gms | German Medical Science

66. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS), 12. Jahreskongress der Technologie- und Methodenplattform für die vernetzte medizinische Forschung e. V. (TMF)

26. - 30.09.2021, online

Qualityreporter – accessible harmonized data quality reporting in cohort studies

Meeting Abstract

  • Carsten Oliver Schmidt - Universitätsmedizin Greifswald, Greifswald, Germany
  • Stephan Struckmann - Universitätsmedizin Greifswald, Greifswald, Germany
  • Adrian Richter - Institut für Community Medicine, Universitätsmedizin Greifswald, Greifswald, Germany
  • Birgit Schauer - Institut für Community Medicine, Universitätsmedizin Greifswald, Greifswald, Germany

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 66. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS), 12. Jahreskongress der Technologie- und Methodenplattform für die vernetzte medizinische Forschung e.V. (TMF). sine loco [digital], 26.-30.09.2021. Düsseldorf: German Medical Science GMS Publishing House; 2021. DocAbstr. 211

doi: 10.3205/21gmds101, urn:nbn:de:0183-21gmds1016

Veröffentlicht: 24. September 2021

© 2021 Schmidt et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe http://creativecommons.org/licenses/by/4.0/.


Gliederung

Text

Introduction: Data quality assessments should be conducted and reported in a harmonized way within and across studies, yet current practice is different and transparency frequently low. An efficient implementation is particular challenging in epidemiologic cohort studies with their high degree of complexity. This talk presents a data quality reporting tool for complex observational studies with multiple examinations along an implementation example based on the Study of Health in Pomerania.

Methods: Qualityreporter was created at the University Medicine Greifswald using the statistical programming environment Stata (Version 15 or higher for full functionality). Qualityreporter consists of more than 50 related functions to enable the generation of extensive data quality reports. Qualityreporter requires two types of data: First, the study data to be assessed. Second, metadata to control the calculation of data quality indicators. The conduct of data quality assessments was guided by a recently published data quality framework with regards to the scope of data quality indicators and the workflow of their computation [1]. Spreadsheet type summaryfiles (overviews) allow for a reuse of assessment results, for example to benchmark data quality across examinations or studies.

Results: Single command calls of Qualityreporter suffice to trigger reports ranging from simple descriptive variable overviews to elaborate quality reports targeting different dimensions as well as multiple report generation with the benchmarking of results. The style and complexity of a report can be accommodated to user needs by selecting one out of several reporting templates. Selection options comprise executive summary of results, and distinct sections on data integrity (e.g. deficient formats), completeness (e.g. item missingness), consistency (e.g. inadmissible values), accuracy (e.g. measurement error related to observers or devices), as well as executive overviews on single variables of interest. The main output formats are Word and PDF with visualization elements to facilitate readability. Qualityreporter has entered routine use within quality assurance processes of the Study of Health in Pomerania (SHIP). Elements from data quality reports will be presented as part of the presentation. Qualityreporter is freely available (https://dfg-qa.ship-med.uni-greifswald.de/).

Discussion: Qualityreporter enables complex data quality reporting using single command calls. The main information to create reports is handled through accessible spreadsheet type documents (e.g. in Excel). Still, several steps are necessary prior to the generation of harmonized data quality reports, particularly related to a consistent setup of metadata. This may be the most time-consuming process for a complex study.

Conclusion: Qualityreporter generates extensive data quality reports without elaborate programming skills. Assessment results may be used to guide quality management processes during the study conduct as well as statistical analyses, once a data collection is completed.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.


References

1.
Schmidt CO, Struckmann S, Enzenbach C, Reinecke A, Stausberg J, Damerow S, et al. Facilitating harmonized data quality assessments. A data quality framework for observational health research data collections with software implementations in R. BMC Med Res Methodol. 2021;21(63).