the intelligent back office in healthcare neo... · ai for prediction ai research for improved...

The Intelligent Back Office in HealthcareNEO HFMA Leadership Institute KPMG Presentation

2© 2019 KPMG LLP, a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International

Cooperative (“KPMG International”), a Swiss entity. All rights reserved.

With you today

Michael Caporusso

Solution Director in KPMG’s

Intelligent Automation

Practice

[email protected]

Rachel Silverman

Director in KPMG’s

Care Continuum

Optimization Practice

[email protected]

mailto:[email protected]



Who we are

The creative

CIOs agenda:

Getting started

with digital labor

KPMG Intelligent Process Automation Thought Leadership

Digital Labor

and the future of

finance

RPA & Cognitive

Automation in

Healthcare

Transforming

business models

with RPA

Bots in the back

office: The

coming Sprint

of digital labor

KPMG

KPMG ranks the highest of the “Big 4” in digital transformation

Forrester

Insights Service Provider Wave

Leading Advisor

Horses for Sources

Intelligent Automation Blueprint

Leading Independent Advisor

With over 200 implemented Use

Cases, KPMG’s global Intelligent

Automation community of

approximately 13,000 people has

broad capabilities to support

projects around the world.

U.S. Healthcare and Life Sciences professionals

2000+

700+

KPMG operates in 155 countries With 700 +offices worldwide

Modernizing records processing in Government

Natural Language Experiences



Intelligent Automation

Optical Character Recognition

Intelligent Automation in Healthcare



The perception of disruption

“While

technology

disruption has

always been

an issue, the

difference today

is speed of this

transformation

and technology

availability”

98

%72%

See

technological disruption

as more of an opportunity

than a threat *

Say that

rather than waiting to be

disrupted by competitors,

their organization is actively

disrupting the sector in

which they operate**

49

%45

%61

%

Are concerned about the

integrity of the data on which

they base decisions **

Say they are not leveraging

digital as a means to connect

to their customers and

vendors/suppliers effectively **

Are concerned about

integrating cognitive processes

and artificial intelligence **

*Source: US CEO Outlook 2018 survey: Growing Pains offers insights into the greatest concerns of CEO’s and how they plan to pursue growth and technology driven disruption. Findings based on a

study of nearly 400 US CEOs, with annual revenues greater than US $500 million; 39% have greater than US $10 billion in revenues.

**Source: US CEO Outlook 2017 survey: Disrupt and grow offers insights into the greatest concerns of CEO’s and how they plan to mobilize for the fourth industrial revolution. Findings based on a

study of the 3-year outlook of nearly 400 US CEOs, with annual revenues greater than US $500 million; 32% have greater than US $10 billion in revenues.



The future of Intelligent Automation

*Source: “LSE - The IT function and Robotic Process Automation” – The London School of Economics and Political Science 2015

“**Source: Artificial Intelligence Market by Offering (Hardware, Software, Services), Technology (Machine Learning, Natural Language Processing, Context-Aware Computing, Computer Vision), End-

User Industry, and Geography - Global Forecast to 2025– Market and Market February 2018

$140M

600% ROI800%AND

50%

Cantor Fitzgerald’s

research suggests that

as many as 110-140 million

FTEs could be replaced

by IA technologies, reducing

costs by 25 – 40%.

Recent research from the London School of

Economics suggests a return on investment in

robotic technologies between 600% and 800%

for specific tasks.*

A recent study by KPMG LLP reports that 50 percent of

respondents would be using Intelligent Automation

technologies at scale within 3 years.

Markets and Markets

estimates that the AI,

or cognitive

computing

marketplace, will be

valued at**billion by 2025

190.61$

26

billion

$

39

billion

$

McKinsey’s 2017

report on the State of

Machine Learning

and AI estimated

that, in 2016 alone,

companies invested

between $26 billion

to $39 billion in

Artificial Intelligence.

Morgan Stanley predicts that 50-60% of white collar work is automatable, and this will have a 30% labor cost reduction.

Gartner predicts that by 2020, smart

machines will be a top five investment

priority for more than 30% of CIOs.

Top

5



Intelligent Automation marketplace is maturing rapidly

Source: 2017 & 2018* CEO Outlook Survey, KPMG LLP (June 2017 & June 2018)

Cognitive technologies

81%Of CEOs are emphasizing trust,

values and strong culture to sustain

the organization’s future

These technologies —from robotic process

automation to cognitive automation—are advancing

at a staggering pace, and will disrupt almost every

business and industry.

Investment in cognitive technologies will

be an area of focus for almost 60% of

CEOs through 2020

45% say they are not effectively

leveraging digital to connect with their

customers

Connecting

with customers45%

60%

60% worry their organizations’ sensory capabilities

and innovative processes will not stand up to rapid

disruption

Active disruption

to gain insightStaying competitive

means embracing digital

86%said their organizations are

actively disrupting their

own sectors*

61% are concerned about integrating

cognitive processes and artificial

intelligence in the workplace.

The concern for

integration

Source: 2017 CEO 12017 anOtk Survey, KPMG LLP (June 2017)

© 2019 KPMG LLP, a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International



Cooperative (“KPMG International”), a Swiss entity. All rights reserved.8© 2019 KPMG LLP, a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International

Cooperative (“KPMG International”), a Swiss entity. All rights reserved. NDPPS 849511

Classification of intelligent automation capabilities

ACT RULES LEARN REASON THINKlike a human Basic process

automation

— Macro-based

applets

— Screen level & OCR

data collection

— Workflow

automation

— Process mapping

— Self executing

Enhanced automation

— Built-in knowledge

repository

— Learning capabilities

— Ability to work with

unstructured data

— Pattern recognition

— Reading source data

manuals

— Natural language

processing

Cognitive automation

— Artificial intelligence

— Natural language

recognition &

understanding

— Self-learning

(sometimes self

optimizing)

— Processing of super

data sets

— Predictive analytics/

hypothesis generation

— Evidence-based

like a human

Class I Class II Class III




Challenges across the care continuum in healthcare

Emergency

Department

Acute Care

Ambulatory CarePatient is readmitted

Patient provided generic

discharge instructions

Improperly staffed

units and poor bed

management. Delays

in care and increased

LOS

Delays in procedures

or treatments

Delay in discharge due

to unaddressed barriers

or post discharge needs

Delay in diagnosis and

DRG determination

Billing delays due to claim

rejections and coding issues

Supplies wasted when

performing procedure

Key

Clinical Challenges

Patient no-show for follow-up

appointments and does not

fill prescriptions.

Readmission risk increases




Applying IA across the care continuum in healthcare

Emergency

Department

Ambulatory Care

Patient provided tailored

instructions based on

personality assessment

Number and frequency of

reminders provided based on

the patient’s patterns and

preferences. Alert sent to

staff if medication was not

filled. At home devices to

aid patients in compliance

Scheduling optimization

of patients and staff to

reduce delays and

case cancellations

Machine learning used to target

key barriers to discharge and

provide EHR CDS alerts

Provider alerted that

medication has not been

administered –

concurrent monitoring of

care variation

Billing errors caught with AI/

machine learning, preventing

rejections and coding issues

Analytics used to identify

resource utilization and

opportunities to identify

wasteKey

Analytics and

automation Use predictive

analytics to support

acuity-based staffing

and dynamic shifting

of staff for current

census.

Acute Care

Patient is identified as high risk for

readmission. Appropriate action taken at

discharge. IA to assess risks for mortality

and complications

Use analytics for

placement of patient in

appropriate bed

IA used to detect benign

or malignant tumors in

radiology images for faster

diagnosis



Targeting healthcare back office functions

— Employee on-boarding and off-boarding

— Payroll

— Time recording and compliance

— Repeatable tasks in ERP

— Email notifications

— Populating/aggregating employee

information— Month-End reporting

— Invoice

processing/exceptions

— AP/AR actions

— Close & reconcile sub-ledgers

— Asset depreciation and

impairment

— Fixed asset reporting

— Financial forecasting

— Invoice validation &

processing

— Tax filings

— Order flow through

— Inventory Mgmt.

