Gain insights from data analytics and take action! Learn why everyone is making a big deal about big data in healthcare and how data analytics creates action.
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
eBook - Data Analytics in Healthcare
1. Gain insights
and take action
Data Analytics in Healthcare
1
2
3
4
5
The right data
The right analysis
The right modeling
The right conclusions
The right actions
The right stuff.
2. NextGen Healthcare puts business
intelligence and analytics at your fingertips.
Harness, aggregate, analyze, and interpret patient data directly from our integrated
NextGen®
Ambulatory EHR and NextGen®
Practice Management solutions.
IDENTIFY
high-risk patients for improved
population health management
and outcomes
ENSURE
a more successful transition from
volume-based to value-based
care and payment
IMPROVE
productivity, increase
reimbursements, and accelerate
cash flow
Watch an online demo | Request a personal demo | Email us at Results@NextGen.com | Call us at 855-510-6398
Ambulatory Practice
Management
AnalyticsPopulation
Health
InteroperabilityInSight
Reporting
3. The right stuff
Data analytics done right is kind of like the Five Rights of
Medication Administration… but with a data analytics twist
Chapter 1 The right data
Chapter 2 The right analysis
Chapter 3 The right modeling
Chapter 4 The right conclusions
Chapter 5 The right actions
…and the right to ask, “Are we done yet?”
What’s the
big deal about
big data in
healthcare?
Find out in this
new eBook.
4. A new study commissioned by EMC
asked federal agencies how big data
can help them. Among the results
published recently:
The healthcare industry is chomping at
the bit for data analytics. Because the
innovative answers needed to improve
patient experiences and the health of
populations, while simultaneously
reducing costs, comes from insights,
trends, and clues hiding in big data.
The right dataand the right to get excited!
How will Big Data Help?
say Big Data will help track and
manage population health more
efficiently
say Big Data will significantly improve
patient care within the military health
and VA systems
say Big Data will enhance the ability to
deliver preventative care
63%
62%
60%
CHAPTER
ONE
5. $450 billionLast year, McKinsey Company
reported that big data could help save
American taxpayers $450 billion in
annual healthcare costs. That’s big.
6. When Dr. Karen DeSalvo took over as
head of the Office of the National
Coordinator (ONC) she said the ONC’s
agenda will launch a new discussion
about interoperability, big data use, and
patient-generated data, plus the security
required to support all three.
High-functioning health information
technology (HIT) analytics can handle
different data formats originating
from scores of different sources.
Which is why “big data” and
interoperability are two health
IT concepts you can’t ignore.
Right from the top
7. “The underpinnings of EHRs need
to be reconfigured to support
the purposes of big data.
”
Dr. Karen DeSalvo
National Coordinator for HIT
8. Please don’t. There’s no reason to. Except if
you’re not preparing properly for big data.
Regardless of your healthcare sector, your
income will be tied to your performance,
which will be evaluated with data analytics
and quality reporting.
The Meaningful Use EHR incentive
program, quality-based reimbursement
models like Patient Centered Medical
Homes (PCMHs) and Accountable Care
Organizations (ACOs), and the Physician
Quality Reporting System (PQRS) all
rely on reporting and healthcare data
analytics output.
With the transformation to value-
based care, health data analytics
is at the heart of accountable,
collaborative care.
The right to panicif you’re not prepared.
10. Ambulatory and
Inpatient EHRs
1
Physical therapy4
pharmacies3
labs/radiology/
ancillary testing
2
extended care
facilities
5
nursing homes6
medical
examiner
8
Data for healthcare
analytics comes from
diverse sources including
but not limited to:
7disease
registr ies
12. New big data sources beyond
the EHR may include genomics,
social determinants of health, and
combining data from multiple
body systems, to name a few.
13. Care for a brontobyte?
Ten to the power of 27 [1+27 zeroes] is
a brontobyte. It’s where big data is
headed. Today, big data is happening on
the planet at the yottabyte level [1024
];
one yottabyte = 250 trillion DVDs.
Today’s data scientist uses Yottabytes to
describe how much government data the
NSA or FBI have on people altogether.
In the near future, Brontobyte will be
the measurement to describe the type of
sensor data that will be generated from
the IoT (Internet of Things).
