case study

administrative-simplification-how-to-save-a-quarter-trillion-dollars-in-us-healthcare.pdf

McKinsey Center for US Health System Reform

Administrative simplification: How to save a quarter-trillion dollars in US healthcare Perspectives on the Productivity Imperative in US Healthcare Delivery

October 2021

McKinsey Center for US Health System Reform

McKinsey Healthcare Systems and Services Practice

October 2021

Cover image: We Are Aphra Ltd

McKinsey.com

@McKinsey

Published by McKinsey & Company, 150 West Jefferson,

Suite 1600, Detroit, Michigan 48226.

Administrative simplification: How to save a quarter-

trillion dollars in US healthcare meets the Forest Steward-

ship Council (FSC) chain of custody standards.

The paper used in Administrative simplification: How to

save a quarter-trillion dollars in US healthcare is certified

as being produced in an environ mentally responsible, socially

beneficial, and economically viable way.

Printed in the United States of America.

Nikhil R. Sahni ([email protected]) is a partner in the Boston office and a fellow in the Economics Department at Harvard University.

Prakriti Mishra ([email protected]) is an associate partner in the Washington, DC, office.

Brandon Carrus ([email protected]) is a senior partner in the Cleveland office.

David M. Cutler is the Otto Eckstein professor of applied economics at Harvard University.

Administrative simplification: How to save

a quarter-trillion dollars in US healthcare is

published by the McKinsey Center for US Health

System Reform, part of McKinsey’s Healthcare

Systems & Services Practice. The McKinsey

Center for US Health System Reform is led by

Stephanie Carlton, David Knott, Edward Levine,

and Nikhil Sahni.

McKinsey & Company

Healthcare Systems & Services Practice

150 West Jefferson, Suite 1600

Detroit, Michigan 48226

This publication is not intended to be used as the basis for trading in the shares of any company or for undertaking any other complex

or significant financial transaction without consulting appropriate professional advisers. No part of this publication may be copied or

redistributed in any form without the prior written consent of McKinsey & Company.

Productivity imperative leaders: David Knott, Edward Levine

Global healthcare practice leader: Shubham Singhal

North America healthcare practice leader: Drew Ungerman

Analysts: Gary Chia, Julius Ewungkem, Nicolas Garcia, Brooke Istvan,

Neda Bassir Kazeruni, Crosbie Marine, Chrissy Meder, Garam Noh,

Rahi Punjabi, Mara Reichle

Editor: Allan Gold, Elizabeth Newman, Mark Staples

Assistant editor: Lyris Autran

Art director: Ginny Hull Hartline

Copy editor: Susan Schwartz, Sarah Smith

External relations: Sharmeen Alam

Lead reach and relevance partner: Scott Blackburn

Practice managers: Julia Barclay, Cameron Kennedy, Heather Megosh

Knowledge manager: Eija Erasmus

Nikhil R. Sahni Prakriti Mishra Brandon Carrus David M. Cutler

McKinsey Center for US Health System Reform

Administrative simplification: How to save a quarter-trillion dollars in US healthcare Perspectives on the Productivity Imperative in US Healthcare Delivery October 2021

The research underlying this report was led by three McKinsey consultants—Nikhil R. Sahni, a partner; Prakriti Mishra, an asso ciate partner; and Brandon Carrus, a senior partner—in conjunction with David M. Cutler, Otto Eckstein Professor of Applied Economics at Harvard University.6 Valua ble perspectives and advice were offered by a distinguished panel of academic and industry experts, including Julia Adler Milstein, Tanya Bentley, David Blumenthal, Melinda Buntin, Michael Chernew, Gaurov Dayal, Wendy Everett, William Frist, Ishani Ganguli, Elizabeth Goodman, Michelle Hood, Rob Huckman, Chip Kahn, Joe Kimura, Bob Kocher, Annie Lamont, Heather McComas, Sandhya Rao, Jaewon Ryu, Mario Schlosser, April Todd, Mike Vennera, and Wendy Warring.7

The report also benefited enormously from the contributions of McKinsey’s global network of industry experts. It drew on McKinsey’s indepth analytical expertise, our work with leading healthcare organiza tions, and our understanding of healthcare systems around the world.

The authors would like to thank the ex ternal and internal advisers for their con tributions, as well as Gary Chia, Julius Ewungkem, Nicolas Garcia, Brooke Istvan, Neda Bassir Kazeruni, Crosbie Marine, Chrissy Meder, Garam Noh, Rahi Punjabi, and Mara Reichle, who helped with analyses. In addition, the authors would like to thank Sharmeen Alam, Lyris Autran, Allan Gold, Ginny Hull, Elizabeth Newman, Sarah Smith, and Susan Schwartz for their help in editing, producing, and dis seminating this report.

This report, “Administrative simplification: How to save a quartertrillion dollars in US healthcare,” lays out a roadmap for how to capture $265 billion of administrative spend ing savings in healthcare. We built what we believe to be a firstofitskind bottomup analysis of how to best allocate the $950 bil lion of ad mini strative spending in healthcare today to stake holder groups such as private payers, hospitals, and physician groups. We translated each stakeholder group’s profit andloss (P&L) statement into functional focus areas, such as a financial transactions ecosystem and admini strative clinical support functions, to align with about 30 known inter ventions that could drive these savings. The aim of this indepen dent report, produced by the McKinsey Center for US Health System Reform, is to arm public and private sector leaders with factbased insights to guide informed decision making.1

This report continues a series of perspectives on the productivity imperative in US health care delivery. This effort began over a decade ago with an investigation of why healthcare spending was higher in the United States than in other wealthy countries.2,3 Following this, “The next imperatives for US healthcare” report laid out three steps the country could take to better control that spending: achieve rapid—and dramatic— productivity improve ments in the delivery of health services, im prove the func tion ing of healthcare markets, and improve population health.4 Most recently, “The productivity imp er ative for healthcare delivery in the United States” reframed the healthcare discussion from waste ful spend ing to improving productivity across labor, capital, and multifactor productivity.5

1 This report was not commissioned or sponsored in any way by a business, government, or other institution. 2 Diana Farrell, Eric Jensen, Bob Kocher, Nick Lovegrove, Fareed Melhem, Lenny Mendonca, and Beth Parish, “Accounting for the cost of US

health care: A new look at why Americans spend more,” December 1, 2008, McKinsey.com. 3 Jesse Bradford, David Knott, Edward Levine, and Rodney Zemmel, “Accounting for the cost of US healthcare: Prereform trends and the

impact of the recession,” December 2011, McKinsey.com. 4 Shubham Singhal and Erica Coe, “The next imperatives for US healthcare,” November 1, 2016, McKinsey.com. 5 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,”

February 27, 2019, McKinsey.com. 6 Nikhil Sahni is also a fellow in the Economics Department at Harvard University. 7 By reviewing this paper, no individual is endorsing its conclusions. All errors remain our own.

