Introduction
Mental health has emerged as a fundamental component of public health, influencing individual well-being, quality of life, and societal functioning [1]. The scale of this challenge is extensive, and in 2019 nearly 970 million people worldwide were affected by mental health conditions [2], with most cases first manifesting during mid-adolescence [Reference Solmi, Radua, Olivola, Croce, Soardo and de Pablo3]. Understanding the prevalence, distribution, and contributing factors of mental health disorders is crucial for developing targeted public health initiatives [Reference Patel, Saxena, Lund, Thornicroft, Baingana and Bolton4], as their burden extends beyond individuals, affecting families, communities, and healthcare systems [Reference Kazdin and Rabbitt5]. Conditions such as depression, anxiety, and schizophrenia consistently rank among the leading causes of disability and economic loss globally [2]. Addressing this public health challenge requires a deep understanding of the interrelated factors that shape mental health outcomes [Reference Lund, Brooke-Sumner, Baingana, Baron, Breuer and Chandra6]. The emergence of large-scale population-based databases has created new opportunities for mental health research. These databases, characterized by their longitudinal design [Reference Vieta and De Prisco7] and large representative sample sizes [Reference De Prisco and Vieta8], provide robust data that enable researchers to identify trends, risk factors, and potential interventions that can inform both public health policies and clinical practices [Reference Oliver, Arribas, Perry, Whiting, Blackman and Krakowski9]. Large population databases have provided valuable insights into mental health outcomes across different healthcare systems. For example, the Swedish National Patient Register, including information on both inpatient and outpatient psychiatric care, has been used to identify mortality patterns among individuals with mental illnesses [Reference de la Cruz, Isomura, Lichtenstein, Larsson, Kuja-Halkola and Chang10, Reference Mataix-Cols, Isomura, Sidorchuk, Rautio, Ivanov and Rück11]. Similarly, Denmark’s nationwide register has been analyzed to explore the complex relationships between physical and mental health conditions [Reference Momen, Østergaard, Heide-Jorgensen, Sørensen, McGrath and Plana-Ripoll12]. In the United States, the Nationwide Inpatient Sample has helped researchers understand how psychiatric comorbidities affect both mental health outcomes and illness severity [Reference Anona, Olaomi, Udegbe, Uwumiro, Tuaka and Okafor13]. While these cohorts provide valuable insights, their findings may not apply uniformly across different regions and countries. Mental health outcomes can vary significantly based on local socio-economic conditions [Reference Evans-Lacko, Aguilar-Gaxiola, Al-Hamzawi, Alonso, Benjet and Bruffaerts14], climate [Reference Radua, De Prisco, Oliva, Fico, Vieta and Fusar-Poli15], and healthcare accessibility [Reference Santomauro, Vos, Whiteford, Chisholm, Saxena and Ferrari16], highlighting the importance of region-specific data collection and analysis.
To address this need, we present a newly established cohort from Catalonia, Spain, that includes all individuals who accessed public specialized mental healthcare between 2015 and 2019, with retrospective follow-up extending to 2010. This cohort enables comparison with a representative sample from the general population. We describe its key features and explore its potential to advance our understanding of mental health patterns, risk factors, and effective interventions. This resource will contribute to improving our understanding of mental disorders and developing evidence-based public health strategies.
Methods
Data for this study were obtained from the “Programa d’analítica de dades per a la recerca i la innovació en salut” (“Public Data Analysis for Health Research and Innovation Programme”, PADRIS) managed by the Agency for Health Quality and Assessment of Catalonia (AQuAS). PADRIS provides the scientific community with access to healthcare data to promote research, innovation, and health evaluation. This is achieved by enabling the reuse and integration of health data generated by the comprehensive public health system of Catalonia (SISCAT), in alignment with legal and regulatory frameworks, ethical principles, and a commitment to transparency with citizens [17]. The use of this data is part of the “PRimary carE digital Support ToOl in mental health” (PRESTO) project [Reference Anmella, Primé-Tous, Segú, Solanes, Ruíz and Martín-Villalba18], which aims to develop an advanced machine learning-driven digital support platform for efficiently screening, assessing, triaging, and delivering personalized treatments for anxiety and depressive symptoms in primary care. This study was approved by the Hospital Clínic Ethics Committee (HCB/2020/0735).
Participants
Participants, referred to hereafter as “cases,” included all individuals who accessed public specialized mental health services (i.e., outpatient clinics, inpatient units, or emergency departments) under SISCAT between January 1, 2015, and December 31, 2019. These individuals were individually matched 1:2 by age, sex, and health region with a comparison group, referred to hereafter as “controls,” comprising individuals who sought primary care attention but had no interactions with public mental health services during the same period. Both cases and controls were retrospectively followed up to January 1, 2010, and data were extracted from public health service records for the entire study period. Information was anonymized and de-identified in strict adherence to the applicable legal and regulatory frameworks, ethical guidelines, and principles of transparency [17].
