Rationale and background
Myocarditis and pericarditis are rare but potentially serious conditions. Evidence on their incidence and presentation remains limited, with reported rates of acute myocarditis ranging from 6.3 to 8.6 per 100,000 individuals and pericarditis ranging from 3 to 32 cases per 100,000 individuals, with higher incidence observed among younger males. Overlap between myocarditis and pericarditis is frequently observed in clinical practice; Inflammatory myopericardial syndrome (IMPS) has been recently defined to describe the diverse spectrum of the potential overlap between these conditions.
More up-to-date estimates of myocarditis and/or pericarditis incidence and on their representation in terms of prodromes, diagnosis (imaging, biopsy, laboratory tests), management, and outcomes (hospitalisation and mortality) in European Real World Data (RWD) are needed to inform future potential vaccine or drug safety studies.
Research question and objectives.
The research questions are:
1. What is the incidence of myocarditis and/or pericarditis over time?
2. How are the presentation, diagnosis, and short-term management and outcomes of myocarditis and/or pericarditis represented in European Real World Data in the DARWIN EU® network?
The aim of this study is therefore to characterise the incidence, presentation, management, and outcomes of myocarditis and/or pericarditis in European Real World Data within the DARWIN EU® network.
The specific objectives of this study are:
1. To estimate the age-standardised incidence rates of myocarditis and/or pericarditis; and the crude incidence rates of myocarditis and/or pericarditis stratified by age and sex, by data source, and by calendar year.
2. To characterise the signs/symptoms, diagnostic procedures (imaging, laboratory measurements, and biopsy), treatment/s, and outcomes (hospitalisation, Intensive Care Unit [ICU] admission, length of hospital stay, mortality) of myocarditis and/or pericarditis as recorded in Real World Data within the DARWIN EU® network during the month before and up to 1 month after diagnosis (of myocarditis and/or pericarditis).
Methods
Study design
• Population-level cohort study to characterise the age-standardised population level incidence rates (to the European population) of myocarditis and/or pericarditis in the general European population from 1st January 2015, or the earliest linked data availability, until the latest data availability. Additionally, crude incidence rates will be estimated stratified by sex and age (sex*age) and further stratified by calendar year and data source. (Objective 1)
• Cohort characterisation study to describe the presentation (signs, symptoms), diagnostic procedures performed (imaging, biopsy), laboratory measurements taken, management (observable medicines), and outcomes (hospitalisation, ICU admission, length of hospital stay, mortality) as and where available in RWD during the month before and up to 30 days after the diagnosis of myocarditis and/or pericarditis. (Objective 2).
Population
The study population for Objective 1 will include all individuals observed in any of the participating data sources with at least 365 days of data availability before cohort entry (1st January 2015 or the earliest date each participant fulfils eligibility criteria). Objective 2 will include all subjects with a recorded diagnosis of myocarditis and/or pericarditis in the study period and with a minimum of 365 days of data visibility before diagnosis and no record of myocarditis and/or pericarditis in the previous 90 days.
Variables Exposure:
None.
Outcomes:
Objective 1: 1) Myocarditis with no record of pericarditis on the same date or previous 7 days, 2) Pericarditis with no record of myocarditis on the same date or previous 7 days, and 3) Myo-pericarditis, defined by the concomitant recording of myocarditis and pericarditis within 7 days. Three additional outcomes will be studied based on the combination of diagnosis and hospitalisation (on the date of diagnosis), named as: 4) Hospitalised myocarditis, 5) Hospitalised pericarditis, and 6) Hospitalised myo-pericarditis.
Objective 2: variables of interest to characterise the diagnosis and management of myocarditis, pericarditis, and myo-pericarditis (each separately), pre-existing comorbidities, and identified as potential confounders in future safety studies: signs/symptoms or comorbidities; diagnostic procedures (imaging, biopsy); laboratory measurements; medicines prescribed/dispensed in the 30 days after; and hospitalisation, ICU admission, length of hospital stay, and mortality in the 1 to 30 days after index date.
Relevant covariates:
None.
