Cost-Effectiveness of App-Guided Self-Management for Posttraumatic Stress: Trial-Based Economic Evaluation

App interventions show promise as effective interventions for trauma-related distress, but evaluations of their cost-effectiveness are scarce.

This study aimed to assess the cost-effectiveness of an app-based intervention for self-management of posttraumatic stress compared to no guided self-management.

An economic evaluation from a Swedish public health care perspective was conducted alongside a randomized controlled trial in which participants (N=179) were randomly assigned to either immediate exposure (intervention group; n=89, 49.7%) or delayed exposure at 3 months (waitlist group; n=90, 50.3%). The number of quality-adjusted life years (QALYs) gained or lost, increases or decreases in the use of different types of health care, and the monetary costs (in SEK; 2023 price level) saved or incurred with the intervention versus the comparator at 9 months after exposure were estimated based on functional disability and health care consumption reported by participants via a web-based written questionnaire at baseline and at 3, 6, and 9 months of follow-up. Estimation was done via linear regression with clustering at the participant level. The probability of the intervention being cost-effective was calculated over a range of cost-effectiveness thresholds up to SEK 1 million per QALY (US $94,225 per QALY; 2023 average exchange rate, US $1=SEK 10.61), and value of information analysis was used to interpret statistical uncertainty in the cost-effectiveness results.

There was no statistically significant difference between the intervention and comparator at 9 months after exposure when QALYs and all categories of health care consumption were analyzed jointly (P=.46). When analyzed separately, there was a significant increase in the number of consultations made in private mental health care (P=.03). The intervention was associated with 0.0065 (95% CI -0.0219 to 0.0349) QALYs gained per user and an increment in costs of SEK -46,359 (95% CI -111,696 to 18,977; US $-4368, 95% CI -10,525 to 1788) per user compared to no guided self-management. Cost savings were due to fewer consultations and care days per user in public health care (-5.50, 95% CI -14.83 to 3.83). The intervention had a 62% probability of both gaining QALYs and saving costs, and the probability that it would be cost-effective remained constant at 92% over our threshold range. The total expected value of perfect information was SEK 5.4 million (US $510,480) and was largely attributable to statistical uncertainty in incremental costs.

The use of a mobile app for self-management of posttraumatic stress was found to be cost-effective in a Swedish setting. A value-of-information analysis suggests that current research is sufficient to support the use of the app in Swedish practice from a cost-effectiveness perspective. However, to support its adoption in other settings or the potential of app-based interventions in general, stronger cost-effectiveness evidence is required.

ClinicalTrials.gov NCT04094922; https://clinicaltrials.gov/ct2/show/NCT04094922.