Preface

2 Preface

McKinsey & Company McKinsey Center for US Health System Reform

Preface 2

Executive summary 4

Thank you to our reviewers 10

Chapter 1. Introduction 11

Chapter 2. Analytical framework 17

Chapter 3. Financial transactions ecosystem 27

Chapter 4. Industry-agnostic corporate functions 39

Chapter 5. Industry-specific operational functions 43

Chapter 6. Customer and patient services 47

Chapter 7. Administrative clinical support functions 53

Chapter 8. Seismic interventions 59

Chapter 9. How to catalyze change 67

Table of contents

3Table of contents

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

clearinghouses; and, in some cases, appeals by providers who disagree with the payment amount must be heard.

Further, the US healthcare system is highly regulated. This leads to more admini strative spending in areas rang ing from adhering to compliance re quirements, such as the Health Insur ance Portability and Accountability Act of 1996 (HIPAA), to participating in new markets like Medicare Advantage. The intent of policymakers is to provide patients with better healthcare; often, for organizations, new administrative expenses are partially the cost of doing business to meet these requirements. But this can also become another layer of expense into which inefficiencies and errors can creep. Other challenges include the need to manage labor dis placement in an industry that is a driver of US workforce growth.3

A new approach Typical approaches to sizing the op portunity for administrative spending reduction tend to compare the United States to other countries in the Orga nisation for Economic CoOperation and Development (OECD). However, the conclusions reached from such an approach may not account for the idio syncrasies of the US health care system and thus may not provide a basis for action. For example, Canada may have lower administrative spending as a percent of total healthcare spending, but it mostly uses a singlepayer sys tem that may not provide the level of choice, access, and innovation that the US sys tem fosters and that some Americans demand.

Instead, we offer a pragmatic perspec tive that addresses how the US health care system could reshape administra tive spending by payers and providers

Every organization or largescale system needs a base of administrative functions to run. As these functions adopt new technologies and innovations, spending typically drops and quality improves. Consider payment processing, which is faster and cheaper than ever, or signing up for a new mortgage, for which you can get preliminary approval on your phone in minutes. Despite gener ations of tech nological advance ments, however, the US healthcare system remains stuck: pro ductivity and quality have stag nated, and change has been slow.1

Of the nearly $4 trillion spent on health care annually in the United States, admin i strative spending is about one quarter of the total; delivery of care is about three quarters. But what portion of that administrative spending is un necessary, and how can it be simplified?

To answer these questions, it is critical to understand what is truly necessary spend ing. The US healthcare sys tem, with thou sands of hospitals and physician groups and more than 900 payers, is geared both to local service and to competition.2 The predominant feefor service payment model puts competitive checks and bal ances on payers, hos pitals, and physician groups. This leads to a number of benefits for the United States, such as being known as a world leader of innovative care delivery. But this fragmentation can also lead to unnecessary spending due to the number of com muni cation and transac tion points among all these organizations. For example, for a healthcare claim to be paid, it must go through multiple hand offs: payers may have to validate the medical necessity of a procedure before authoriz ing physicians to provide the service; phy sicians and members must submit claims to payers; payers need to review and then contact providers to con firm details; pay ments have to flow through multiple

Executive summary

4 Executive summary

McKinsey & Company McKinsey Center for US Health System Reform

functions that are mostly industry agnostic, such as finance and human resources

— Industry-specific operational functions: Backoffice, nonclinical functions that are mostly industry specific, such as underwriting, en rollment, quality reporting, and accreditation

— Customer and patient services: The set of activities and processes that provide services to customers, typically done via call centers and increasingly moving toward digital and selfservice functions

— Administrative clinical support functions: Activities that have a clinical component (for example, nursing administration, case man agement), which can be customer facing and require some clinical expertise but are not related to the handson care of patients

within the current system (Exhibit A). The goal is not to reduce administrative spending to zero but rather to gain the highest value for each administrative dollar spent without sacrificing quality or access.

Too often, payers’ and providers’ profit andloss (P&L) statements do not pro vide enough detail to estimate what is necessary and unnecessary spending. Even when they do, the data are not broken down in a way that mimics how the organization operates. From our experience, administrative spending can instead be reorganized into five functional focus areas (Exhibit B):

— Financial transactions ecosystem: The movement of all payments, claims, and billing throughout the healthcare ecosystem among payers, hospitals, phy sician groups, and customers

— Industry-agnostic corporate func- tions: Backoffice, nonclinical

Exhibit A

Breakdown of administrative spending by stakeholder group

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 1.1 and Exhibit A of 9.2 (total 22 exhibits)

US healthcare spending by type of spending

% of total, 2019 (100% = $3.8 trillion)

Breakdown by stakeholder group

$ billion, 2019 (percent of total administrative spending)

Total administrative spending 950 (100%)

180 (19%)

250 (26%)

205 (22%)

80 (9%)

235 (24%)

Private payers

Hospitals

Physician groups¹

Public payers²

Other sites of care³

Note: Medical spending is not within the scope of this report. ¹ Hospital-a�liated and independent physician groups; employed physician groups included in hospitals. ² Includes administrative spending for fee-for-service Medicare and Medicaid, Children’s Health Insurance Program (CHIP), Department of Defense, Department of Veterans A�airs, and other federal programs.

³ Includes, for example, dental services, home healthcare, and nursing care facilities. Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Administrative spending: All activities in support of the delivery of care, includ- ing services like payment transactions, back-o�ce corporate and operational functions, customer and patient services, and ad- ministrative clinical support

Medical spending: Costs incurred for direct delivery of care, including time spent by physicians and clinical nurses on direct patient care, prescription drugs, and clinical IT

2019

5Executive summary

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

tions that nurse managers use to man age staffing and budgeting.

Some other interventions can be made “between” organizations. These require agreement and collaboration between organizations but not broader, industry wide change; they could deliver about $35 billion in annual savings, or 4 percent of total ad mini s trative spending. Building payer– provider communications plat forms that unify messaging to customers is one example.

All the within and between interventions have a positive return on investment and, in our experience, can be deployed using current technology and nominal invest ment (that is, onetime spending of 0.7 to 1.0 times the annual runrate savings).

The third intervention type is “seismic” and requires broad, structural agreement and changes across the US healthcare system.5 These interventions could de liver about $105 billion in annual savings, or 11 percent of total administrative spend ing. Seismic interventions—including those that require technology platforms,

Saving a quarter-trillion dollars To our knowledge, this approach to cate gorizing administrative spending is the first of its kind. It allows us to break up an administrative function into two parts: what work is necessary, and what could be eliminated in the next three years through proven techniques while holding or improving access and quality at today’s levels.4 By identifying simpli fication op portunities for each functional focus area, we were able to build a road map of about 30 interventions that could deliver up to $265 billion in annual savings (Exhibit C). This is based on three types of interven tions: “within,” “between,” and “seismic.”

The first type is “within” interventions, which can be controlled and implement ed by individual organizations. These within interventions could deliver about $175 billion in annual savings, or 18 percent of total administrative spending. Some examples include automating repetitive work in back office functions, such as human resources and finance, and integrating a suite of tools and solu

Exhibit B

Breakdown of administrative spending by functional focus area

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.6 and Exhibit B of 9.2 (total 22 exhibits)

Note: Figures may not sum to 100%, because of rounding. ¹ Stakeholder groups not shown include public payers ($80B) and other sites of care ($235B). ² Hospital-a�liated and independent physician groups; employed physician groups included in hospitals. Source: Centers for Medicare & Medicaid Services; McKinsey analysis

$ billion, 2019

Financial transactions ecosystem

Industry-agnostic corporate functions

Industry-speci�c operational functions

Customer and patient services

Administrative clinical support functions Other

$950 billion $180 billion $250 billion $205 billion

Total¹ Private payers Hospitals Physician groups²

21%

14%

11%

39%

23%

28%

10%

25%

17%

16%

46%

24%

44%

13%

6 Executive summary

McKinsey & Company McKinsey Center for US Health System Reform

healthcare organizations must be vigilant to avoid biases, such as algorithms built on skewed data that could adversely affect equity or access for vulnerable populations. In addition, many interventions that rely on auto mation should be coupled with re skilling programs that allow existing talent to be placed in highervalue roles.

A roadmap for action Administrative simplification may not be at the top of stakeholders’ priority lists, but the potential to save $265 billion could be compelling to leaders across healthcare. Even better, these savings are available

operational alignment, or payment de sign—generally benefit from partnerships between the public and private sectors to align incentives for change.

Many seismic interventions address the same sources of spending as the within and between ones but take the savings a step further. Accounting for this over lap, we estimate total savings across all three types of interventions at about $265 billion, or 28 percent of total administrative spending.6

Furthermore, all interventions come with some specific limitations: when deploying these interventions, especially automation,

Exhibit C

Savings opportunities across known intervention types

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit C (total 22 exhibits)

“Within” Interventions that can be controlled and implemented by individual organizations

“Between” Interventions that require agreement and collaboration between organiza- tions but not broader, industry-wide change

• Financial transactions ecosystem (prior authorization): Align jointly on PA criteria such as medical necessity or required documentation • Customer and patient services: Build strategic payer- provider platforms to reduce demand by proactively sharing data (for example, providing list of in-network specialists to physicians)

“Seismic” Interventions that require broad, struc- tural agreement and changes across the US healthcare system

• Technology platforms: Adopt a centralized, automated claims clearinghouse; prioritize high-value interoperability use cases • Operational alignment: Standardize medical policies; standardize physician licensure; streamline quality reporting • Payment design: Modularize product design; adopt globally capitated payment models for segments of the care delivery system

• Financial transactions ecosystem (claims processing): Streamline claims submission process through simpli�ed provider platforms; clarify Explanation of Bene�ts • Industry-agnostic corporate functions: Automate repetitive work in human resources and �nance; build functions of the future leveraging new technologies, such as analytics and cloud computing • Administrative clinical support functions: Remove manual work for nursing managers through automated tools for scheduling and sta�ng; integrate suite of tools and solutions to communicate 360-degree view of patients to case managers

~$175 18

~$35 4

~$105 11

~$265 28After accounting for overlap¹

Type of known intervention Example interventions

Savings, $ billion

Total administrative spending, %

¹ We estimated $50 billion of overlap across within and between interventions and seismic interventions. As a result, the total estimate is not fully additive. Source: Centers for Medicare & Medicaid Services; McKinsey analysis

7Executive summary

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

to identify and streamline to the highest value measures, could be a seismic way to unlock this opportunity by accel erating technology modernization in organizations (for ex ample, digitizing sources of data).

Apart from the outsize potential for savings, external forces are also creating pressure for organizations to act. Across the US economy, the COVID-19 pandemic and subsequent economic downturn have prompted organizations to rethink opera tions and invest in digital transformations. Indeed, research has shown that organiza tions that aggressively pursue industry leading productivity programs are twice as likely to be in the top quintile of their peers as measured by economic profit.12

To galvanize the seismic opportunity, we see actions for three sets of stakeholders:

— Government could set the framework in which other organizations operate. Federal and state bodies can set guardrails for payers, hospitals, and physician groups.

— Investors can prove ideas with pilots. They might create public– private partnerships to test interventions within a state and then scale up success stories nationally.

— Third parties, such as foundations and bipartisan groups, can conduct objec tive fact gathering and analyses. An arbiter of facts can gal vanize action.

There is an opportunity to capture over a quartertrillion dollars in savings in the next few years without compromising care delivery in the current US healthcare system. There is a clear roadmap ahead with proven solutions; the choice to act is upon everyone.

today. If fully realized, these savings would be more than three times the combined budgets of the National Institutes of Health ($39 billion), the Health Resources and Services Administration ($12 billion), the Substance Abuse and Mental Health Ser vices Administration ($6 billion), and the Centers for Disease Control and Prevention ($12 billon).7 Put another way, $265 billion is greater than Medicare Part A spending ($201 billion in 2019) and is equi valent to $1,300 for each American adult.8

Some organizations have made impres sive progress on administrative simpli fica tion by deploying within and between interventions. At these organizations we found a set of common denomina tors of success. These include the following:

— Prioritizing administrative simpli fication as a strategic initiative

— Committing to transformational change versus incre mental steps

— Engaging the broader partnership ecosystem on the right capabilities and investments

— Disproportionally allocating resourc es, such as capital and talent, to the under lying drivers of productivity

Seismic interventions are more difficult, largely because they are generally needed due to a lack of motivation to innovate at the organization level.9 For example, today, the Centers for Medicare & Medi caid Ser vices (CMS) requires reporting on more than 1,700 quality measures.10 Physicians spend the time equivalent to seeing nine patients reporting on such measures weekly.11 Laying out mechanisms that could promote standardization, such as convening a public–private partnership

$265 billion is greater than Medicare Part A spending ($201 billion in 2019) and is equivalent to $1,300 for each American adult.