— Exceptions/fallout

— Research/document

review

— Document preparation

— Controls automation

— Virtual agents (chat bots)

— Call center “agent assist”

— Task execution

— NLP enabled analytics

— Social media mining/monitoring

— Predicting high value sales leads

— Manual CRM updates

— Virtual sales agents

Human Resources

Legal/Compliance

Finance & Accounting

Sales & Marketing

Supply Chain

Customer Support

He

alt

hc

are

Ba

ck

off

ice

Fu

nc

tio

ns



Applying Intelligent Automation to healthcare

Best

Practice:

Start with a

Strategy

AI for Detection

AI research for detection: Radiology,

Pathology, Dermatology, Ophthalmology,

Gastroenterology, Cardiology.

Incorporated into workflows to improve

time from detection to treatment.

AI for Prediction

AI research for improved predictive

accuracy, examples include: in-hospital

mortality, readmission, complications in

ICU patients. Speed and accuracy tested

to support clinical decision making.

AI for Patient Interaction

AI research for voice assistance: patients

in acute setting and home care.

AI is currently used to support call centers

for non-clinical patient interactions –

scheduling, bill payment, questions

AI for Operational Efficiency

AI algorithms used for improved

efficiencies, examples: voice-enabled

workflows, contract, invoice, payroll, and

expense management, chart abstraction,

predictive claims denials and submittals, .

Incorporation into EHRs for improved

‘usability’



Case Study: Improving payment accuracy

Modernizing and accelerating a time-consuming and manual

documentation intake validation process at a healthcare agency with:

― Robotic process automation (RPA)

― Optical character recognition (OCR)

― Machine learning (ML)

― Natural language processing (NLP)



Case Study: Improving payment accuracy



To successfully

incorporate Intelligent

Automation within

processes and teams,

organizations must

proactively address the

impacts to their people

& the overall organization

in order to minimize

business disruption and

expedite the timing of

benefits realization.

Organizational & people impacts – our point of view

Changing Behaviors

Adopting and adapting

the new ways of working

Leadership Vision

Agreeing on future state

vision for the

organization

Talent

Hiring, reskilling

and exiting talent

Culture Shift

Overcoming the

fear factor

Workforce Shaping

Adaptive workforce realignment for

evolving digital labor needs

Organizational

& People Impacts

Speed of

ImplementationThe rate of change is faster

than traditional process and

system implementations

Constant ChangeIntelligent Automation

implementations will be iterative

and constantly evolving to

develop optimal workforce

productivity and ROI

Demands a Higher Purpose

ConversationEmployers will need to

understand and engage with the

impact they will have on society

Unique Characteristics of Intelligent Automation Implementations



Getting started

“Size the Prize” – Evaluate processes by suitability for automation and effort to estimate overall

benefit potential

Conduct a Proof-of-Concept – Demonstrate technology effectiveness and validate performance

Define a Deployment Roadmap – Outline steps to stand up an Intelligent Automation capability and

begin to capture the benefits



Additional considerations and lessons learned

Establish an

enterprise-wide

capability

Partner with your

technology function

Strike the balance

of your digital

transformation

Protect your

business case

Select vendors

aligned with your

ambition

Set your priorities

and the rest will

follow

Build solid

foundations

Identify and

incentivize talent

Start small; deliver

swiftly

Consider business

scalability

Evolve your

analytics capability

Automation ‘horses

for courses’



Questions to start thinking about todayIt’s no longer business as usual

— Where can we streamline and enhance our patient or provider experience?

— Where can we reduce the number of mundane and respective tasks our workforce

performs?

— What message will automation bring to our workforce and how would the future look?

— What are the risks and costs we face when we have rework or corrections?

— Is our data telling us all we need to know?

— Can we offer more within our current footprint? Could we expand our scope/market

share of services we offer?

— Where are we with automation now and where can we expand?

— What if our competitors automate and we don’t?

© 2018 KPMG LLP, a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent

member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.

The KPMG name and logo are registered trademarks or trademarks of KPMG International.

The information contained herein is of a general nature and is not intended to address the circumstances of any particular

individual or entity. Although we endeavor to provide accurate and timely information, there can be no guarantee that such

information is accurate as of the date it is received or that it will continue to be accurate in the future. No one should act on

such information without appropriate professional advice after a thorough examination of the particular situation.

kpmg.com/socialmedia

Some or all of the services described herein may not be

permissible for KPMG audit clients and their affiliates.

the intelligent back office in healthcare neo... · ai for prediction ai research for improved...

Documents