Harmonization of the Definition of Sudden Cardiac Death in Longitudinal Cohorts of the ESCAPE-NET Consortium
Creators
- 1. Department of Cardiology, University of Copenhagen, Rigshospitalet, Copenhagen, Denmark
- 2. Department of Public Health, Amsterdam Public Health Research Institute, University of Amsterdam, Amsterdam, The Netherlands
- 3. Department of General Practice, Amsterdam University Medical Center, Location VUmc, Amsterdam, The Netherlands
- 4. Department of Cardiology, Amsterdam UMC (location AMC), University of Amsterdam, The Netherlands
- 5. Université de Paris, INSERM UMRS-970, Paris Cardiovascular Research Center, Paris, France
- 6. Istituto Auxologico Italiano IRCCS, Center for Cardiac Arrhythmias of Genetic Origin and Laboratory of Cardiovascular Genetics, Via Pier Lombardo 22 - 20135 Milan, Italy
- 7. Cedars-Sinai Medical Center, Los Angeles, California, USA
- 8. Department of Epidemiology and Data Science, Amsterdam University Medical Center, Location VUmc, Amsterdam, The Netherlands
Description
Background
The burden of sudden cardiac death (SCD) in the general population is substantial and SCD frequently occurs among people with few or no known risk factors for cardiac disease. Reported incidences of SCD vary due to differences in definitions and methodology between cohorts. This study aimed to develop a method for adjudicating SCD cases in research settings and to describe uniform case definitions of SCD in an international consortium harmonizing multiple longitudinal study cohorts.
Methods
The harmonized SCD definitions include both case definitions using data from multiple sources (e.g. autopsy reports, medical history, eyewitnesses) as well as a method using only information from registers (e.g. cause of death registers, ICD-10 codes). Validation of the register-based method was done within the consortium using the multiple sources definition as gold standard and presenting sensitivity, specificity, accuracy and positive predictive value (PPV).
Results
Consensus definitions of 'definite', 'possible' and 'probable' SCD for longitudinal study cohorts were reached. The definitions are based on a stratified approach to reflect the level of certainty of diagnosis and degree of information. The definitions can be applied to both multisource and register-based methods. Validation of the method using register-information in a cohort comprising 1335 cases yielded a sensitivity of 74%, specificity of 88%, accuracy of 86%, and PPV of 54%.
Conclusions
This study demonstrated that a harmonization of SCD classification across different methodological approaches is feasible. The developed classification can be used to study SCD in longitudinal cohorts and to merge cohorts with different levels of information.
Files
1-s2.0-S0002870321004816-main.pdf
Files
(816.3 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:c334dd662a00de3a3bd0838904032f80
|
816.3 kB | Preview Download |