WHITE PAPER How a Modern Data Architecture Will ... Papers... · financial, research, population health, actuarial and other pertinent data. It resides in various departments, databases,

© 2020 Clarity Insights

WHITE PAPER

How a Modern Data Architecture Will Revolutionize the Healthcare Industry

cl-0034-912029-WP-Revolutionizing HC industry.indd 1cl-0034-912029-WP-Revolutionizing HC industry.indd 1 18/12/2019 15:4618/12/2019 15:46

Introduction

The role of data in the healthcare landscape has grown

exponentially in recent years and is advancing rapidly.

Data is no longer the purview of actuaries and medical

researchers; it’s becoming the backbone of the healthcare

industry. Unfortunately, many organizations are ill-

equipped to take advantage of all the benefits their data

has to offer and most can’t even access or process all of

the data they’ve already amassed. Traditional data solutions

were built based on existing demands and technologies,

but the ever-growing amount of data and the insights

that can now be extracted from it have rendered these

solutions obsolete.

Data Challenges in Healthcare

Healthcare is a heavily regulated industry and comprises

organizations with vastly different and, sometimes

opposing goals. Healthcare data is incredibly siloed

including patient, outcome, diagnostic, administrative,

financial, research, population health, actuarial and

other pertinent data. It resides in various departments,

databases, electronic health records (EHRs), provider

records, billing companies, essentially all created as a by-

product of an operational or functional process that must

then be manipulated into information that has business

value. Understandably, the integrity, consistency and

structure of this data vary dramatically, frequently resulting

in errors or redundancies which is not ideal, especially

when making vital healthcare and operational decisions.

Further complicating this morass of data is the deluge of

new information from medical devices, fitness and health

trackers, and mobile applications. The picture below

discusses some of the major data related challenges facing

the healthcare industry.

Traditional data infrastructures simply cannot keep

up with current demands. They struggle to handle the

enormity of all this data, unable to store and process

the incomprehensible amount of information available

to and accumulated by healthcare firms. As companies

spend money and IT resources simply trying to keep up,

the opportunity costs of using outdated architectures

accumulate quickly. Traditional systems are inhibiting the

growth and the potential of healthcare organizations, and

those that are slow to adopt a modern data architecture

will fall far behind their competition.

Current data challenges within healthcare

Lack of Business Insights

• Business tied up with

descriptive analytics

• No focus on

predictive and

prescriptive analytics

to support key

business decisions

Siloed Data Stores

• Data Spread across

multiple data stores

• Inconsistent

outcomes

• Maintenance hassle

Expensive Platforms

• Expensive appliance

databases: more than

$100/TB of storage in

some cases

Massive Data Growth

• Data growing from

terabytes to

petabytes to exabytes

• Unable to manage

with traditional

databases

$ $

$

$Current

DataChallenges

2 www.clarityinsights.com © 2020 Clarity InsightsREAL. CONTINUOUS. INTELLIGENCE.


What is Modern Data Architecture?

Where traditional systems flounder, outpaced by data

growth and processing needs, a modern data architecture

excels. Advanced data solutions not only handle current

demands but are built to be future-ready, and they enable

you to connect multiple hardware or software devices to

work as a single entity. By taking advantage of the cloud,

these innovative systems deliver on-demand storage

(vertical scaling) and expand to accommodate increased

processing loads (horizontal scaling).

Unlike the multiple, disparate sources used by traditional

platforms, a single data lake is the cornerstone of the

modern data architecture. Creating this lake begins with

a clear data governance program that ensures only the

highest quality data is gathered, processed and funneled

into your lake. Once the single source of data is established

and governance guidelines are in place, the focus turns to

ingestion and cataloging of metadata—mapping fields and

defining how to parse data from all sources.

Another important feature of this type of data architecture

is a robust search function, which facilitates enterprise-

wide data accessibility and empowers data stewards—

employees who act as gatekeepers to maintain the

stringent quality of the data lake. For example, if the

decision is made to migrate member data into the lake, but

some member data already exists, data stewards make

sure there are no errors or duplication of data (or effort).

