Abstract
Objectives: The E-COMPARED trials aimed to evaluate the cost-effectiveness of blended (face-to-face and internet-based) cognitive behavioural therapy (bCBT) compared to treatment-as-usual (TAU) for major depressive disorder (MDD). Five randomised controlled trials (RCTs) were conducted over 12 months in primary care (Germany, UK, Poland, Spain and Sweden), and three in specialised mental health care (France, Netherlands and Switzerland). The current studies objective was to use Discrete Event Simulation (DES) to extrapolate the cost-effectiveness of bCBT versus TAU over five years. Interim E-COMPARED RCT results are used at present.
Methods: Participants were aged 18-65 years with DSM-IV diagnosed MDD. Interim within trial results were available for 412 patients (206 patients each arm) at baseline, 3, 6 and 12 months. For years 2-5, the DES model was calibrated using within trial data and data from the literature. Individual patient history was stored with patient characteristics from within trial data (e.g. PHQ9, demographics). Outcomes include QALYs (EQ-5D-5L) and societal costs (healthcare, patient and lost productivity costs) using Dutch costs. Missing within trial data were imputed using multiple imputation by chained equations. Uncertainty is estimated using cost-effectiveness acceptability curves.
Results: Long-term interim results show that the difference in QALYs (bCBT - TAU) is 0.07 (95% CI -2.23, 2.59) and the difference in costs is €7,678 (95% CI -€192,290, €212,482). The cost-effectiveness acceptability curve shows a 50% probability that bCBT is cost-effective compared to TAU for willingness to pay values between e 0 to €50,000/QALY.
Conclusions: Long term extrapolation of the E-COMPARED trial results show that there are no significant differences in QALYs or societal costs between bCBT and TAU, and that bCBT cannot be considered cost-effective as compared to TAU. Results should be interpreted with caution given the high uncertainty shown by the wide confidence intervals and the fact that only interim results are reported now.
Methods: Participants were aged 18-65 years with DSM-IV diagnosed MDD. Interim within trial results were available for 412 patients (206 patients each arm) at baseline, 3, 6 and 12 months. For years 2-5, the DES model was calibrated using within trial data and data from the literature. Individual patient history was stored with patient characteristics from within trial data (e.g. PHQ9, demographics). Outcomes include QALYs (EQ-5D-5L) and societal costs (healthcare, patient and lost productivity costs) using Dutch costs. Missing within trial data were imputed using multiple imputation by chained equations. Uncertainty is estimated using cost-effectiveness acceptability curves.
Results: Long-term interim results show that the difference in QALYs (bCBT - TAU) is 0.07 (95% CI -2.23, 2.59) and the difference in costs is €7,678 (95% CI -€192,290, €212,482). The cost-effectiveness acceptability curve shows a 50% probability that bCBT is cost-effective compared to TAU for willingness to pay values between e 0 to €50,000/QALY.
Conclusions: Long term extrapolation of the E-COMPARED trial results show that there are no significant differences in QALYs or societal costs between bCBT and TAU, and that bCBT cannot be considered cost-effective as compared to TAU. Results should be interpreted with caution given the high uncertainty shown by the wide confidence intervals and the fact that only interim results are reported now.
| Original language | English |
|---|---|
| Article number | MH1 |
| Pages (from-to) | A403 |
| Number of pages | 1 |
| Journal | Value in Health |
| Volume | 20 |
| Issue number | 9 |
| DOIs | |
| Publication status | Published - 1 Oct 2017 |