Epidemiology Surveillance is a Lever of Change - for implementation of National Digital Health Blueprint and adoption of National Digital Health Ecosystem.
Given the vast and variable nature and quality of legacy and operational data being generated in real time which requires to be captured, any solution adopted requires to essentially ingest these disparate forms, allowing meaningful options in its use. This becomes a tall order for traditional architectures like Data Warehouses, that constrain the types of data that can be stored in them, both in terms of type and quality.
Data lakes are the leading edge and evolving architecture that can help store, share and use electronic health records and other patient data in its ever-expanding variety. Data Lakes open the possibility of taking Healthcare Analytics to its Next Level by keeping pace with the rapid growth in types and magnitude of data that needs to be harnessed and made use of. It is important here, to understand the differences between Data Lakes vs. Data Warehouses. Data Lakes store raw data on the one hand, while Warehouses store current and historical data in an organized fashion. Data warehouses are best for analyzing structured data quickly with great accuracy and transparency for managerial and regulatory purposes.
A federated data lake is essentially an architecture to store high-volume, high-velocity, high-variety, as-is, data, as it gets rolled up from State-level Health Information Exchanges. State-level Health Information Exchanges will pull in vast amounts of data — structured, semi-structured, or unstructured — in real-time, from local healthcare facilities.
Data Lakes can additionally, also ingest data from Internet-of-Things sensors, clickstream activity on public health websites, log files, social media feeds, videos and online transaction processing (OLTP) systems. Data Lakes are used for Big Data and real-time analytics. To ensure that a lake doesn’t become a mere swamp, it is very helpful to provide for a catalog that makes data visible and accessible to the business teams, as well as to IT and data-management professionals.
A healthy Data Lake requires maintenance. There are no constraints on where the data originates from, but it is a good practice to use metadata tagging, to add some level of organization to the data ingested. This will allow for relevant data to surface for queries and analysis. The value of a well-maintained data lake dealing with large volumes of disparate data, pivots on the links to metadata and ontologies. It requires consistent management of metadata, terminology management, ontology management, linked open data and modeling, and the kinds of automated algorithms that can be deployed to use these resources efficiently, to crack difficult problems such as epidemic surveillance.
The Provider eObjects can gather statistics for the appropriate authorities to allow calculation of burden of disease, epidemiological Studies for epidemic surveillance, monitor for incipient epidemics and compare outcomes across facilities. They will also play a vital role in providing epidemiological, utilization and quality data for analysis and action. Additionally, these eObjects can serve as a powerful means of recording and observing the natural history of a disease, determining the effectiveness of various clinical treatment protocols and assessing the quality of care being dispersed at different levels of care, facilitating biomedical research, while conducting analysis on disease trends.
Read the Reports:
- Data Insights Hub Report - ACCESS Health Digital
- Health Information Exchange Analytics Framework HIEAF Report - ACCESS Health Digital