The Positive Predictive Value of Pediatric Stroke Diagnoses in Administrative Data: A Retrospective Validation Study

Abstract

Background

This retrospective cohort study aimed to examine the positive predictive value (PPV) of pediatric stroke diagnoses in the Danish National Registry of Patients (DNRP) and the impact of different stroke definitions on the PPV.

Methods

We included children registered with a stroke or stroke-related diagnosis in the DNRP between January 2017 through December 2020. Two assessors reviewed medical records and validated cases according to the American Heart and American Stroke Association (AHA/ASA) stroke definition. The level of interrater agreement was examined using kappa statistics. Validation by the AHA/ASA definition was compared with validation according to the definition in the International Classification of Disease 11th version (ICD-11) and the World Health Organization’s definition.

Results

Stroke was confirmed in 120 of 309 included children, yielding an overall PPV of 0.39 (95% CI: 0.33–0.45). PPV varied across stroke subtypes from 0.83 (95% CI: 0.71–0.92) for ischemic stroke (AIS), 0.57 (95% CI: 0.37–0.76) for unspecified stroke, 0.42 (95% CI: 0.33–0.52) for intracerebral hemorrhage (ICH) to 0.31 (95% CI: 0.55–0.98) and 0.07 (95% CI: 0.01–0.22) for cerebral venous thrombosis and subarachnoid hemorrhage (SAH), respectively. Most non-confirmed ICH and SAH diagnoses were in children with traumatic intracranial hemorrhages (36 and 66% respectively). Among 70 confirmed AIS cases, 25 (36%) were identified in non-AIS code groups. PPV varied significantly across stroke definitions with the highest for the AHA/ASA definition (PPV = 0.39, 95% CI: 0.34–0.45) and the lowest for the WHO definition (PPV = 0.29, 95% CI: 0.24–0.34). Correspondingly, the incidence of pediatric AIS per 100.000 person-years changed from 1.5 for the AHA/ASA definition to 1.2 for ICD-11 and 1.0 for the WHO-definition. The overall interrater agreement was considered excellent (κ=0.85).

Conclusion

After validation, stroke was confirmed in only half of the children registered in the DNRP with a stroke-specific diagnosis. Non-validated administrative data should be used with caution in pediatric stroke research. Pediatric stroke incidence rates may vary markedly depending on which stroke definition is used.

Keywords:

• positive predictive value
• stroke diagnosis
• stroke definition
• children
• interrater reliability
• Danish National Registry of Patients

Data Sharing Statement

The data that support the findings of this study are available on request from the corresponding author (JBB). The data are not publicly available as their containing information could compromise the privacy of research participants.

Ethics

The study was approved by the Danish Data Protection Agency (J. no. P-2021-539) and the Centre for Regional Development, Health Research and Innovation, the Capital Region of Denmark (J. no. R-21048871). Registry-based studies are exempted from ethical approval in Denmark.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors have no conflicts of interest to declare.

Additional information

Funding

This study was supported by grants from the Novo Nordisk Foundation (0065517), the Elsass Foundation (21-B01-1477) and the Institute of Clinical Medicine, Copenhagen University. The funders had no role in the design and conduct of the study; in the collection, analysis and interpretation of data; or in the preparation or approval of the manuscript.