Real-Time Analytics: The Key to Keeping Pace with Big

1
2
REAL-TIME ANALYTICS: THE KEY TO
KEEPING PACE WITH BIG HEALTH DATA
FOR FEDERAL AGENCIES USING BIG
HEALTH DATA TO PIONEER THE NEXT
MEDICAL BREAKTHROUGH, REAL-TIME
ANALYTIC TOOLS COULD BE THE
DIFFERENCE MAKER
The last decade has witnessed a veritable
explosion in the amount of data made available to
healthcare providers and federal agencies. In
2012, experts estimated that more than 500
petabytes (1 petabyte = 1 thousand terabytes) of
health data was created – and that figure is
projected to grow to 25,000 by 2020.1 The
proliferation of health data from electronic health
records (EHR), research trials, physician notes,
medical devices, insurance claims, and social
media feeds is driving a paradigm shift toward
more innovative models of health service delivery,
promising to reduce costs and improve the quality
of care for all Americans.
Big Opportunities for Big Health Data
The key challenge facing federal health agencies is
deploying the right set of tools to allow them to
sift through immense targets of structured and
unstructured data. Real-time analytics enabled by
in-memory data platforms could provide
government data scientists with the capabilities
they need to separate the signal from the noise
and derive actionable insights in areas ranging
from public health to personalized care.
Public Health Research
Budgetary investment in the current fiscal climate
proves the value that federal leaders are placing
on big health data. In July 2013, the National
Institutes of Health (NIH) announced a
commitment of $96 million over four years to
establish Big Data to Knowledge (BD2K) Centers of
Excellence aimed at supporting data science to
advance biomedical research. BD2K is a top NIH
priority according to Director Francis S. Collins,
who stated that its “goal is to help researchers
translate data into knowledge that will advance
discoveries and improve health, while reducing
costs and redundancy.”2
A number of federal agencies are already putting
big data analytics to use to improve public health
outcomes. For instance, after a series of highprofile pharmaceutical recalls, the Food and Drug
Administration (FDA) deployed Sentinel, a system
that identifies latent safety risks in post-marketed
medications by cross-referencing 125 million
Americans’ EHRs with pharmaceutical trials and
insurance claims databases.3 Between November
2012 and July 2013, Sentinel helped the FDA reissue safety guidelines that are estimated to have
saved hundreds of lives.4 In another part of the
Department of Health and Human Services,
officials from the Centers for Disease Control and
Prevention (CDC) are pioneering ways to leverage
1
4
3
“THE GOAL IS TO HELP RESEARCHERS
TRANSLATE DATA INTO KNOWLEDGE
THAT WILL ADVANCE DISCOVERIES
AND IMPROVE HEALTH, WHILE
REDUCING COSTS AND REDUNDANCY.”
- FRANCIS S. COLLINS, DIRECTOR, NIH
data from medical records, social media, and overthe-counter antiviral medication sales to track the
spread of influenza and allocate vaccine supplies
accordingly.5
But few initiatives can match the scale of the
Congressionally-launched Patient-Centered
Outcomes Research Institute (PCORI), which
aims to aggregate 30 million Americans’ complete
medical records by September 2015. By tapping
into data from millions of potential samples,
PCORI researchers hope to compare the
effectiveness of different treatments for conditions
like cystic fibrosis, multiple sclerosis, and certain
cancers.6
launch of a public cloud to house its 2.5-petabyte
centralized database, the Cancer Genome Atlas,
for further collaborative research. This data has
already proven incredibly useful; NCI data was at
the center of recent findings that cancer treatment
may be more effective when targeting the genetic
cause of mutation rather than its tissue type.8
Medical genetics can also lay the foundation for
more personalized and preventive – and therefore
less expensive – care. In an article in the New
England Journal of Medicine, leaders from NIH
and FDA described a not-so-distant future in
which patients’ entire genomes could be
integrated into their EHR, allowing doctors to
more accurately prescribe medication and replace
expensive DNA testing with a simple electronic
query.9
Genomics and Personalized Medicine
Significant technological advances have reduced
the cost of sequencing the human genome by a
factor of a thousand, creating opportunities for
both public and private research institutions to
uncover the root causes of chronic conditions like
diabetes and Alzheimer’s disease.7 However,
unlocking the human genome’s power to improve
healthcare quality and accelerate innovations in
personalized medicine will require powerful
analytic tools capable of recognizing molecular
patterns in thousands of samples, each containing
more than 3 billion base pairs of genetic code.
Applying new technologies to genome sequencing
could exponentially reduce diagnostic timelines as
well as identify targeted protocols for treatment.
One field with perhaps the greatest potential for
innovation is the mapping of genetic signatures of
aggressive cancer strains. In January 2014, the
National Cancer Institute (NCI) announced the
Evidence-Based Medicine
Garnering better insight from harnessing data can
also help federal health agencies adopt the
practice of evidence-based medicine (EBM). EBM
involves supplementing physicians’ medical
expertise with real-time access to clinical data
from hundreds of thousands of similar cases to
pinpoint the most accurate diagnosis and effective
course of treatment for each patient.10 For
example, the Department of Veterans Affairs (VA),
the largest primary healthcare provider in the
federal government, is working to develop
interoperable data models for the Veterans Health
2
6
5
Agency and Defense Health Agency to provide
doctors with a complete picture of veterans’ medical
histories spanning the length of their service
careers.11 Data shared between the VA and DoD
could be a valuable asset in cataloguing and
monitoring chronic conditions like traumatic brain
injury and post-traumatic stress disorder, allowing
physicians to better employ EBM to improve longterm outcomes.
processing.14 Coupled with an architecture that
combines predictive text analytics, spatial
processing, and data virtualization, this
translates to the ability to process information
at speeds never before possible. Petabytes of
health data produced by EHR, genome
sequencing, clinical trials, and many other
sources can be analyzed in a fraction of the
time taken with traditional analytic models.
