- Open Access
Cost-utility analysis of the UPRIGHT intervention promoting resilience in adolescents
BMC Psychiatry volume 23, Article number: 178 (2023)
As mental health in adulthood is related to mental status during adolescence, school-based interventions have been proposed to improve resilience. The objective of this study was to build a simulation model representing the natural history of mental disorders in childhood, adolescence and youth to estimate the cost-effectiveness of the UPRIGHT school-based intervention in promoting resilience and mental health in adolescence.
We built a discrete event simulation model fed with real-world data (cumulative incidence disaggregated into eight clusters) from the Basque Health Service database (609,381 individuals) 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 wellbeing of adolescents into different risks of mental illnesses for a time horizon of 30 years.
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).
Although the primary analysis of the trial did not did not detect significant differences, the UPRIGHT intervention promoting positive mental health was dominant in the economic evaluation from the societal perspective. Promoting resilience was more cost-effective in the most deprived group. Despite a lack of information about the spillover effect in some sectors, the economic evaluation framework developed principally for pharmacoeconomics can be applied to interventions to promote resilience in adolescents. As prevention of mental health disorders is even more necessary in the post-coronavirus disease-19 era, such evaluation is essential to assess whether investment in mental health promotion would be good value for money by avoiding costs for healthcare providers and other stakeholders.
As mental health in adulthood is strongly dependent on mental status during childhood and adolescence , various school-based interventions have been proposed to promote complete mental health during youth [2, 3]. Among other aspects of mental health, these interventions address resilience, which is a general concept defined as the ability of an individual to adapt to life challenges or adversities while maintaining mental health and wellbeing [4, 5]. Although the literature contains numerous examples of school-based interventions aimed to boost adolescents’ resilience and wellbeing , they have rarely undergone economic evaluation . Such evaluation is essential to assess whether investment in the promotion of mental health would be good value for money by avoiding costs for healthcare providers and other stakeholders associated with future mental disorders . Should interventions be shown to be good value, this would indicate a need to overturn the current view that decision-makers do not need to prioritize mental health promotion . Moreover, economic evaluations to date have aimed to measure the health benefits and costs of interventions versus the standard alternative only for the duration of trials [9, 10]. Within the INCLUSIVE trial, given the difficulty of assessing cost-utility, a cost-consequence analysis was performed without expressing the results as cost-effectiveness or cost-utility ratios . Given this, the evaluation did not address the challenge of projecting the effectiveness of the intervention to medium- and long-term time horizons in terms of the prevention of mental disorders [12, 13]. Short-term assessments within the duration of trials explore improvements in positive mental health through scores on wellbeing scales [9, 10]. In contrast, the measurement of long-term outcomes implies investigating whether interventions succeed in preventing mental health problems. A conceptual model to bridge between short- and long-term focuses should be grounded on the dual-continua model of mental health taking into account the intertwined relationship between mental health and mental disease [14, 15]. Traditionally, mental health was reduced to the absence of mental illness . Instead, the dual-continua model distinguishes between these concepts, representing them on different dimensions [14, 16].
The present study is framed within the UPRIGHT research project funded by the European Union’s Horizon 2020 programme (No. 754919) which developed a universal preventive resilience intervention to promote mental health among adolescents in schools . It was designed as a whole school approach (for the school community including staff, adolescent students, and their families) to boost a culture of mental wellbeing by improving resilience and preventing mental disorders. To tackle the challenge of the long-term economic evaluation of this intervention, we have employed simulation modelling which is more routinely used in fields like pharmacoeconomics and cancer prevention.
The objective of this study was to build a simulation model representing the natural history of mental disorders in childhood, adolescence and youth to estimate the cost-effectiveness of the UPRIGHT intervention promoting resilience and mental health in adolescence.
The study consisted of the comparison of two scenarios (usual care, implying delivery of the school curricula, and the UPRIGHT intervention including delivery of both the school curricula and the UPRIGHT intervention, a proactive program targeting all adolescents to improve their resilience) in terms of utility (quality-adjusted life years [QALYs]) and resource use (costs) by the mathematical representation of the natural history of mental disorders in adolescence and youth. For that purpose, we used information from the patient record system of the Basque Health Service (Spain), which is an anonymized well documented and maintained database of a whole population (real-world data). Based on these data, we built a discrete event simulation (DES) model  using the Arena® simulation tool.
DES is a specific technique that constructs a conceptual model which incorporates entities in a mathematical system and assigns them attributes or features. Entities represent individuals constituting a population to be followed along their pathway through the system generating the results required to understand the system and answer research questions [17, 18]. We chose to measure QALYs instead of disability-adjusted life years  or years lived with disability  because the cost per QALY or incremental cost-utility ratio (ICUR) is the format used to establish standard willingness-to-pay thresholds in the economic evaluation of interventions [21, 22]. The study took a societal perspective which implied identifying direct healthcare costs, direct non-medical costs and indirect costs and intervention costs. Impacts on mortality and health-related quality of life were also addressed. The cost-utility analysis included an annual discount of 3% for survival and costs . The protocol of the study was approved by the Clinical Research Ethics Committee of the Basque Country (ref: PI2019078). UPRIGHT researchers, in collaboration with schools, obtained written informed consent from all participants, including teachers, adolescents and families (legal tutors also signing consent forms for adolescents’ participation).
The benefit of improving the resilience of adolescents is projected into adulthood in reductions in the risk of mental illness and increases in wellbeing . The problem is that the literature does not provide data on the longitudinal evolution of this relationship in a way that it could be incorporated into the model. Recognizing this limitation, we projected the effect of the intervention only to the incidence of mental disorders for a time horizon of 30 years. Our rationale was that mental health is a state that can be assessed in each adolescent through questionnaires and that mental illnesses are events diagnosed following clinical criteria by general practitioners and psychiatrists during childhood, adolescence and youth . Among other risk and protective factors, the development of mental disorders is partially related to mental wellbeing in early years as defined by effective coping skills, mindfulness, and resilience to external stressors [1, 4, 26].
The natural history of mental disorders is graphically represented in Fig. 1. As resilience does not prevent all conditions , its protective effect was disaggregated by type of mental disorder. For this task, eight diagnostic clusters were defined: anxiety, attention deficit hyperactivity disorder (ADHD), conduct disorders, depression, substance use, psychosis and personality disorders, eating disorders, and self-harm. International Classification of Diseases (ICD)-9-Clinical Modification, ICD-10 and Anatomical, Therapeutic, Chemical classification system (ATC) codes were used to identify corresponding diagnoses (see Supplementary material, Table SM1). Additionally, individuals who had any prescriptions for antidepressants (ATC N06A group) or antipsychotics (ATC N05A group) were included in the depression and psychosis categories respectively. Although the intervention was not expected to change the natural history of conditions in all these clusters, they were all included to represent the whole spectrum of mental disorders.
The study population was the 609,381 individuals who, as of December 31, 2018, were between 1 and 30 years old and were registered in the Basque Health Service (Table 1). At that time, the total Basque population was 2,180,449. All the data on demographic characteristics, mental disorders and healthcare resource use associated with mental disorders were extracted from the health service record system or dynamic population registry of the Basque Health Service’s institutional database, Oracle Business Intelligence (OBI)  which has been used to store clinical and administrative data since 2004 and is updated daily. A limitation of OBI is that it does not contain data from private practice records, and even though access to the health system is nearly universal for all residents, 20% of the population has double or complementary coverage through private insurance. This insurance does not, however, reimburse pharmacy prescriptions and that is why we also gathered data on prescriptions for antidepressants or antipsychotics, as mentioned above. Additional information needed to define simulation parameters was obtained from other official sources or the literature. We identified individuals with a diagnosis of mental health problems considering all the episodes of primary, emergency, outpatient and in-hospital care to measure the cumulative incidence of the eight diagnostic clusters.
Statistical analyses for obtaining various parameters for the model were performed using Stata (version 14.0) or R (version 4.0.1). The cumulative incidence was calculated considering the date of onset of symptoms for each individual in the target population, for each cluster using the Aalen-Johansen estimator for competing risks (death and file closure due to individuals moving away) (function plotCIF from the R package Epi in CRAN) . We also retrieved individuals’ socioeconomic status (SES) from their pharmacy co-payment category, which is based on household income , allowing cumulative incidence to be disaggregated by SES as well as gender. In this way, the model could also be employed for evaluating specific programs for reducing inequalities in high-risk populations. Nevertheless, in this study, we evaluated a universal intervention targeting the whole adolescent population.
The UPRIGHT universal preventive intervention to strengthen resilience was grounded on an individual, family and school staff framework (Figure SM1) and has been fully described elsewhere . The economic evaluation sought to compare the epidemiological and economic impacts of the base case scenario with those of the alternative scenario of implementing the UPRIGHT intervention. The general population risk of mental disorders shaped the base case, and for the intervention scenario, the model captured the changes observed with specific questionnaires after deployment of the UPRIGHT intervention. Specifically, changes in questionnaire scores were translated to changes in disease incidence and the epidemiological and economic impacts could then be automatically calculated. Table SM2 shows the relationship between diagnostic clusters and scales used in UPRIGHT.
General simulation model
The simulation model presented reproduces trajectories of 30 cohorts of the Basque general population up to 30 years old, one for each age group between 1 and 30 years of age (609,381 entities or individuals) according to the appearance of mental disorders across this range of ages and implementation of the UPRIGHT intervention. To relate SES to the risk of mental disorders in the model, the cumulative incidence of the eight clusters of mental disorder diagnoses considered was calculated separately for each sex-SES combination. The model started from 1 year of age to provide a complete representation of the natural history of mental disorders in childhood, adolescence and youth. Although the first 13 years have not been used in this work, it serves as a framework for possible studies analysing interventions that modify mental disorders that begin before the age of 13. Figure SM2 shows the model flow diagram, building on the conceptual model (Fig. 1). Table 2 summarizes the sources of the parameters.
The development of the general simulation model was broken down into four sub-steps: (1) calculation of the SES- and gender-specific cumulative incidence rates for the diagnostic groups in the general population between 1 and 30 years of age to construct empirical distributions for time until the onset of mental disorders (Tables SM3-SM6), (2) retrieval of data on utilities and costs from available databases and literature, (3) estimation of the UPRIGHT intervention effectiveness comparing pre- and post-intervention data to modify times until events in a cloned population, and (4) construction and validation of the general simulation model combining all the information collected. Finally, both general and UPRIGHT populations (Fig. 1) were run from birth to 30 years to obtain outputs. As UPRIGHT was scheduled to be implemented when individuals were 12 years of age and it lasted 2 years, both costs and QALYs were recorded from 14 years old until the end of the time horizon (30 years).
