Methods and dimensions of electronic health record data quality assessment: enabling reuse for clinical research

Abstract

ObjectiveTo review the methods and dimensions of data quality assessment in the context of electronic health record (EHR) data reuse for research.Materials and methodsA review of the clinical research literature discussing data quality assessment methodology for EHR data was performed. Using an iterative process, the aspects of data quality being measured were abstracted and categorized, as well as the methods of assessment used.ResultsFive dimensions of data quality were identified, which are completeness, correctness, concordance, plausibility, and currency, and seven broad categories of data quality assessment methods: comparison with gold standards, data element agreement, data source agreement, distribution comparison, validity checks, log review, and element presence.DiscussionExamination of the methods by which clinical researchers have investigated the quality and suitability of EHR data for research shows that there are fundamental features of data quality, which may be difficult to measure, as well as proxy dimensions. Researchers interested in the reuse of EHR data for clinical research are recommended to consider the adoption of a consistent taxonomy of EHR data quality, to remain aware of the task-dependence of data quality, to integrate work on data quality assessment from other fields, and to adopt systematic, empirically driven, statistically based methods of data quality assessment.ConclusionThere is currently little consistency or potential generalizability in the methods used to assess EHR data quality. If the reuse of EHR data for clinical research is to become accepted, researchers should adopt validated, systematic methods of EHR data quality assessment.

NETIMIS: Dynamic Simulation of Health Economics Outcomes Using Big Data

“…It follows that the provenance of EHR data matters when re-using the data for research [32]. Weiskopf and Weng [16] reviewed 95 journal papers which discussed EHR data quality and their thematic analysis identified five common quality dimensions (completeness, correctness, concordance, plausibility and currency) and seven quality assessment methods including comparison between data sources and "gold standards" determined through reviews with patients or clinicians.…”

“…The team included health economists, software engineers, data miners, business analysts and clinicians from the study areas. The approach combined dynamic simulation modelling good practice [18], data mining methods (including statistical analysis, extract transform load, data cleansing), with modelling methods (care pathways [40], process modelling, economic modelling), primary data collection (including observation and interviews) [37] and data quality methods [16] (notably comparison between sources, validity checks and "gold standard" clinical review). The end of each iteration was marked by a team meeting to review the model with the focus moving from plausibility, to completeness to correctness with each session driving questions and suggested methods for the next iteration of investigation.…”

“…In health, the big data in EHRs has the potential to be re-used to reduce the cost and transform the nature of research [12]. There are major challenges however -EHRs are still relatively new, their adoption has frequently been problematic [13], there are issues associated with research access to sensitive and confidential patient data [14] and there are concerns about the quality of this data [15] and therefore its suitability for research [16].…”

Challenges in Outcome Measurement: Clinical Research Perspective

“…These databases and systems often lack standard content, structure, or format across providers, facilities, and systems, introducing substantial barriers to aggregating data for larger-scale analyses. The data in these systems are often not constructed for research-related data abstraction and may be insufficient or incomplete for research purposes, making retrieval tedious, error-prone, and costly [37,38]. For example, many of these systems require significant reprogramming to create a single, detailed data matrix of variables in columns indexed by patients in rows, as is typically needed for statistical analyses.…”

“…Practical Barriers: Rethink Current Clinical Processes Practical challenges to wide-scale outcome measurement exist, but with the continuing trend toward electronic medical records and interconnected administrative systems [27,37,38], these issues are ultimately unlikely to be the limiting factor. Some recent systems have been developed and implemented using standard measures and connecting across different data sources (eg, medical records and financial databases) [8,10].…”