Validation of automated data abstraction for sccm discovery virus covid19 registry: practical ehr export pathway

SESSION TITLE: Practice Management and Administration Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: 10/10/2023 12:00 pm - 12:45 pm PURPOSE: The gold standard for gathering data from the electronic health records (EHR) has been manual data extraction; however, this requires vast resources and personnel. Automation of this process reduces resource burdens and expands research opportunities.We aimed to determine the feasibility and reliability of automated data extraction in a large registry of adult COVID-19 patients. METHODS: This observational study included data from sites participating in the SCCM Discovery VIRUS COVID-19 registry. For feasibility dataset, important demographic, comorbidity, and outcome variables were chosen for manual and automated extraction. We quantified the degree of agreement with Cohen’s kappa statistics for categorical variables. The sensitivity and specificity were also assessed. Correlations for continuous variables were assessed with Pearson’s correlation coefficient and Bland-Altman plots. The strength of agreement was defined as almost perfect (0.81-1.00), substantial (0.61-0.80), and moderate (0.41-0.60) based on kappa statistics. Pearson correlations were classified as trivial (0.00–0.30), low (0.30–0.50), moderate (0.50–0.70), high (0.70–0.90), and extremely high (0.90–1.00). RESULTS: Total cohort included 652 patients from 11 sites. The agreement between manual and automated extraction for categorical variables was almost perfect in 13 (72.2%) variables (Race, Ethnicity, Sex, Coronary Artery Disease, Hypertension, Congestive Heart Failure, Asthma, Diabetes Mellitus, ICU admission rate, IMV rate, HFNC rate, ICU and Hospital Discharge Status), and substantial in five (27.8%) (COPD, CKD, Dyslipidemia/Hyperlipidemia, NIMV, and ECMO rate. The correlations were extremely high in three (42.9%) variables (age, weight, and hospital LOS) and high in four (57.1%) of the continuous variable (Height, Day to ICU, ICU LOS and IMV days). The average sensitivity and specificity for the categorical data was 90.7% and 96.9%. CONCLUSIONS: Our study confirms the feasibility and validity of an automated process to gather data from the EHR. CLINICAL IMPLICATIONS: The surge waves of the COVID-19 pandemic and the resulting increased workload and staff pressures have made it difficult for clinical researchers to perform rapid and reliable manual data extraction from electronic health records (EHRs). Therefore, this study aimed on determining the feasibility and reliability of automated data extraction in a large registry of adult COVID-19 patients. This study found significant and near-perfect agreement between automated and manual data extraction. Two-thirds of the categorical variables had nearly perfect agreement, while all continuous variables had Extremely High or High agreement. This study validates the feasibility, reliability, and validity of an automated data collection approach from the EHR. The automation of data collection is equivalent to the gold standard method of data collection. The automated data extraction gives an additional alternative approach when large amounts of patient data need to be extracted quickly. DISCLOSURES: No relevant relationships by Vikas Bansal No relevant relationships by Katherine Belden No relevant relationships by Marija Bogojevic No relevant relationships by Rodrigo Cartin-Ceba No relevant relationships by Janna Castro No relevant relationships by Jen-Ting Chen No disclosure on file for Juan Pablo Domecq Garces No relevant relationships by Smith Heavner No relevant relationships by Vitaly Herasevich No relevant relationships by Rahul Kashyap No relevant relationships by Syed Khan No relevant relationships by Vishakha Kumar No relevant relationships by Abigail La Nou No relevant relationships by Roman Melamed No relevant relationships by Rahul Nanchal No relevant relationships by Ronald Reilkoff No relevant relationships by Devang Sanghavi No relevant relationships by Imran Sayed No relevant relationships by Nikhil Sharma No relevant relationships by Mayank Sharma No relevant relationships by Aysun Tekin No relevant relationships by Diana Valencia Morales No relevant relationships by Allan Walkey No relevant relationships by Simon Zec