When individuals were first entered into the model at age zero, they were assigned characteristics or personal attributes (gender and SES), and random numbers that made each individual’s life course different . To guarantee that the two populations (general and UPRIGHT) had the same characteristics, once the attributes had been assigned, the population was cloned to produce two identical copies. In this way, the difference in the risk of mental disorders (as reflected in the cumulative incidences) due to the intervention was the only determinant of between-population differences in the trajectories of individuals. Second, the model calculated times until the onset of anxiety, ADHD, conduct disorders, depression, substance use, psychosis and personality disorders, eating disorders, and self-harm. Stochastic (or first-order) uncertainty was incorporated into the model by the aforementioned random numbers that acknowledged that individuals facing the same probabilities according to their attributes behave differently in the process of assigning events . The main competing risks of the model were the onset of disease for each diagnostic cluster and death. It was assumed that after the onset of the mental disorders their consequences remained. Figure 2 illustrates the assignment of time until the onset of mental disorders, based on the empirical distribution of the cumulative incidence for the risk in each population until 30 years of age (the time horizon). In the examples, the intervention effect delays the onset of a mental disorder in individual 1, prevents the disorder in individual 2 and does not change the trajectory for individual 3. Finally, each individual was assigned costs and utilities according to the date of onset of the mental disorders. QALYs were obtained by multiplying the utilities by survival in years.
The time until death was obtained through a parametric survival analysis adjusted for independent variables (gender and presence of mental disorders) of the Basque population previously described using the Akaike Information Criterion . Given the small number of deaths before 30 years old, mortality was not disaggregated by diagnostic cluster. After that, the Gompertz function was used to determine the time of death by gender and the presence of mental disorders (Table SM7). In this way, the reduction in mental disorder incidence rates due to the intervention also diminished the risk of death.
Utilities for individuals with and without mental disorders were extracted from a study  based on the 2012 Spanish National Health Survey (Spanish Ministry of Health, Social Services and Equality) , a nationwide survey conducted periodically in a representative sample of the non-institutionalized Spanish population aged ≥ 15 years. The 2012 survey included information from 21,007 individuals who responded to EuroQol's 5-level 5-dimension health status questionnaire, allowing the calculation of utility values for the Spanish general population as a function of various characteristics . Table SM8 lists the utility values used for the general simulation model. We applied those values for each group according to gender, and socioeconomic status, from birth until 30 years old and differentiating between individuals with and without mental disorders.
Direct healthcare costs
Direct healthcare costs were estimated by searching the aforementioned health service database for data on all resource use during 2018, including all contacts with nurses and general practitioners at healthcare centres, at home or by telephone in the case of primary care, and in the case of hospital care, all contacts with outpatient services and emergency services as well as hospitalizations. All the drugs prescribed to patients were also considered. Unit costs of healthcare resources for 2021 in euros (EUR, €) were taken from the Basque Health Service accounting system. Table SM9 lists the unit costs and Table SM10 the annual healthcare costs per patient updated to 2021.
Direct non-medical costs and indirect costs
Direct non-medical and indirect costs were obtained from a systematic review, which provided specific estimates of the economic costs of brain disorders in Spain . That study took a societal perspective and combined the methods and data retrieved from previous European-level research . As ICD-10 codes were used to define the mental health categories, it was possible to relate them to the diagnostic groups defined for the UPRIGHT project in terms of cost per patient-year as direct non-medical and indirect costs. We did not consider costs due to premature mortality, intangible costs and costs of crime given a lack of data or appropriate assessment methods. Table SM10 presents the annual direct non-medical costs and indirect costs per patient that were updated to 2021 using the consumer price index.
The cost of the intervention per student was estimated by a microcosting approach. The Basque Education Department provided unit costs for teachers’ time and the staff training time unit cost was retrieved from a Basque Health Service list of official salaries. The time that each professional spent on intervention training and delivery was converted into training costs, implementation costs and material costs which were distributed among the mean number of students to obtain the total cost per student (€135.70). Table SM11 reports the disaggregated costs per student.
Effectiveness of the intervention
The UPRIGHT intervention effect was estimated by comparing pre- and post-intervention scores on questionnaires administered during the cluster trial: the General Anxiety Disorder-7  for anxiety, the Patient Health Questionnaire  for depression, and specific items of WHO's Health Behavior in School-Aged Children survey (HBSC)  for substance use and behavioral disorders (Table SM2). Given a lack of evidence about the duration of the intervention effect, two options were analyzed considering a waning effect after 2 and after 5 years.
The main analysis of the UPRIGHT intervention performed using mixed models did not detect significant differences between the two arms of the cluster trial in questionnaire scores indicating improvements in mental health associated with the intervention. As it was implemented from September 2018 to June 2021, the base case questionnaires were answered before the beginning of the coronavirus disease 2019 (COVID-19) pandemic and the intermediate and final ones during it. Given the added stress experienced by adolescents under conditions of exposure to COVID-19, lockdown and other measures to reduce social interaction, it did not seem appropriate to evaluate the intervention based on these findings. Therefore, the design of this study focused on developing the evaluation framework rather than measure the cost-utility of the UPRIGHT intervention. Nonetheless, seeking to test the capacity of the model to evaluate resilience interventions, the differences between intervention and control groups were analyzed by logistic regression considering only the individuals from the sample whose scores worsened from baseline to final measurements. Specifically, we found that the likelihood of worsening during the pandemic or changing from a low- to a high-risk group was significantly lower in the intervention group for anxiety (OR = 0.71; CIs: 0.50–1.00) and depression (OR = 0.66; CIs:0.47–0.91). On the other hand, UPRIGHT was not associated with similar changes in HBSC scores (Table SM12). These ORs were converted to relative risks (RRs) and multiplied by the base case incidence during the duration of the effect from the 14 years old of the intervention implementation (2 or 5 years) to estimate the cumulative incidence in the UPRIGHT arm adjusted for each cluster, gender and SES. Changing the incidence of one, two or five years also changes the cumulative incidence curve from that age up to 30 years. This property of empirical distributions is what we used to extend the time horizon up to 30 years even though the effect of the intervention wanes after one, two or five years.