Data sources
1. Denmark: Danish Data Health Registries (DK-DHR)
2. Finland: Finnish Care Register for Health Care (FinOMOP-THL)
3. Germany: InGef Research Database (InGef RDB)
4. The Netherlands: Integrated Primary Care Information (IPCI)
5. Norway: Norwegian Linked Health Registry data (NLHR)
6. Spain: Base de Datos para la Investigación Farmacoepidemiológica en el Ámbito Público (BIFAP)
7. Spain: The Information System for Research on Primary Care (SIDIAP)
8. Spain: Valencia Health System Integrated Dataset (VID)
9. Sweden: Health Impact – Swedish Population Evidence Enabling Data-linkage (HI-SPEED)
10. The United Kingdom: Clinical Practice Research Datalink GOLD (CPRD GOLD).
Statistical analysis
For objective 1, incidence rates per 100,000 person-years and corresponding 95% confidence intervals will be reported, standardised by age based on the European Standard Population 2013. Additionally, crude incidence rates will be estimated, stratified by age * sex within each data source and calendar year. Findings from each data sources will be reported separately and pooled using a random-effects meta-analysis model.
For objective 2, characterisation analyses will include: count of individuals with a record of myocarditis, pericarditis, or myo-pericarditis per data source. Index date will be the date of diagnosis. Cohort characterisation includes diagnoses, signs/symptoms (chest pain, dyspnoea, palpitations/arrhythmia), diagnostic procedures (chest/thorax/cardiac imaging, biopsy), laboratory measurements, and medicines recorded in the 30 days before to 30 days after index date. Hospitalisation, ICU admission, and mortality (all-cause) will be ascertained from day 1 to 30 days following index date using cumulative incidence functions.

Obesity is a rapidly growing global public health challenge imposing significant clinical and economic burdens on healthcare systems. Excess body weight is a well established driver of chronic complications. The burden of obesity-driven chronic complications translates into escalating healthcare costs and healthcare resource utilisation (HCRU) as Body Mass Index (BMI) increases.
Objectives:
Primary objectives:
1.To estimate, for each calendar-year (from 2021 to 2025), the annual HCRU, annual HCRU-related costs, annual incidence of obesity-related complications (ORCs), and annual all-cause mortality, stratified by the baseline BMI category of the calendar-year of interest, and to compare these outcomes across BMI categories.
2.To estimate, for each follow-up year (up to 5 years), the annual HCRU, annual HCRU-related costs, annual incidence of ORCs, and annual all-cause mortality, stratified by BMI category (registered at baseline and, if available, updated annually), and to compare these outcomes across BMI categories.
3.To estimate cumulative HCRU and HCRU-related costs over up to 5 years of follow-up, stratified by baseline BMI category, and to compare these outcomes across BMI categories.
Secondary objectives:
1.To estimate the incremental annual HCRU, annual HCRU-related costs, annual incidence of ORCs, and annual all-cause mortality according to knee osteoarthritis (OA) status (yes/no) among adults with baseline overweight or obesity (combined and separately), and to compare these outcomes between knee OA status.
2.To estimate the incremental annual HCRU, annual HCRU-related costs, annual incidence of ORCs and annual all-cause mortality according to overweight/obesity status (yes/no) among adults with baseline knee OA, and to compare these outcomes between overweight/obesity status.
Exploratory objectives:
1.To describe frequency and patterns of BMI category transitions (e.g., BMI improvement category: transition from obesity to overweight) over follow-up among adults with overweight or obesity.
2.To estimate annual HCRU, annual HCRU-related costs, and annual incidence of ORCs during follow-up among adults with overweight or obesity, stratified by BMI category transitions (e.g., BMI category improvement: transition from obesity to overweight) and to compare these outcomes across BMI category transitions (e.g., using stable BMI as the reference group).
3.To estimate annual HCRU, annual HCRU-related costs, and annual incidence of ORCs during follow-up among adults with overweight or obesity, stratified by BMI unit percentage change (e.g., =20% decrease in BMI) and to compare these outcomes across BMI percentage change categories.
4.To estimate annual HCRU, annual HCRU-related costs, and annual incidence of ORCs during follow-up among adults with overweight or obesity who achieved weight loss (defined as =5% BMI unit reduction), stratified by weight regain status (i.e., maintained the weight loss or continued to lose weight versus regained weight), and to compare these outcomes across weight regain status using maintained weight loss or further weight loss as reference.
