gms | German Medical Science

Background: Globally, depressive disorders have a high prevalence and cause a large burden of disease. Depression screening is already widely applied, feedback interventions in combination with screening methods are still to be investigated to a broad extent. The DISCOVER trial was a web-based, three-armed randomized controlled trial (RCT) on patient-tailored feedback after online depression screening.

Objective: The objective of this work was to conduct the cost-effectiveness analysis (CEA) of the DISCOVER trial from a societal perspective.

Methods: In the RCT, participants underwent online depression screening and were immediately randomized into one of three study arms if screened positive. NO FEEDBACK (NF) participants received no feedback on their screening results, STANDARD FEEDBACK (SF) participants received generic feedback, and TAILORED FEEDBACK (TF) participants were presented with patient-tailored feedback including further information based on their responses. The time horizon of the trial was six months.

Main outcome was the incremental cost-effectiveness ratio (ICER), defined as the ratio of incremental costs (healthcare costs and productivity losses) and quality-adjusted life years (QALY). Health-related quality of life (HRQL) was measured by means of the EQ-5D-5L. To assess the uncertainty around the ICER, cost-effectiveness acceptability curves were constructed based on the net monetary benefit regression approach, applying linear OLS regressions. Furthermore, four different scenarios of sensitivity analyses were conducted, one of which considered depression-free-days (DFD) as an outcome. Explorative subgroup analyses were conducted.

Results: 1,012 participants (NF: 343; SF: 338; TF: 331) were included at baseline. During follow-up, HRQL increased for all study arms. NF participants experienced the most QALY, SF the least. Total costs were lowest for TF and highest for SF. All differences were statistically not significant. NF and TF both exhibited dominance over SF. The ICER of TF compared to NF was € 109,730/QALY. Cost-effectiveness probability of TF against NF ranged between 47% and 76% (willingness-to-pay range: € 0/QALY-€ 160,000/QALY). Sensitivity analyses showed robustness of results. DFD during follow-up were highest for TF and lowest for NF. In this case, TF was dominant over both NF and SF. At a WTP of € 15/DFD, its cost-effectiveness probability was approx. 80% compared to NF and 91% compared to SF. Subgroup analyses revealed high cost-effectiveness probabilities for the patient-tailored intervention for people who were not sure at baseline if they suffered from depression and those, whose depressive symptoms were confirmed at baseline.

Implication for research: Regarding QALY, both TF and NF are highly likely to be cost-effective compared to SF. The cost-effectiveness between TF and NF highly depends on assumed WTP margins. Considering DFD, TF is highly probable to be cost-effective. Subgroup analyses showed cost-effectiveness for the target group of screening. Long-term effects of feedback interventions remain to be investigated.

Funding: Individual funding (BMG, DRV, BMBF, DFG, etc); Project name: DISCOVER; Grant number: PV7039