instructions
Health Information Systems
Directions: Answer the questions below. Your answers should demonstrate critical thought and application of the
course material.
Chapter 11: Analytics, Business Intelligence and Clinical Intelligence
1. Answer the following questions about Figure 11.1
a. In your own words, describe and interpret Figure 11.1
. What does the figure tell you about adoption of analytics, Business Intelligence and Clinical
Intelligence within healthcare organizations? How do you think healthcare organizations can
overcome ba
iers to adopting more robust data analytics and BI / CI within their organization?
(answer both parts of the question)
c. Review the case example ‘A Well-designed BI/CI solution for Multi-provider OBGYN Care’. Where
does this case / situation fit within Figure 11.1? Why? (answer both – where and why)
HIS and Digital Health
2. Describe Telehealth and Telemedicine. Discuss how they are alike and how they differ.
3. Review the approaches to Digital Health in HIS described in the course material.
a. Of the approaches discussed in the course material, which one of the approaches seems most
useful in today’s healthcare world? In your opinion, why is the approach you chose most useful?
. Select two approaches. For each approach, answer the following questions:
Approach 1: ______________________ (list the approach)
» Define and describe the approach
» Give an example of the approach
» Discuss how the approach might by applied in a healthcare organizational care setting such as a
hospital, clinic or provider practice.
Approach 2: ______________________ (list the approach)
» Define and describe the approach
» Give an example of the approach
» Discuss how the approach might by applied in a healthcare organizational care setting such as a
hospital, clinic or provider practice.
Understanding Health Information Systems: For The Health Professions
© Kheng Guan Toh/ShutterStock, Inc.
CHAPTER 11
Analytics, Business Intelligence
and Clinical Intelligence
Ric Speake
LEARNING OBJECTIVES
By the end of this chapter, the student will be able to:
■ Appreciate the historical foundations of healthcare business intelligence (BI) and clinical
intelligence (CI).
■ Understand the concept, purpose, and potential of healthcare analytics.
■ Describe the myriad stakeholders who need BI and CI information to perform their jobs in the
healthcare arena.
■ Identify the methods for receiving, organizing, storing, mining, and formatting data for BI and
CI purposes.
■ Be introduced to the technological advances that may benefit health care’s dire need for useful
analytics.
▸ Introduction
Analytics is the art and science of apply-ing information, derived from data, to a given situation. This is where data
gets exciting and extremely relevant. Addition-
ally, at the time of this writing, the field of ana-
lytics is on the verge of changing the healthcare
industry for the better in major ways.
The generic definition of business intel-
ligence (BI) is a set of theories, methodolo-
gies, processes, architectures, and technologies
that transform raw data into meaningful and
useful information for business purposes. This
term—and the practice—is applied widely
throughout various industries. In line with
health care’s deserved reputation of slow adop-
tion and implementation of integrated data,
purposeful data mining lags behind other
industries. Like BI, the term clinical intelli-
gence (CI) refers to data sets turned into use-
ful information; unlike BI, CI is healthcare
specific. For a host of reasons—review Chap-
ter 10: Data—the entire healthcare industry
must accelerate the use of both clinical and
usiness data.
312
The urgent tone in this chapter reflects the
climate in the healthcare industry at large. The
Centers for Medicare and Medicaid Services
(CMS) Administrator Seema Verma wrote,
“We must shift away from a fee-for-service sys-
tem that reimburses only on volume and move
toward a system that holds providers account-
able for outcomes and allows them to inno-
vate. Providers need the freedom to design and
offer new approaches to delivering care.” Key
notions suggested by Verma are “allows them
to innovate” and “providers need the freedom
to design and use new approaches.” Healthcare
providers must lead in a transformation of
their industry. It should not be payers, the gov-
ernment, or the profit-taking suppliers who
do not actually practice patient care. Provid-
ers will never accomplish meaningful change
until they can use and value the prodigious
amount of historical (and cu
ently escalating
production of) data. The imminent adoption
of the market’s value-based financial risk par-
adigm shift, clinician compensation incentives
and disincentives, cost transparency, popula-
tion health management, and other industry
changes that are affecting all stakeholders will
force leveraging historical and real-time data.
Further, advances in software, hardware, and
data storage methods will facilitate data min-
ing. Technology innovations such as artificial
intelligence (AI), machine learning (ML),
cloud-based data storage, deep learning
(DL), and application software capabilities are
all promising developments that will assist in
health care’s quest to leverage data for the pur-
pose of analytics and innovation of care.
▸ Health Care’s BI and CI
A Definition and Overview
of BI/CI Complexity
BI handles large amounts of data and informa-
tion to help identify and develop new oppor-
tunities. Making use of new opportunities and
implementing an effective strategy can provide
a competitive market advantage and long-term
stability. BI technologies provide historical,
cu
ent, and predictive views of business oper-
ations. Common functions of BI technologies
include reporting, online analytical process-
ing, analytics, data mining, complex event
processing, business performance manage-
ment benchmarking, text mining, predictive
mining, predictive analytics, and prescriptive
analytics (Mulcahy, n.d.).