Conceptual modern data architecture

Data Sources Information Delivery

Search & DataDiscovery

OperationalReports

Dashboards

Self-ServiceReporting

Enterprise Information Management

Data Governance and Security

Enterprise Data

Data Lake

Environment Management

MetadataManagement

(Catalog &Search)

EnterpriseService Bus

BusinessRules Engine

Master Data Management

Elastic Storage& Processing

AutomatedAnalysis

User/AnalyticsSandbox

Active storageIn-memory storage

Archive storageETL/Branch serviceReal-time service

ETL Framework w ABC

SQL serviceSpeech analysis

Visualization & reportingData analytics

Machine learningAnalytical models for insights

Ad-Hoc analyticsWhat-if analytics

User Sandbox

Resource analysis& allocation

Backup & disasterrecovery

Configurationmanagement

Deploymentmanagement

EnterpriseData Warehouse

Reference Data(Master Data)

Dat

a In

gest

ion

Fram

ewo

rk (

Rea

l-Ti

me/

Bat

ch)

Structured Data

Claim data:proceedure &

diagnosis codes, drug info, paid

amounts info, etc

Member data:Sex, age, location,

etc

Provider data:State, City, ZIP, etc

Public data: Procedure codes,

conversion to physical time, bundling rules,

social media, etc

Clinical data: HL7, FHIR

Other

Unstructured Data

Speech-to-text conversion data

Web logs

Other



Advantages for the Healthcare Industry

A modern data architecture empowers every employee

to extract the valuable insights from data that are

necessary to achieve business goals and propel your

organization’s success.

Fraud, Waste and Abuse Detection:

Machine learning recognizes patterns, trends and

anomalies across all datasets. This serves two main

purposes. First, it facilitates predictive analytics, a powerful

tool in forecasting trends among your membership and

providers. Second, it helps detect and combat fraud, waste

and abuse—for example, catching an orthopedic claim

submitted by an ophthalmologist.

Population Health Analytics:

By merging claim, EHR and demographic data with

information from your company’s mobile apps and medical

and consumer device data, you create a holistic view of

your patients or members. You’re then able to determine

where your efforts to improve population health are

succeeding and which areas need improvement. You can

also use these insights to transmit health screening or

fitness tracker data to physicians and alert them if a health

event is imminent, which results in improved outcomes.

Cost Reduction:

Predictive analytics assists providers and payers alike

in anticipating high-cost activities, such as unnecessary

tests, chronic conditions and delayed care. Data-fueled

insights hold enormous potential for your organization,

from reducing disenrollment to calculating premiums for a

particular population.

Compliance:

With a modern data architecture, you can separate data

from the lake and anonymize it, allowing your teams

to preserve PHI and comply with HIPAA guidelines. An

advanced data platform also enables compliance with

Meaningful Use Stage 3 objectives, which mandate

access to EHRs, immunization status reporting and secure

messaging between patients and providers.

Operational Improvements:

Your business can realize significant cost-savings with

a modern data system. A cloud-based platform offers

extensibility and on-demand storage. Implementation

costs a fraction of traditional on-premises systems, and

a full rip-and-replace is usually not necessary—a modern

architecture bridges your existing infrastructure to augment

what you already have in place. And by eliminating data

silos and creating a “single source of truth,” data is available

to all teams to drive decision-making, reduce expenses

and increase efficiency.



What to Look for in a Modern Data Architecture

A powerful, modern solution that can keep up with current

and exponentially increasing data demands is foundational

to your organization’s success.

When searching for a solution that’s designed specifically

for your business needs, seek a modern data architecture

with these key features:

• Cloud-based architecture that offers horizontal

scaling and on-demand storage and is able to

be implemented quickly

• Automation tools, such as scripts and

frameworks, to expedite the setup of your end-

to-end environment

• Robust ingestion framework that will funnel,

catalog and enable searching of all your data

and metadata

• Comprehensive data governance program,

coupled with human data stewards, to ensure

data quality and reduce inefficiency and errors

• Most important thing is both IT and business

own this now, which lends itself to a cultural

or organizational change that must also take

place in order to take full advantage of this new

technology and process

The picture below shows the progression of a modern

data architecture initiative from conceptualizing the

overall strategy and roadmap to align with business goals,

to the creation of an architectural blueprint that tailors

to the business needs and leverages modern tools and

technology platforms to build an enterprise data lake that

consolidates multiple disparate data sources. This data lake

could be built on a cloud platform and enables machine

learning and advanced analytics to deliver deep insights

into your business.

Establishing a modern data architecture

Strategy and Roadmap | Establish a strategy and

roadmap to meet business goals.

Architectural Blueprint | Establish an end-to-end architectural blueprint

tailored to your business objectives that leverages modern technologies and

scalable platforms.

Enterprise Data Lake | Establish an enterprise data lake to consolidate

all disperate data sources and process all data

needs for EDW. BI and advanced analytics using

scalable platforms.

Cloud Solutions | Establish a scalable cloud platform

MDA

Machine Learning | Establish

machine-learning algorithms to perform

predictive analytics and gain deep insights into

your business.

Advanced Analytics | Establish reactive

approach to data, and proactively leverage

analytics to manage risks and improve

decision-making.