In much the same way, the Centers for Medicare and
Medicaid Services (CMS) is collaborating with other
public and private institutions to explore the
application of advanced analytics in comparativeeffectiveness research.12 By pooling its claims
database alongside surgical records or
pharmaceutical trials, CMS can compare alternative
procedures or medications longitudinally and
evaluate relative risks and benefits before issuing
best practice recommendations.13
The big data revolution in healthcare is here
and here to stay. For federal agencies, the
move to a next-generation real-time data
platform could mean the difference between
being overwhelmed with health data and being
part of the next major medical breakthrough.
Translating Health Data into Health Knowledge with InMemory Data Platforms
The cases noted above illustrate the opportunities
available for federal agencies that adopt a forward
thinking, data-centric mindset and invest in
analytics. However, big data is only as powerful as
the platform it runs on. For true real-time analytics
capabilities, federal agencies will need the speed and
computing power supplied by in-memory database
(IMDB) technologies. Unlike traditional database
management systems, IMDBs such as SAP HANA
can store massive data sets in the form of RAM and
eliminate the bottlenecks that occur with disk
About GBC
Government Business Council (GBC), the research arm of
Government Executive Media Group, is dedicated to advancing
the business of government through analysis and insight. GBC
partners with industry to share best practices with top
government decision-makers, understanding the deep value
inherent in industry’s experience engaging and supporting
federal agencies.
About SAP
SAP sets a new standard for innovation and performance in the
federal government with integrated solutions that apply
database technology, business analytics, applications, cloud
computing, and mobility to solve their toughest challenges.
Whether they are managing core government functions or
delivering mission-critical operations, SAP brings intelligent
ideas to the government with integrated solutions that
deliver results.
33
Sources
1. John Andrews, “Clouds Roll In To Handle Stratospheric Capacity Needs”. Healthcare IT News:
September 30, 2011 http://www.healthcareitnews.com/news/clouds-roll-handle-stratosphericcapacity-needs
2. “NIH Commits $24 Million Annually for Big Data Centers for Excellence”. National Institutes of Health,
Office of Communications and Public Liaison: July 22, 2013
http://www.nih.gov/news/health/jul2013/nih-22.htm
3. Merrill Goozer, “Big Data and Big Government: Strong Federal Role Needed to Organize Productive Use
of Patient Data”. Modern Healthcare: June 22, 2013
http://www.modernhealthcare.com/article/20130622/MAGAZINE/306229988
4. “Health Technology Trends”. Emergency Care Research Institute: December 2013 p. 8
5. Julie Bort, “How the CDC Is Using Big Data to Save You From the Flu”. Business Insider: December 13,
2013 http://www.businessinsider.com/the-cdc-is-using-big-data-to-combat-flu-2012-12
6. Ariana Eunjung Cha, “Scientists Embark on Unprecedented Effort to Connect Millions of Patient
Medical Records”. The Washington Post: April 15, 2014
http://www.washingtonpost.com/national/health-science/scientists-embark-on-unprecedented-effortto-connect-millions-of-patient-medical-records/2014/04/15/ea7c966a-b12e-11e3-9627c65021d6d572_story.html
7. Ashlee Vance, “Human Gene Mapping Price to Drop to $1000, Illumina Says”. Bloomberg News:
January 15, 2014 http://www.bloomberg.com/news/2014-01-15/human-gene-mapping-price-to-dropto-1-000-illumina-says.html
8. Anthony Brino, “FDA Approves Next Gen Sequencers
in Watershed for Personalized Med”. Government
Health IT: November 20, 2013 http://www.govhealthit.com/news/next-gen-sequencers-get-ok watershed-personalized-med
9. Francis S. Collins, MD PhD, and Margaret A. Hamburg, MD, “First FDA Authorization for Next Generation Sequencer”. New England Journal of Medicine:
December 19, 2013 p. 2370
10. Peter Groves, Basel Kayyali, David Knott, and Steve Van Kuiken, “The ‘Big Data’ Revolution in
Healthcare: Accelerating Value and Innovation”. McKinsey
& Company Center for U.S. Health System
Reform: January 2013 p. 3
11. Bob Brewin, “VA Is Competing For the Pentagon’s Electronic
Health Record Contract”. Nextgov: March
13, 2014 http://www.nextgov.com/defense/2014/03/va-competing-pentagons-electronic-health record-contract/80485/
12. Vicki Fung, PhD, Richard Brand, PhD, Joseph Newhouse, PhD, and John Hsu, MD, “Using Medicare
Data for Comparative Effectiveness Research -- Opportunities and Challenges”. American Journal of
Managed Care: July 9, 2013 p. 490
13. Jennifer Bresnick, “CMS Claims, Clinical Data Team Up In New Analytics Database”. Health IT
Analytics: March 13, 2014 http://healthitanalytics.com/2014/03/13/cms-claims-clinical-data-team-up in-new-analytics-database/
14. Michael Vizard, “The Rise of In-Memory Databases”.
July 13, 2012:
http://slashdot.org/topic/datacenter/the-rise-of-in-memory-databases/
Images: National Institutes of Health and Wikipedia Creative Commons user Calleamanecer
44