Complexities and challenges for healthcare data

CHAPTER 9
Data
LEARNING OBJECTIVES
By the end of this chapter, the student will be able to:
Appreciate the myriad and competing sources for healthcare data.
Understand the size and exponential growth in healthcare data.
Describe the complexities and challenges for healthcare data.
Identify methods and priorities that health care may adopt to
leverage both clinical and business data.
INTRODUCTION
The study of healthcare data has two dimensions. The first dimension is
that of technical, pedantic knowledge. This technical dimension is
paramount to data organizational bias, security, integrity of data, and
system performance. Without highly skilled personnel, a poor
infrastructure for data will prevent the true utility of these data from being
realized.
The second dimension is the need to imaginatively understand data and
its properties for analysis as the foundation for organizational growth. This
dimension is less technical and more connective in nature. It is the
responsibility of a nontechnical participant in business intelligence
(BI)/clinical intelligence (CI) to contribute to the translation of data for
386
technical professionals into the form needed for analytics. Finding ways
for data to support their respective needs is a creative and rewarding
process.
Health care has a history of reinventing itself to adapt to
macroeconomic and political influences. It is now adopting new
technologies and care delivery methods that are profoundly increasing the
volumes of data to levels heretofore unimagined. The ultimate goal is to
leverage these data to provide more affordable and quality care. For the
purposes of this chapter, we will define data (with a tweak for health care)
as follows:
Base-level computer information that may consist of
numerical or word elements, facts, values, or combinations of
stored information that can be either qualitative or
quantitative, and from which knowledge is derived and
decision making may be made better and more logical.
Implicit in this definition is the message that data can be valuable. If an
industry may make better decisions based on data, then it is incumbent
upon that industry to capitalize on its use. In the case of health care, data
have always had some value, but due to the recent acceleration of
development and adoption of technology, massive changes in how data are
sourced, assimilated, integrated, stored, and mined as assets are beginning
to significantly, and positively, affect the healthcare industry. The
landscape and complexity of health care have changed due to national and
state economic politics, operational introspection from both clinicians and
business leaders, and rapid evolution of technology in medicine.
Health cares data may take varying forms. Simple data such as patient
demographics, including date of birth, sex, and so forth, are generally
formatted and similarly available across computer software systems. (Note
that this is a generalization; later we will discuss issues of data
normalization, including varying structures in different systems for the
same data element and referential integrity, including incorrect and/or
inconsistent spelling and referencing for the same name, code, and other
seemingly unambiguous terms that are simply entered improperly.)
Conversely, clinical datasuch as lab results, previous or ongoing
conditions, and so onmay be of a confusingly disparate nature requiring
normalization and, as a result, may be more difficult to access. For
example, an older medical office practice management system may have
387
1000 unique data elements, whereas a contemporary electronic health
record solution may have more than 5000 discrete data elements. Trying to
merge data from these two disparate data sources would be challenging
indeed.
The vast sums of data collected over the years in health care have
principally served the internal and operational purposes of separate
healthcare organizations; that is, these myriad players have processed the
data for their patients and employed administrative functions for their own
purposes. With the new political attention and intense focus on
consistently lowering costs, however, the healthcare industry is now
looking introspectively into how it might use these data to perform better.
With better dataand while remaining aware of the political
consequencesthe healthcare industry can measure quality improvements
that were previously only theoretical. In this way, healthcare organizations
can make an effort to better analyze both business and clinical metrics.
In comparison to other industries, health care is great at adopting
acronyms and buzzwords but often not so great at achieving real
movement. The industry is notorious for exceedingly slow adoption of
new health information technology (IT) methods. Yet, clinical tools and
methods and todays cellular phone technology are to some extent eroding
this relunctance to change. In relation to data, healthcare organizations are
now talking about Big Data and velocity, volume, and variety (the
Three Vsmore on that later). Major companies with expertise in
working with massive data sets in other industries, such as mobile
communications, retail, petroleum, and banking, are circling around to
health care and providing previously unavailable data storage and access
methods. However, the healthcare environment is complicated by heavy
federal and state regulation. Just recently, Amazonone of the leading
vendors in this businesselected to depart from healthcare data storage
due to the data security liability issues associated with the Health
Insurance Portability and Accountability Act (HIPAA) and the need for
high degrees of security, privacy, and confidentiality of protected health
information (PHI). Soon after, however, Amazon reentered the healthcare
data storage market, but only after performing the technical and
organizational work necessary to meet the healthcare data management
requirements described earlier.
Compared to other industries, health care has more than its fair share
388
of challenges in leveraging the incredible asset represented by data. The
notion that health care is local is most assuredly true in many ways; in
regard to data, it is profoundly true. By this, we mean that not only are
people treated in their neighborhood, but also that their neighborhood
healthcare organization approaches and records their data differently than
a healthcare organization in another region might. Healthcare
organizations do the same thing differently, including generation,
storage, accessing, and mining of data elements. While business-oriented
data have significant overlap and commonality, clinical data are siloed and
not communicated between the hundreds of commercially dissimilar
software systems. In other words, the emergency room on 5th Avenue may
collect and catalog its data differently than the surgery center on 8th
Avenue, or the clinic on 45th Avenue, and so on. However politically
convenient, the expectation of aggregating data for public health purposes
or having electronic health records communicate with one another
whether across the nation or just across townis fairly spurious and
remains far from reality.
The current state of data within health care is problematic. Many
theories have been advanced regarding data safety matters, data organizing
methods, medical lexicon standardization, and the ways in which access
and availability should work. Overlay this with arguably the most stringent
privacy legislation within any industry, and the result is an area fraught
with intense intellectual challenge and opportunity. With improved
methods for data generation, capture, storage, analysis, and other
competing practical uses, proper data management may produce
tremendous clinical value and business process efficiencies to an industry
desperately in need of both.
DATA SOURCES
The 2011 U.S. gross domestic product (GDP) was approximately $17.6
trillion: Health cares share was approximately 18% of this amount and
growing proportionately faster than the other larger segments of the U.S.
economy. It is forecasted that health care may represent 20% of the GDP
by 2020.1 Commensurate with these statistics, the data generated and the
data necessary for the operation of the U.S. healthcare industry are
proportionately outpacing other industries for a host of reasons.
389
At a macro level, an estimated 2.5 quintillion bytes of new data is
created daily worldwide across all industries and uses. Further, 90% of all
the digital data created worldwide has been created within the last 2 years.2
Mobile technology takes the credit for most of this proliferation of data.
Smartphones, tablets, and other mobile devices are growing in exponential
popularity within health care and are being used by clinicians, patients,
and other industry players. In addition, countries with little traditional
infrastructure (e.g., India) have jumped into mobile and wireless
computing, skipping the development of many of the more traditional
types of system platforms requiring a more significant infrastructure
investment that other countries have invested in over the last 25 years.
As a result of several converging trendsthe adoption of mobile
devices, the governments financial incentives fueling proliferation and
expansion of physician adoption of electronic health records (EHRs),
imaging technologies, and privacy legislationdatas preeminence is
finally an imperative for the entire healthcare industry. Much of this
movement has compelled healthcare organizations to adopt a got to do
something attitude versus the older ambivalence of want to do
something.
Historically, most of the focus on healthcare data was on information
in revenue-cycle (i.e., billing and claims) systems along with bits and
pieces of fragmented clinical data (Exhibit 9.1). Living within modified
managed care and fee-for-service environments without an electronic
clinical record allowed providers and payers to principally survive on
claims and billing data as means to assess their respective financial status,
acumen, and success. Under the provisions of the Affordable Care Act
(ACA), however, economic risk is being rebalanced with financial
incentives for providers to implement EHR systems and to be measured
and compensated for improved, calculable patient outcomes.
Within health care today, hundreds of discrete revenuecycle and health
record operational systems or data sources are used among payers,
hospitals, physician groups, clinics, and other provider settings. Some
systems may be very small, used in a single department for a specific
application that produces small packets of data, creating an information
silo. At the other extreme, enterprise-wide and missioncritical
applications are commonly used to accomplish multiple functions,
applications that are used constantly all day long by a majority of
390
employees. Enterprise systems produce prolific volumes of data and may
communicate these data to third parties such as insurance companies,
governmental agencies, and others, as well as retain the information for
internal use (Figure 9.1). These very different data sourcesplaces from
which data emanatemay have applications for both clinical or business
processes and analyses.
EXHIBIT 9.1 Revenue Cycle Management
In the healthcare industry, financial issues related to revenue cycle
management (RCM) are often complicated by changes to insurance
products and contracting terms with providers. These insurancerelated
changes are somewhat fluid but mostly involve tactical, daily activities,
including claims preparation, billing for services, and accounts
receivable. In spite of these changes, the patientprovider visit process
has changed little and still generally involves the how much, who, what,
when, and where of the procedural and diagnostic experience. The data
packets derived from these visits are relatively static in data sizeyet
overall volumes of data are increasing due to the growing number of
medical visits associated with the growing U.S. population and the
aging of the baby boomers.
In many business contexts, the phrase Follow the money has
relevance. The U.S. healthcare system is capitalistic in nature and,
therefore, no exception to this notion. Health care is provided through
an extremely complex environment of procedures, codes, diseases,
materials, tools, facilities, and human resources. Unaligned contractual
relationships between providers, insurers, suppliers, government, and
employers are omnipresent at all levels. In turn, healthcare providers at
all levels must document and report their services to payers, which then
pay them through the contracts they have with employers and
individually insured patients. Federal and state entities follow similar
processes through insurance claims for the Medicare and Medicaid