Study design:
This is an observational retrospective multi-country cohort study using existing healthcare databases (i.e. Electronic Medical Record [EMR], Electronic Health Record [EHR], and claims data). Participating countries will be from e.g. Europe, Asia-Pacific, and other regions, with the final database selection to be confirmed based on data availability and regulatory feasibility.he overall study period will span from 01 January 2018 to 31 December 2025 or until the latest year with data available in each database (including the look-back period and follow-up); while the study period will span from 01 January 2021 to 31 December 2025.
The source population will include adults with =1 BMI record during the overall study period, with the first BMI record being at age 18 years or older. Multiple cohorts will be derived from the source population to address each study objective.
Index date will be defined separately for each cohort.
Source population:
Inclusion Criteria: individuals meeting all inclusion criteria will be included in the study:
• 1 BMI record during the overall study period
• First BMI record recorded at age 18 or after
Exclusion Criteria: individuals meeting the exclusion criterion will be excluded from the study:
• Missing gender
• History of bariatric surgery at the time or any time prior to the first BMI record in the overall study period.
Cohorts created for each objective are:
• Cohorts for the primary objective 1. Five cohorts, 1 for each calendar-year (from year 2021 to 2025), will be derived from the source population. These cohorts will include adults with minimum available BMI data (see Section10.4.1).
• Cohort for the primary objective 2 and 3. Derived from the source population, and it will include adults minimum available BMI data (see Section 10.4.2).
• Cohort for the secondary objective 1. Derived from the cohort for the primary objective 2, and it will include adults with minimum available BMI data (see Section 10.4.2) and baseline overweight or obesity.
• Cohort for the secondary objective 2. Derived from the source population, and it will include adults with minimum available BMI data (see Section 10.4.2) and baseline knee OA.
• Cohort for the exploratory objective 1. Derived from the cohort for the primary objective 3 and will include adults with minimum available BMI data (see Section 10.4.2) and overweight or obesity at baseline.
• Cohort for the exploratory objectives 2 and 3. Derived from the cohort for the exploratory objective 1 and will include adults with minimum available BMI data (see Section 10.4.1) with baseline overweight or obesity.
• Cohort for the exploratory objective 4. Derived from the cohort for exploratory objective 1 and will include adults with minimum available BMI data (see Section 10.4.2) with baseline overweight or obesity followed by =5% BMI reduction.
In all cohorts, patients with a pregnancy record and patients with =1 condition related to potential unintentional weight loss (e.g., active malignancy excluding benign skin cancer, HIV/AIDS, and limb amputation) in periods predefined for each study objective will be excluded.
Variables:
Outcome variables: HCRU counts, including hospitalisations, emergency department visits, outpatient and specialist visits, and prescriptions; HCRU-related costs (total and component-specific); all-cause mortality; incidence of pre-specified ORCs, including type 2 diabetes, cardiovascular disease (heart failure, myocardial infarction, stroke), chronic kidney disease, obstructive sleep apnoea, knee OA, and metabolic dysfunction-associated steatohepatitis; cumulative ORC count; BMI category transitions; and weight regain status.
Grouping variables: BMI category defined according to WHO, Japan Society for the Study of Obesity, and Working Group on Obesity in China classifications, as applicable per data source; knee OA status (yes/no) defined from diagnosis records; BMI percentage change; and weight regain (=5% BMI increase following initial =5% reduction).
Demographic and clinical variables: Age, sex, index year, baseline ORCs, obesity management medications (including GLP-1 receptor agonists, orlistat, and bariatric surgery), concomitant treatments, and cardiometabolic parameters (HbA1c, blood pressure, lipids — captured for descriptive characterisation only, not included as regression covariates as they represent mediators on the causal pathway from obesity to healthcare costs).
Data sources/Data Collection:
Existing EMR, EHR, and claims databases from the ENGINE-CMH federated network will be included. Participating countries will span Europe, Asia-Pacific, and other regions, with the final database selection confirmed based on data availability and regulatory feasibility.
Statistical Analysis:
Analyses will follow a hierarchical framework aligned with the study objectives. Details and analytical approaches will be specified in the statistical analysis plan (SAP). For all objectives, baseline characteristics will be summarised by BMI category. Incidence of ORCs and all-cause mortality will be estimated using person-time methods and compared by BMI classes recorded at baseline (or updated at different follow-up timepoints, as per study objective).