Measures
For each participant, sociodemographic and clinical characteristics were provided.
Sociodemographic characteristics included: age group, sex, country of birth, socioeconomic level, healthcare territorial unit, and date of death.
To ensure anonymization and de-identification, the exact age of the participants was not provided; instead, their age was reported within a five-year range. Based on this information, we estimated each participant’s age by calculating the average of the lower and upper limits of the range.
The date of death was available only for the cases, as the controls were selected from individuals who were alive during the period considered.
Clinical characteristics included: psychiatric diagnosis, medical comorbidities, smoking status, body mass index (BMI), adjusted morbidity groups, number of visits to public specialized mental health services, general medicine services, primary care services, and pharmacological interventions prescribed.
Psychiatric diagnoses were recorded based on both specialized mental health services and primary care services, and were reported using International Classification of Diseases (ICD) codes, either the 9th edition [19] or the 10th edition [20]. To ensure consistency, we converted all ICD-9 diagnoses to ICD-10 using a conversion table. In addition to the specific diagnosis, we also reported the main ICD diagnostic category to which it belonged.
Medical comorbidities were determined based on the physical diagnoses participants received during the last two years of follow-up. These diagnoses were provided by PADRIS and were grouped into broader categories, including acquired immunodeficiency syndrome, arthritis, arthrosis, asthma, chronic kidney disease, cirrhosis, chronic obstructive pulmonary disease, dementia, diabetes, heart failure, hypertension, ischemic stroke, musculoskeletal disorders, neoplasms, and osteoporosis.
Regarding smoking status, individuals who were classified as smokers at least once during the study period were considered “ever smokers,” while those who were never classified as smokers during the same period were considered “never smokers.”
BMI data were available only for the final year of the follow-up period. In cases where multiple measurements were collected during the same year, we averaged them to obtain a single value. Additionally, since some values appeared unusually low or high, suggesting potential reporting errors, we decided to remove these outliers based on clinical judgment, excluding values greater than 65 or less than 10.
Adjusted morbidity groups were assigned to each participant to classify their health conditions based on the severity of their morbidity. These groups were designed to reflect the complexity of individuals’ health conditions, accounting for multimorbidity and the level of medical intervention required [21]. This data was available from 2014 to 2019, with participants categorized into five risk groups, ranging from “Very Low Risk” to “Very High Risk”.
Statistical analyses
Data cleaning, data harmonization, and all statistical analyses were conducted using RStudio, R version 4.3.1 [22]. Due to the large sample size, we assumed normality of the data based on the Central Limit Theorem, which states that, for sufficiently large sample sizes, the distribution of the sample mean approaches normality [Reference Kwak and Kim23]. Continuous variables were reported as means and standard deviations, and categorical variables were presented as counts and percentages. Group comparisons for specific continuous variables (i.e., mean age and BMI) were conducted using independent samples t-tests. The magnitude of the effect for continuous variables was assessed using Cohen’s d, with interpretations as follows: small (d = 0.2), moderate (d = 0.5), and large (d = 0.8) [Reference Cohen24]. Group comparisons for specific categorical variables (i.e., age group, sex, country of birth, socioeconomic level, medical comorbidities, and smoking status) were made using the chi-square test. Cramer’s V was used to assess the strength of the association, with interpretations as follows: small (V = 0.1), moderate (V = 0.3), and large (V = 0.5) [Reference Kakudji, Mwila, Burger and Du Plessis25]. For each variable, we reported information based on all available data (available-case analysis). Specifically, individuals with missing values were excluded only from descriptions and analyses involving the corresponding measure, while being retained in all others where data were complete. No imputation procedures were applied. Statistical significance was set at p<0.05. To account for multiple comparisons, Bonferroni correction was applied, and adjusted p-values are reported.
Results
Cohort description and sample characteristics
Data from a total of 1,421,510 individuals (age: 41.6±22.1, females: n = 762,558; 53.6%) were obtained from the PADRIS. Data on sociodemographic characteristics was complete. Most individuals were born in Europe (90.5%), followed by America (4%), Africa (3.6%), Asia (1.9%), and Oceania (<0.1%). Most individuals had an annual income below €18,000 (50.9%), followed by those with an income between €18,000 and €100,000 (32.3%), and those earning more than €100,000 (0.9%); the remaining group was exempt from income reporting (15.9%). Data on clinical characteristics was incomplete, with medical comorbidities, smoking status, and BMI reported for 98.7% (n = 1,403,566), 54.3% (n = 772,093), and 10.6% (n = 150,636) of the sample, respectively. Additional details are presented in Table 1. Among the individuals included in the whole sample, 473,812 (mean age: 41.6±22.1, females: n = 254,133; 53.6%) were identified as cases, and 947,698 (mean age: 41.6±22.1, females: n = 508,425; 53.7%) as controls. Within the cases, 19,917 individuals died during the follow-up period, while 453,895 individuals (mean age: 40.2±21.3, females: n = 243,877; 53.8%) remained alive until the end of the follow-up.
