Ensuring high accuracy of data abstracted from patient charts: the use of a standardized medical record as a training tool

The quality and integrity of information is pivotal to the validity and reliability of inferences drawn in research. The aim of this study is to demonstrate that standardized medical records can be used as a data abstraction training tool and a quality control measure to assess the validity of medic...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of clinical epidemiology 2005-09, Vol.58 (9), p.918-923
Hauptverfasser: Pan, Larry, Fergusson, Dean, Schweitzer, Irwin, Hebert, Paul C.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The quality and integrity of information is pivotal to the validity and reliability of inferences drawn in research. The aim of this study is to demonstrate that standardized medical records can be used as a data abstraction training tool and a quality control measure to assess the validity of medical record data abstraction. Sixteen hospitals participating in a large multicenter study completed standardized data abstraction forms for three representative patient charts, one in each of the clinical areas of postoperative critical care and trauma, cardiac surgery, and repair of hip fracture. The completed forms were then compared to an established gold standard. The mean level of accuracy of the completed data abstraction forms in each of the above three clinical areas were 91.8, 77.5, and 91.5%, respectively. Missing data accounted for 19% of all discrepancies between the abstracted information and the gold standard. If queries and amendments were made by the study's coordinating center, the mean level of accuracy increased to 94.5, 82.5, and 92.9%, respectively. The present study stressed the need for quality control measures in abstracting information from medical records to ensure the accuracy and completeness of the data abstracted.
ISSN:0895-4356
1878-5921
DOI:10.1016/j.jclinepi.2005.02.004