Written by Harshad Patil for CHIME CIO Connection.
Harshad is Senior Consultant - Healthcare Informatics at CitiusTech.
Driven by healthcare reform and consumer driven
technology (mobile health), we are seeing an exponential growth in the volume
and types of healthcare data. Healthcare decision makers have the opportunity
to enhance clinical outcomes by unlocking this data to generate actionable
intelligence. Analysts point out that the value derived from big data could
potentially be worth $300 billion per year in the US healthcare industry.
However, in comparison to other industries where big
data has already become mainstream, the healthcare industry is still behind the
curve, as a majority of healthcare systems are not yet equipped to handle
real-time data processing. To effectively make use of big data for clinical
analytics, healthcare organizations today need to address a few key challenges:
With a large part of the healthcare industry, both on
the provider and payer side, still using siloed, legacy healthcare systems, they
are limited in their ability to manage, integrate and process a wide variety of
structured and unstructured data. The primary challenge for organizations is to
make current systems ready for future use, and be able to handle big data requirements,
in terms of scale as well as the variability for structure and format.
on Unstructured Data
Organizations need to invest in sophisticated BI and
analytics that can combine clinical data with unstructured data (e.g. clinicians
notes), non-clinical data (patient responses on a portal), or even analog data
(diagnostic results, medical images).
The technology should be able to process extremely
high volumes of data (as much as 5000 events / second) in real time. Also,
traditional clinical BI/analytics tools have generally focused around
structured patient and population data, and use predefined measures or KPIs for
benchmarking and reporting. With unstructured data, users have multiple new
analytics capabilities, e.g. trend analysis to identify clinical or lifestyle patterns
within identified populations, locating "hotspots” and "outliers” to help
providers control avoidable costs and ADEs. Decision makers will have to adopt a
different approach on how they leverage reporting/analytics tools.
Data Use Cases
Big data has the potential to impact multiple areas
across the continuum of care, such as improving clinical, financial and operational
decision making, reducing cost of care, greater savings under federal and pay-for-performance
incentive programs. Organizations would need to identify and prioritize
use-cases based on their near-term and long-term objectives, and develop responsive
architectures that can be easily used by identified stakeholders with basic technological
There is no quick-fix solution to address these and
many other challenges that the healthcare industry needs to address to maximize
value from the healthcare data deluge. Organizations need to develop time phased, scale-as-you
grow, cross enterprise strategic solutions, to integrate, manage, analyse and draw
actionable intelligence from big data, to effectively improve care quality and derive
significant monetary benefits.