A comparison of cost-effectiveness between offering antidepressant–CBT combinations first or second, for moderate to severe depression in Japan

Background: It is not clear which method is more cost-effective: To initially provide all depressed patients with combination therapy (COMB; i.e. cognitive behavioural therapy plus pharmacotherapy), followed by antidepressant treatment (AD) for those still in depression; or, to first provide AD for all patients, followed by COMB for non-remission patients. The aim is to investigate whether a COMB-first strategy would be more cost-effective than an AD-first strategy, in treating depression. Methods: A Markov model was developed to perform the analysis. The primary outcome was the incremental cost-effectiveness ratio (ICER) per quality-adjusted life year (QALY) at 104 weeks. Probabilistic sensitivity analysis and scenario analysis were performed, to investigate the uncertainty associated with the clinical parameters and the impact of CBT's cost on the results, respectively. Results: The ICER per QALY at 104 week, was JPY 591,822 (USD 5,725) for moderate depression and JPY 499,487 (USD 4,832) for severe one. The scenario analysis revealed the ICER became JPY 1,147,518 (USD 11,101) for moderate and JPY 968,484 (USD 9,369) for severe when the CBT cost was set as JPY 14,400 (USD 139)(i.e. GBP 96: the unit cost of CBT in UK), which is well below the threshold recommended by NICE (i.e. GBP 20,000–30,000). Limitations: This is a model-based analysis which was conducted from the health insurance perspective. Then, the analysis from the societal perspective would generate different results. Conclusions: The results suggest that a COMB-first strategy would be more cost effective than an AD-first strategy.