Table 1. Overall characteristics of the whole sample
Abbreviations: AIDS, acquired immunodeficiency syndrome; BMI, body mass index; COPD, chronic obstructive pulmonary disease; SD, Standard deviation.
Comparative analysis of sociodemographic and clinical variables
When comparing these surviving cases to the control group, significant differences were observed across all sociodemographic and clinical characteristics, except for sex (p = 0.37; Bonferroni-corrected p = 1). Variables reaching at least a small effect size in the comparison between cases and controls included cases being more likely to be born in Europe (94.7% vs. 88.3%; p<0.001; V = 0.10), being exempt from income reporting (25.2% vs. 11.6%; p<0.001; V = 0.17), and were less likely in the €18,000 to €100,000 income range (22.9% vs. 37.1%; p<0.001; V = 0.14). Cases were also more frequently ever-smokers (48.6% vs. 37.2%; p<0.001; V = 0.11) and had a higher prevalence of musculoskeletal disorders as a comorbidity (51.4% vs. 38.1%; p<0.001; V = 0.13), compared to controls. Additional details are presented in Table 2. When deceased individuals were included, no differences were observed between cases and controls in terms of age mean, age groups, or sex. However, comparisons of other sociodemographic and clinical characteristics showed differences of similar magnitude to those previously described. Further details can be found in the Supplementary Materials, eTable 1.
Table 2. Comparison between people affected by mental disorders and general population, excluding deceased individuals
Abbreviations: AIDS, acquired immunodeficiency syndrome; BMI, body mass index; COPD, chronic obstructive pulmonary disease; d, Cohen’s d; df, degrees of freedom; SD, standard deviation; V, Cramer’s V.
* p-value after Bonferroni correction for multiple comparisons.
Distribution of psychiatric diagnoses
A total of 1,547,374 psychiatric diagnoses were recorded during the follow-up period across various settings, including outpatient clinics, inpatient units, emergency departments, and primary care services, encompassing 716 unique ICD-10 codes. The most common diagnosis was “Anxiety disorder, unspecified” (F41.9; 10.46%), followed by “Major depressive disorder, single episode, unspecified” (F32.9; 5.63%), “Nicotine dependence, unspecified, uncomplicated” (F17.200; 3.86%), “Adjustment disorder with mixed anxiety and depressed mood” (F43.23; 3.76%), and “Dysthymic disorder” (F34.1; 3.29%). The complete list is provided in the Supplementary Materials, eTable 2. When considering the main diagnostic categories, “Anxiety, dissociative, stress-related, somatoform, and other nonpsychotic mental disorders” (F40–F48) accounted for the highest percentage of diagnoses at 31.38%, followed by “Mood disorders” (F30–F39; 18.83%), “Mental and behavioral disorders due to psychoactive substance use” (F10–F19; 13.59%), “Behavioral and emotional disorders with onset usually occurring in childhood and adolescence” (F90–F98; 11.07%), “Disorders of adult personality and behavior” (F60–F69; 6.51%), “Schizophrenia, schizotypal, delusional, and other non-mood psychotic disorders” (F20–F29; 6.09%), “Behavioral syndromes associated with physiological disturbances and physical factors” (F50–F59; 5.23%), “Pervasive and specific developmental disorders” (F80–F89; 4.26%), “Mental disorders due to known physiological conditions” (F01–F09; 1.81%), and “Intellectual disabilities” (F70–F79; 1.09%). Lastly, “Unspecified mental disorder” (F99) made up 0.14% of the diagnoses. The distribution of main diagnostic categories is represented in Figure 1, and the top five categories for each year between 2010 and 2019 are shown in the Supplementary Materials, eFigure 2.
Figure 1. Distribution of the 1,547,374 psychiatric diagnoses recorded during the follow-up period and grouped according to the ICD-10 categories. F01–F09, Mental disorders due to known physiological conditions, F10–F19, Mental and behavioral disorders due to psychoactive substance use, F20–F29, Schizophrenia, schizotypal, delusional, and other non-mood psychotic disorders, F30–F39, Mood Disorders, F40–F48, Anxiety, dissociative, stress-related, somatoform, and other nonpsychotic mental disorders, F50–F59, Behavioral syndromes associated with physiological disturbances and physical factors, F60–F69, Disorders of adult personality and behavior, F70–F79, Intellectual disabilities, F80–F89, Pervasive and specific developmental disorders, F90–F98, Behavioral and emotional disorders with onset usually occurring in childhood and adolescence, F99, Unspecified mental disorder.
