OBJECTIVES: As mental health in adulthood is largely dependent on mental status during adolescence, various school-based interventions have been proposed to promote resilience. Economic evaluations to date have aimed to measure the health benefits and costs of interventions only for the duration of trials. The objectives of this study were to develop a conceptual model and build a simulation model representing the natural history of mental disorders in childhood, adolescence and youth to estimate the cost-utility of the UPRIGHT intervention promoting mental health in adolescence.

METHODS: We built a discrete event simulation model using the Arena® simulation tool fed with real-world data (cumulative incidence disaggregated into eight clusters) from the Basque Health Service database (609,381 individuals database) to calculate utilities (quality-adjusted life years [QALYs]) and costs for the general population in two scenarios (base case and intervention). The model translated changes in the well-being of adolescents into different risks of mental illnesses for a time horizon of 30 years.

RESULTS: The number of cases of anxiety was estimated to fall by 5,125 or 9,592 and those of depression by 1,269 and 2,165 if the effect of the intervention lasted 2 or 5 years respectively. From a healthcare system perspective, the intervention was cost-effective for all cases considered with incremental cost-utility ratios always lower than €10,000/QALY and dominant for some subgroups. The intervention was always dominant when including indirect and non-medical costs (societal perspective).

CONCLUSIONS: UPRIGHT Intervention to promote resilience in adolescents was cost-effective. Moreover, as the cumulative incidence rates were disaggregated by gender and socioeconomic status, the model also evidenced more efficiency in high-risk groups. Our approach relies on the use of an epidemiological model to build the intertwined natural history of mental health and mental disorders based on the whole registry for the Basque population (real-world data).