programs. The unfortunate overabundance of uninsured patients must
also be processed. The bureaucracies and administrative requirements
for all of these activities are intricate and create an unnecessary
inefficiency in the U.S. healthcare system. Notwithstanding this fact,
providers must manage this cycle of procedural reporting and spend
excessive resources to assure the sufficient collection of revenues so
391
that their practices can survive and grow. Like any other organizations,
healthcare organizations must be able to predict streams of revenue
accurately to stay financially viable. With the ever-growing complexity
described previously, this task is especially challenging for healthcare
organizations of all sizes.
FIGURE 9.1 Organizational Chart of the House Democrats Health
Plan
Reproduced from the Joint Economic Committee, Republican Staff. Congressman Kevin
Brady, Ranking House Republican Member
Some of the most significant and voluminous sources of healthcare
data are profiled here:
Clinical
Electronic health records/electronic medical records/personal health
records (EHRs/EMRs/PHRs)
Images and image management systems (e.g., picture archiving and
392
communication systems [PACSs], digitized X-rays, CT scans, PET
scans)
Case mix, care management, and disease management systems
Independent laboratory and other clinical results (e.g., blood, tissue,
fluids)
Monitoring systems (e.g., maternity, cardiology, ICU)
Transactional/Operations
Hospital information systems (e.g., admissions, emergency
department visits)
Hospital departmental systems (e.g., radiology, laboratory,
pharmacy, surgery, emergency department, order entry)
Materials management, supplies, and cost accounting systems
Physician practice management systems (e.g., scheduling, billing)
Revenue-cycle processes (e.g., provider billing, claims, patient
accounting)
Post-acute clinical and billing systems (e.g., skilled nursing, home
care)
Payer
Payer claims and contracting systems (e.g., benefit rules, risk
calculations, claims adjudication)
Care management systems (for coordinating transitions of care and
discharges to home or other facilities)
Third Party
Research systems (e.g., universities, human, animal) Clinical trials
systems (e.g., pharmaceutical companies, universities)
Satisfaction surveying systems (e.g., patients, providers, staff)
External
Internet resources
393
Registries, population management, statistics, and risk adjustments
Industry reporting (e.g., benchmarks, score cards, report cards)
Cellular devices and applications
Government
Federal government programs (e.g., Centers for Medicare and
Medicaid Services [CMS], State Childrens Health Insurance
Program [SCHIP], Department of Defense TRICARE and TRICARE
for Life programs, Veterans Health Administration [VHA] program,
and Indian Health Service [IHS] program)
State and local government programs (e.g., Medicaid, MediCal, State
Health Insurance Assistance Program [SHIP], Childrens Health
Insurance Program [CHIP], Health Resources and Services
Administration Primary Care: The Health Center Program,
healthcare marketplace regulatory programs)
Types of Organizations
Listing the principal participants in creating healthcare data will set the
stage for understanding the kinds of data sources they require. The
following groups all generate and process data in different intervals,
including real time, daily, and monthly:
Providers (e.g., physicians, nurses, and other clinicians)
Hospitals and hospital systems
Outpatient care facilities (e.g., imaging, urgent care, physical
therapy)
Payers and third-party administrators
Government, including military organizations
Post-acute care facilities (e.g., skilled nursing, hospice)
Home health
Pharmaceutical companies and laboratories
Research centers (e.g., universities, government)
Public health organizations
394
With more than 850,000 licensed physicians in the United States,3 3.1
million registered nurses, more than 500 healthcare insurance companies
(and many more health plans offered by provider groups), nearly 6000
registered hospitals, and a multitude of government programs, one can
begin to imagine the interrelated dependencies and the complexities of
healthcare data.4 From each system and in each participant environment,
vast sums of information are generated and subsequently stored in the
complex practice of medicine and delivery of health services. Examples
for discrete data elements include a Current Procedural Terminology
(CPT) code (Exhibit 9.2), an incremental lab result, an insurance payment
denial code, the sex identifier or resident ZIP code of a patient, the name
of a prescription drug, and a chronic disease code. While these are rather
straightforward and simple data elements, a patient might have a single
digital image result stored in his or her medical record that takes up 200
megabytes of data; a chronically ill patient may have gigabytes of
historical data associated with his or her care.
EXHIBIT 9.2 Complexity in Coding Healthcare
Diagnoses and Procedures
The complexity of health care becomes clear when one considers that
the kernels for healthcare billing and revenue-cycle management
include International Statistical Classification of Diseases and
Related Health Problems (ICD-9 and ICD-10) codes, Current
Procedural Terminology (CPT) codes, and diagnosticrelated groups
(DRGs) coding. There are approximately 14,000 ICD-9 and 68,000
ICD-10 unique codes, each three to seven characters in length, along
with potential modifiers that must be used by providers to attain
contractual reimbursement for their services. Currently, ICD-9 codes
are being replaced by ICD-10 codes. This process is leading to
tremendous changes in personnel training and software systemsthe
transition will have a deleterious cash-flow effect on providers, while
the greater specificity of ICD-10 codes will allow for more precise
matching of codes to complex diagnoses and conditions. There are
7800 CPT codes with even more potential modifiers. DRGs are used in
acute care as a bundling method for multiple patient care activities
grouped based on similar consumption of resources. All of these codes
may be used by employers as methods to compensate providers, project
395
income, plan for facilities and materials, and serve as input for
population management evaluation and trending. Tens of thousands of
coders must translate what providers do into these diagnostic and
procedural codes to properly complete and adjudicate an insurance
form. These codes must be retained and available for extensive periods
of time for a host of analytical and legal reasons.
The magnitude and the growth of types and volume of data in health
care are staggering. To quantify some of the issues, consider the following
sizing example:
One digital image approximates 200 (or more) megabytes (about 60
MP3 songs).
Each image may be redundantly stored at the imaging site, within the
patients EHR, and/or by the PACS vendor site.
When extended to the maximum scope, this one type of datathe
imagemight be present as more than 50 billion images worldwide,
or a total worldwide data requirement of 70 exabytes (18 zeroes) of
data (more than 3.3 trillion MP3 songs).
Importantly, in contrast to data in other business environments, good
data in health care are essential in making clinical decisions that improve
the health of patients and may even save lives. Conversely, untimely or
wrong data may contribute to inaccurate decisions with potential
implications for patient mortality. Unlike employees in other data-rich
industries, healthcare providers make real-time decisions based on
laboratory results, images, trends in patient histories and across
communities, and other data points. As undesirable as it may be to
miscount inventory or overbill someone in the retail environment, it is
quite another matter to conclude a course of medical action based upon
misinterpreted data that leads to a negative outcome.
VELOCITY, VOLUME, AND VARIETY
(THREE VS) AND BIG DATA
The importance of the velocity, volume, and variety of data is endemic
throughout the world and not exclusive to health care.
396
Velocity
Velocity suggests that data have momentum that is accelerating through
increased applications and consumer and business uses. This momentum is
virtually exponential. Everyone who has used cellular technology has
experienced this velocity of data through their phones enhanced
technology and the Internets growth. Just a few years ago, consumers
could not track their exercise performance, research extensive disease
narratives, or communicate with their providers by digital means with
ease. Now consumers can perform all these tasks with personal electronic
devices, portals, and websites.
Volume
Volume suggests similar exponential growth of accessible and seemingly
necessary data. Earlier in this chapter, we discussed a volume example
based on imaging; it is extremely relevant for understanding the
ramifications of the increasing amounts of data in health care. Data growth
in the future will likely be larger in health care than in any sector other
than global security.
Variety
Variety suggests that data are associated with, and will continue to take on,
seemingly limitless descriptions. The use of new personal electronic
devices will make fluid, organ, and almost any other bodily function
measurement feasible. Cellular devices will serve up various health data
such as blood glucose, blood pressure, cholesterol levels, and other
essential measures in an attempt to improve the quality of medicine and
disease management. Data reflecting issues of public health concern not
only personal data, but also environmental, social, and other systemic data.
In concert with the notion of the Three Vs impacting health care, three
types of software applications and technologies, along with their
associated data sources, hold out the promise of improving health care. At
the same time, these three areas present some critical challenges related to
data management. When these areas are combined with health cares
exacting security and safety requirements regarding data, one can see a
whole new operational frontier opening. These three areas are:
1. Imaging
397
2. EHR/EMR systems
3. Mobile communications and devices
Imaging
Imagings uses are expanding as it evolves into an enhanced and
noninvasive diagnostic utility for continuous care management. In fact,
more and more test results are taking the form of digital images in health
care; for instance, with biopsy results, a picture may accompany the
written pathology report. The use of digital imaging is precise: It better
leverages physician time and, compared to radiological films of the past,
its portability, storage, and retrieval are more convenient. Moreover,
digital imaging is very popular among clinicians. Predictably, there is a
direct correlation between enhanced imaging technology and an
exponential increase in data requirements. As explained elsewhere in the
text, this important medical tool has created a challenging data source for
analytics.
As imaging technologies are enhanced such that the granularity and
acuity of images increase further, there will be a proportionately greater
data volume problem: More detailed images necessarily take up more
space. More importantly, images or image readings may be digitized
simply as singular data files. Unfortunately, the translation of the images
detail into any sort of a granular bit-mapped data set can currently be
stored only as a retrievable location of the image in aggregate. In other
words, an X-ray image or scan is stored as a single picture, but you cannot
extract any data other than that picture or image stream as a whole. For
example, an X-ray of a foot No.7.24.13.XYZ is without judgment, bone
description, dimension, or any data narrative. The X-ray data storage and
retrieval would be made based on the code No.7.24.13.XYZ. In the near
future, an exciting development will make these static data references
(images) available with more data analytics based on new technology. In
turn, a very large data volume will be required for a single diagnostic
element such as a diabetic glucose reading.
EHR/EMR Systems
The second area is of tremendous importance to data storage: the
proliferation of EHR or EMR systems. Part of the American Recovery and
Reinvestment Act of 2009 (ARRA) includes financial incentives paid to
398
hospitals or physicians or their practices for the various stages of
implementation for an EHR. As discussed in other chapters, before
ARRAs investment incentives were made available, healthcare
organizations embrace of EHRs was much more tepid. The typical EHR
contains an incredible amount of data; easy, fast, and relevant retrieval of a
patients data is a highly desirable result of an EHR, but it comes at the
price of extensive clinician data entry and large storage requirements. The