Patterns of healthcare service use
A total of 687,051 individuals (48.3%) received at least one psychiatric diagnosis in either specialized mental health services or primary care services (mean: 5.33±6.57; range: 1–361), and 492,866 individuals (34.7%) received more than one psychiatric diagnosis.
Regarding the number of visits, a total of 76,241,239 visits (mean: 6.3±8.57; range: 0–1,227; 5.36 visits/person/year) were conducted in primary care services during the follow-up period, and 9,318,815 visits to specialized mental health outpatient services (mean: 0.78±2.93; range: 0–427) were recorded. Emergency visits amounted to 5,113,577 (mean: 0.43±1.21; range: 0–358), while psychiatric hospitalizations reached 158,972 (mean: 0.01±0.19; range: 0–51). Day hospital services accounted for 440,771 visits (mean: 0.16±1.54; range: 0–179). A total of 1,068,470 hospitalizations (mean: 0.09±0.39; range: 0–42) were conducted in general medical settings, while visits to emergency departments specifically for general medical issues totaled 1,410,515 (mean: 0.51±1.35; range: 0–318). Regarding inpatient care, the total number of days spent in psychiatric inpatient care was 1,025,679 (mean: 0.37±6.60; range: 0–366), in general medical units was 1,064,758 (mean: 0.38±3.42; range: 0–366), and in emergency-related general hospitalizations was 781,279 (mean: 0.28±2.76; range: 0–366). Additionally, 330,819 days (mean: 0.12±4.11; range: 0–366) were spent in long-term sociosanitary facilities and 431,144 days (mean: 0.15±3.52; range: 0–366) in medium-term sociosanitary facilities.
Pharmacological treatment profiles
Regarding pharmacological interventions, a total of 67,086,050 prescriptions were recorded during the follow-up period (4.72 prescriptions/person/year), covering 217 unique interventions. Figure 2 illustrates the distribution of interventions that account for at least 1% of the total prescriptions. Of these, 45,995,358 prescriptions (68.6%) were for medications with Anatomical Therapeutic Chemical (ATC) codes starting with N03 (antiepileptics), N05 (psycholeptics), and N06 (psychoanaleptics), corresponding to 89 unique interventions. A total of 368,217 individuals received at least one prescription for these interventions (mean: 124.91±137.68; range: 1–1309), with 356,671 individuals receiving more than one prescription. Further details on the medications can be found in the Supplementary Materials, eTable n.3.
Figure 2. Distribution of interventions that account for at least 1% of the total number of prescriptions.
Adjusted morbidity groups and risk stratification
Regarding the adjusted morbidity groups, data from six years were considered. The full population distribution for each year is provided in the Supplementary Materials, eTable 4, and graphically represented in eFigure n.1.
Discussion
This paper provides a comprehensive overview of the PADRIS-PRESTO cohort, which contains sociodemographic and clinical information for all individuals who accessed specialized mental health services in Catalonia during the period from 2010 to 2019, along with a representative sample of matched individuals from the general population. With 1,421,510 individuals, including 473,812 cases and 947,698 controls, this represents the largest population-based cohort focused on mental health in Catalonia and, more broadly, in Spain. This resource will enable researchers to study patterns of access to public health facilities, medication prescription patterns, diagnostic practices across different mental health settings, and other important aspects such as the role of psychiatric and physical comorbidities in the progression of primary disorders.
The descriptive analysis revealed disparities between cases and controls, and differences that reached at least small magnitude are discussed. Individuals accessing specialized mental health services exhibited greater economic vulnerability, as indicated by their higher rates of exemption from healthcare copayments and lower representation in the 18,000–100,000-euro income bracket. This socioeconomic disparity reflects what previous literature has documented extensively: the relationship between financial difficulties and mental disorders is complex and bidirectional [Reference Lund, Brooke-Sumner, Baingana, Baron, Breuer and Chandra6]. Mental health conditions may increase the risk of unemployment and subsequent financial strain [Reference Mojtabai, Stuart, Hwang, Susukida, Eaton and Sampson26], while economic instability simultaneously undermines mental well-being [Reference Lund, Breen, Flisher, Kakuma, Corrigall and Joska27]. Additionally, previous research has linked lower economic status to suicidal ideation and behavior [Reference Iemmi, Bantjes, Coast, Channer, Leone and McDaid28], increasing the public health significance of these findings. Health behavior assessments revealed a significantly higher prevalence of ever smoking among cases. Individuals with psychiatric conditions have high rates of nicotine dependence [Reference Fornaro, Carvalho, De Prisco, Mondin, Billeci and Selby29], and while cigarette smoking has declined in the general population over recent decades [30], those with severe psychiatric conditions may not have experienced the same reductions [Reference Prochaska, Das and Young-Wolff31]. These differences are particularly concerning as tobacco use may amplify existing vulnerabilities, contributing to the increased susceptibility to physical disorders commonly observed in psychiatric populations, including cardiovascular diseases and neoplastic disorders [Reference Dragioti, Radua, Solmi, Gosling, Oliver and Lascialfari32]. The interaction between mental health conditions and physical comorbidities was further evidenced in our findings. Among medical comorbidities, musculoskeletal disorders showed the largest difference between cases and controls. Conditions such as low back pain are among the leading causes of disability worldwide [Reference Vos, Lim, Abbafati, Abbas, Abbasi and Abbasifard33], and prior research has documented associations between musculoskeletal disorders and mental health conditions [Reference Heikkinen, Honkanen, Williams, Leung, Rauma and Quirk34]. Additionally, these conditions may be associated to increased BMI [Reference Heuch, Heuch, Hagen and Zwart35], a common side effect of many psychotropic drugs [Reference Wu, Siafis, Hamza, Schneider-Thoma, Davis and Salanti36], as well as a result of sedentary lifestyle and unhealthy eating habits.
