Quality of Electronic Emergency Department (ED) Data: How Good Are They?

Background: Electronic ED data are currently mandated for submission to the New York State Department of Health (NYSDOH). However, the quality of the data is unknown. The purpose of this study is to estimate the accuracy of the ED data focusing on asthma & respiratory diagnoses.

Methods: Electronic ED records from October 2003 to September 2004 were randomly selected with asthma as the primary diagnosis (type1), asthma as other diagnosis (type2) or respiratory disease (excluding asthma) as primary diagnosis (type3) from the NYSDOH ED data set. Matched ED medical charts were collected from hospitals in NYS and compared to the records for diagnosis and demographic variables. Sensitivity, positive predictive value (PPV) and 95% CI were calculated.

Results: The hospital response rate was 99.5%; 1,360 charts were collected. Among type1, sensitivity was 92.6%, PPV was 94.5%; type2: sensitivity was 79.3%, PPV was 76.4%; and type3: sensitivity was 99.4%, PPV was 96.5%. Overall, of the 7.4% type1 ED records that were misclassified by the electronic ED data, the majority (6%) were classified as type2. Sensitivity of age(+ 2), gender, race, payment source, patient's zip code and disposition variables were 96.6%, 98.8%, 57.2%, 56.7%, 93.3% and 97.2%, respectively.

Conclusions: The electronic ED data appears to have good quality with a high sensitivity and PPV especially for primary diagnoses. This supports utilization of this type of data for surveillance, interventions, evaluation and research.