The EHDEN Academy, in collaboration with our guest lecturer, Clare Blacketer, Janssen Epidemiology, Erasmus Medical Centre and OHDSI, is pleased to announce the launch of our next course, ‘Building confidence in real-world data: data quality reporting’, as an introduction to data quality considerations, concepts, skills and tools. The Academy is free, used in more than 60 countries, and only requires a participant to set up an account and log in.
The course, our fourteenth, precedes a more in-depth course scheduled for 2022, and supports all those wanting to gain deeper insights in the growing field of data quality in RWD/RWE, and the developing tool and methods work in the OHDSI community, alongside the OMOP common data model and standardised analytical tools.
Course Summary:
Data quality and its evaluation is an integral skill and applied method to working with real world data (RWD) in generating real world evidence (RWE) in order that there is confidence in the RWE, the analytical methods involved, and the provenance of the source data used. Brennen, et al (JAMIA, 2000) stated that data quality in and across diverse data sources (e.g., electronic health records, claims), ‘[is] the problem of ensuring the validity of the clinical record as a representation of the true state of the patient.’ In this course, using a mixture of video, audio and publications, this subject will be more deeply explored.
Starting with common issues, in section (1), such as implausibility of values, errors in values, and evaluating for instance records mapped to the OMOP common data model (CDM), we can already start to see problems. Importantly, in this example, it is not just the source data evaluation, but also the evaluation of the quality of the mapping conversion to the CDM too. Using the OMOP CDM/OHDSI example, integral quality steps are important in ensuring the quality of source data (which may be behind firewalls in a federated/distributed network model), and the standardisation via the CDM mapping, inclusive of standardised analytical tools and methods.
Quality is critically important, discussed in section (2), as evidenced by Regulatory Authorities want to use more RWE, but with concerns about the RWD being used, with increasing transparency needed on its quality to support the evidence submitted. In the course the consequences of impaired quality, bias and issues with RWD are explored further. Kahn, et al (eGEMS, 2016) provide a framework for describing data quality, which is being utilised by the OHDSI community in its own framework and development of the data quality dashboard (DQD) tool. Course participants will be guided through this section (3) of the course on the application of the framework and the underlying thinking in how you use this approach to transparently evaluate data quality, utilising the DQD.
Lastly, in the final section (4), further development of the DQD and underlying methodology will focus on better understanding how data quality impacts evidence generation, and on the quantification of data quality. In the latter aspect, being able to share the quantification transparently more widely, inclusive of in publication of evidence, will be elaborated. Further parameters, such as data quality over time, the development of a learning system, and ‘fitness for use’ are expanded on.
We hope many more of you will find this course beneficial, and will add this to your existing portfolio of EHDEN Academy courses, or if new, add more to this one.
On the behalf of the Academy Project Management Team, the very best wishes to all for the festive season, may you be safe and well, and we look forward to our expanding curriculum in the New Year:
