Data completeness is consistently described as essential to real-world data (RWD) quality. But how exactly can data completeness be measured and improved?
We consider data to be ‘complete’ if a patient’s record contains all of the observations made during a clinical encounter. For many research questions, however, gaining the true picture of a patient's healthcare journey requires stitching together clinical encounters over time often from multiple care sites.
In the real world, there are numerous barriers placed on the pathway to creating complete data, including:
- an inconsistency in the level of detail and format of information providers document on patient interactions
- the lack of interoperability of electronic medical record systems
- multiple factors that lead patients to receive care at different organizations like changes in insurance, geographic location, or other life factors.
Gaining the true picture of a patient's healthcare journey requires stitching together clinical encounters over time often from multiple care sites.
Recently, we set out to measure completeness, taking a deep dive into a Multiple Sclerosis research cohort. This infographic, built in partnership with The Evidence Base® explains the PicnicHealth approach to breaking down the barriers to collecting complete real-world data.
We explore how completeness is improved by:
- Capturing data across all sites of care, including outpatient and inpatient and all relevant specialities and provider types
- Abstracting data from both structured and unstructured portions of the medical records
- Leveraging machine learning techniques and expert human review to increase accuracy in data interpretations.
With patient-centric data collection we’re increasing the accuracy and reliability of real-world data and expanding the scope of research questions our partners can answer.
Download the infographic today to learn more!