Examination of diagnostic distributions within the PADRIS-PRESTO cohort reveals anxiety disorders, mood disorders, and behavioral and emotional disorders with childhood and adolescent onset as the most prevalent diagnostic categories in the Catalan population. This pattern aligns with the latest Global Burden of Disease Study [2], where anxiety and depressive disorders showed the highest disability-adjusted life-years and years lived with disability across most age groups, while conduct and anxiety disorders ranked highest during childhood and early adolescence. This distribution highlights the significant challenges that mental disorders pose to both individuals and public health systems. In fact, anxiety disorders can significantly impair cognitive performances and reduce quality of life, leading to cascading consequences including academic underachievement, underemployment, and interpersonal difficulties [Reference Craske, Stein, Eley, Milad, Holmes and Rapee37]. These outcomes generate both direct and indirect healthcare costs [Reference Konnopka and König38]. Similarly, depressive disorders are associated with cognitive impairments, reduced quality of life, higher morbidity and mortality, and adverse functional outcomes through their effects on academic and vocational achievement, economic status, and relationships [Reference Marx, Penninx, Solmi, Furukawa, Firth and Carvalho39]. From a societal perspective, the economic burden of depressive disorders has been increasing in recent years, with workplace costs representing the largest growth component [Reference Greenberg, Fournier, Sisitsky, Simes, Berman and Koenigsberg40]. It is important to highlight that the diagnostic patterns observed in this cohort reflect routine clinical practice, where variability in ICD coding among providers and settings is shaped by both clinical practices and the complexity of mental health presentations. Factors such as clinician workload, time constraints, and the nature of the healthcare setting may lead clinicians to prioritize diagnoses with clearer or more acute symptoms, potentially resulting in the overrepresentation of certain conditions and the inadequate recognition of others requiring more detailed clinical assessment.
The substantial healthcare use and burden associated with these conditions is reflected in the more than 76 million clinical visits conducted and 67 million pharmacological interventions prescribed during the 10-year study period. This extensive prescription pattern warrants careful examination, as it may indicate changing clinical practices with significant public health implications. For instance, a previous analysis of the PADRIS-PRESTO cohort focusing solely on antidepressant prescriptions in primary care revealed an increasing prescription rate that outpaced the incidence of mental disorders with established antidepressant indications [Reference Anmella, Sanabra, Primé-Tous, Segú, Solanes and Ruíz41]. This phenomenon, observed in other countries as well [Reference Gualano, Bert, Mannocci, La Torre, Zeppegno and Siliquini42, Reference Huijbregts, Hoogendoorn, Slottje, van Balkom and Batelaan43], carries additional implications beyond clinical outcomes. For example, the increased production and distribution of psychotropic drugs contributes a significant carbon footprint [Reference Argaluza, Domingo-Echaburu, Orive, Medrano, Hernandez and Lertxundi44], with environmental consequences that could potentially exacerbate mental health conditions through ecological degradation and climate change effects [Reference Radua, De Prisco, Oliva, Fico, Vieta and Fusar-Poli15], creating a concerning feedback loop between treatment approaches and environmental determinants of mental health. Additionally, the rising prevalence of anxiety and depressive symptoms may also be linked to systemic shortcomings in psychological care resources, highlighting the need for a more integrated approach [Reference Anmella, Primé-Tous, Segú, Solanes, Ruíz and Martín-Villalba18].
