SAS® Real World Evidence Features
- Ability to collect internal or external data from point-of-care systems, electronic health records, insurance claims, patient-reported outcomes, trusted third-party data providers and others.
- Built-in capabilities for cleansing, standardizing, loading and integrating real-world data prior to using it.
- Intuitive interface for profiling, integrating and moving data stored in Hadoop without having to code.
- Process automation maps data to a common data model and refreshes cohorts and outputs as new data arrives.
- Interactive drag-and-drop cohort creation for identifying research cohorts without coding.
- Ability to run complex queries that go beyond simple subsetting to selecting criteria with multiple temporal relationships and Boolean logic.
- Progressive patient counts and graphics for showing the effects of each inclusion/exclusion criterion on the patient population to determine study feasibility.
- Reusable, editable cohort queries for improving productivity and efficiency.
- Prebuilt analytical models, including predictive models for health costs, utilization and outcomes.
- Analytics library of methodologies includes simple descriptive statistics, predictive analytics and machine learning methods.
- Ability to use third-party analytics and visualization tools on defined cohorts.
- Intuitive interface for including SAS or R programs and driving parameters.
Data visualization and exploration
- Explore, visualize and report on real-world data sources to generate insights to support decisions on treatment regimens, gaps in care, reimbursement, formulary access or support clinical development decisions.
- Predefined cohort characterization.
- Ad hoc exploration capabilities.
- Advanced visual analytics for understanding therapeutic areas, medical products or devices, and longitudinal effects of therapies on patients.
Speed and agility
- Point-and-click interface for navigating and exploring massive data sources with little or no lag time.
- High-performance analytics run calculations on millions of rows of data in seconds, rather than minutes, hours or days.