Practical Guide to Clinical Big Data Sources

Ce dossier présente un ensemble de guides pratiques concernant l'intelligence artificielle et les sources de données massives dans la recherche en chirurgie

Résumé en anglais

Clinical “big data” sources are composed of patient-level data generated from interaction with and assessment of patients and are typically abstracted from institutional electronic health records (EHRs) (Box). Here, we focus on datasets not previously included in the 2018 JAMA Surgery Practical Guide to Surgical Data Sets Series. A more expansive list is available on the website of the Surgical Outcomes Club.Clinical data sources can be collated by health systems, by government agencies (eg, Surveillance, Epidemiology, and End Results [SEER] by the National Cancer Institute), by professional societies through participation that may be tied to accreditation or data access (eg, the American College of Surgeons [ACS] and the Society of Thoracic Surgeons [STS]), and by vendors that source data from multiple institutions sharing a common EHR platform (eg, Epic Cosmos). In addition, EHR-based networks such as All of Us by the National Institutes of Health and PCORnet by the Patient-Centered Outcomes Research Institute work across different EHR platforms using a common data model for reporting. These sources are distinct from claims-based datasets, which are generated from billing and insurance-related transactions rather than interactions with people, and are addressed elsewhere in this series. Depending on the variables available, clinical datasets can be used to conduct a wide variety of health services research, including observational cohort studies, comparative effectiveness research (CER), and economic evaluations (Table). Clinical data sources with prospectively collected, centrally coordinated data can also be used to conduct pragmatic clinical trials (eg, ACS National Surgical Quality Improvement Program [NSQIP], Abdominal Core Health Quality Collaborative [ACHQC]).