As a sensitivity analysis, we ran the model incorporating the effectiveness of universal resilience-focused interventions targeting child and adolescent mental health in a school setting estimated in a systematic review by Dray et al. . The meta-analysis showed significant effectiveness only for depression and anxiety and over a 1-year follow-up. Specifically, the cumulative incidence for the intervention arm was reduced with a standardized mean difference of -0.13 for depression and -0.18 for anxiety and only for 1 year .
To validate the model, we used the AdViSHE tool, assessing four parts : Part A: Validation of the conceptual model; Part B: Input data validation; Part C: Validation of the computerized model; and Part D: Operational validation. To test the external validity of the model outcomes, we assessed whether there were differences between the simulated age-specific cumulative incidences (from 1 to 30 years of age) and the observed ones. For that purpose, goodness-of-fit testing was conducted to obtain the correlation coefficient, normalised mean square error, fractional bias, fractional variance and the fraction of predictions within a factor of two.
Medium-long term impacts
The model produced three types of results: (1) estimation of the medium-to-long term epidemiological and economic impacts of UPRIGHT, (2) cost-utility analysis to assess its cost-effectiveness from health system and societal perspectives, and (3) sensitivity analysis to test the robustness of the assumptions made about the effect size and duration.
Part A: Validation of the conceptual model. Within the UPRIGHT Consortium, experts were asked to judge the appropriateness of the conceptual model. The expert panel, including psychiatrists, psychologists and mental health epidemiologists, reviewed the model and provided feedback helping to improve the model through an interactive process. Further, a cross-comparison was performed, comparing our approach with the conceptual models described by Le et al. in their systematic review  and this confirmed the comprehensiveness of our approach to representing the natural history of mental health in children, adolescents and young adults. Part B: Input data validation. The same experts judged the input parameters used to build the model to be appropriate. The agreement between our results and the cumulative incidence at 18 years measured by Dalsgaard  evidenced the goodness of fit of the model with external data on a key parameter. Other parameters, like mortality rates, utilities and costs, were taken from validated sources. Part C: Validation of the computerized model. To confirm that there were no substantive differences between the conceptual model and the DES model, an expert on model simulation not involved who had previously applied the AdViSHE tool to an obesity simulation model  reviewed the consistency of general and submodule architecture. Part D: Operational validation. All the goodness-of-fit statistics were within the established criteria when the simulated age-specific cumulative incidences were compared with the observed ones for each diagnostic cluster disaggregated by gender and SES (Table SM13).
Epidemiological and economic impacts
As shown in Table 3 for the whole population included in the model (609,381 individuals), the number of cases of anxiety was estimated to fall by 5,125 if the intervention effect lasted 2 years, 9,592 if it lasted 5 years and 2,792 considering the meta-analysis result. Similarly, the number of cases of depression fell by 1,269, 2,165 and 442 respectively. In terms of mortality, the implementation of the intervention could be expected to avoid 14, 19 and 11 deaths respectively.
Applying a discount of 3%, the intervention reduced the total costs by €69.16, €120.95 and €20.72 million depending on whether the effect lasted 2 years, 5 years or as in the meta-analysis (Table SM14). Considering only healthcare costs, the differences between the two scenarios were €50.01, €58.41 and €39.91 million respectively, the intervention costs being higher (€82.69 million). Table SM15 presents the same results but without discount.
When the intervention cost-effectiveness was measured considering only healthcare costs (Table 4 with discount and SM16 without discount), the intervention was cost-effective in all cases with ICURs always lower than €10,000/QALY and dominant for some subgroups. In the discounted analysis the ICUR for the total population was €3,231/QALY for a 2-year effect, €1,939/QALY for a 5-year effect and €5,440/QALY with the effectiveness found in the systematic review. For all low-SES subgroups, the UPRIGHT scenario was dominant, i.e., it was associated with lower costs and higher QALYs than the usual care scenario. The pattern was the same for the undiscounted analyses, the intervention being dominant in more subgroups.
When the cost-effectiveness of UPRIGHT was measured from the societal perspective including indirect and non-medical costs, the intervention was always dominant for discounted (Table SM17) and undiscounted (Table SM18) analyses. The saving generated for different stakeholders far outweighed the costs of the intervention.
The findings of this study show that the UPRIGHT psychoeducational intervention was associated with cost savings and health benefits when the societal costs were included, and therefore, it may be described as dominant. From the healthcare payer perspective, the intervention was cost-effective as the ICUR was always below the cost-effectiveness threshold defined in Spain . Moreover, our research highlights that a framework traditionally applied in pharmacoeconomics  can also be employed for the economic evaluation of psychoeducational interventions to promote resilience in adolescents. In practical terms, it underlines that the prevention of mental disorders can be implemented as a sustainable public health policy based on an economic assessment. Moreover, as the cumulative incidence rates were disaggregated by gender and SES, the model could also be used to assess specific interventions targeting high-risk groups.