addition of personal genomic data to EHRs is an even more momentous
undertaking, considering the fact that we as a society are only just learning
how to use such data for patient care, predictive purposes, and other
analytic purposes.
Mobile Communications and Devices
The third area of challenge for data concerns mobile technology.
Excitingly, we are at the early stages of mobile health, more succinctly
called mHealth. There is a frenzy of development and idea promotion
around the propagation of applications that are focused on enhanced
communications through cellular devices. This includes personal health
management and even diagnostic capabilities. Further, one can now use
handheld or small body-attached informationsensing wireless medical
devices to measure blood glucose, blood chemistry, heart monitoring, and
other Star Treklike diagnostic workups.
In contrast to other historically transient technology applications such
as universal patient identification cards (e.g., credit cardlike items
containing a persons medical history), developments in these three areas
imaging, EHR/EMR systems, and mobile communications and devices
will likely be sustainable and are already bringing environmental and
cultural changes to healthcare organizations. As in other facets of their
lives, consumers are demanding responsiveness and breadth in all forms of
electronic communications related to health care. To some extent, these
demands are putting strains on the traditional structures of health care.
Healthcare providers are now struggling to manage the consequences of
making data accessible to consumers via cellular phone technologies.
Nevertheless, consumers demand for access to their own and their
familys medical data is here to stay, as is the movement by healthcare
organizations and providers to engage patients in their care processes.
399
DATA CHALLENGES
All healthcare institutionssmall and largehave a growing obligation
and need to develop, adopt, and manage their data strategy (i.e., their HIS
plan). Each institution will face philosophical differences, budgetary
constraints, lack of experience, resource limitations, and a multitude of
other challenges as it works through its respective needs.
Data Ownership
The importance of the healthcare organization retaining the ownership of
its data is discussed in the Implementation chapter in the section about
contracting with software vendors. Yet, within the organization, who
owns or is assigned responsibility for the accuracy and quality of the
data? The ownership and stewardship of data are critically important, yet
these may be contentious issues. Traditionally, one would assume that a
CIO or lead IT person would be the logical owner. Those personnel may,
in fact, be well suited for this purpose. However, there is a difference
between datas technical nature and the real purpose of data. A simple
analogy is that a person may store a car in a full-service parking garage
where it is kept clean, safe, maintained, and always available, but it is the
driver who uses the automobile. And so it is with data: IT is responsible
for the physical storage and safety of data, while end users are responsible
for the quality of data. Each institution should dedicate an individual
and/or group to the single responsibility of data management.
Data Delivery and Translation
The data management team will be responsible for architecting the data
environment, cataloging the data, securing data integrity, managing data
definitions and normalization, sanctioning data sources, and generally
ensuring the reliable availability of data for their constituents (or
customers). A critical skill set is necessary to bridge the inherent gap
between the data management teams incredibly technical environment
and its generally nontechnical customersnot everyone makes a good
translator. Another convenient analogy is that of a homeowner (the data
customer) whose attitude about the homes infrastructure is that he or she
cares little about the source, dimensions, or material composition of the
homes plumbing, but rather just wants clean, reliable water on demand.
400
An essential part of the data management role is translation. This task
is addressed in more detail in the chapter on business and clinical
intelligence, but basically it means that the data management team
members need to possess the ability to understand what their customers
need and work backward to the data architecture to ensure data delivery.
Frequency, presentation, modifications, integration, and other data
requirements require tremendous listening skills and empathy for whatever
purpose the customer has for the data. Ways of connecting with customers
often clinicians providing health care to patientsinclude daily use of
key performance indicators (KPIs) that measure anomalies relative to
standard performances, dashboard presentations based on the users skills
and role, and appropriate data available for the ambiguity of ad-hoc
analytics. Data need to be made available to the customer set in a relatively
easy and understandable manner. Moreover, data need to be packaged
for role-based utility, as different customers need very different
manifestations of data to perform their respective roles.
Data Storage
On the Big Data continuum with regard to the three most voluminous
healthcare application environments (imaging, EHR/EMR systems, and
mobile communications and devices), a disproportionate share
approximately 70%of the current global data storage is consumed by
those three elements alone. Moreover, as imaging is enhanced and EHRs
and mHealth mature, these three applications will contribute
disproportionately more new data than all the other categories combined,
requiring more incremental storage. Imaging data are challenging to
analyze, and EHRs and mobile devices are fragmented. These data sources
may be valuable in their purpose, but they are difficult to integrate within
data storage and retrieval strategies.
Software Openness
In the production of data, one can find hundreds of commercial EHR,
practice management system (PMS), hospital management, and PACS
software packages that are written in myriad programming languages and
may be decades old. Because data in one system may be formatted or
identified differently than data in another system, the challenge of
combining or communicating among similarly purposed computer systems
401
(being open)whether nationally or locallyis significant. For systems
whose roots are embedded in traditional, less flexible technologies and
languages, this integration is more challenging and creates major barriers
to the current goal of interoperability between systems from different
organizations. There are inherent conflicts for vendors who choose to
make their systems truly open when others are unable or unwilling to do
so. As with any business, one must quantify and assert a motivation to
participate in an investment to be open to others. For software vendors to
be open, they must first define with whom and what they are open to,
who is willing to financially support or mandate this openness, and
whether it is a level playing field for all to make such an investment.
Currently, the EHR market is an extraordinarily competitive environment
in which software application features are quickly changing, thereby
requiring vendors to invest significant development dollars to keep pace
with industry changes, much less meet the complex and ever-changing
requirements to be open and interoperable. Because data in one system
may be formatted or identified differently than in another, the challenge of
combining or communicating among similarly purposed but different
systemswhether nationally or locallyis significant.
Technical Lexicon and Lack of Universal Terminology
The healthcare industry has a complicated lexicon for everything it does.
Providers may use highly technical terminology that is not understandable
by patients. For example, one physician may use the term high blood
pressure while another may use hypertension to characterize the same
chronic patient disease. Within the healthcare supply chain, a
differentiated organizational bias in terminology is apparent: Vendors that
build and supply materials used in the industry often use different
nomenclatures, while academic reporting and analytical organizations may
enforce their own imprimatur for medical terms. Further, with healthcare
delivery being local, differences in terminology may make data sharing
more complicated.
An important element of data transmission, storage, and retrieval has to
do with normalizing information. A data element needs to represent what it
says it is. The reliability of data includes the notion that the same data
element, when measured or recorded more than once, will always be the
same.
402
Data Sharing
As use of EHR or EMR systems proliferates within a given community,
multiple methods for sharing data among those systems may emerge.
Individual patients may keep records of their activities, diets, and personal
biological testing by using personal computer or cellular methods to create
a personal health record (PHR). Their physicians are likely to have EMRs
for their offices purposes in caring for each patient. The local hospital
may have an inpatient EHR that tracks testing ordered by each physician
and for hospital-based procedures for each patient. If patients have used
community-based physical therapy, psychiatric, dental, or other caregiving
modalities, disparate data may be present in those siloed medical recordkeeping systems that may have different data management architectures
and terminology. One can easily imagine the overlapping and unaligned
longitudinal care difficulties, which result in substantial data
inconsistencies and coordination challenges.
Attempts at communication standards have offered some help in
navigating these often incompatible systems, but universal adoption has
not occurred. Health information exchanges (HIEs) are environments for
the purpose of sharing data that communicate or aggregate disparate
healthcarerelated data into a single, universally accessible mechanism or
location. These exchanges have been introduced at the state and regional
levels and within small and large urban areas. They are relatively new and
have experienced mixed reviews and results: Some have already failed,
whereas others are demonstrating some efficiency. HIE organizations and
initiatives are terrifically ambitious and involve the alignment of often
strange bedfellows. The bottom line today, however, is that data remain
siloed and the jury is still out on how successful the healthcare industry
will be at electronic data exchange, notwithstanding the inclusion of HIE
and system interoperability in the governments Meaningful Use criteria.
Another invention of recent healthcare law is the accountable care
organization (ACO), which is addressed elsewhere in this text in the
context of care providers. ACOs look much like capitated care, managed
care, and other models seen in the past. The difference is that health
information systems and technology infrastructure are now available to
integrate care across the continuum, so this model is now actually feasible
to accomplish. Applicant healthcare provider organizations were granted
Pioneer ACO status by the federal government based on a multitude of
403
qualifying criteria. In relation to their data, these ACOs will be taking on
financial riskin the form of sharing in the rewards or lossesbased on
performance, cost, quality, appropriateness, efficiency, and other
measurements. Risk, in this context, means being responsible for the
financial consequences of providing cost-effective and contractually viable
health care. This involves a very real operational change for the healthcare
organizations that have joined this movement, and many ACOs will be
woefully tested in managing their data and operationalizing decisions.
Current Pioneer ACOs are achieving a variety of results; some are making
their way, and others are leaving the program (Exhibit 9.3).
Interfacing Data Within an Organization
Data are resident within any number of disparately purposed software
applications. It would not be unusual for a large hospital to have licensed,
purchased, or developed 50 to 100 different software applications, all
concurrently used to run its business. It is important to understand that
between these software applications, data are never fully integrated. True
integration using interfacing software is almost impossible. Aggregating,
normalizing, and applying other methods to combine data from software
applications constitute an interfacing process that does not lead to true
integration. (Integration and interfacing are discussed in more detail in the
Managing HIS and Technology Services: Delivering the Goods chapter.)
