How 196,000 Spanish participants are helping decode heart disease risk

By Hugo Francisco de SouzaReviewed by Susha Cheriyedath, M.Sc.May 14 2025

Researchers unite 35 Spanish population cohorts to uncover why some people are more vulnerable to heart disease, and how precision medicine can change the odds.

Study: Cohort profile: the CORDELIA study (Collaborative cOhorts Reassembled Data to study mEchanisms and Longterm Incidence of chronic diseAses). Image Credit: ridersuperone / Shutterstock

In a recent study in the European Journal of Epidemiology, researchers summarise the Collaborative cOhorts Reassembled Data to study mEchanisms and Longterm Incidence of chronic diseAses (CORDELIA) study, its proposed contributions to cardiovascular science, and its methodology. The study pools data from 35 independent Spanish population cohorts (n = 196,632, mean age = 52 yrs, 54% female) to unravel the variables contributing to cardiovascular disease in this ethnic group.

The study aims to carry out the most extensive cardiovascular disease (CVD)-centric genome-wide association study (GWAS) yet in Southern Europe, with the goal of improving prediction and understanding of CVD rather than direct mitigation, within the region. CORDELIA represents a significant step forward in medicine's understanding of CVD risk factors and has the potential to spearhead advances in precision medicine across Southern Europe.

Background

Atherosclerotic cardiovascular diseases (ASCVDs) are age-associated chronic diseases caused by cholesterol, fats, and other insoluble substances that coat the lining of artery walls, eventually resulting in diminished or even halted blood flow. While the miracles of modern science have helped curb ASCVD incidence and mortality rates globally, they have also extended human lifespans, inadvertently contributing to an aging world and the consequent increase in absolute ASCVD cases.

Decades of research have attempted to unravel the risk factors contributing to ASCVD genesis and have thus far observed that genetic, environmental, and socioeconomic factors all contribute to the condition and must be closely monitored. While healthy lifestyles can mitigate ASCVD risk, risk factors demonstrate a curious locally divergent feature wherein variables strongly predicting ASCVD prevalence in one region may not necessarily play a vital role in ASCVD prevalence elsewhere.

"...justifies the creation of large population cohorts in specific regions in the world. Very few large cohorts are available in Southern European and Mediterranean populations. In Spain, the recently launched IMPaCT Precision Medicine initiative is the only cohort study of its kind, aiming to recruit approximately 200,000 participants by 2028 and complete a 10-year follow-up period by 2038."

Southern Europe is a severely understudied population from the lens of ASCVD risk (only a few studies have included samples from the region) and GWAS (scarce representation in GWAS databases). Given the inadequacies of current predictive models in predicting ASCVD risk in South European peoples, establishing a GWAS-focused ASCVD cohort in the region is essential.

About CORDELIA

The CORDELIA study aims to address this need by comprehensively assessing ASCVD risk factors (environmental, clinical, socioeconomic, genetic, etc.) in Southern European peoples by combining data from 35 already existing Spanish cohorts. It also aims to carry out the most extensive ASCVD-centric GWAS in the region and use this data to generate polygenic risk scores for at-risk populations to predict their 10-year incidence probability of contracting ASCVDs.

"CORDELIA also aims to identify subgroups of the population according to their genetic predisposition to disease that would benefit the most from the available pharmacological treatments and lifestyle modifications. All the previous results will be stratified by sex, highlighting the relevance of sex-specific differences in ASCVD, which can lead to more precise and effective prevention strategies."

Methodology

The 35 Spanish cohorts comprising the CORDELIA dataset cumulatively include 196,632 participants between 18 and 84 years old. DNA samples were available for 60% of participants, serum/plasma samples for 50%, and genotype data was already available for five cohorts, each genotyped using different platforms and reference genomes. For the remaining cohorts, standardization of genotyping is being conducted using the Axiom™ Spain Biobank Array-2 to facilitate data harmonization.

All participants were provided with questionnaires (World Health Organization [WHO] standardized surveys), which included a personalized code, study inclusion date, age, sex, region of residence, and civil status. These variables were synthesized to comprise the administrative variables/demographic data of the CORDELIA study.

Clinical outcomes were evaluated using event-based follow-up over periods ranging from 5 to 30 years depending on the cohort, with any medical emergency (or CVD event) reported using International Classification of Diseases (ICD-10 and ICD-9) codes. Datasets are further linked with those of the official Spanish National Institute of Statistics' Mortality Registry and the Spanish hospital discharges datasets, facilitating future comparisons of fatal versus non-fatal events.

Additionally, all participants underwent baseline clinical examinations, with follow-up data collected over time primarily for outcome ascertainment rather than repeated full risk factor measurement for all participants. These examinations included blood tests (following overnight fasting) and questionnaires to record recent health behaviours (sleep, exercise, smoking status, etc.).

It is important to note that while several data points, such as anthropometric, lifestyle, and socioeconomic variables, were collected, in some cohorts these were self-reported rather than measured by trained personnel, which may increase the possibility of measurement error.

Furthermore, the CORDELIA team developed harmonized algorithms for conditions such as hypertension and diabetes to ensure consistency across the different cohorts, as definitions and measurement approaches sometimes varied over the study’s 30-year recruitment window.

Preliminary findings

Thus far, the study has revealed that participants are 52 years old (average), 54% female, and 96% were born in Spain. The cohort is relatively well educated, with 20% having received a university degree. Encouragingly, 52% of participants had never smoked, but 24% reported consistent current smoking.

It should be clarified that this paper is the first joint publication of the CORDELIA meta-cohort. Previous scientific publications have made use of data from the individual cohorts that are now integrated into CORDELIA, rather than from the new, harmonized meta-dataset itself. These publications often utilized data from the individual cohorts now integrated into CORDELIA, reflecting the broad utility and scientific impact of the resource.

Conclusions

The CORDELIA study demonstrates great promise as the next step in South Europe's fight against ASCVD. On completion, it will comprise the largest ASCVD cohort in the region and the most extensive and detailed GWAS on individuals of South European ancestry worldwide. The project's in-depth data collection methodologies facilitate research in non-ASCVD-associated fields, such as COVID-19, cancers, and sleep.

"By integrating comprehensive clinical, environmental, genetic, and socioeconomic data, CORDELIA exemplifies how large-scale collaborative research can advance precision medicine."

However, researchers caution that the CORDELIA Study faces several limitations. The data originate from cohorts that used different recruitment methods, time periods, and variable definitions, leading to heterogeneity in the dataset. Not all variables were collected in the same way—some data are self-reported, and recruitment strategies varied (including population-based, hospital-based, and worker cohorts).

DNA and biological samples were available for only a subset of participants (60% and 50%, respectively). The power to detect associations for rare genetic variants is limited, and the cohort may not be fully representative of the broader Spanish or Southern European population.

The CORDELIA team has implemented statistical and methodological approaches to address these issues (such as data harmonization and mixed-effects models), but these challenges should be considered when interpreting the study’s findings.