8 Executive summary

McKinsey & Company McKinsey Center for US Health System Reform

1 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,” February 27, 2019, McKinsey.com.

2 We defined physician groups as hospitalaffiliated and independent physician groups with five or more doctors. There are 136,000 active physician groups in the United States ranging in size from solo practices to physician practices with 8,700 members. From “Top physician groups by size and Medicare charges,” Definitive Healthcare, Healthcare Insights, 2021, definitivehc.com.

3 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,” February 27, 2019, McKinsey.com.

4 We used financial and operational lenses in our analysis but acknowledge the broader benefits these interventions can have on outcomes such as access, quality, patient experience, physician experience, and equity, which we did not focus on or quantify in this report.

5 We do not propose a comprehensive list of all seismic interventions. We identified a few examples based on analogs from other industries where such interventions delivered a discontinuous but substantial improvement. These example interventions are meant to show the potential in US healthcare but are not a specific pointofview of what is best or should be pursued.

6 We estimated $50 billion of overlap across within and between interventions and seismic interventions. 7 Office of Budget, “Putting America’s health first: FY 2021 President’s budget to HHS,” Department of Health & Human Services, June 2021,

hhs.gov. 8 Monthly Federal Spending/Revenue/Deficit Charts, US Government Spending, 2021, usgovernmentspending.com. 9 Nikhil Sahni, Maxwell Wessel, and Clayton Christensen, “Unleashing breakthrough innovation in government,” Stanford Social Innovation

Review, Summer 2013, ssir.org. 10 Gail Wilensky, “The need to simplify measuring quality in health care,” Journal of the American Medical Association, June 19, 2018, Volume

319, Number 23, pp. 2369–70, jamanetwork.com. 11 Lawrence Casalino et al., “US physician practices spend more than $15.4 billion annually to report quality measures,” Health Affairs, March

2016, Volume 35, Number 3, healthaffairs.org. 12 Chris Bradley, Martin Hirt, and Sven Smit, Strategy Beyond the Hockey Stick: People, Probabilities, and Big Moves to Beat the Odds,

Hoboken, NJ: Wiley, 2018.

9Executive summary

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

Thank you to our reviewers

Ishani Ganguli, MD

Elizabeth Goodman, PhD

Michelle Hood

Rob Huckman, PhD

Chip Kahn

Joe Kimura, MD

Bob Kocher, MD

Annie Lamont

Sarah Calkins Holloway

Jess Kahn

Basel Kayyali

Paul Kline

David Knott, PhD

Adi Kumar

Alok Ladsariya

Meredith LaPointe

Rob Levin

Ed Levine, MD

Meera Mani, MD, PhD

External experts

Julia Adler-Milstein, PhD

Tanya Bentley, PhD

David Blumenthal, MD

Melinda Buntin, PhD

Michael Chernew, PhD

Gaurov Dayal, MD

Wendy Everett, ScD

William Frist, MD

McKinsey experts

Munmun Balshya

Gretchen Berlin, RN

Stephanie Carlton, RN

Sameer Chowdhary

Deniz Cultu

Philippa Duffy

Michael Elliot, PhD

Patrick Finn

Yuri Goryunov

Zach Greenberg

Pranay Gupta

The following people provided invaluable insights and advice as we were preparing this report. We thank them for their help.

Heather McComas

Sandhya Rao, MD

Jaewon Ryu, MD, JD

Mario Schlosser

April Todd

Mike Vennera

Wendy Warring, JD

Jan Mischke, PhD

Mhoire Murphy

Fatima Paruk, MD

Bridget Pollak, JD

Neil Rao, JD

Prashanth Reddy

Shubham Singhal

Hrishika Vuppala

Kyle Weber, JD

Rodney Zemmel, PhD

Note: By reviewing this paper, no individual is endorsing its conclusions. All errors remain our own.

10 Thank you to our reviewers

McKinsey & Company McKinsey Center for US Health System Reform

Administrative simplification has been a long sought-after goal in US healthcare, but there has been limited movement in the past few decades. It is clear that some administrative spending is necessary to keep the system running, but it is unclear how much is unnecessary and what could be done to simplify administrative processes. In this report, we lay out a first-of-its-kind analytical framework and break down the $950 billion in admini strative spending in US healthcare into functional focus areas such as the financial transactions ecosystem and customer and patient services. We do so by using average profit-and-loss statements for different stakeholder groups, including private payers, hospitals, and physician groups. For each area, we identify about 30 known interven- tions that could support simplification. The interventions fall into three types: those that can be achieved “within” each individual organization, those that can be carried out “between” a few organizations, or those that require “seismic” interventions, including public–private partnerships. We estimated that $265 billion, or 28 percent, of administrative spending could be reduced without affecting quality or access. Our aim is to arm public and private sector leaders with these fact-based insights to guide informed decision making.