The use of economic evaluation to inform policy-makers about mental health promotion from the societal perspective implies examining the resource requirements and outcomes for different stakeholders . Some of them, identified as non-healthcare sectors by the Second Panel recommendations for conducting cost-effectiveness analysis, are key for the assessment of mental illness costs like loss of productivity, social services, legal or criminal justice and education . To complement the data on healthcare resource consumption retrieved from Basque real-world data, we used data on costs from the review by Parés-Badell et al., which defined direct non-medical costs as informal care, adaptation costs and transportation costs and indirect costs as those related to absence from work and early retirement . Therefore, the scope of our approach based on considering only some stakeholders is a somewhat limited societal approach, as it did not include social services, legal or criminal justice, or education systems. The lack of information about the spillover effect that prevention of mental disorders actually has on social services, education or criminal justice represents a hurdle for a comprehensive economic evaluation of interventions promoting mental health. Despite this limitation, the UPRIGHT intervention gained health and saved costs .
The stress and measures to restrict social mobility associated with the COVID-19 pandemic hampered proper assessment of the effectiveness of the UPRIGHT intervention to strengthen resilience in the study (2020–2021) [46, 47]. Specifically, the repeated measures approach used to estimate the effectiveness was implemented in the context of adolescents experiencing added difficulties in their social lives and greater anxiety and stress. Nonetheless, in the case of mental wellbeing, while all participants showed deterioration from baseline, this was more marked in the control participants, and the magnitude of the reduction in mental wellbeing decreased over time for the intervention group. Therefore, our framework relies on the well-evidenced phenomenon that better coping with early adversity by improving resilience is an effective measure for preventing mental disorders . Within the economic evaluation, we did not address the difficulties associated with the large-scale implementation of mental health promotion interventions beyond cost-effectiveness outcomes as highlighted by Le et al. in a systematic review . They reported high dropout rates, indicating problems with acceptability, adherence, and feasibility regarding the interventions evaluated. Within the implementation of the UPRIGHT project, we identified some barriers such as a lack of commitment among school headteachers and problems with the participation of foster families. Teachers’ involvement was hampered by their job insecurity and work overload. Some saw the intervention as an extra strain on their working hours. These obstacles make it difficult to maintain the intervention over time. Its effectiveness might be enhanced if it were integrated into the school program in more year groups, but that would mean more pressure on schools.
We used real-world data to estimate the natural history of mental disorders by disaggregating them into clusters of diagnoses by gender and SES. The shaping of the model by multiple cohorts and clusters allowed us to obtain not only economic results but also epidemiological outcomes associated with the intervention in the 30 cohorts of the adolescent population at 13 years of age. The core of the model was the representation of the trajectories followed by individuals according to the risk of mental disorders adjusted for gender and SES. The cumulative incidence of mental disorders was calculated following the approach described by Dalsgaard et al., who estimated the same epidemiological indicators from a Danish registry . The agreement between our results and the cumulative incidence at 18 years measured by Dalsgaard  is striking. The fact that two studies analysing different populations (Danish and Basque) found similar cumulative incidence rates of diagnosed mental disorders should strengthen confidence in real-world data as a source of parameters for modelling . Studies from different countries also found similar incidence rates of dementia from registries based on electronic health records [28, 49, 50]. However, estimates of the prevalence of mental disorders collecting the presence of self-reported symptoms by adolescents through questionnaires resulted in higher figures than the cumulative incidence recorded in clinical databases [51,52,53]. These differences could be explained by the different methodology used, since surveys proactively ask for the presence of symptoms, while clinical databases include cases that contact the health system . In our case, the public system only.
The main limitation of our study was the way the results of the UPRIGHT trial were incorporated into the model. The model would have been more robust if longitudinal data had been available relating mental health in terms of wellbeing and resilience in adolescence to the diagnosis of mental disorders in adulthood . The questionnaires used measure the perception of adolescents of their interactions with the situations they are experiencing in family and school settings and other stressful events. It seems plausible that providing them with skills to cope with such events would improve their scores on these questionnaires. Nonetheless, they are surrogate outcomes that we have projected to the risk of mental disorders recorded as events in the health system. As underlined elsewhere, more long-term studies are needed to evidence this correlation between intermediate perceptions and final outcomes in terms of preventing mental disorders .
Resilience is a general concept defined as the ability of an individual to adapt to life challenges or adversities while maintaining mental health and wellbeing [4, 5]. Therefore, to move forward, we need to specify which clusters of mental disorders have lower incidence rates when resilience is improved. Our finding of anxiety and depression as the types of mental health problems that can be delayed or prevented by the intervention is consistent with the conclusion in a systematic review that universal resilience-focused interventions targeting adolescents are effective for depression and anxiety .
Our sensitivity analysis also suggested the intervention was cost-effective, finding an ICUR of €5,440/QALY. Another systematic review focused on school-based depression and anxiety prevention programs for young people also found small but significant effect sizes . We overcame the lack of information about the waning effect by analysing different duration times and projecting them over the time horizon. Psychoeducational interventions require stable implementation in schools to maintain their effectiveness but the evaluation of such programs is beyond the scope of our research group’s activity.
The UPRIGHT intervention promoting positive mental health was dominant from the societal perspective. Promoting resilience was more cost-effective in the most deprived group. Despite the lack of information about the spillover effect in some sectors, the economic evaluation framework developed principally for pharmacoeconomics can be applied to interventions to promote resilience in adolescents. The strength of the approach taken lies in the use of an epidemiological model built on a whole population registry to represent the intertwined natural history of mental health and mental illness. Given that economic models are still newcomers in mental health, more research is required to develop this approach to the same level as in other fields like cancer or cardiovascular disease. As prevention of mental health disorders is, if anything, even more necessary in the post-COVID-19 era, this type of evaluation is essential to assess whether investment in the promotion of mental health would be good value for money by avoiding costs for healthcare providers and other stakeholders.