Another key consideration is choosing a solution provider

that offers industry experience, technical mastery and

expertise in healthcare and related industries. Clarity

Insights is peerless in this regard. We offer repeatable data

ingestion, data lake transformation and machine-learning

frameworks that have already processed petabytes of

information from your industry—data from providers,

insurers, pharmaceutical firms and other healthcare and

medical businesses.

Clarity’s expertise coupled with their agnostic approach to

technology also positions them to extract the maximum

value from the client’s existing world/investments,

while moving forward over time to the newer, required

approaches already mentioned above. There’s no need to

reinvent the wheel, because when you partner with Clarity,

you’re already outpacing the competition.

Clarity’s Ingestion Framework

Data Sources Data Ingestion Framework

Unstructured Data

Clinical Data | Doctor’s Notes |

Wearables | Medical Docs | Labs/Images

internal Data

Enrollment Claims | Provider | Payers |

Medicare/Medicaid | Products |

Adjudications | Patients

Common Components

Template Design (Real-Time)

Template Design (Batch)

Real-Time Data Storage

Audit, Control, Balance

Enterprise MetadataManager (EMM)

TechnicalMetadata

DataLineage

Data Lake (RAW Zone)

RAW data

Data Profiling

Clarity’s Transformation Framework

Data Lake (RAW Zone) Data Transformation Framework

Template Design (Batch) Data Lake (Final Zone)

Hive/Spark

EDW

MPP DatabaseFACT Tables

Dimension Tables(type 1 & type 2)

Common Components

ParallelExecution

ConfigurableRules &Maping

Restartable,Fault-Tolerant

Auditable

Job Control &SLA Monitoring



Conclusion

In today’s market, data sources and technologies are

rapidly increasing. Massive volumes of data, stringent

regulations and tightening competition mean that

healthcare organizations must evolve to survive. Flexibility

and adaptability are vital to your success, and a modern,

cloud-based data architecture delivers both with less

investment and more capabilities than traditional, on-

premises systems.

Data belongs to everyone within your enterprise, and with

a modern solution that starts with a pristine data lake and

includes strong governance, flexible processing power and

data stewardship, your data becomes democratized: able

to be applied to all business problems, large and small.

By investing now in your company’s future and adopting

a long-term strategy that focuses on the horizon, your

organization will be driven by data and insights and will

outpace the competition.

Clarity Insights has executed numerous data

modernization projects, and as part of this process, we

have built repeatable frameworks for data ingestion, data

consumption and advanced analytics. These frameworks

accelerate the modernization of modern data initiatives

by ingesting and transforming data with minimal code. To

learn more about how we can help modernize your data

architecture, contact us at [email protected] or

800-920-6648.



Why Clarity

At Clarity Insights, we use data strategy, engineering,

science and visualization to help companies action

their insights.

Why? Because a majority of Chief Analytics Officers say

that their biggest challenge is overcoming cultural barriers

to new insights, as well as getting business buy-in. It is

no longer enough to take an “if we build they will come”

approach when it comes to insight systems—be it a data

lake or a machine learning model. It takes an approach that

will help build an insights driven culture.

How do we help clients do this?

We start any project by understanding our client’s business

strategy, then understand how data can make it a success.

This way we are always focused on the business outcomes

and not what data actually exists. This approach also

ensures business buy-in.

When we help you find the insights to achieve those

business outcomes, we don’t down our tools. Rather,

we help you embed them in your processes, and use

change management to obtain adoption. We also focus

heavily on knowledge transfer to our clients, ensuring

they are empowered to take action faster and with more

confidence.

These are just some of the reasons that more than 80% of

Clarity’s customers hire us for additional engagements. We

have been trusted partners to the most exacting, data-

intensive organizations in the nation for years.

Contact Us Today www.ClarityInsights.com

800.920.6648

Ramu Kalvakuntla

Chief Enterprise Architect

[email protected]

About the Author

Ramu is the Chief Enterprise Architect for modern data architecture

within the Healthcare Practice at Clarity Insights. He helps clients develop

strategies and roadmaps for their data needs and is a pioneer in building

reusable solutions that enable companies to quickly adopt modern data

architectures, facilitate advanced analytics and gain deep insights.

Ramu has strong technical expertise in the architecture and management

of big data, IoT, data warehouse, data integration and business intelligence

systems in the healthcare industry. His solid management track record and

ability to bridge technology and business goals complement his 20 years of

experience with emerging technologies including Hadoop, NoSQL, Azure,

AWS, GCP, Talend, Informatica BDE, Spark, HD Insights, RedShift and more.



Documents

WHITE PAPER How a Modern Data Architecture Will ... Papers... · financial, research, population health, actuarial and other pertinent data. It resides in various departments, databases,