Cost-effectiveness of transdiagnostic group cognitive behavioural therapy for anxiety disorders v. treatment as usual: economic evaluation of a pragmatic randomized controlled trial over an 8-month time horizon using self-reported data

Background:
This economic evaluation supplements a pragmatic randomized controlled trial conducted in community care settings, which showed superior improvement in the symptoms of adults with anxiety disorders who received 12 sessions of transdiagnostic cognitive-behavioural group therapy in addition to treatment as usual (tCBT + TAU) compared to TAU alone.

Methods:
This study evaluates the cost-utility and cost-effectiveness of tCBT + TAU over an 8-month time horizon. For the reference case, quality-adjusted life years (QALYs) obtained using the EQ-5D-5L, and the health system perspective were chosen. Alternatively, anxiety-free days (AFDs), derived from the Beck Anxiety Inventory, and the limited societal perspective were considered. Unadjusted incremental cost-effectiveness/utility ratios were calculated. Net-benefit regressions were done for a willingness-to-pay (WTP) thresholds range to build cost-effectiveness acceptability curves (CEAC). Sensitivity analyses were included.

Results:
Compared to TAU (n = 114), tCBT + TAU (n = 117) generated additional QALYs, AFDs, and higher mental health care costs from the health system perspective. From the health system and the limited societal perspectives, at a WTP of Can 50 000/QALY, the CEACs showed that the probability of tCBT + TAU v. TAU being cost-effective was 97 and 89%. Promising cost-effectiveness results using AFDs are also presented. The participation of therapists from the public health sector could increase cost-effectiveness.

Conclusions:
From the limited societal and health system perspectives, this first economic evaluation of tCBT shows favourable cost-effectiveness results at a WTP threshold of Can 50 000/QALY. Future research is needed to replicate findings in longer follow-up studies and different health system contexts to better inform decision-makers for a full-scale implementation.