Reducing suicidal ideation: Cost-effectiveness analysis of a randomized controlled trial of unguided web-based self-help

Abstract

Background: Suicidal ideation is highly prevalent, but often remains untreated. The Internet can be used to provide accessible interventions. Objective: To evaluate the cost-effectiveness of an online, unguided, self-help intervention for reducing suicidal ideation. Methods: A total of 236 adults with mild to moderate suicidal thoughts, defined as scores between 1-26 on the Beck Scale for Suicide Ideation (BSS), were recruited in the general population and randomized to the intervention (n = 116) or to a waitlist, information-only, control group (n = 120). The intervention aimed to decrease the frequency and intensity of suicidal ideation and consisted of 6 modules based on cognitive behavioral techniques. Participants in both groups had unrestricted access to care as usual. Assessments took place at baseline and 6 weeks later (post-test). All questionnaires were self-report and administered via the Internet. Treatment response was defined as a clinically significant decrease in suicidal ideation on the BSS. Total per-participant costs encompassed costs of health service uptake, participants' out-of-pocket expenses, costs stemming from production losses, and intervention costs. These were expressed in Euros (€) for the reference year 2009. Results: At post-test, treatment response was 35.3% and 20.8% in the experimental and control conditions, respectively. The incremental effectiveness was 0.35 - 0.21 = 0.15 (SE 0.06, P = .01). The annualized incremental costs were -€5039 per participant. Therefore, the mean incremental cost-effectiveness ratio (ICER) was estimated to be -€5039/0.15 = -€34,727 after rounding (US -$41,325) for an additional treatment response, indicating annual cost savings per treatment responder. Conclusions: This is the first trial to indicate that online self-help to reduce suicidal ideation is feasible, effective, and cost saving. Limitations included reliance on self-report and a short timeframe (6 weeks). Therefore, replication with a longer follow-up period is recommended.