CI is emerging adjunct to BI and is
focused on the healthcare industry. With the
mandated proliferation of various forms of
electronic clinical data use, as well as indus-
try and political pressures to obtain and uti-
lize clinical measurements, BI is the obvious
technological foundation for CI. The afore-
mentioned BI definition works perfectly as
CI’s mechanics and functionality underpin-
ning. Thus, a generic definition of CI is a set
of theories, methodologies, processes, archi-
tectures, and technologies that transform raw
data into meaningful and useful information
for clinical purposes. Specific uses for health
care’s BI and CI include statistics, scorecards,
quality metrics and reporting, multipurpose
presentation dashboards, outcomes-based
compensation, longitudinal care management,
key performance indicators (KPIs), alerts,
supply-chain analysis, experience-based rating
engines, and population management.
There is no convenient singular descrip-
tion to create a cogent definition for the many
meanings and types of health care’s BI and CI.
Health care’s BI and CI are very subjective—
emember, “Health care is local,” and so is data
use. As with the practice of medicine and the
extremely competitive environment for opera-
tional software, there is a paucity of standards
that affects data, its organization, and its min-
ing. The application of an intelligence process
may be static and trivial, or it may be dynamic
and extremely complicated. Thousands of dis-
crete software applications, cellular software
tools, and data collection devices, and tens of
Health Care’s BI and CI 313
thousands of discrete data elements are used
daily by providers, payers, and related health
organizations, all of which would have their
own description for what and how data intelli-
gence is relevant for them.
There are a nearly infinite number of cur-
ent and imagined health care’s BI and CI con-
tent examples. We will review some next.
Assembling and preparing all these data
for specific subsequent use is hard, techni-
cal work. Adoption of BI/CI has been slow—
slower even than electronic health record
(EHR) adoption. (It did not hurt to have signif-
icant financial incentives from CMS to accel-
erate the latter.) It is estimated that 30 percent
of U.S. healthcare providers have implemented
data warehouses and BI/CI software to support
clinical decision making, pay-for-performance
(P4P) programs, comparative effectiveness
esearch (CER), and to track clinical outcomes
as part of healthcare quality measurement ini-
tiatives (Agosta, XXXXXXXXXXGiven the government
mandates for information from providers,
increasing sources of clinical data from EHR
systems, reimbursement increasingly tied to
value, needs to manage the health of popu-
lations, greater transparency through public
eporting of healthcare outcomes, and a host
of other circumstances, in the near term BI/CI
will become more pervasive and vendor tools
will become easier and more robust to use.
Healthcare organizations have had very
mixed experiences in regard to BI/CI. With
the support of sophisticated internal informa-
tion technology (IT) departments, the more
advanced healthcare providers have created
valuable central data repositories (CDRs),
or data warehouses, through diligent attention
to quality, organization, and maintenance of
those new data storehouses. Where nascent
CASE EXAMPLE: HYPOTHETICAL BI
A CEO wants a dashboard that is refreshed nightly. A single dashboard screen presentation includes
all of the organization’s profit and loss data, accounts receivable (A/R) status, insurance payment
denials, patient throughput volumes, prospectively booked appointments, and additional KPIs. The
CEO will be able to review the insurance payment denials graphic and hover her computer mouse
over the bar chart for the Blue Cross/Blue Shield (BCBS) payer indicator, double-click, and open the
details for all the claims without outstanding denied payments. She may then tag this detailed report
and send it to her chief financial officer (CFO) with questions and a requested response expectation.
Case Example by Ric Speaker.
CASE EXAMPLE: HYPOTHETICAL INTEGRATED BI/CI
A clinic administrator is negotiating with an insurance company over an at-risk contract for the
insurer’s largest business client’s employee population. The administrator searches the BI/CI data for
the same population (including subscriber family dependents) cu
ently seen in the practice. Then
the administrator sorts the list of patients by diagnostic procedural codes, stratifying them by age,
weight, ethnicity, como
idities, charges and payment history, IDC-10 and CPT codes (International
Classification of Diseases, Tenth Revision, and Cu
ent Procedural Terminology), and prescriptions. The
administrator produces the same summary for three other payers and large employer contracts and
subsequently produces a comparison to these existing contracts and the proposed compensation
y the new insurer.
Case Example by Ric Speaker.
314 Chapter 11 Analytics, Business Intelligence and Clinical Intelligence
BI/CI solutions were designed for accu-
acy, performance, and
eadth of purpose,
these organizations have achieved marked
successes.
Conversely, where data have been con-
tained