CHAPTER 1

Introduction

11Chapter 1. Introduction

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

For clarity, medical spending—or costs incurred for direct delivery of care—were not within the scope of this report. This meant that our analysis excluded areas such as time spent by physicians and clinical nurses on direct patient care, prescription drugs, and clinical IT. Further, we used financial and operational lenses in our analysis, but we acknowledge the broader benefits these interventions can have for outcomes such as access, quali- ty, patient experience, physician experi- ence, and equity, which we did not focus on or quantify in our work.5-7

We estimated that approximately $950 billion (or 25 percent) of total healthcare spending in 2019 was administrative ( Exhibit 1.1).8,9 This spending was spread across multiple stakeholder groups: pri- vate payers (19 percent), hospitals (26 percent), physician groups (22 percent), public payers (9 percent), and other sites of care (24 percent).10-12

For years, researchers and policymakers have asserted that a large portion of this spending could be removed without affecting quality or access. Using various methodologies (for example, comparisons with other countries’ healthcare systems or with other industries), previous analyses estimated that approximately 40 percent of this spending could be eliminated.13-18

However, these analyses miss important considerations about the US healthcare system. For example, unlike other coun- tries, the United States has a healthcare system that is multi-provider (more than 6,000 hospitals and 11,000 non- employed physician groups with more than five physicians) and multi-payer (more than 900 private payers) to encourage com- petition.19-21 Doing so also increases the complexity of the system due to its greater number of communication and transac- tion nodes. Underlying this structure is a predominantly fee-for-service payment model, although there has been uptake of value- based models.22 No matter which payment model is used, stakeholder

In 2019, healthcare spending in the United States reached $3.8 trillion and comprised 18 percent of US GDP.1,2 Over the past 15 years, while US healthcare delivery comprised 9 percent of US GDP growth, the sector represented 29 percent of workforce growth.3 This imbalance suggests major productivity issues in the healthcare system. We defined productivi- ty in previous work in this series as output per given unit of input. In this definition, the outputs in healthcare delivery are largely the serv ices delivered and out- comes achieved; the inputs include the workforce, invested capital, and new technologies. An advantage of looking at healthcare delivery this way is that it puts the focus not on spending minimization, but on long-term growth and overall spending trajectory.4

Administrative spending is a subset of total US healthcare spending. We define it as all activities in support of the delivery of care, including services like payment transactions, back-office corporate and operational functions, customer and patient services, and administrative clinical support. Specific functions using traditional terminology include claims proces sing, billing, accounting, prior au- thorization, and payment integrity within payments; industry-agnostic corporate functions such as human resources, sales and marketing, finance, and procurement; industry- specific operational functions such as medical records, quality reporting, clinician credentialing, underwriting, transparency tools, and broker manage- ment; call centers and medical reception- ists within customer and patient services; and nursing administration management tasks (for example, scheduling) and care management teams within administrative clinical support functions. Given the size and neces sity of administrative spending in the US healthcare system, we set out to identify opportunities to improve the productivity of administrative functions and to “bend the cost curve” for this bucket of spending.

12 Chapter 1. Introduction

McKinsey & Company McKinsey Center for US Health System Reform

regulations issued by the Department of Health and Human Services are second only to those imposed by the Department of the Treasury in terms of time spent by the private sector.24 Major moments of change in the healthcare system spurred by new policies can substantially affect administrative spending in both positive and negative ways (Exhibit 1.2).25-28 For example, the creation of the Affordable Care Act required states to build their own exchanges, including online marketplaces, and payers to invest in reporting on em- ployer size and the extent to which their employees were covered.29 When estab- lishing these regulations, policymakers may generally focus on the impact of a given change on the population, rather than on its effect on the bottom line of payers, hospitals, and physician groups. For example, MA created more competi- tion in the Medicare market, which should be better for members. But the healthcare organization may bear the operational expense of implementing the new policies.

groups continue to put checks and bal- ances on each other to ensure the other party is acting appropriately. The implica- tion is that certain portions of administra- tive spending are necessary (for example, for service delivery and technology) and others unnecessary (such as excess spending on antiquated systems).

Furthermore, the US healthcare system is highly regulated, requiring administrative spending by organizations to comply with the rules. These regulations range from compliance requirements such as the Health Insurance Portability and Account- ability Act of 1996 (HIPAA) to markets such as Medicare Advantage (MA), a private- sector alternative to traditional Medicare. For example, research has found that physicians spent 2.6 hours per week on quality measure reporting, much of which is not synchronized across payers.23 That is the equivalent of caring for nine patients. Data released in 2016 by the Information Collection Budget found that

Exhibit 1.1

Breakdown of administrative spending by stakeholder group

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 1.1 and Exhibit A of 9.2 (total 22 exhibits)

US healthcare spending by type of spending

% of total, 2019 (100% = $3.8 trillion)

Breakdown by stakeholder group

$ billion, 2019 (percent of total administrative spending)

Total administrative spending 950 (100%)

180 (19%)

250 (26%)

205 (22%)

80 (9%)

235 (24%)

Private payers

Hospitals

Physician groups¹

Public payers²

Other sites of care³

³ Includes, for example, dental services, home healthcare, and nursing care facilities. Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Medical spending: Costs incurred for direct delivery of care, including time spent by physicians and clinical nurses on direct patient care, prescription drugs, and clinical IT

2019

13Chapter 1. Introduction

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

Exhibit 1.2

Major policy changes and evolution in administrative spending over the last 40 years

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 1.2 of 9.2 (total 22 exhibits)

Nixon executive order freezing

prices and wages in 1971

Healthcare industry

voluntary cost containment e�ort in 1977

Introduction of Medicare DRG¹ payment system

in 1983

Rise of managed care plans and

Balanced Budget Act of 1996

Children’s Health Insurance Program (CHIP)

of 1997

Great Recession

(2007–09) and A�ordable Care

Act (ACA)

Downward pressure

As all prices were frozen, payers likely

looked to reduce administrative

spending in order to stabilize pro�ts

Downward pressure

Healthcare groups

(eg, AMA, AHA, FAH, BCBSA)¹

agreed to voluntary cost containment

Downward pressure

Flat reimburse- ment based on Medicare DRG code increased administrative

simplicity in Medicare

No pressure Provisions aimed to reduce federal expenditures by

enrolling Medicaid bene�ciaries

in managed care and creating

Medicare Advantage

Upward pressure

Administrative complexity

increased as new program

was implemented

Upward pressure

Substantial administrative confusion as healthcare

organizations learned to adapt to new provisions

of the ACA

Policy change

Methodology Private payers, Commercial: Divided “Net Cost of Health Insurance Expenditures: Private Health Insurance” by “Total National Health Expenditures: Private Health insurance” Federal payers: Divided “Federal Administration Expenditures” and “Net Cost of Health Insurance Expenditures: Medicare, Federal Medicaid, Federal CHIP” by “Total National Health Expenditures: Federal Expenditures, Medicare, Federal Medicaid, Federal CHIP, DoD, and DVA” State and local payers: Divided “State and Local Administration Expenditures” and “Net Cost of Health Insurance Expenditures: State and Local Medicaid, State and Local CHIP” by “Total National Health Expenditures: State Expenditures, State and Local Medicaid, State and Local CHIP, Other State and Local Programs” Note: Similar �gures not available for providers.