The PADRIS-PRESTO cohort represents the largest population-based cohort in Catalonia dedicated to mental health research, providing an important resource for researchers and public health authorities to gather insights into regional mental health trends and challenges [Reference Lousdal and Plana-Ripoll45]. By including all individuals who accessed specialized mental health care facilities during the study period alongside a matched sample, the findings derived from this cohort may reasonably be generalized to the broader Catalan population and potentially extended to other regions in Spain and countries with similar demographic and healthcare characteristics. While this cohort is rooted in the Catalan healthcare system, which provides public access to healthcare services, the structure of this system is broadly aligned with models in other European countries such as Italy, Portugal, and the United Kingdom [46]. In terms of mental health services specifically, the Catalan system is characterized by a community-oriented approach, similar to those observed in other European contexts. It also shows parallels in how local decision-makers implement protocols to guide system reforms, allocate resources, and promote collaboration between research teams and public health agencies [Reference Gutiérrez-Colosía, Salvador-Carulla, Salinas-Pérez, García-Alonso, Cid and Salazzari47]. Nevertheless, broader differences in local policies, resource availability, and population characteristics should be considered when extrapolating these findings to other settings. The PADRIS-PRESTO database opens unprecedented possibilities to explore in detail factors associated with the whole mental health spectrum and the need for specialized mental health care in each case. For policymakers, this cohort may offer evidence-based insights to inform resource allocation, service planning, and targeted intervention strategies [Reference Urban, Haller, Pieper and Mathes48]. The diagnostic distribution, for instance, highlights the predominance of anxiety and mood disorders, allowing authorities to prioritize interventions targeting these conditions that significantly impact quality of life and healthcare use [Reference Mas, Clougher, Anmella, Valenzuela-Pascual, De Prisco and Oliva49]. The detailed prescription data can help monitor adherence to clinical guidelines, identify potential prescribing disparities across demographic groups, and evaluate the economic impact of various treatment approaches. As mental health continues to gain recognition as a global public health priority, population-based resources like the PADRIS-PRESTO cohort will be instrumental in advancing our understanding of mental health conditions and developing effective interventions. The insights generated from this cohort will contribute to international comparative research, identifying both universal aspects of mental health challenges and the region-specific considerations necessary for developing more precise healthcare solutions. We already have and continuously work on each of these aspects as subprojects to enable the dissemination of these results toward these aims.
The present work has some limitations. First, our cases include individuals who had at least one contact with specialized public mental health services, but it is possible that some subjects, despite having psychiatric diagnoses provided by specialists, accessed private facilities only and therefore were not included in our cohort. As a result, our estimates may slightly underestimate the true prevalence of mental disorders or differ from patterns in populations with higher use of private services. Despite this limitation, Catalonia operates within a public healthcare system where care is provided free of charge, which minimizes the proportion of people exclusively using private services. Second, mortality data are available only for cases, while controls were selected among individuals who remained alive throughout the study period. Although this prevents us from making mortality comparisons between cases and controls, it remains possible to compare mortality outcomes across different subpopulations of cases. Third, not all data are available for the entire population, with missing values particularly common for smoking status and BMI. We addressed this by using an available-case analysis, which assumes that data are missing at random, which may not always hold true. As such, findings involving variables with high levels of missingness should be interpreted with caution. Nonetheless, these variables are still available for a substantial number of participants, and most other data, including number of visits, diagnoses, and prescription types, are available for each year of follow-up, allowing for detailed observation of longitudinal trends in this cohort. Fourth, to ensure anonymization of each participant, some data, such as age or income levels, were not provided precisely but rather as ranges. This approach, while limiting certain analyses, strengthens the ethical foundations of the database and ensures compliance with privacy regulations while still enabling meaningful population-level insights. Fifth, the nature of registry-based data collection means that clinical information was recorded in routine clinical practice by various healthcare professionals across different settings, rather than through standardized research protocols. This introduces several potential concerns, including variability in diagnostic practices, inconsistencies in the application of ICD coding, and the absence of structured or validated clinical assessment tools. Such heterogeneity may lead to classification bias and reduce the internal consistency of certain measures. These limitations are common to all large population-based datasets but should be weighed against the ecological validity of our findings, which offer valuable insights into real-world clinical practices, treatment patterns, and outcomes that may not be captured in more controlled research environments.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1192/j.eurpsy.2025.10103.
Data availability statement
The dataset used and analyzed in this study will be available upon reasonable request from the corresponding author, in accordance with the agreement between PADRIS-AQuAS and Hospital Clínic of Barcelona.