For the cross-sectional burden analysis (primary objective 1), annual HCRU counts will be modelled using Poisson or negative binomial regression (depending on overdispersion), and HCRU-related costs using generalised linear models with Gamma family and log link. BMI category, calendar-year, and their interaction will be included to estimate year-specific cost and HCRU ratios by BMI class relative to normal weight
For the longitudinal incidence-based analysis (primary objective 2), BMI will be treated as a time-varying categorical exposure, updated at annual landmarks using the most recent available measurement. Person-year-level data will be analysed using generalised estimating equations to account for repeated measures within patients. For cumulative analyses (primary objective 3), baseline BMI will be held fixed and total follow-up costs and HCRU aggregated over up to 5 years, with inverse probability of censoring weights (IPCW) as the primary method to account for incomplete follow-up.
Secondary objectives will assess the reciprocal burden of knee OA and obesity using interaction models (knee OA × BMI category) within the respective populations, and population attributable fractions will quantify the population-level impact under stated assumptions.
Exploratory objectives will employ a landmark design at 3 years to classify BMI transitions, percentage BMI change, and weight regain status prior to assessing post-landmark HCRU, costs, and ORC incidence. A continuous-time multi-state Markov model will describe BMI category transition patterns. IPCW will be applied as a sensitivity analysis for annual outcomes to address potential informative censoring. All models will be adjusted for pre-specified demographic and clinical covariates (age, sex, index year, and database-specific baseline comorbidity indicators). This study is sponsored by a private organisation and conducted in collaboration with multiple independent data partners. Analyses will be carried out locally within each site’s Observational Medical Outcomes Partnership (OMOP)–standardised database using a federated analytical framework, ensuring that only aggregate, non identifiable results are shared centrally.

Rationale and background
The DARWIN EU® network consists of heterogeneous real-world data sources with varying population characteristics, which may affect study feasibility, analytical approaches, and the interpretation and generalisability of results. Understanding clinical heterogeneity is important when conducting studies in the network.
Research question and objectives
Research questions
The aim of this study is to characterise the DARWIN EU® network in terms of demographics and common clinical conditions.
Objectives
1. To describe age and sex of the populations under observation in the DARWIN EU® data sources.
2. To develop reusable standardised phenotypes for selected common clinical conditions.
3. To estimate the period prevalence of selected common clinical conditions.
4. To describe the median age at first record and sex distribution of selected common clinical conditions.
Methods
Study design
A descriptive cohort study will be conducted.
The index date will be the start of observation during the study period (1 January–31 December 2023) for Objectives 1 and 3 or the date of first record of the clinical condition for Objective 4. Individuals will be followed until the end of the study period for Objective 3.
Population
Inclusion criteria:
• Under observation at any time during the study period.
• At least 365 days of available history before index date, except for children aged <1 year at the start of observation during the study period and for hospital-based data sources (APHM, PGH, SUCD, EMDB-ULSEDV, EMDB-ULSGE, and ULSM-RT). Exclusion criteria: • Missing information on age or sex. Variables Outcomes: • Selected cancers and conditions across diseases of the circulatory, digestive, endocrine, genitourinary, and respiratory systems, haematological, infectious, mental health, musculoskeletal, neurological, and skin conditions. Relevant covariates: • Age • Sex Data sources This study will be conducted using routinely collected data from 19 data sources within the DARWIN EU® network, covering 13 European countries. These include: i) four nation-wide primary care data sources; ii) four nation-wide data sources covering primary and secondary outpatient care, as well as hospital inpatient care; iii) three regional data sources covering primary and secondary outpatient care, as well as hospital inpatient care; iv) six hospital-based data sources; v) one regional primary care data source linked to hospital discharge data; and vi) one biobank including primary and secondary outpatient care, as well as hospital inpatient care data. All data were a priori mapped to the OMOP CDM. Statistical analysis All analyses will be conducted using Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) mapped data. A minimum cell count of 5 will be used when reporting results, with any smaller count reported as “<5”. Objectives 1 and 4: Age will be summarised using medians and interquartile ranges at the index date, and sex using proportions. The age distribution will be presented as proportions across predefined age categories and stratified by sex. Objective 3: Annual period prevalence will be defined as the total number of individuals in observation during the study period with the clinical condition of interest, divided by the entire eligible population in observation during the same period. Binomial 95% confidence intervals will be calculated.