In the medical care or in the clinical research organizations obtain a large amount of clinical data. As in the process of data acquisition many people are involved, different human errors influence negatively the quality of the medical data. Such data anomalies as sloppiness, fraud introduce risks for patient safety and non-compliance.
What can be done against fraud & misconducts?
A protection against dirty data is a continuous statistical data quality assurance (Data QA), which is able to detect intentional and unintentional fraud, carelessness, and sloppiness.
It is vital to adopt zero tolerance to all suspected misconduct. Application of specialized statistical and data mining tools helps in identification of inconsistent data. Continuous acceptability & sanity data checks provide an objective quality control, highlighting suspicious defects.
How the fraud is detected by Cyntegrity’s Data QA engine?
The deviation from the norm is a signal for detection of suspicious data snippets. Cyntegrity’s statistical apparatus and rules-engine is able to identify data defects of all nature: coming from human or errors or resulted by erroneous software. Cyntegrity’s Data QA engine compares data points among each other and finds out those, which are not of the same statistical and mathematical nature. When a human artificially creates a data set, it is near to impossible to repeat the same distribution pattern as it would be data from natural sources (like, e.g., measurements).
The statistical surveillance approach and unique algorithmic engine was developed in a 2-year research project in cooperation with Institute of Bioinformatics and Mathematical Modeling from Goethe University and Fraunhofer Institute. Cyntegrity, in partnership with Widler and Schiemann AG, offers for everybody to evaluate the power of statistical data quality assurance and protect your data-sets from mistakes and anomalies.