Availability of data and materials
The data that support the findings of this study are available from the Basque Health Service but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the Basque Health Service.
Quality-adjusted life years
Discrete event simulation
Incremental cost-utility ratio
Attention deficit hyperactivity disorder
International Classification of Diseases
Anatomical, Therapeutic, Chemical classification system
Health Behavior in School-Aged Children survey
Mesman E, Vreeker A, Hillegers M. Resilience and mental health in children and adolescents: an update of the recent literature and future directions. Curr Opin Psychiatry. 2021;34:586–92.
O’Neil A, Jacka FN, Quirk SE, Cocker F, Taylor CB, Oldenburg B, et al. A shared framework for the common mental disorders and Non-Communicable Disease: key considerations for disease prevention and control. BMC Psychiatry. 2015;15:15.
Wykes T, Haro JM, Belli SR, Obradors-Tarragó C, Arango C, Ayuso-Mateos JL, et al. Mental health research priorities for Europe. Lancet Psychiatry. 2015;2:1036–42.
Las Hayas C, Izco-Basurko I, Fullaondo A, Gabrielli S, Zwiefka A, Hjemdal O, et al. UPRIGHT, a resilience-based intervention to promote mental well-being in schools: study rationale and methodology for a European randomized controlled trial. BMC Public Health. 2019;19:1413.
Windle G. What is resilience? A review and concept analysis. Rev Clin Gerontol. 2011;21:152–69.
van Agteren J, Iasiello M, Lo L, Bartholomaeus J, Kopsaftis Z, Carey M, et al. A systematic review and meta-analysis of psychological interventions to improve mental wellbeing. Nat Hum Behav. 2021;5:631–52.
Mcdaid D, Park A-L. Investing in mental health and well-being: findings from the DataPrev project. Health Promot Int. 2011;26(Suppl 1):i108-139.
Le LK-D, Esturas AC, Mihalopoulos C, Chiotelis O, Bucholc J, Chatterton ML, et al. Cost-effectiveness evidence of mental health prevention and promotion interventions: A systematic review of economic evaluations. PLOS Med. 2021;18:e1003606.
Bonell C, Allen E, Warren E, McGowan J, Bevilacqua L, Jamal F, et al. Modifying the secondary school environment to reduce bullying and aggression: the INCLUSIVE cluster RCT. Southampton: NIHR Journals Library; 2019.
Connolly P, Miller S, Kee F, Sloan S, Gildea A, McIntosh E, et al. A cluster randomised controlled trial and evaluation and cost-effectiveness analysis of the Roots of Empathy schools-based programme for improving social and emotional well-being outcomes among 8- to 9-year-olds in Northern Ireland. Southampton: NIHR Journals Library; 2018.
Bonell C, Allen E, Warren E, McGowan J, Bevilacqua L, Jamal F, et al. Effects of the Learning Together intervention on bullying and aggression in English secondary schools (INCLUSIVE): a cluster randomised controlled trial. Lancet Lond Engl. 2018;392:2452–64.
Stahl JE. Modelling methods for pharmacoeconomics and health technology assessment: an overview and guide. Pharmacoeconomics. 2008;26:131–48.
Knapp M. Economic evaluations and interventions for children and adolescents with mental health problems. J Child Psychol Psychiatry. 1997;38:3–25.
Iasiello M, van Agteren J, Muir-Cochrane E. Mental Health and/or Mental Illness: A Scoping Review of the Evidence and Implications of the Dual-Continua Model of Mental Health. Evid Base. 2020;1:1–45.
Payton AR. Mental health, mental illness, and psychological distress: same continuum or distinct phenomena? J Health Soc Behav. 2009;50:213–27.
Keyes CLM. Mental illness and/or mental health? Investigating axioms of the complete state model of health. J Consult Clin Psychol. 2005;73:539–48.
Karnon J, Stahl J, Brennan A, Caro JJ, Mar J, Möller J. Modeling using discrete event simulation: a report of the ISPOR-SMDM Modeling Good Research Practices Task Force-4. Med Decis Mak Int J Soc. 2012;32:701–11.
Caro JJ, Möller J, Karnon J, Stahl J, Ishak J. Discrete Event Simulation for Health Technology Assessment. Boca Raton: CRC Press; 2015.
Lee YY, Barendregt JJ, Stockings EA, Ferrari AJ, Whiteford HA, Patton GA, et al. The population cost-effectiveness of delivering universal and indicated school-based interventions to prevent the onset of major depression among youth in Australia. Epidemiol Psychiatr Sci. 2017;26:545–64.
Issakidis C, Sanderson K, Corry J, Andrews G, Lapsley H. Modelling the population cost-effectiveness of current and evidence-based optimal treatment for anxiety disorders. Psychol Med. 2004;34:19–35.
Drummond MF, Sculpher MJ, Torrance GW, O’Brien BJ, Stoddart GL. Methods for the Economic Evaluation of Health Care Programmes. Oxford: Oxford University Press; 2005.
Knapp M, Wong G. Economics and mental health: the current scenario. World Psychiatry Off J World Psychiatr Assoc WPA. 2020;19:3–14.