Acknowledgements
The PRESTO project was supported by the Fundació Clínic per a la Recerca Biomèdica through the Pons Bartran 2020 grant (PI046549); the Spanish Foundation for Psychiatry and Mental Health, the Spanish Psychiatric Society, and the Spanish Society of Biological Psychiatry (PI046813); the Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR) – PANDÈMIES 2020 grant (2020PANDE00081), Generalitat de Catalunya; the Ministerio de Sanidad through the Pla Director de Salut Mental i Addiccions and the Direcció General de Planificació i Recerca en Salut, Departament de Salut, Generalitat de Catalunya; and by the Ministerio de Ciencia, Innovación y Universidades (MCIN) and the Agencia Estatal de Investigación (AEI, project TED2021-131999BI00, Strategic Projects Oriented to the Ecological Transition and the Digital Transition 2021), with funding from the European Union NextGenerationEU/PRTR. MDP is supported by the Translational Research Programme for Brain Disorders, IDIBAPS. VO is supported by a Rio Hortega 2024 grant (CM24/00143) from the Spanish Ministry of Science, Innovation and Universities financed by the Instituto de Salud Carlos III (ISCIII) and co-financed by the Fondo Social Europeo Plus (FSE+). CVP is supported by a FPU grant for University Teaching Training (FPU23/01555) from the Spanish Ministry of Science, Innovation, and Universities and would like to thank the Ministry for its support. MBo thanks the support of a Marató-TV3 Foundation grant 202230-31. AB thanks the support MCIU/AEI/FEDER, UE grant (PID2022-141700OB-I00, MCIN/AEI/10.13039/501100011033); Spanish Stress Research Network, MCIN/AEI /10.13039/501100011033; and Fundación La Marató-TV3 and Catalan Government (Grant 202207). EVil thanks the support of the Catalan Agency of Research and Universities (2021SGR01065). MFF received the support of “Contratos predoctorales de formación en investigación en salud" (PFIS22) (FI22/00185) from the Instituto de Salud Carlos III (ISCIII) with European funds from the Recovery, Transformation and Resilience Plan, by virtue of the Resolution of the Directorate of the Carlos III Health Institute, O.A., M.P. of December 14, 2022, granting Predoctoral Research Training Contracts in Health (PFIS Contracts). Funded by the European Union Next Generation EU. She has also been granted by M-AES mobility fellowship (MV24/00050). SM has received support from the “Fundacio Clinic de Recerca Biomedica” through the Emili Letang-Josep Font Scholarship program. VLB is supported by a Rio Hortega 2024 grant (CM24/00074) from the Spanish Ministry of Science, financed by the Instituto de Salud Carlos III (ISCIII) and co-financed by the Fondo Social Europeo Plus (FSE+). MBi thanks the support from the Spanish Ministry of Health, Instituto de Salud Carlos III (PI20/01066), Fundació La Marató de TVV3 (202206-30-31), and the Pons-Bartran legacy (FCRB_IPB1_2023). AMA thanks the support of the Spanish Ministry of Science, Innovation and Universities (PI21/00787, PI24/00432), integrated into the Plan Nacional de I+D+I and co-financed by the Instituto de Salud Carlos III -Subdirección General de Evaluación and the Fondo Europeo de Desarrollo Regional (FEDER); the Generalitat de Catalunya and Secretaria d’Universitats i Recerca del Departament d’Economia i Coneixement (2021 SGR 01128), CERCA Programme, Generalitat de Catalunya; La Marató-TV3 Foundation grants 202235 33. JR thanks the support of the Spanish Ministry of Science, Innovation and Universities (CPII19/00009, PI22/00261), integrated into the Plan Nacional de I + D+I and co-financed by the ISCIII-Subdirección General de Evaluación and the Fondo Europeo de Desarrollo Regional (FEDER) and the Instituto de Salud Carlos III; and the Secretaria d’Universitats i Recerca del Departament d’Economia i Coneixement (2021-SGR-01128). EVie thanks the support of the Spanish Ministry of Science, Innovation and Universities (PI21/00787, PI24/00432) integrated into the Plan Nacional de I+D+I and co-financed by the Instituto de Salud Carlos III -Subdirección General de Evaluación and the Fondo Europeo de Desarrollo Regional (FEDER); the Generalitat de Catalunya and Secretaria d’Universitats i Recerca del Departament d’Economia i Coneixement (2021 SGR 01128), CERCA Programme, Generalitat de Catalunya; La Marató-TV3 Foundation grants 202234-30; the European Union Horizon 2020 research and innovation program (H2020-EU.3.1.1. - Understanding health, wellbeing and disease, H2020-EU.3.1.3. Treating and managing disease: Grant 945151, HORIZON.2.1.1 - Health throughout the Life Course: Grant 101057454 and EIT Health (EDIT-B project). DHM gratefully acknowledges the support of the Spanish Ministry of Health, Instituto