Resource:
http://www.theregister.co.uk/2012/12/04/
hp_discover_autonomy_vertica_big_data/
Analytics 101:
How big is big?
Brontobyte
This will be our digital
universe tomorrow...
1027
1024
Yottabyte
This is our digital
universe today
1018
Exabyte
1EB of data is created
on the Internet each
day - 250 million DVDs
1015
Petabyte
The CERN Large Hadron
Collider generates
1PB per second
1012
Terabyte
500TB of new data per
day are ingested in
Facebook databases
109
Gigabyte
106
Megabyte
1021
Zetabyte
1.3 ZB network
traffic by 2016
14. Data analyticsdrives population health.
Integrated HIT with data analytics
functionality. That’s your goal.
You’ll need data analytics functionality in
your HIT system to implement population
health properly… and profitably. Same
with coordinated care. Ditto for new
reimbursement models. Ditto to:
• track and manage population health
more efficiently
• enhance preventive care
• reduce per capita cost of patient care
• enhance progress in diagnostics and
medical research
• understand retail healthcare trends
• negotiate properly with payers
15. The right modelingWhat is predictive analytics?
It’s when you extract information
from existing data sets in order
to determine patterns and predict
potential future outcomes and
trends. Predictive analytics will not tell you
what will happen in the future. It helps you
forecast what might happen and includes
what-if scenarios and risk assessments.
In Gartner’s IT Glossary, among the
characteristics of predictive analytics most
important to healthcare reform is rapid
analysis of massive quantities of data (real-
time/hours/day… not months); emphasis
on the relevance of resulting insights; and
an emphasis on ease of use.
CHAPTER
THREE
16. We just covered predictive
analytics. How about descriptive
and prescriptive analytics?
Descriptive analytics is the simplest
form of analytics. It’s the easiest to do
because it’s using data to describe what
happened to patients in the past. It’s the
most common form of data analytics being
used in healthcare today.
Predictive analytics is in the middle of
this descriptive, predictive, and prescriptive
analytics triad. It has the potential to
improve healthcare delivery by analyzing
all aggregated current and historical
patient data to identify high-risk patients
and opportunities for intervention
and treatment.
Prescriptive analytics is the most
advanced of these three types of data
analytics. In healthcare, prescriptive
analytics is what’s growing clinical decision
support platforms. It goes beyond
descriptive and predictive analytics by
recommending one or more courses of
action – and including the likely outcome
of each decision or action.
What’s so great about
predicitive analytics?
BIGDATA
ANALYTICS
17. Predictive analytics can significantly increase the potential
to improve care and population health. By analyzing all
aggregated current and historical patient data, providers
can identify high-risk patients and opportunities for
intervention and treatment. Providers assess risk level based
on a particular set of health conditions and clinical decision
making to develop an effective care plan.
The goal of predictive modeling is to identify and actively
manage high-risk patients, intervene before they become
critical, and reduce or eliminate unnecessary ED visits and
hospital admissions. Each of these steps can drive down
healthcare costs, improve clinical outcomes for patients,
and promote a healthier patient panel.
Data analytics functionality
creates models used to predict
scenarios and probable trends.
The analytics triad
for healthcare.
Descriptive
analytics
Predictive
analytics
Prescriptive
analytics
18. The right conclusionsWhat’s the secret?
It’s not a secret.
It’s the patient registry.
A patient registry (also called a central data
repository or master patient index “MPI”)
is a centralized database that aggregates
patient data from multiple healthcare
providers and organizations (disparate
data sets – see page 23.
Providers and authorized users can
identify and query patient groups through
myriad segmentations and relational
database functions. For example, treatment
queries can target patients by specific
diagnosis or conditions (e.g., a risk factor)
that predispose them for a health-related
event. These patient groups are called
patient cohorts.
CHAPTER
FOUR
19. The patient registry seamlessly
aggregates multiple disparate data
sources, payer data, preventative,
and clinical quality scores to improve
clinical and financial outcomes
across the practice.
20. And why shouldn’t they? Public and private payers are using
their analytics expertise to mine data for the answers they need to
build new pay for performance provider reimbursement models.