López-Bastida J, Oliva J, Antoñanzas F, García-Altés A, Gisbert R, Mar J, et al. Spanish recommendations on economic evaluation of health technologies. Eur J Health Econ HEPAC Health Econ Prev Care. 2010;11:513–20.
Las-Hayas C, Mateo-Abad M, Vergara I, Izco-Basurko I, González-Pinto A, Gabrielli S, et al. Relevance of well-being, resilience, and health-related quality of life to mental health profiles of European adolescents: results from a cross-sectional analysis of the school-based multinational UPRIGHT project. Soc Psychiatry Psychiatr Epidemiol. 2022;57:279–91.
Dalsgaard S, Thorsteinsson E, Trabjerg BB, Schullehner J, Plana-Ripoll O, Brikell I, et al. Incidence Rates and Cumulative Incidences of the Full Spectrum of Diagnosed Mental Disorders in Childhood and Adolescence. JAMA Psychiat. 2020;77:155–64.
Gee DG. Early Adversity and Development: Parsing Heterogeneity and Identifying Pathways of Risk and Resilience. Am J Psychiatry. 2021;178:998–1013.
Dray J, Bowman J, Campbell E, Freund M, Wolfenden L, Hodder RK, et al. Systematic Review of Universal Resilience-Focused Interventions Targeting Child and Adolescent Mental Health in the School Setting. J Am Acad Child Adolesc Psychiatry. 2017;56:813–24.
Mar J, Gorostiza A, Arrospide A, Larrañaga I, Alberdi A, Cernuda C, et al. Estimation of the epidemiology of dementia and associated neuropsychiatric symptoms by applying machine learning to real-world data. Rev Psiquiatr Salud Ment. 2021;:S1888–9891(21)00032-X.
Aalen OO, Johansen S. An empirical transition matrix for non-homogeneous Markov chains based on censored observations. Scand J Stat. 1978;5:141–50.
Oliver-Parra A, Dalmau-Bueno A, Ruiz-Muñoz D, García-Altés A. Relationship between parents’ mental disorders and socioeconomic status and offspring’s psychopathology: A cross-sectional study. PLoS ONE. 2020;15:e0240681.
Arrospide A, Machón M, Ramos-Goñi JM, Ibarrondo O, Mar J. Inequalities in health-related quality of life according to age, gender, educational level, social class, body mass index and chronic diseases using the Spanish value set for Euroquol 5D–5L questionnaire. Health Qual Life Outcomes. 2019;17:69.
Spanish National Statistics Institute. Spanish National Health Survey 2011–2012. Madrid: Spanish National Health Statistics 2017; 2017.
Gustavsson A, Svensson M, Jacobi F, Allgulander C, Alonso J, Beghi E, et al. Cost of disorders of the brain in Europe 2010. Eur Neuropsychopharmacol J Eur Coll Neuropsychopharmacol. 2011;21:718–79.
Briggs AH, Weinstein MC, Fenwick EAL, Karnon J, Sculpher MJ, Paltiel AD, et al. Model parameter estimation and uncertainty analysis: a report of the ISPOR-SMDM Modeling Good Research Practices Task Force Working Group-6. Med Decis Making. 2012;32:722–32.
Latimer NR. Survival analysis for economic evaluations alongside clinical trials–extrapolation with patient-level data: inconsistencies, limitations, and a practical guide. Med Decis Mak Int J Soc Med Decis Mak. 2013;33:743–54.
Ramos-Goñi JM, Craig BM, Oppe M, Ramallo-Fariña Y, Pinto-Prades JL, Luo N, et al. Handling Data Quality Issues to Estimate the Spanish EQ-5D-5L Value Set Using a Hybrid Interval Regression Approach. Value Health. 2018;21:596–604.
Parés-Badell O, Barbaglia G, Jerinic P, Gustavsson A, Salvador-Carulla L, Alonso J. Cost of disorders of the brain in Spain. PLoS ONE. 2014;9:e105471.
Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166:1092–7.
Johnson JG, Harris ES, Spitzer RL, Williams JBW. The patient health questionnaire for adolescents: validation of an instrument for the assessment of mental disorders among adolescent primary care patients. J Adolesc Health Off Publ Soc Adolesc Med. 2002;30:196–204.
Leal-López E, Sánchez-Queija I, Vieno A, Currie D, Torsheim T, Pavlova D, et al. Cross-national time trends in adolescent alcohol use from 2002 to 2014. Eur J Public Health. 2021;31:859–66.
Vemer P, Corro Ramos I, van Voorn G a. K, Al MJ, Feenstra TL. AdViSHE: A Validation-Assessment Tool of Health-Economic Models for Decision Makers and Model Users. PharmacoEconomics. 2016;34:349–61.
Sanders GD, Neumann PJ, Basu A, Brock DW, Feeny D, Krahn M, et al. Recommendations for Conduct, Methodological Practices, and Reporting of Cost-effectiveness Analyses: Second Panel on Cost-Effectiveness in Health and Medicine. JAMA. 2016;316:1093–103.
Vallejo-Torres L, García-Lorenzo B, Serrano-Aguilar P. Estimating a cost-effectiveness threshold for the Spanish NHS. Health Econ. 2018;27:746–61.
Walker S, Griffin S, Asaria M, Tsuchiya A, Sculpher M. Striving for a Societal Perspective: A Framework for Economic Evaluations When Costs and Effects Fall on Multiple Sectors and Decision Makers. Appl Health Econ Health Policy. 2019;17:577–90.