Interfacing these disparate data sources involves internal software and
infrastructure requirements for such uses as data warehouses, portals,
departmentally used applications, and more. Further, with third-party
application updates and software changes, these systems data interfacing
tools must be modified on a regular basis. A whole world of vendoroffered interfacing hubs and networks with tools attempting to enhance
these laborious and challenging responsibilities has emerged.
EXHIBIT 9.3 Kaiser Permanente
For decades, the quintessential ACO in the United States has been
Kaiser Permanente. It has been successfully taking financial and
actuarial risks with a truly integrated healthcare delivery and insurance
model. With its often-criticized disciplines surrounding managed care,
Kaiser Permanente is also thought to be exacting in its patient
relationships and employment and contracting methods: It is now a
404
bastion of what is needed for a successful and well-functioning ACO.
One of Kaiser Permanentes greatest accomplishments is that it
understands and uses its own data at least as well as any other similar
organization.
Even a small physician practice must invest in software that supports
sophisticated patient scheduling, claims processing, and patient billing that
often involves different formatting for each insurance company covering
patients (clearinghouses attempt to simplify this effort), procedure coding
validation, authorization and referral authentication, and patient billing
functionality. In a medical community of 500 physicians, there may be a
dozen or more different PMS software packages being used.
DATA SECURITY AND PROTECTION
Many industriesbanking, security, and communications, for example
face challenging data security issues. The healthcare industry deals with
similar demographic and financial data, yet its environment is also
complicated by special circumstances owing to the laws governing the
protection and privacy of patients clinical information. As a consequence,
healthcare data are subject to not only the same federal protections as
banking (or any other industry), but also a host of additional regulations
specific to health care. With so many users of patient information, there is
tremendous data protection exposure and risk for the myriad provider,
insurance, and supporting healthcare entities. Of course, protecting the
security, privacy, and confidentiality of patient data has always been
considered a hallmark of good medical record management, even when all
records were based on paper. In 1996, however, in recognition of the
exponentially increasing automation and data availability and
communication in health care, HIPAA was passed by Congress to
distinguish PHI as a special category of data warranting extra safeguards.
The purpose of this law was to establish methods and safeguards for
nondisclosure of clinical and demographic patient data other than for use
in care delivery and healthcare processes. HIPAA has not always been
carefully enforced in spite of severe mandatory penalties imposed for
willful neglect (fines between $250,000 and $1.5 million), although this
situation is changing (see recent examples and recent upgrades to sharpen
HIPAAs teeth). These laws have included, or have been expanded to
405
include, both civil and criminal penalties for business associates (e.g.,
vendors, consultants) that may peripherally have access to the same data.
A singular data element may need to be protected, yet available, within
the same employer. For example, a physician may need to see the history
of a patients HIV testing, yet this same clinical finding should not be
accessible by all other institutional employees. Thus granular data
elements need to be tagged for their appropriate use. This differentiation
based on the reason for the use of the datamakes accomplishing good
data management and protection difficult, given the complexities of data
structures and disparate systems in healthcare entities. Further, according
to HIPAA regulations, an audit capability must be integrated into systems
at this granular data level to ensure protection and discovery of attempted
inappropriate access. HIPAA requires an electronic audit trail for each and
every transaction within a healthcare systemone of the many HIPAA
requirements that healthcare software vendors are legally required to meet
as a standard part of their products.
Laws also protect patient medical histories by describing the term for
retention of medical data. For example, federal law requires a hospital to
retain a patients data for 5 years and 3 years after the patients death. The
American Health Information Management Association suggests retention
for a period of 10 years after the patient encounter.5 Most organizations
save medical records forever, a practice that has further implications for
data security and storage. It is almost as if keeping records is
simultaneously a security measure (protection against litigation) and a risk
(more fodder for data breach).
SUMMARY
Data have become an increasingly valuable resource for healthcare
organizations and providers. Data support clinical work and business
processes, and they provide fuel for information creation and analytical
opportunities. Data are created and captured in transaction systems that
support the daily activities of health care, both clinically and from a
business perspective. Healthcare data continue to exponentially increase in
terms of their velocity (usefulness), volume (amount), and variety (types).
Ultimately, these combine to form a resource yet to be defined, called Big
Data, which creates notions of limitless possibilities for insight,
406
information, imagination, and intelligence. None of these goals will be
reached easily, as the challenges and barriers to intelligent information
usage are many. Disparate source systems and silo processes continue to
plague healthcare organizations, and it will be many years before this
fragmentation is overcome. Data must be kept secure, private, and
confidential. To not do so defies long-standing principles of health care
and respect for the patient, but to do so requires persistent human and
technological efforts.
By considering the technical and analytical needs and uses of data, the
healthcare industry is attempting to accelerate implementation of the longstanding belief that the economics and efficacy of health care may be
enhanced through evidence-based medicine. The exponential growth of
data production will be both the foundation to measure this belief and a
technological challenge to manage. Systems must be developed to more
efficiently collect, normalize, and integrate data to be analyzed on both a
real-time and a retrospective basis.
A growing number of healthcare institutions, however nascent, have
undertaken the monumental task of providing a solution for the data
explosion. In the long run, this discipline will provide an opportunity for
growth of organizational cultures into learning organizations, as well as
individual employee development and career growth. Technical positions
for data architecture and management will proliferate.
KEY TERMS
Affordable Care Act (ACA)
Big Data
Clinical data
Current Procedural Terminology (CPT)
Data delivery and translation
Data security
Data sharing
Data source
Diagnostic-related group (DRG)
407
International Statistical Classification of Diseases and Related Health
Problems
Key performance indicator (KPI)
mHealth
Normalization
Practice management system (PMS)
Revenue cycle management (RCM)
Velocity, volume, and variety (Three Vs)
Discussion Questions
1. Is this subject too technical for you to fathom? Describe the areas that
you do not understand.
2. Can you envision any other new data source that might become part
of the healthcare arena?
3. Are there any roles described within this discussion on data that are of
interest to you?
4. If sharing of data between healthcare organizations may have a macro
industry benefit, how do you think this could happen?
5. Have you worked or are you currently working in a healthcare
organization that has emphasized data as being strategic? How has the
organization communicated this idea throughout its entire team?
6. Describe how disparate data might be converted into a single
vocabulary.
7. Have you personally experienced an encounter with a provider and
felt it beneficial that the provider might have access to or use realtime data to help your care?
8. What strikes you as the three biggest data challenges in health care?
9. Do you believe that other industries that have been using data for
their own operational efficiencies may help the healthcare industry
better navigate the adoption and use of data for both business and
clinical benefit? If yes, how so
4
?
08
10. Is there a single best source of data for health care to start with in
terms of managing the inherent responsibilities to collect, aggregate,
and store data? Or do all the data sources need to be tapped at once?
REFERENCES
1. Kaiser Health News. (2012, June). Health care costs to reach nearly
onefifth of GDP by 2021. http://www.kaiserhealthnews.org/dailyreports/2012/june/13/health-care-costs.aspx
2. IBM. (n.d.) Big data at the speed of business. http://www01.ibm.com/software/data/bigdata/
3. Young, A., Chaudhry, H. J., Rhyne, J., & Dugan, M. (2011). A
census of actively licensed physicians in the U.S. in 2010. Journal of
Medical Regulation.
http://www.nationalahec.org/pdfs/FSMBPhysicianCensus.pdf
4. AFL-CIO, Department for Professional Employees. (2012, April
12). Nursing: A profile of the profession: Fact sheet 2012.
http://dpeaflcio.org/wp-content/uploads/Nursing-A-Profile-of-theProfession-2012.pdf
5. Enterprise content and record management for healthcare. (2008).
Journal of AHIMA, 79(10), 9198.
http://library.ahima.org/xpedio/groups/public/documents/ahima/bok1_040405.hcsp?
dDocName=bok1
_
040405
409
CHAPTER 10
Business and Clinical Intelligence
(BI/CI)
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 sources and the evolution of data.
Describe the myriad stakeholders who need BI and CI information to
perform their jobs in the healthcare arena.
Identify methods for receiving, organizing, storing, mining, and
formatting data for BI and CI purposes.
INTRODUCTION
The generic definition of business intelligence (BI) is a set of theories,
methodologies, processes, architectures, and technologies that transform
raw data into meaningful and useful information for business purposes.
This termand the practiceis applied widely throughout various
industries. BI handles large amounts of data and information to help
identify and develop new opportunities. Making use of new opportunities
and implementing an effective strategy can provide a competitive market
410
advantage and long-term stability. BI technologies provide historical,
current, and predictive views of business operations. Common functions of
BI technologies include reporting, online analytical processing, analytics,
data mining, complex event processing, business performance
management benchmarking, text mining, predictive mining, predictive
analytics, and prescriptive analytics.1
Clinical intelligence (CI) is an emerging adjunct to BI that is focused
on the healthcare industry. With the proliferation of various forms of
electronic clinical data use, as well as industry and political pressures to
obtain and utilize clinical measurements, BI is the obvious technological
foundation for CI. The aforementioned BI definition works perfectly as
CIs mechanics and functionality underpinning. Thus a generic definition
of CI is a set of theories, methodologies, processes, architectures, and
technologies that transform raw data into meaningful and useful
information for clinical purposes. Specific uses for healthcare 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 description or collections of words to
create a cogent definition for the many meanings and types of healthcare
BI and CI. Healthcare BI and CI are very subjectiveremember, Health
care is local, and so is data use. The application of an intelligence process
may be static and trivial, or it may be dynamic and extremely complicated.
Thousands of discrete software applications and tens of 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 intelligence is relevant for them.
There are a nearly infinite number of current and imagined healthcare
BI and CI content examples. We will review some next.
Business/BI Example
A CEO wants a dashboard that is refreshed nightly. A single dashboard
screen presentation includes all of the organizations 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
411
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 CFO with questions and a
requested response expectation.
Clinical/CI Example
In the purchasing department conference room, the director of materials
management for a mid-sized hospital is listening to a pharmaceutical sales