Administrative spending as a percent of payer healthcare spending

¹ AHA, American Hospital Association; AMA, American Medical Association; BCBSA, Blue Cross Blue Shield Association; DRG, diagnosis-related group; FAH, Federation of American Hospitals. Source: Centers for Medicare & Medicaid Services; Marsha Gold et al., “E�ects of selected cost-containment e�orts: 1971–1993,” Health Care Financ Rev, 1993, 14(3): 183–225; Stuart Guterman and Allen Dobson, “Impact of the Medicare prospective payment system for hospitals,” Health Care Financ Rev, 1986, 7(3): 97–114; Andy Schneider, “Overview of Medicaid Provisions in the Balanced Budget Act of 1997, P.L. 105–33,” CBPP, 1997; McKinsey analysis

Commercial private payer administrative spending as a percent of commercial private payer healthcare spending

State and local government payer administrative spending as a percent of state and local government payer healthcare spending

Federal payer administrative spending as a percent of federal payer healthcare spending

20181970 1975 1980 1985 1990 1995 2000 2005 2010 2015 0

Expected e�ect on administrative spending as percent of payer healthcare spend

14 Chapter 1. Introduction

McKinsey & Company McKinsey Center for US Health System Reform

— “Within”: Interventions that can be controlled and implemented by indi - vidual organizations

— “Between”: Interventions that require agreement and collaboration between organizations, but not broader, industry- wide change

— “Seismic”: Interventions that require broad, structural agreement and changes across the US healthcare system

We used this construct because it allows for the development of an actionable roadmap and more accurately identifies the burden of responsibility for each stake- holder group in the healthcare system. For each functional focus area, we discuss the within and between interventions in its respective chapter. Seismic interventions are discussed separately (see chapter 8). Our list of seismic interventions is not com- prehensive. We identified a few examples based on analogs from other industries where such interventions delivered a dis- continuous but substantial improvement. These example interventions are meant to show the potential in US healthcare but are not a specific point-of-view on what is best or should be pursued.

We found that approximately $175 billion (or 18 percent of the $950 billion total administrative spending) could be saved through within inter ventions (for example, automating rules-based tasks in back-office functions) and another $35 billion (or 4 percent) through between interventions (for example, setting up a joint claims- status- tracking workflow between payers and providers). This overall total of $210 billion (or 22 percent) represented net savings after accounting for ongoing operating expenses; based on our experience, they require a one-time investment of 0.7 to 1.0 times the annual run-rate savings. About $105 billion (or 11 percent) could be saved through seismic interventions that require foundational shifts in how the US health- care system operates (for example, adopt- ing a centralized, automated claims clear-

In the case of MA, the change led many payers to hire a larger administrative workforce to code and audit patient risk scores.30

Given this background, what portion of administrative spending is actually neces- sary? What amount could be reduced through simplification, and how might that happen without compromising access and quality? To help move the national discus- sion forward, we set forth to build a granu- lar, bottom-up breakdown of administra- tive spending (see more detail in chapter 2) for three key private stakeholder groups— private payers, hospitals, and physician groups. To the best of our knowledge, this is the first-of-its-kind analysis of US healthcare administrative spending. These three stakeholder groups accounted for approximately $635 billion of the $950 billion of US healthcare administrative spending in 2019. By compiling these figures from average profit-and-loss statements, we were able to systematic- ally categorize interventions and estimate capturable savings in five functional focus areas: the financial transactions ecosys- tem, industry- agnostic corporate functions, industry- specific operational functions, customer and patient services, and admini- strative clinical support functions. These comprised 94 percent of total US admini- strative spending.

Chapters 3 through 7 of this report explore each of the five functional focus areas in detail. We then identify known interven- tions that could simplify administrative processes and reduce spending without sacrificing qual ity or access. To do so, we used three criteria: proven but not fully scaled changes across US healthcare, changes related to technology that will fully come to market within the next three years, and transformational changes that are analogous to those implemented in other US industries. We grouped these criteria into one of three categories based on where stakeholders must reach “agree- ment” to effectuate change:

15Chapter 1. Introduction

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

return on investment for a given organiza- tion, such as payment integrity for payers or revenue cycle management for providers.

In the rest of this report, we will review each functional focus area and break down the spending by each of the three stakeholder groups (private payers, hospitals, and phy- sician groups), explain the major pain points that result in unnecessary administrative spending, and describe interventions that could be used to capture savings. Finally, we will discuss how leaders across sectors might catalyze this change (see chapter 9) and share specific actions for each stake- holder group to consider.

inghouse or standardizing medical policies). Many of these seismic interventions could replace certain within and between interventions, resulting in total identified savings of $265 billion (or 28 percent).

This amount is less than the 40 percent, top-down estimate of unnecessary admini- strative spending cited from pre vious litera- ture above. By contrast, our estimate reflects a unique methodology offering a bottom- up roadmap of what savings could be practi- cally and realistically captured in the next three years without affecting quality or ac- cess. Furthermore, the focus on practicality helped to prioritize interventions with a high

1 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. 2 United States GDP (current US dollars), The World Bank Group, 2021, data.worldbank.org. 3 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,”

February 27, 2019, McKinsey.com. 4 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,”

February 27, 2019, McKinsey.com. 5 Michael Kyle and Austin Frakt, “Patient administrative burden in the US health care system,” Health Services Research, September 8, 2021,

pp. 1–11, onlinelibrary.wiley.com. 6 David Cutler and Dan Ly, “The (paper) work of medicine: understanding international medical costs,” Journal of Economic Perspectives,

Spring 2011, Volume 25, Number 2, pp. 3–25, aeaweb.org. 7 Lawrence Casalino et al., “US physician practices spend more than $15.4 billion annually to report quality measures,” Health Affairs, March

2016, Volume 35, Number 3, pp. 401–6, healthaffairs.org. 8 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. 9 James Kahn et al., “The cost of health insurance administration in California: Estimates for insurers, physicians, and hospitals,” Health Affairs,

November 2005, Volume 24, Number 6, pp, 1629–39, healthaffairs.org. 10 Includes categories such as dental services, home health, nursing care facilities, other professional services, and other health, personal,

and residential care. 11 “National health expenditure accounts: Methodology paper, 2019,” Centers for Medicare & Medicaid Services, 2019, cms.gov. 12 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. 13 Natasha Parekh, Teresa Rogstad, and William Shrank, “Waste in the US health care system: Estimated costs and potential for savings,”

