Predictive values of acute coronary syndrome discharge diagnoses differed in the Danish National Patient Registry

Area of Science:

• Cardiology
• Public Health
• Health Informatics

Background:

• The Danish National Patient Registry is a valuable resource for epidemiological studies.
• Accurate diagnostic coding is crucial for the validity of registry-based research.
• Acute coronary syndrome (ACS) encompasses serious cardiac events requiring timely and precise diagnosis.

Purpose of the Study:

• To evaluate the accuracy of acute coronary syndrome (ACS) diagnoses recorded in the Danish National Patient Registry.
• To determine the positive predictive value (PPV) of ACS diagnoses, including unstable angina pectoris, myocardial infarction, and cardiac arrest.
• To assess potential improvements in predictive value by stratifying diagnoses.

Main Methods:

• A cohort of participants from the "Diet, Cancer and Health" study with first-time ACS diagnoses in the Danish National Patient Registry (up to 2003) was identified.
• Medical records of 1,577 patients were reviewed against European Society of Cardiology criteria for ACS.
• Positive predictive values were calculated for overall ACS and stratified by sub-diagnosis and hospital department.

Main Results:

• The overall positive predictive value for ACS diagnoses in the registry was 65.5%.
• Myocardial infarction diagnoses showed a higher PPV of 81.9%.
• Patients diagnosed with ACS in a ward setting had a PPV of 80.1%.

Conclusions:

• ACS diagnoses in hospital discharge registries should be interpreted with caution due to moderate overall accuracy.
• For more reliable research, consider validating ACS diagnoses or focusing analyses on myocardial infarction.
• Restricting analyses to patients diagnosed in wards may enhance the precision of registry data for ACS research.