Al-Janabi H, van Exel J, Brouwer W, Coast J. A Framework for Including Family Health Spillovers in Economic Evaluation. Med Decis Mak Int J Soc Med Decis Mak. 2016;36:176–86.
Pan K-Y, Kok AAL, Eikelenboom M, Horsfall M, Jörg F, Luteijn RA, et al. The mental health impact of the COVID-19 pandemic on people with and without depressive, anxiety, or obsessive-compulsive disorders: a longitudinal study of three Dutch case-control cohorts. Lancet Psychiatry. 2021;8:121–9.
Schelhorn I, Ecker A, Lüdtke MN, Rehm S, Tran T, Bereznai JL, et al. Psychological Burden During the COVID-19 Pandemic in Germany. Front Psychol. 2021;12:640518.
Orsini LS, Berger M, Crown W, Daniel G, Eichler H-G, Goettsch W, et al. Improving Transparency to Build Trust in Real-World Secondary Data Studies for Hypothesis Testing-Why, What, and How: Recommendations and a Road Map from the Real-World Evidence Transparency Initiative. Value Health J Int Soc Pharmacoeconomics Outcomes Res. 2020;23:1128–36.
Ponjoan A, Garre-Olmo J, Blanch J, Fages E, Alves-Cabratosa L, Martí-Lluch R, et al. Epidemiology of dementia: prevalence and incidence estimates using validated electronic health records from primary care. Clin Epidemiol. 2019;11:217–28.
Perera G, Pedersen L, Ansel D, Alexander M, Arrighi HM, Avillach P, et al. Dementia prevalence and incidence in a federation of European Electronic Health Record databases: The European Medical Informatics Framework resource. Alzheimers Dement J Alzheimers Assoc. 2018;14:130–9.
Tørmoen AJ, Myhre M, Walby FA, Grøholt B, Rossow I. Change in prevalence of self-harm from 2002 to 2018 among Norwegian adolescents. Eur J Public Health. 2020;30:688–92.
Winsper C, Bilgin A, Thompson A, Marwaha S, Chanen AM, Singh SP, et al. The prevalence of personality disorders in the community: a global systematic review and meta-analysis. Br J Psychiatry. 2020;216:69–78.
Hoek HW. Incidence, prevalence and mortality of anorexia nervosa and other eating disorders. Curr Opin Psychiatry. 2006;19:389–94.
Violán C, Foguet-Boreu Q, Hermosilla-Pérez E, Valderas JM, Bolíbar B, Fàbregas-Escurriola M, et al. Comparison of the information provided by electronic health records data and a population health survey to estimate prevalence of selected health conditions and multimorbidity. BMC Public Health. 2013;13:251.
Werner-Seidler A, Perry Y, Calear AL, Newby JM, Christensen H. School-based depression and anxiety prevention programs for young people: A systematic review and meta-analysis. Clin Psychol Rev. 2017;51:30–47.
We acknowledge the help of Ideas Need Communicating Language Services in improving the use of English in the manuscript and the support of this work from the entire UPRIGHT group: Kronikgune (Spain): Nerea González and Maider Mateo-Abad; Osakidetza (Spain): Iñaki Zorrilla, Patricia Pérez Martínez and Itziar Vergara; UPV/EHU (Spain): Jessica Fernández-Sevillano; Bruno Kessler Foundation (Italy): Silvia Gabrielli, and Silvia Rizzi; Marshal's Office of Lower Silesia (Poland): Antoni Zwiefka, and Dominik Krzyżanowski; Daily Centre for Psychiatry and Speech Disorders A. (Poland): Iwona Mazur, Luba Jakubowska, Renata Poteralska, Piotr Czyż, and Urszula Andruszko; Falkiewicz Specialist Hospital (Poland): Paweł Błasiak, Katarzyna Krajewska, Grzegorz Pytlarz, and Ilona Szczygieł-Grüdl; Norwegian University of Science and Technology (Norway): Odin Hjemdal, Roxanna Morote, and Frederick Anyan; Directorate of Health in Iceland (Iceland): Dora Gudrun Gudmundsdottir, and Solveig Karlsdottir; Aarhus University (Denmark): Hans Henrik Knoop, Mette Marie Ledertoug, and Louise Tidmand; and University of Iceland (Iceland): Anna Sigridur Olafsdottir, Unnur B. Arnfjord, and Bryndis Jona Jonsdottir.
This work has been conducted within the UPRIGHT project, which is funded by the European Union’s Horizon 2020 framework programme for research and innovation under grant agreement No. 754919. This paper reflects only the views of the authors, and the European Union is not responsible for any use that may be made of the information it contains. The funding body has had no role in the study design, writing of the protocol or the decision to submit the paper for publication.
Ethics approval and consent to participate
This study was performed in line with the principles of the Declaration of Helsinki and received approval from the Ethics and Clinical Research Committee of Euskadi (reference number: PI2019078). UPRIGHT researchers, in collaboration with schools, obtained signed informed consent from all participants, including teachers, families (legal tutors also signed informed consent forms for adolescents’ participation) and adolescents.
Consent for publication
The authors have no biomedical financial interests or potential conflicts of interest to declare.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
About this article
Cite this article
Mar, J., Larrañaga, I., Ibarrondo, O. et al. Cost-utility analysis of the UPRIGHT intervention promoting resilience in adolescents. BMC Psychiatry 23, 178 (2023). https://0-doi-org.brum.beds.ac.uk/10.1186/s12888-023-04665-4
- Adolescents health
- Mental health
- Mental disorders
- Simulation model
- Health policy