Journal of the American Medical Association, October 2019, Volume 322, Number 15, pp. 1501–9, jamanetwork.com. 14 Tanya Bentley et al., “Waste in the U.S. health care system: A conceptual framework,” The Milbank Quarterly, December 2008, Volume 86,

Number 4, pp. 629–59, doi:10.1111/j.1468-0009.2008.00537.x. 15 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,”

February 27, 2019, McKinsey.com. 16 Institute of Medicine, The Healthcare Imperative: Lowering Costs and Improving Outcomes: Workshop Series Summary, Washington, DC:

The National Academies Press, 2010. 17 Donald Berwick and Andrew Hackbarth, “Eliminating waste in US health care,” Journal of the American Medical Association, April 2012,

Volume 307, Number 14, pp. 1513–6, jamanetwork.com. 18 Terry Campbell, David Himmelstein, and Steffie Woolhandler, “Health care administrative costs in the United States and Canada, 2017,”

Annals of Internal Medicine, January 2020, Volume 172, Number 2, pp. 134–42, acpjournals.org. 19 American Hospital Association, “Fast facts on U.S. hospitals, 2021,” partial data from 2019 survey, updated January 2021, aha.org. 20 We defined physician groups as hospital-affiliated and independent physician groups with five or more doctors. There are 136,000 active

physician groups in the United States ranging in size from solo practices to physician practices with 8,700 members. From “Top physician groups by size and Medicare charges,” Definitive Healthcare, Healthcare Insights, 2021, definitivehc.com.

21 Facts + statistics: Industry overview, Insurance Information Institute, August 2021, iii.org. 22 A fee-for-service model is one in which hospitals and physician groups are paid based on the amount of healthcare services they deliver.

A value-based care model is one where they are paid based on patient outcomes. 23 Lawrence Casalino et al., “US physician practices spend more than $15.4 billion annually to report quality measures,” Health Affairs, March

2016, Volume 35, Number 3, pp. 401–6, healthaffairs.org. 24 Nikhil Sahni, Pooja Kumar, Edward Levine, and Shubham Singhal, “The productivity imperative for healthcare delivery in the United States,”

February 27, 2019, McKinsey.com. 25 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. 26 Marsha Gold et al., “Effects of selected cost-containment efforts: 1971–1993,” Health Care Financing Review, 1993, Volume 14, Number 3,

pp. 183–225. 27 Stuart Guterman and Allen Dobson, “Impact of the Medicare prospective payment system for hospitals,” Health Care Financing Review,

1986, Volume 7, Number 3, pp. 97–114. 28 Andy Schneider, “Overview of Medicaid provisions in the Balanced Budget Act of 1997, P.L. 105–33,” Center on Budget and Policy Priorities,

September 1997, cbpp.org. 29 “Read the Affordable Care Act,” Centers for Medicare & Medicaid Services, 2021, healthcare.gov. 30 Office of the Actuary, “Risk score credibility guidelines,” Centers for Medicare & Medicaid Services, April 2015, cms.gov.

16 Chapter 1. Introduction

McKinsey & Company McKinsey Center for US Health System Reform

Historical analyses of the savings potential from administrative simpli fication typically use topdown comparisons of the United States and other countries such as Canada. Our goal was to connect the macro, an estimated $950 billion in administrative spending in 2019, to the micro, the average profitandloss (P&L) for private payers, hospitals, and phy sician groups. We did this by creating five functional focus areas of spending that cut across all stakeholder groups: the financial trans actions ecosystem ($200 billion), industryagnostic corporate functions ($375 billion), industryspecific operational functions ($135 billion), customer and patient services ($80 billion), and administrative clinical support functions ($105 billion).* By doing this, we can isolate what spending is necessary versus what could be removed through known interventions and scaled in the next three years while maintaining or improving access and quality.

CHAPTER 2

Analytical framework

* The remaining $55 billion was categorized as “other” and includes spending such as public relations and charitable giving.

17Chapter 2. Analytical framework

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

to 2019 and found that the categories of healthcare spending in the United States have not shifted substantially.4-6 Second, increased administrative spending re flecting innovation and responses to regulations likely counteracted potential savings from any new interventions.

We further confirmed the overall esti mate by focusing on three individual stakeholder groups—private payers, hospitals, and physician groups. Using these results, we estimated administra tive spending for the two remaining stakeholder groups of public payers and other sites of care (Exhibit 2.1).

For hospitals and physician groups, we reviewed blinded data for more than 50 individual organizations and found ad ministrative spending equaled 20 to 25 percent of revenue for hospitals and 25 to 30 percent for physician groups. Us ing 2019 National Health Expenditure Accounts (NHEA) data, we found that

Our goal was to answer the core ques- tions of what portion of administrative spending is actually necessary and what amount could be reduced without compromising access and quality. To do so, we needed to develop estimates of overall administrative spending and then break that spending down into functional focus areas.

Estimation of overall administrative spending Over the past two decades, research has consistently found that about 25 percent, ranging from 15 to 35 percent, of US healthcare spending was for administra tive functions.1-3 We sought to refresh this research and confirm the validity of earlier estimates. We assumed that the overall percentage of 25 percent has remained constant for two reasons. First, we ran a series of regressions of “excess spending adjusted for wealth” from 2010

Exhibit 2.1

Key stakeholder groups in analytical focus for this report

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.1 of 9.2 (total 22 exhibits)

¹ Includes administrative spending for fee-for-service Medicare and Medicaid, Children’s Health Insurance Program (CHIP), Department of Defense, Department of Veterans A�airs, and other federal programs.

² Hospital-a�liated and independent physician groups; employed physician groups included in hospitals. ³ Includes, for example, dental services, home healthcare, and nursing care facilities. ⁴ Includes miscellaneous activities such as charitable giving, community health education, and public relations. Source: McKinsey analysis

Analytical focus (67% of total administrative spending)

Stakeholder groups

Financial transactions ecosystem

Customer and patient

services

Industry- agnostic

corporate functions

Industry- speci�c

operational functions

Administrative clinical support

functions Other⁴

Providers

Payers Public¹

Private

Hospitals

Physician groups²

Other sites of care³

Extrapolated full savings potential (100% of total administrative spending)

Functional focus areas

18 Chapter 2. Analytical framework

McKinsey & Company McKinsey Center for US Health System Reform

federal administration spending ($49 billion in 2019) and approximately 40 percent of the net cost of health insur ance expenditures on Medicare, Medi caid, and CHIP, which represents the feeforservice portion of Medicare and Medicaid enrollment ($32 billion in 2019).9 This calculation produced an estimate of approximately $80 billion in total administrative spending.

