Healthcare Data Management and Analytics
Healthcare data management and analytics is the area of health informatics concerned with how the data generated across clinical, administrative, and public-health systems are stored, integrated, governed, and turned into useful knowledge. It spans the engineering of repositories that consolidate heterogeneous health data, the disciplines that keep those data trustworthy, and the analytic methods that mine them to answer questions about effectiveness, populations, and operations.
Definition
Healthcare data management and analytics is the set of methods and infrastructures for organizing health-related data and applying statistical, computational, and data-mining techniques to them in order to generate evidence and support decisions at the levels of research, populations, and health systems.
Scope
This area orients the reader to the lifecycle of health data: capture and integration, storage in clinical data warehouses, governance and quality assurance, and downstream analytics for research, population measurement, and operations. It gathers five topics that move from infrastructure (warehouse design) through stewardship (governance and quality) to use (comparative effectiveness, population health, and big-data applications). It is a reference overview, not a build guide or clinical decision tool.
Sub-topics
Key concepts
- Clinical data warehouse
- Data integration and extract-transform-load (ETL)
- Common data models
- Data governance and stewardship
- Data quality dimensions
- Secondary use of clinical data
- Comparative effectiveness research
- Population health measurement
- Big data analytics
- Predictive modeling and data mining
Mechanisms
Health data originate in electronic health records, claims systems, registries, devices, and surveillance feeds. To be reusable, these heterogeneous streams are extracted, transformed, and loaded into integrated repositories such as clinical data warehouses, often mapped to a common data model so that queries are portable across institutions. Governance structures assign accountability for the data, and quality assessment evaluates dimensions such as completeness, correctness, and plausibility before the data are analyzed. Analytic methods then range from descriptive measurement to data mining and predictive modeling, with the analytic question determining which design and which data are appropriate.
Clinical relevance
The infrastructures and analytic methods described here underpin much of the secondary evidence used in health care, including quality measurement, comparative effectiveness studies, and population surveillance. Understanding them helps clinicians and researchers judge how data-derived evidence is produced and what its limitations are. This area describes how evidence and decision support are generated; it is not itself a source of individual diagnostic or treatment instructions.
History
As electronic health records and administrative systems accumulated large volumes of routinely collected data, attention shifted from primary data capture to the secondary use of those data for research and management. Initiatives to build shareable research repositories, such as the i2b2 platform, and the growth of big-data analytics in health care during the 2010s, established data management and analytics as a distinct strand of health informatics with its own concerns about quality, governance, and reproducibility.
Key figures
- David W. Bates
- Shawn N. Murphy
- Isaac Kohane
Related topics
Seminal works
- murphy-2010
- weiskopf-weng-2013
- bates-2014
Frequently asked questions
- What is the difference between health data management and health data analytics?
- Data management covers how health data are captured, integrated, stored, and governed so that they are trustworthy and reusable; analytics covers the statistical and computational methods applied to those data to produce knowledge. The two are interdependent: analytics is only as reliable as the management and quality of the underlying data.
- What does 'secondary use' of clinical data mean?
- It refers to using data that were originally collected for clinical care or billing for additional purposes such as research, quality measurement, or population surveillance. Because the data were not collected for these purposes, governance and quality assessment are central to using them responsibly.