representative propose certain purchases of disposable surgical items and
associated pharmaceuticals. From his laptop, the hospital executive selects
multiple items within his CI system to display in a count, cost, shelf-life,
surgeon preference, and surgical room scheduling pie chart. He compares
these metrics with the recommendations from the pharmaceutical
representative and places a very focused order.
Integrated BI/CI Example
A clinic administrator is negotiating with an insurance company over an
at-risk contract for the insurers largest business clients employee
population. The administrator searches the BI/CI data for the same
population (including subscriber family dependents) currently seen in the
practice. Then the administrator sorts the list of patients by diagnostic
procedural codes, stratifying them by age, weight, ethnicity, comorbidities,
charges and payment history, IDC-9 and CPT codes, 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 by the new insurer.
Assembling and preparing all these data for specific subsequent use is
hard, technical work. Adoption of BI/CI has been slowslower even than
electronic health record (EHR) adoption. Perhaps 20% of U.S. healthcare
providers have implemented data warehouses and BI/CI software to
support clinical decision making, pay-for-performance (P4P) programs,
and comparative effectiveness research (CER) and to track clinical
outcomes as part of healthcare quality measurement initiatives.2 Given the
government mandates for information from providers, increasing sources
of clinical data from EHR systems, greater transparency through public
reporting of healthcare outcomes, and for a host of other circumstances, in
412
the future BI/CI will become more pervasive and vendor tools will become
easier and more robust to use.
HEALTHCARE BUSINESS AND CLINICAL
INTELLIGENCE
Healthcare organizations have had very mixed experiences in regard to
BI/CI. With the support of sophisticated internal information 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 BI/CI solutions were designed for
accuracy, performance, and breadth of purpose, these organizations have
achieved marked successes. Conversely, where data have been contained
in silos and where either an IT or finance department has restricted use of
those data to a chosen few parties, there have been some black eyes for
vendors and for BI/CI initiatives. Expectations often are not met by actual
outcomes of those projects: Expectations may be inflated due to the vendor
overselling its methods, an organization misunderstanding the level of
expertise, investment and commitment necessary, or both parties failing to
appreciate the complexity of BI/CI (Figure 10.1).
CI is a term that has only recently been coined, in conjunction with the
increased adoption and use of the EHR, electronic medical record (EMR),
and personal health record (PHR), all of which overlap to some extent in
terms of their functionality and usage. Their differences generally reflect
who collects the data, the setting where this occurs, and the
comprehensiveness of the application.3 An EHR system comprises the
broadest description of a patients medical and health history, covering a
larger collection of clinical data including ambulatory, acute care, and
holistic care information. It is used to facilitate clinical care processes on a
real-time basis. An EMR is often employed in a singular ambulatory or
acute-care setting, such as a free-standing physicians office or hospital,
and is not connected to another facility such as an ambulatory or hospital
setting. EMRs lack functionality with regard to offering a longitudinal
view across multiple settings of care. The PHR is managed by the patient
(or a parent or guardian) and populated with personally acquired medical
and health-related information. The varying terminology and need to
integrate these dissimilar platforms create a significant challenge when
413
attempting to implement analytics using data from these and other
systems.
FIgure 10.1 Relationship of Complexity for Content Use and Creation
There is a correlation between the number of users for intelligence and analytics and the
complexities for creating content. The vast number of users generally requires more mundane
and more easily created content. More complex content is utilized by smaller audiences created
by more technical personnel.
CI, when properly implemented, not only will support more focused
clinical analysis than BI, but also should be designed and managed to
integrate business data for broader analysis, including, for instance, cost
accounting data within a clinical analysis. A longitudinal patient medical
history combined with the business history (e.g., charges, payments, or
disposable medical equipment consumption) for the same patient would
provide invaluable insights for outcomes measurement and monitoring as
well as process effectiveness. Consider the following example of BI/CI
integration to see how this works: Suppose a patient with cardiovascular
disease receives a stent in a catheterization procedure. Future and
concurrent caregivers might like to know the stents source, cost, SKU
identification, and date of manufacture and installation (as well as by
whom and where); provider follow-up observations; patient-reported
experience; related diagnoses and other therapies provided; patient
414
outcome including complications; and other related medical issues.
Hypothetically, in the event of the patients continued complaints or a
manufacturers recall, the organization could search the patients
integrated data and take appropriate action.
In spite of tremendous industry and governmental attentionas well as
technology advanceshealthcare organizations adoption of
comprehensive BI/CI has been disappointingly slow. Some components of
reporting, analytics, and data presentation are being used, but usually in
siloed application environments. Siloed application environments such as
revenue-cycle or laboratory systems that pull data from separate
repositories tell only part of the story of what is going on for the patient or
the organization, rather than providing a more holistic or comprehensive
viewpoint. These silos can deter progress and syphon off resources that
could be devoted toward more integrated BI and CI implementation
projects.
Cost, limited experience, restricted implementation resources,
numerous mandates and industry requirements (e.g., the proliferation of
EHRs and ICD-10 conversions), and a host of other considerations are
some of the valid reasons for lowering the prioritization for BI and CI.
Nevertheless, surveys of CIOs and other organizational executives by the
Healthcare Information and Management Systems Society (HIMSS), the
Medical Group Management Association (MGMA), and other member
groups consistently indicate that healthcare organizations plan to become
proactive in tackling more relevant and dramatic projects within their
landscape of data. Such expanded capabilities would be consistent with the
burgeoning development of data warehousing and BI initiatives in other
industries, a movement that creates greater overall availability of BI and
CI technology product and talent. Challenging as this work might be, it is
both exciting and rewarding. Given the current crises in cost and quality,
health care stands to gain as much as, if not more than, any industry by
increasing its secondary use of data for BI and CI insights and
improvements. These data analytics platforms will offer essential
capabilities to help better manage and improve the quality of care
delivered by healthcare organizations.
HISTORY OF BI AND CI
415
The healthcare industry has come a long way in data collection and
processing: It has built a massive collection of business and clinical data.
Nevertheless, healthcare organizations have a long way to go in effectively
organizing and analyzing these data to realize workplace efficiencies and
better medical outcomes.
When computers entered the marketplace in the late 1960s and early
1970s, healthcare-related data were generally focused on processing
billing and automating the byzantine world of insurance claims. At that
time, the landscape for the for charges,
payments, and adjustments was straightforward. Getting paid as providers
and paying as payers according to the appropriate procedure and
diagnostic codes worked rather well. The financial environment was
characterized by rather than
todays complex risk arrangements, with concomitant detailed contracting
and productivity measures. Interestingly, the notion of using dormitory
refrigeratorsized computers for coding and sending off insurance claims
both stressed and fascinated healthcare providers during that era.
Principally, the adoption rate for computerization was slow because it
required a significant resource investmenttoo much for many
selfemployed physicians and investment risk-adverse hospitals.
Pharmaceutical companies just needed to count things they sold,
laboratories sent out results and bills on paper, and payers used massive
computers to enroll subscribers and adjudicate claims with fewer
contracting rules than today. Hospitals had paper census lists, paper bills,
and paper medical records (and overused photocopy machines).
As transactional software began to capture data sufficient for providers
to file claims and for payers to adjudicate those claims, the art of
functional reporting emerged. Reporting became an important part of the
promise of health information systems (HIS) through the advent of early
healthcare software. Reports were the now ancient predecessors of
analytical intelligence. Limited tabular screen shots provided some
breakthrough efficiencies in an environment that was conditioned to use
paper and pencil. For example, aged trial balance reports and computerbased appointment registries empowered physicians and became the staple
reports for receivables management.
These reports also amplified the ensuing friction between payers and
providers arguing about whose pocket the money should be in and when.
416
Insurance claims began being printed by providers as a significant time
saver. In contrast, insurance company personnel manually entered the
claims. When providers and payers conducted contract negotiations, weeks
might pass while IT departments tried to figure out inputted (but difficult
to retrieve) costs and payment history. Few data processing directors (the
forerunners of todays chief information officers) spent any time with
information, but rather treaded water while trying to keep hardware
working and prevent software and hardware from crashing. Accessing data
was profoundly difficult.
Reports were typically structured or canned. User requests for report
changes joined a queue of reporting enhancements to be eventually
delivered by IT departments and software vendors. With fewer software
offerings available, reports were produced for specific applications for
specific operational audiences. If some executive at hospital X wanted to
see information across her fiefdom of departments and within her niche
computer applications, she would likely call upon someone from her IT or
accounting team to populate a basic spreadsheet. Over time, the
spreadsheet started to overtake reports as an analytical tool.
During the early days of computing, multitasking software and PCs
were not available. Terminal screens were gray, orange, or green. A
storage capacity of 10 megabytes might suffice for a busy 10-provider
group practice; 200 megabytes would carry the day for a small hospital.
Larger institutions often used service bureaus with larger computers along
with slow phone line connections to run their businesses. To meet the need
for reporting, a programmer who knew COBOL, FORTRAN, or early
MUMPS (early software programming languages still in use in many
systems worldwide) would have to manually program each report. Adding
an unforeseen data element to the application at a later date meant going
back to the programming cubicle and rewriting code. There was no such
thing as email or the Internet; likewise, networks (either hard-wired or
wireless), now such a familiar part of life, were not broadly available.
As with many innovations in the computer and software world, an
evolutionary development occurred in the mid-1990s. The notions of
executive information systems (EISs) and decision support systems
(DSSs) were introduced in the healthcare industry. These types of
application software brought together data from multiple sources. As their
names suggest, EISs and DSSs were designed to help management teams
417
at healthcare organizations make decisions by providing relevant
information from data acquired from transaction systems. In spite of the
fact that this BI effort was still generally static, it represented a quantum
leap forward. Unfortunately, producing these reports took an inordinate
amount of time, and they were primarily focused on BI rather than CI.
Universities, pharmaceutical companies, and laboratory testing companies
started to provide some analysis for limited clinical measures, mostly for
internal knowledge. For example, disease management gained some
theoretical traction in trying to predict better clinical care and apply early
cost-control measures. Managed care had progressed to the point that
payers wanted better information principally for the purpose of actuarial
and underwriting for risk evaluationa relatively narrow area of
concentration.
All of these efforts were limited by the computer power of the time and
constraints associated with the older programming languages. However,
this was also a point in history when an uncomfortable political focus on
quickly rising healthcare costs began, which helped prepare the way for
early EHR systems, increased computing power, use of Internet
technologies, and less expensive data storage. Little could these early
adopters and innovators possibly imagine the world of healthcare
computing today!
CURRENT CHALLENGES FOR ANALYTICS
Most healthcare processes can be described and managed at a very
scientific level. Chemistry, biology, radiology, and other macro disciplines
in the practice of medicine should be black and white. By extension, one
might assume that science would breed objectivity. However, there is a
great deal of subjectivity in the delivery of health care and BI/CI. Reasons
for this subjectivity and variability include the constantly growing nature
of evidence-based medicine, opinionated and differently trained human
practitioners and managers, inherent patient uniqueness, and challenges in
communication. This conundrum is reflected in the computer and software
systems that attempt to support these divergences.
No single computer hardware or software system has taken the
healthcare world by storm, becoming the standard, dominant selection. For
almost any application, there is a handful of leading competitive systems
418
that vie for customers in the niche. Available to any healthcare
organization are well in excess of 1000 vendors espousing their singular
superiority while running on different hardware platforms, relying on
different operating systems, and using a host of programming languages,
many of them proprietary and all of which are modified, enhanced, and
upgraded on a regular basis, thereby affecting the continuity of data,
interfaces, and operational use. One could walk through any metropolitan
citys hospitals and medical clinics and find disparate vendor software
packages, computer configurations, and strategies for IT and business
management. These organizations views of the world and the purpose of
HIS are manifested in very different ways. As a result, it might be argued
that many of health cares macro inefficiencies are inherent in the
confusing landscape of software and derived from the independent
methods of delivery utilized by the healthcare industrys participants.
In the retail, manufacturing, and banking industries, for example, the
adoption of business intelligence, analytics, modeling, and forecasting has
outpaced that in health care. None of these other industries is modest in its
complexity, and information system implementation and data use are
certainly no small tasks. In defense of the healthcare industrys relatively
slow embrace of CI/BI, its amorphous and complex nature brings about a
number of profoundly challenging issues when it comes to the subject of
analytics. Health care is extremely fragmented. As stated elsewhere in the
text, common entities, professionals, and components in this industry more
often than not do the same thing differently. Following are some major
reasons why healthcare analytics can be a challenging environment:
Health care is a nearly $3 trillion industry, fast on its way to
accounting for almost 20% of the U.S. gross domestic product, and
with lots of moving parts.4
Little financial incentive exists for coordination of medical care
among competitors.
More than 3 million healthcare providers of various types provide
services in the United States.
In the United States, there are more than 6000 hospitals,5 more than
200 insurance payers,6 141 medical schools,7 and more than 16,000
nursing homes 8 all doing the same thing differently.
Dozens of legacy (meaning retired) and contemporary
419
programming languages and operating systems are
contemporaneously in use throughout the healthcare industry.
From single departments to enterprise-wide applications, there are
more than 1000 commercially sold and home-grown software
offerings for healthcare practices, including nearly 200 disparate
EHR/EMR/PHR solutions with potentially 10 times more discrete
data elements than a revenue-cycle software system.
Several (but no single, de facto) communications or formatting
standards exist.
Disparate terminologies and entry errors in data entry are
commonplace.
Software systems must adapt to ever-changing healthcare laws and
industry operations, which affect data elements, functionality,
security and interoperability requirements, interfaces, and
operational use.
Public health and population managements access to
nongovernmental health delivery is limited.
Despite these hurdles, healthcare organizations must use their data to
help solve and improve all facets of the industry. A plethora of solutions
must be adopted or invented, with their varying degrees of success and
failure only then becoming clear. Next, we walk through some models for
an enterprise architecture and strategy that might lead to better healthcare
analytics.
MODELS FOR DATA ARCHITECTURE AND
STRATEGY
Figure 10.2 illustrates the critical components necessary to arrive at a
usable analytics solution. A few terms and acronyms require definition:
Source data. Operational or transactional software applications may
be used as a point solution (laboratory, radiology, or materials
management) or may serve as a multipurpose, mission-critical
solution such as the hospitals admit, discharge, and transfer (ADT)
system; revenue cycle management (RCM); or EHR. Data are
entered for that applications operational use, with those data then
420
becoming available for subsequent extraction to a CDR (discussed
later) or BI/CI system. Materials management involves purchasing,
inventorying, and pricing such materials. E-Apps are any of the
myriad web-based software applications. These systems
acknowledge that the Internet is a source of much valuable, usable
data.
Extraction, transformation, and load (ETL). ETL is a generic
term used across industries that refers to the process for creating data
repositories for analytical purposes. This process starts with a host
software application into which data have been entered through use
of that application. These data can be formatted for extraction by the
developer or a third party. In either case, the data need to be
transformed, which may include normalization, organization,
redefinition to ensure aligned vocabulary, and assurance of
referential integrity. The source data may then be imported or loaded
into the target repository, typically a BI/CI solution or other
applications software. New extraction methods are being developed,
but ETL is the most widely used method at the current time.
Metadata. The literal definition of metadata is data about data. In
the study of BI/CI, metadata refers to data identifying where the
source data were created, when (time and date), by whom, and for
what purpose; where they are located within the computer
architecture; and whether standards were used in the data creation
process. Metadata may be modified and created by nontechnical
personnel and may provide grouping and logical similarities and/or
differences among the data.
Central data repository (CDR). In years past, the CDR might have
been called a data warehouse. It stores larger amounts of
information, provides a replication of data from the source systems,
organizes the data for extraction for analytics, and may provide an
environment for disaster recovery (DR). The CDR supports a
systems ability to replicate and rebuild data if the original hardware
is destroyed.
Business and/or clinical intelligence (BI/CI). These terms are
synonymous with analytics. BI has been a practice throughout
various industries for more than decade and more recently has been
embraced in health care. CI is newer and parallels the proliferation of
421
electronic health information, where clinical data are both
voluminous and have tremendous mining value.
FIgure 10.2 Example of an Architectural Map. This illustrates critical
components necessary for any usable analytics solution.
Courtesy of Yale University School of Medicine.
Utilities and rules. Within the IT solution, one will find
administrative utilities, object management, and rules such as those
governing distribution or content scheduling. The functionality for
analytics, ad-hoc reporting, dashboard tools, KPIs, score card
methods, and other application software capabilities are programmed
here.
As healthcare analytics matures and is more broadly adopted, certain
outcomes will be accepted as means to measure their value. Following is a
list of requirements and functional attributes for a viable BI/CI solution.
Requirements for BI/CI Solutions
Secure. Whether combined within a CDR or in a BI/CI solution, all
422
data must be secure. For PHI and HIPAA compliance, the systems
architectural environment must ensure that each data element
maintains a secured utility to limit access, use, and improper
exposure. Thus each data element is tagged to control its use and to
provide an essential audit trail indicating when, how, and by whom
the data have been accessed.
Extensible/scalable. Considering the expanding direction of
healthcare data, the BI/CI solution must be able to scale up and
extend to new areas. Just as the adoption of EHRs has proliferated
over the last decade, so newly expanded uses for the Internet and
mobile technologies will require more data as they are developed.
Additionally, governments role is still rather fluid, but it seems
relatively safe to assume that government agencies will demand
increased care measurements and reporting in the future. Lastly, like
most maturing industries, the healthcare industry is consolidating:
Smaller shops are coming together to create mega-organizations that
will be combining huge amounts of historical data during their
mergers.
Integrity. Health care is a professional industry populated by a highly
educated workforce. Science and pragmatism are pervasive
throughout the healthcare arena. Nevertheless, there is always a level
of subjectivity and variability for the reasons discussed earlier;
additionally, there is great emphasis on humanity in sustaining a
patients health. From both of these vantage points, accuracy and
integrity are imperative. Data must have high integrity. For users to
trust the system there must be a high degree of veracity and
authenticity across software applications, the resulting data sets, and
any analytics and reporting. You do not have to look far to find
voluminous anecdotal history of software systems failing to be
adopted because providers did not trust the output.
Performance. EHR, laboratory, and other clinical point of care
systems must provide real-time information for healthcare providers.
Their clinical decision making depends on the timeliness of this
information. Conversely, BI/CI is inherently a retrospective analysis
of transactional and clinical systems. Different data feeds to the
BI/CI system may be real time, daily, monthly, or even quarterly.
Well-timed performance is a subjective observation: Each output
423
must be timely (whether the information is needed immediately or
periodically). When an output from a BI/CI system is required, it
must be readily accessible. Depending on the form of output, the
system should provide the information in a few seconds to a few
minutes. A responsive solution will have automated utilities for the
scheduled delivery of many objects such as KPIs, dashboards,
statistics, or other predetermined requirements. Ad-hoc demands will
take longer in the first run of an inquiry.
Functional Attributes of BI/CI Solutions
Accessibility. All modern workers are knowledge workers who
require information to perform their jobs most effectively, and a
properly designed BI/CI system should make access to data
relatively easy for those who have been trained in its use. Most
internal consumers or users will never create intelligence content, but
they or their managers should understand the concept of how the
system works if they are to have properly trained personnel develop
content to their needs. From a control standpoint and to satisfy
general security concerns, a healthcare organization may choose to
program utilities that identify improper commands or inordinate
demands from the central processing unit (CPU) to control illogical
or improper personnel access. In environments where operational
personnel are dependent upon a small IT or analytics group to
develop content, there will be an untenable bottleneck.
Usability. One reason for the relatively slow adoption of BI/CI
solutions is that many are still rather complex to use. The more
content that can be built or modified by nontechnical personnel, the
greater the likelihood of success. For example, a clinician doing the
hands-on work of health care may appreciate which kind of BI/CI is
usable in the field better than an IT staff member: If they can drill
down into the information themselves, the BI/CI system can provide
users with information agility that would not be possible otherwise.
In addition to tables and spreadsheets, graphic and visual
representations of data are invaluable. Because of its functional
value, a BI/CI solution may regularly generate hundreds of unique
content displays at any one hospital.
Actionable. If the content from a BI/CI solution is not actionable
424
that is, if it does not improve decision making, enhance day-to-day
performance, solve relevant problems, or improve processesthen it
will undoubtedly be discarded onto the giant heap of unrealized
clever ideas. A BI/CI solution can only provide visibility into an
organizations data; by itself, it will not make decisions. The BI/CI
solution must provide clear evidence so that appropriate personnel
feel comfortable in changing methods, managing their function,
doing their work, and making decisions to improve the organization
in ways that matter based on those data.
Asking from the answers. This principle is based on the adage, You
dont know what you dont know. Once a BI/CI solution is in the
hands of healthcare workers, they may know better which kind of
questions to ask. A competent analytics solution will allow an
inquirys answer to be queried time and again. In the world of
BI/CI, you do not know the next question until you see the answer to
the first. In theory, this may occur ad infinitum, especially as the
questioning audience grows and multiple thoughts begin to challenge
the results. In turn, an analytical result may promote a follow-up
question or the addition or deletion of data to better refine the
original inquiry. It becomes a circle of better investigation of data
based on digging and creatively using stages of analytical output.
Roles-based use. Every employee has one or more work roles. DSS
and EIS are antiquated concepts, as every workernot just
executivesneeds some level of actionable information to perform
his or her job or role. The dissemination of information is imperative
and no longer valuable only to management and executives. Such
distribution may consist of the presentation of small, daily refreshed
tactical data such an insurance denial report or an appointment
follow-up list. Also appropriate to her or his role, an executive may
want to review more sweeping views of different tactical or strategic
business operations, with the ability to drill down into highlighted
areas of interest or concern. There may conceivably be as many
daily, weekly, and monthly analytical outputs as there are users in
the enterprise.
Dashboards. Dashboards allow often used data sets to be graphically
presented in one or a few locations on a regular basis, where the user
has unlimited access and may toggle between information with just a
425
few keyboard or mouse clicks. The idea is to centrally locate lots of
information, according to agreed-upon KPIs. A first glance gives a
summary level, which can be clicked on to access supporting
detailed data.
Retrospective nature of BI and CI. Real-time clinical decision
making is based on a combination of the clinicians experience and
the information available within the EHR. While more real-time
analytic innovations are under development, with todays
technology, the traditional frequency of updating a CDR or pure BI
solution is daily. Therefore, analytical reports, dashboards, and other
outputs are generally produced for the various users the morning
after systems have been updated on the prior evening. This gives end
users data to inform decisions going forward in the new day.
EXAMPLES OF BI/CI AT WORK
Example 1
A relatively small obstetrics/gynecology practice had problems tracking
compliance in its administration department and inventorying Gardasil.
The Gardasil vaccine is used to prevent human papillomavirus infection,
which may lead to cervical cancer; for maximum effectiveness, it needs to
be administered as a series of doses in the same patient. This drug has a
limited shelf life, so an expensive inventory must be destroyed if not used.
Using HIS and technology systems to provide a BI/CI solution, the
practice combined data from its purchasing, scheduling, EMR, and
provider productivity systems to stratify by provider which patients were
receiving Gardasil. To better manage the inventory of this vaccine, it then
calculated the drug volumes that would be necessary based on trending
appointments and the targeted patient population demographics. The data
provided (1) a list of patients for scheduling and (2) an order process for
Gardasil. The resulting analytics provided a quality factor to ensure proper
patient scheduling and administration of the drug (including a potential
recall) while saving the practice hundreds of thousands of dollars in
wasted inventory. This example demonstrates the many different
disciplines involved in improving a single clinical process, along with the
resulting impacts on cost management and clinical quality.
426
Example 2
A busy, university-based practice combined a dozen data sets relating to
provider productivity, comparative patient satisfaction feedback,
benchmarked relative value unit (RVU) measures, and other data sets. The
ultimate goal was to deliver a month-end summary of each providers
compensation. Heretofore, this exercise necessitated more than eight
unique report formats, took many employee hours, and was delivered more
than a week after the month-end closing.
The new BI/CI-driven single-page summary included an information
dashboard with nine different windows combining all of the former
discrete reports. The process was completely automated, was deliverable
in the providers preferred modality (email, fax, or hard copy), and was
customized for each department reflecting the specialty nuances of
measures. This combination of data to present timely, actionable
information was challenging to create, but once it was in place, it became a
valuable management tool, useful to many.
THE FUTURE OF BI/CI
There will undoubtedly be tremendous opportunities for individuals who
understand the complexities of health cares BI/CI. Prospective provider
organizations and BI/CI application vendor roles will include the
following:
BI/CI executives and SMEs: operations, quality, planning, medical
staff leadership, nursing leadership, strategy, and management who
understand the problems inherent in each discipline and, therefore,
what the information solutions might be
Systems architects: personnel who are generally highly technical and
create the environment for acquiring, storing, and accessing the data
necessary for BI/CI
Programmers: personnel who program or create software
applications and then support and enhance these programs once the
software is developed and used in production
Data analysts: personnel who work with data and often act as the
liaisons between technical and non-technical users of data and
information
427
Content creators: personnel who format and structure data in a
desired or logically deliverable design for all roles in the
organization
Implementation: trainers and personnel who deal with interface
creation, data translations, roll-out to users, updates, and issues
resolution
Project managers
The notion of intelligence and analytics is profoundly useful and
beneficial at so many levels. Considering the healthcare industrys size and
influence on the U.S. economy as well as its dependency on the
voluminous amounts of data and types of BI/CI for its operational
efficiency and clinical performance, it must and will accelerate the
adoption of better methods in leveraging those data. Talented project
managers are the key to organizing the detailed tasks into projects,
persevering to pull together the many threads inherent in BI/CI solutions,
and helping users adopt their use. As in other healthcare IT pursuits, BI/CI
work is not without significant challenges, but it is greatly rewarding once
accomplished. Despite the structural and logistical limitations that must be
overcome to reach this goal, there is now enough political, consumerbased, and industry demand to support its achievement.
SUMMARY
If we all agree that making current, practicable information available to
workers will likely improve performance within the healthcare workplace
environment, then BI/CI is a viable concept. Health care has an inordinate
number of moving parts and an incredible plethora of raw and
uncoordinated data. In lay terms, bringing all those data into a single
location, harmonizing them, and making the results available to the masses
of potential users is a very good thing.
Beyond good, it is necessary. Health cares unsynchronized and
unaligned methods of providing care have become untenable and
unaffordable as organizations attempt to effectively manage their
businesses. As the government, quality watch groups, employers, and
patients as consumers of health care prescribe necessary change, the
healthcare industry must put into practice proven methods of gathering and
distilling information into intelligence. Other industries have already
428
demonstrated the value of BI, and so, in growing ways, has health care.
All healthcare organizations have limited capacity to implement IT
investments. Currently, there are mandated priorities, such as with EHRs
and ICD-10 conversions; organizations are developing Internet strategies,
e-mobile integration, security solutions, and other competing initiatives as
well. However, recent surveys of C-level and IT executives, as well as
industry member groups such as the Healthcare Information and
Management Systems Society and the Medical Group Management
Association, consistently espouse the need for better BI.9 There are no
precise data on the use of BI in health care, but best estimates would
suggest fewer than 20% of all provider organizations have implemented a
true BI/CI solution and even fewer have done so on an enterprise-wide
level.
In addition to the pure logic of using analytics, the BI/CI paradigm has
a very intellectually rewarding and creative side. The process of
identifying what needs better workflow process and obtaining distinct
business and clinical answers, plus the ability to dig around and present
data solutions, can be deeply fulfilling. With properly designed and
implemented data architecture and intelligence mining software, one can
discover empirical answers to both mundane and extraordinarily complex
healthcare questions.
KEY TERMS
Canned reports
Central data repository (CDR)
Central processing unit (CPU)
Dashboard
Decision support system (DSS)
Disaster recovery (DR)
Executive information system (EIS)
Extraction, transformation, and load (ETL)
Medical Group Management Association (MGMA)
Metadata
429
Pay-for-performance (P4P)
Roles-based use
Source data
Discussion Questions
1. Who might use and benefit from healthcare BI/CI?
2. What do you think might be the preferred way of structuring a BI/CI
group within a hospital? For a multispecialty ambulatory group?
3. How might you describe the differences between non-healthcare
industries and the healthcare industrys use of analytics?
4. If applicable, how have you used analytics as part of your work in a
healthcare organization? Did you use BI or CI?
5. Do you believe that healthcare organizations might push personal
analytics out to patients for their own monitoring and improvement of
health and quality of life issues?
6. Would you use more personalized health information in your own
life? How might you do so? What would be your preferred method of
communication?
7. Do any of the roles described in relation to the use of BI/CI interest
you, and why?
8. Are the existing HIPAA, PHI, and other security measures sufficient
to allow you to trust healthcare organizations in aggregating, mining,
and analyzing data?
9. Do you believe that BI/CI should be mandated and controlled by the
various government agencies that are involved in health care? Why or
why not?
REFERENCES
1. Mulcahy, R. (n.d.). Business intelligence definition and solutions
business intelligence topics covering definition, objectives, systems and
solutions. CIO.
430

Get Professional Assignment Help Cheaply

Are you busy and do not have time to handle your assignment? Are you scared that your paper will not make the grade? Do you have responsibilities that may hinder you from turning in your assignment on time? Are you tired and can barely handle your assignment? Are your grades inconsistent?

Whichever your reason is, it is valid! You can get professional academic help from our service at affordable rates. We have a team of professional academic writers who can handle all your assignments.

Why Choose Our Academic Writing Service?

• Plagiarism free papers
• Timely delivery
• Any deadline
• Skilled, Experienced Native English Writers
• Subject-relevant academic writer
• Adherence to paper instructions
• Ability to tackle bulk assignments
• Reasonable prices
• 24/7 Customer Support
• Get superb grades consistently

Online Academic Help With Different Subjects

Literature

Students barely have time to read. We got you! Have your literature essay or book review written without having the hassle of reading the book. You can get your literature paper custom-written for you by our literature specialists.

Finance

Do you struggle with finance? No need to torture yourself if finance is not your cup of tea. You can order your finance paper from our academic writing service and get 100% original work from competent finance experts.

Computer science

Computer science is a tough subject. Fortunately, our computer science experts are up to the match. No need to stress and have sleepless nights. Our academic writers will tackle all your computer science assignments and deliver them on time. Let us handle all your python, java, ruby, JavaScript, php , C+ assignments!

Psychology

While psychology may be an interesting subject, you may lack sufficient time to handle your assignments. Don’t despair; by using our academic writing service, you can be assured of perfect grades. Moreover, your grades will be consistent.

Engineering

Engineering is quite a demanding subject. Students face a lot of pressure and barely have enough time to do what they love to do. Our academic writing service got you covered! Our engineering specialists follow the paper instructions and ensure timely delivery of the paper.

Nursing

In the nursing course, you may have difficulties with literature reviews, annotated bibliographies, critical essays, and other assignments. Our nursing assignment writers will offer you professional nursing paper help at low prices.

Sociology

Truth be told, sociology papers can be quite exhausting. Our academic writing service relieves you of fatigue, pressure, and stress. You can relax and have peace of mind as our academic writers handle your sociology assignment.

Business

We take pride in having some of the best business writers in the industry. Our business writers have a lot of experience in the field. They are reliable, and you can be assured of a high-grade paper. They are able to handle business papers of any subject, length, deadline, and difficulty!

Statistics

We boast of having some of the most experienced statistics experts in the industry. Our statistics experts have diverse skills, expertise, and knowledge to handle any kind of assignment. They have access to all kinds of software to get your assignment done.

Law

Writing a law essay may prove to be an insurmountable obstacle, especially when you need to know the peculiarities of the legislative framework. Take advantage of our top-notch law specialists and get superb grades and 100% satisfaction.

What discipline/subjects do you deal in?

We have highlighted some of the most popular subjects we handle above. Those are just a tip of the iceberg. We deal in all academic disciplines since our writers are as diverse. They have been drawn from across all disciplines, and orders are assigned to those writers believed to be the best in the field. In a nutshell, there is no task we cannot handle; all you need to do is place your order with us. As long as your instructions are clear, just trust we shall deliver irrespective of the discipline.

Are your writers competent enough to handle my paper?

Our essay writers are graduates with bachelor's, masters, Ph.D., and doctorate degrees in various subjects. The minimum requirement to be an essay writer with our essay writing service is to have a college degree. All our academic writers have a minimum of two years of academic writing. We have a stringent recruitment process to ensure that we get only the most competent essay writers in the industry. We also ensure that the writers are handsomely compensated for their value. The majority of our writers are native English speakers. As such, the fluency of language and grammar is impeccable.

What if I don’t like the paper?

There is a very low likelihood that you won’t like the paper.

Reasons being:

• When assigning your order, we match the paper’s discipline with the writer’s field/specialization. Since all our writers are graduates, we match the paper’s subject with the field the writer studied. For instance, if it’s a nursing paper, only a nursing graduate and writer will handle it. Furthermore, all our writers have academic writing experience and top-notch research skills.
• We have a quality assurance that reviews the paper before it gets to you. As such, we ensure that you get a paper that meets the required standard and will most definitely make the grade.

In the event that you don’t like your paper:

• The writer will revise the paper up to your pleasing. You have unlimited revisions. You simply need to highlight what specifically you don’t like about the paper, and the writer will make the amendments. The paper will be revised until you are satisfied. Revisions are free of charge
• We will have a different writer write the paper from scratch.
• Last resort, if the above does not work, we will refund your money.

Will the professor find out I didn’t write the paper myself?

Not at all. All papers are written from scratch. There is no way your tutor or instructor will realize that you did not write the paper yourself. In fact, we recommend using our assignment help services for consistent results.

What if the paper is plagiarized?

We check all papers for plagiarism before we submit them. We use powerful plagiarism checking software such as SafeAssign, LopesWrite, and Turnitin. We also upload the plagiarism report so that you can review it. We understand that plagiarism is academic suicide. We would not take the risk of submitting plagiarized work and jeopardize your academic journey. Furthermore, we do not sell or use prewritten papers, and each paper is written from scratch.

When will I get my paper?

You determine when you get the paper by setting the deadline when placing the order. All papers are delivered within the deadline. We are well aware that we operate in a time-sensitive industry. As such, we have laid out strategies to ensure that the client receives the paper on time and they never miss the deadline. We understand that papers that are submitted late have some points deducted. We do not want you to miss any points due to late submission. We work on beating deadlines by huge margins in order to ensure that you have ample time to review the paper before you submit it.

Will anyone find out that I used your services?

We have a privacy and confidentiality policy that guides our work. We NEVER share any customer information with third parties. Noone will ever know that you used our assignment help services. It’s only between you and us. We are bound by our policies to protect the customer’s identity and information. All your information, such as your names, phone number, email, order information, and so on, are protected. We have robust security systems that ensure that your data is protected. Hacking our systems is close to impossible, and it has never happened.

How our Assignment Help Service Works

1. Place an order

You fill all the paper instructions in the order form. Make sure you include all the helpful materials so that our academic writers can deliver the perfect paper. It will also help to eliminate unnecessary revisions.

2. Pay for the order

Proceed to pay for the paper so that it can be assigned to one of our expert academic writers. The paper subject is matched with the writer’s area of specialization.

3. Track the progress

You communicate with the writer and know about the progress of the paper. The client can ask the writer for drafts of the paper. The client can upload extra material and include additional instructions from the lecturer. Receive a paper.

4. Download the paper

The paper is sent to your email and uploaded to your personal account. You also get a plagiarism report attached to your paper.

GET A PERFECT SCORE!!!