Phenomic Analytics and Clinical Data Core
The Phenomic Analytics and Clinical Data Core (PACDC) provides data content expertise, support and summarization of findings for research studies to Geisinger investigators, clinicians, residents and medical student learners.
- Data development and modeling from multiple data sources
- Development of validated phenotypes and analytic variables for population cohort algorithms
- REDCap survey data capture
- Data visualization
- Study feasibility in the form of Preparatory to Research (PTR)
Data Modeling and Dataset Development
PACDC provides data support in the form of data management, programming and summarization of large databases to create analysis-ready datasets, data models and data visualization tools.
To create these data repositories that are used to participate in research networks and answer study questions, PACDC builds and develops data from our EPIC Electronic Health Record (EHR) data, claims, billing and the Phenomics Initiative Database (PIDB), our custom architected data model derived from our clinical, claims, billing source systems.
Currently, PACDC manages the development of the data models for the Health Care Systems Research Network (HCSRN) Virtual Data Warehouse (VDW) and the National Patient-Centered Clinical Research Network (PCORnet) Common Data Model (CDM).
Development of Phenotypes and Analytic Variables
Defining patient populations and their treatment and outcomes is an iterative and highly complex collaborative process between PACDC analysts, investigators and/or clinicians. Multiple clinical criteria (i.e. applicable diagnoses, labs, medications, procedures, etc.) are evaluated in accordance with clinical guidelines and content expertise when formulating these data variables and patient population specific datasets. The phenotypes and data variables are rigorously tested and validated through a variety of quality assurance techniques, which include clinician input and chart review to increase quality and accuracy.