Payers want to know everything. They monitor, track, measure,
manage, and report healthcare services, workflows, and outcomes
using state-of-the-art data analytics. And they know a healthier
population means lower costs for both payers and patients.
Payers just love, Love,
LOVE data analytics.
21. The right actionsHow do answers from data analytics create action?
Use results from thoughtful
healthcare data analytics programs
to help create innovative
approaches that enable you
to continually improve your
performance, your other providers’
performances, or the performance
of your practice or facility.
• Evaluate provider performance in managing disease(s)
• Adjust treatment plans in accordance with evidence-based guidelines
• Better understand and treat diseases that influence multiple body systems
• Identify a patient’s risk level through a hybrid data assessment – clinical, social, cultural
• Develop treatment programs that align with recommended clinical guidelines
• Engage patients in a meaningful care transition program to ensure continuity of care
• Create care coordination protocols driven by evidence-based medicine
and personalized care
• Cultivate better transition of care to help reduce readmissions and decrease costs
• Evaluate patient outcome trends to negotiate fair reimbursement for patient cohorts
• Rank yourself against your peers and national healthcare benchmarks; know where
you stand, be a savvy healthcare reform provider
CHAPTER
FIVE
22. Do more with lessAnalytics makes it happen
Like we said at the beginning of
this eBook: You want answers.
But you’re searching for them in a
healthcare setting that demands
doing more with less, every day.
Only sophisticated analytics can create
the insights and data patterns you need
to create new actions that’ll get your
toughest questions answered. It’s the way
to intelligently leverage your data.
Payers can figure out which patients are
most likely to generate the highest costs.
Providers will discover which of their
patients aren’t taking their meds. Hospital
executives can better understand the
probabilities of relapse and readmission.
That’s why more and more healthcare
professionals are interested in using big
data and analytics to prevent problems
before they occur in healthy patients.
23. “Advanced analytics [in healthcare]
allows you to be much more
sophisticated in where you
intervene and with what.
”
Dr. Bob Nease
Chief Scientist, Express Scripts
24. Are we done yet?Almost. But we need to mention interoperability.
Without interoperability, big data
and data analytics are useless.
HIT systems must achieve high degrees of
interoperability and data sharing for big
data to impact real-time clinical decision
making across the nation. Disparate
systems need to work together. Seamlessly.
We’re not there yet, but like Dr. DeSalvo’s
quote on page 6 of this eBook, the use of
big data across interoperable HIT systems
is the essence of ONC’s new 10-year plan.
(Told you it was quick!)
25. When data resides in
multiple disparate silos,
payers and providers cannot
cost-effectively aggregate,
analyze, and assess risk.
26. Hint: It’s a trick question.
Here’s a not-so-secret secret: Lots of
providers vote “yes” for data analytics and
“no” for wanting to do it. They want the
value; the new insights and answers. But
they don’t want the deep data dive for fear
of not understanding what to do or how to
do it and for wasting a lot of time trying to
figure it out.
That’s where your HIT vendor can help.
Don’t try to figure this out on your own.
You’re a medical professional, not a
data scientist.
Work with a committed, long-term HIT
partner. They’ll have a better understanding
of how to integrate and leverage data
analytics into your daily EHR and practice
management workflows.
And remember: A data analytics initiative
without an interoperability strategy is
like writing a book that no one can read.
Ask your vendor to share their long term
interoperability road map.
“Yes!” or “No!”
for data analytics?
27. 1 Gain insights and take __________________.
2 The healthcare industry is chomping at the bit for__________________ __________________.
3 Dr. Karen DeSalvo said the underpinnings of EHRs need to be reconfigured to support the purposes of __________ __________.
4 A brontobyte is ten to the power of __________________.
5 Our digital universe today is happening at the __________________ level. One of these = 250 trillion DVDs.
6 A central repository or master patient index is called a __________________ __________________.
7 Patient groups are called __________________.
8 Predictive analytics increases the potential to __________________ __________________.
9 HIT systems must achieve high degrees of __________________.
10 Data analytics without interoperability is like ____________________________________________________.
*Answer key next page
Pop Quiz!
Go ahead. Surprise yourself with how much you now know about data analytics!