Concordance in the recording of stroke across UK primary and secondary care datasets: a population-based cohort study
View Full Paper
Ask AI
Bookmark
Share
Why the study?
Previous work demonstrated that recording of acute health outcomes like MI may be suboptimal in primary healthcare databases, prompting an evaluation of stroke recording completeness and accuracy in UK primary care.
Population
Patients with stroke identified in CPRD primary care records and linked HES data
Comparison
Stroke recording in CPRD primary care records vs linked Hospital Episode Statistics
Design
Population-based longitudinal cohort study
Authors
Discussion
Loading...
Member takes
Implication
CPRD stroke subtype data warrant cautious use in observational research; leaves open need for improved validation in subtype-specific analyses.
Structured PICO
P
Population
72,298 adults aged ≥18 years registered at a HES or ONS-linked GP practice contributing to CPRD, with at least one stroke record in CPRD, HES, or ONS between January 1, 2004, and December 31, 2016.
O
Outcome
Concordance in the recording of stroke events between CPRD primary care records and linked HES hospital records
Reliance on a single dataset to identify stroke underestimates cases, highlighting the need for linked primary and secondary care records for accurate stroke identification in research.
Limitations
- Lack of linkage to the Sentinel Stroke National Audit Programme
- Inherent uncertainty in the diagnosis of stroke relative to other acute cardiovascular diseases
- Assumptions and compromises required when defining appropriate time scales for distinguishing multiple events in the same patient
Cite This Study
Morgan et al. (2020) studied this question.
View Full Paper
Ask AI
Bookmark
Share