de Salud Carlos III (PI049759) and the Pons-Bartran Legacy grant 2024 (FCRB_IPB1_2024). GA thanks the support of the Spanish Ministry of Science, Innovation and Universities financed by the Instituto de Salud Carlos III (ISCIII) and co-financed by the European Social Fund+ (ESF+) (JR23/00050, MV22/00058, CM21/00017); the ISCIII (PI24/00584, PI24/01051, PI21/00340, PI21/00169); the Milken Family Foundation (PI046998); the Fundació Clínic per a la Recerca Biomèdica (FCRB) – Pons Bartan 2020 grant (PI04/6549), the Sociedad Española de Psiquiatría y Salud Mental (SEPSM); the Fundació Vila Saborit; the Societat Catalana de Psiquiatria i Salut Mental (SCPiSM); and the Translational Research Programme for Brain Disorders, IDIBAPS. IG has received support from the Spanish Ministry of Science, Innovation and Universities (MCIN) (PI23/00822) integrated into the Plan Nacional de I+D+I and cofinanced by the ISCIII-Subdirección General de Evaluación and European Union (FEDER, FSE, Next Generation EU/Plan de Recuperación Transformación y Resiliencia_PRTR); the Instituto de Salud Carlos III; the CIBER of Mental Health (CIBERSAM); and Generalitat de Catalunya and Secretaria d’Universitats i Recerca del Departament d’Economia i Coneixement (2021 SGR 01128), CERCA Programme/Generalitat de Catalunya as well as the Fundació Clínic per la Recerca Biomèdica (Pons Bartran 2022-FRCB_PB1_2022). SM has received support from the “Fundacio Clinic de Recerca Biomedica” through the Emili Letang-Josep Font Scholarship program.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
GF has received CME-related honoraria, or consulting fees from Angelini, Janssen-Cilag and Lundbeck. AB is an inventor of the issued patents WO2011131693 AND WO2014064258 for ligand-conjugated oligonucleotides molecules (ASO, miRNA, siRNA). AGP has received CME-related honoraria, or consulting fees from Janssen-Cilag, Lundbeck, Casen Recordati, LCN, Rovi and Angelini. MSV has received financial support for CME activities and travel funds from Janssen-Cilag, Lundbeck and Rovi. SM has received CME-related honoraria, or consulting fees from BeckleyPsytech, Janssen-Cilag, Lundbeck, Lundbeck/Otsuka, and Viatris, with no financial or other relationship relevant to the subject of this article. MBi has been a consultant for, received grant/research support and honoraria from, and been on the speakers/advisory board of has received honoraria from talks and/or consultancy of Adamed, Angelini, Casen-Recordati, Exeltis, Ferrer, Janssen, Lundbeck, Neuraxpharm, Otsuka, Pfizer, Rovi and Sanofi, and grants from Spanish Ministry of Health, Instituto de Salud Carlos III (PI20/01066), Fundació La Marató de TV3 (202206-30-31) and Pons-Bartran legacy (FCRB_IPB1_2023). JR has received CME-related honoraria from Adamed, outside the submitted work. EVie has received grants and served as consultant, advisor, or CME speaker for the following entities: AB-Biotics, AbbVie, Angelini, Biogen, Biohaven, Boehringer-Ingelheim, Celon Pharma, Compass, Dainippon Sumitomo Pharma, Ethypharm, Ferrer, Gedeon Richter, GH Research, Glaxo-Smith Kline, Idorsia, Janssen, Lundbeck, Medincell, Neuraxpharm, Newron, Novartis, Orion Corporation, Organon, Otsuka, Rovi, Sage, Sanofi-Aventis, Sunovion, Takeda, Teva, and Viatris, outside the submitted work. GA has received CME-related honoraria, or consulting fees from Adamed, Angelini, Casen Recordati, Johnson & Johnson, Lundbeck, Lundbeck/Otsuka, Rovi, and Viatris, with no financial or other relationship relevant to the subject of this article., with no financial or other relationship relevant to the subject of this article. IG has received grants and has served as a consultant, advisor or CME speaker for the following entities: ADAMED, Angelini, Casen Recordati, Esteve, Ferrer, Gedeon Richter, Janssen Cilag, Lundbeck, Lundbeck-Otsuka, Luye, SEI Healthcare, Viatris outside the submitted work. She also receives royalties from Oxford University Press, Elsevier, Editorial Médica Panamericana. SM has received CME-related honoraria, or consulting fees from BeckleyPsytech, Janssen-Cilag, Lundbeck, Lundbeck/Otsuka, and Viatris, with no financial or other relationship relevant to the subject of this article. DHM has received CME-related honoraria and served as a consultant for Abbott, Angelini, Neuraxpharm, Ethypharm Digital Therapy, Lundbeck and Viatris.
All the other authors have no conflict to declare.
Declaration of Generative AI and AI-assisted technologies in the writing process
During the preparation of this work, the authors used ChatGPT to improve readability and language. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Open access
Comments
No Comments have been published for this article.