Key result
Integrating clinical text and echo data boosts HF detection F1 score to 86% vs ICD codes.
Why the study?
Does an advanced automated detection algorithm integrating multiple electronic health record sources improve the detection of heart failure compared to ICD codes alone in hospital inpatients?
Population
71,625 inpatients treated at the Medical Department I of the Würzburg University Hospital between 2000 and…
Comparison
Advanced automated detection algorithms… vs ICD-10 codes alone.
Design
Cross-sectional
Authors
Loading...
Estimating heart failure prevalence using ICD codes alone significantly underestimates the true burden, which can be markedly improved by using dedicated search algorithms integrating multiple electronic health record sources.
Observational (n=71,625)
No
Does an advanced automated detection algorithm integrating multiple electronic health record sources improve the detection of heart failure compared to ICD codes alone in hospital inpatients?
Absolute Event Rate: 86% vs 65%
Estimating heart failure prevalence using ICD codes alone significantly underestimates the true burden, which can be markedly improved by using dedicated search algorithms integrating multiple electronic health record sources.
Kaspar et al. (2018) conducted an observational in Heart failure (n=71,625). Advanced data warehouse algorithm (AF1) vs. ICD-based diagnosis (MICD) was evaluated on F1 score for detecting heart failure compared to a physician-based reference standard. An advanced data warehouse algorithm integrating clinical text and echocardiography data improved the F1 score for detecting heart failure from 65% to 86% compared to relying solely on ICD codes.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: