2001 values). Incremental cost-effectiveness analyses showed a treatment strategy of using venlafaxine and switching if necessary to an SSRI was dominant over all other strategies considered. Sensitivity testing demonstrated that the cost of an SSRI could be reduced to 4 pence daily and amitriptyline to zero before the expected 6-monthly cost of venlafaxine ceased to be the lowest. CONCLUSION: The SNRI, venlafaxine, may be a cost-effective option compared with the SSRIs and TCAs when used as a first-line drug for depression in primary care in the UK. As this is a model, cost effectiveness can be suggested but not proven.