In addition to these four stakeholder groups, we created an “other sites of care” group representing the remainder of administrative spending. Examples in this group included dental services, home healthcare, and nursing care fa cilities. In total, administrative spending for this group was estimated to be $235 billion. This value was about 30 percent of total revenue in these other sites of care, an estimate that corresponded with our review of blinded data from more than 40 individual organizations across these categories.

our estimate indicated total administra tive spending of $250 billion and $205 billion, respectively.7

For private payers, we reviewed blinded data for more than 30 individual organiza tions and found administrative spending was equal to roughly 15 percent of reve nue. Using 2019 NHEA data, this figure resulted in total administrative spending of $180 billion. We compared this total to the sum of the net cost of health insur ance expenditures on private health in surance ($131 billion in 2019) and approxi mately 60 percent of the expenditures on Medicare, Medicaid, and the Children's Health Insurance Program (CHIP), a per centage which represents the managed care portion of Medicare and Medicaid enrollment ($48 billion in 2019).8 We found that the NHEA and net cost totals were about the same.

For public payers, NHEA reports the following data. We summed state and

Exhibit 2.2

Deriving administrative spending from who is paying and where it is being spent

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.2 of 9.2 (total 22 exhibits)

¹ Based on 2019 National Health Expenditure Accounts (NHEA) “Source of Funds” table. ² Based on NHEA “Type of Expenditure” table. ³ Sum of 2019 NHEA Medicare ($799B), Medicaid ($614B) and other health insurance programs (for example, CHIP) spending ($145B). ⁴ Hospital-a�liated and independent physician groups; employed physician groups included in hospitals. ⁵ Sum of 2019 NHEA dental services ($143B); home healthcare ($114B); nursing facilities ($174B); other health, residential, and personal care ($194B); and other professional services ($111B).

⁶ Includes government public health activities ($98B), investment ($202B), out-of-pocket spending ($407B), and third-party payers (for example, workforce comp; $336B). ⁷ Includes government administration ($49B), government public health activities ($98B), investment ($202B), net cost of health insurance ($240B), and retail outlet sales of medical products ($509B).

Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Total healthcare spending, $ billion

Who is paying?¹Stakeholder group

Where it is being spent²

Administrative spending

Administrative spending as a percent of total spending

Public payers

Private payers

Hospitals

1,558³

1,195

—

1,192

180

250

15%

21%

Physician groups⁴

Other sites of care⁵

Not included in analysis

Total

—

1,043⁶

3,795

772

734

1,097⁷

3,795

205

235

N/A

950

27%

32%

N/A

25%

19Chapter 2. Analytical framework

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

such as underwriting, enrollment, quality reporting, and accreditation

— Administrative clinical support functions: Activities that have a clinical component (for example, nursing administration, case man agement), which can be customer facing and require some clinical expertise, but are not related to the handson care of patients

Breaking down a P&L statement into these functional focus areas may look familiar to payers that already examine their finan ces in terms of administrative versus medical spending and also already use some of this termin ology. However, these functional focus areas will likely be unfamiliar to hospitals and physician groups as they may not differ entiate explicitly between administrative versus medical spending. They also typically assess their P&L statements in terms of “Net Patient Service revenue,” or perpatient metrics of revenue (in stead of total spending). For our pur poses, we needed to use common terminology and definitions across all stakeholder groups so we could identify crossstakeholder interventions.

Private payers Three functional focus areas accounted for about 76 percent of the $180 billion that private payers spent on administra tive functions in 2019: industryspecific oper ational functions ($50 billion), industry

Our analyses showed the administra tive spending baseline across the five stakeholder groups to be $950 billion (Exhibit 2.2).

Functional breakdown of administrative spending As a next step, we developed a set of five functional focus areas that cut across the stakeholder groups; these helped us align known interventions that could deliver savings (see chapters 3–7). Given data limitations, we focused our analysis on three stakeholder groups—private payers (including pharmacy benefit managers, or PBMs), hospitals, and physician groups— which represented 70 percent of all ad ministrative spending.10

For these stakeholder groups, we used profitandloss (P&L) statements based on blinded data from private payers, hospitals, and physician groups. These P&Ls use standard accounting defini tions. To make the standard accounting groupings actionable, we converted them into functional focus areas based on actual operational activities (see below for details by stakeholder group):

— Financial transactions ecosystem: The movement of all payments, claims, and billing throughout the healthcare ecosystem among payers, hospitals, phy sician groups, and customers

— Industry-agnostic corporate functions: Backoffice, nonclinical functions that are mostly industry agnostic, such as finance or human resources

— Industry-specific operational func- tions: Backoffice, nonclinical func tions that are mostly industryspecific,

Three stakeholder groups—private payers, hospitals, and physician groups—represented 70 percent of all administrative spending.

20 Chapter 2. Analytical framework

McKinsey & Company McKinsey Center for US Health System Reform

tions, such as general administration, sales and marketing support, and IT. The financial transactions ecosystem, which forms the backbone of payments for private payers, includes claims and utili zation management. Beyond these three functional focus areas, customer and patient services ($20 billion) and adminis trative clinical support functions ($20 bil lion) each accounted for approximately 10 percent of total private payer administra

agnostic corporate functions ($45 billion), and the financial transactions ecosystem ($40 billion) (Exhibit 2.3).

Industryspecific operational functions for private payers include specialized broker based sales; underwriting, actu arial and pricing; clinician credentialing; and membership and billing. Industry agnostic corporate functions include many standard operations for organiza

Exhibit 2.3

Private payers: Representative operating activity pro�t-and-loss statements

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.3 of 9.2 (total 22 exhibits)

Note: Figures may not sum to 100%, because of rounding. ¹ IT spending for utilization management/quality review and case management are represented in those specific line items, and not in the IT line item. ² Contains mix of clinical support operations (for example, behavioral health, wellness) and other vendor spending (for example, subrogation). Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Functional focus area, $ billion

Financial transactions ecosystem

Industry- agnostic corporate functions

Total administrative spending

Total administrative spending, $ billion, 2019

Industry- speci�c operational functions

Administrative clinical support functions Other

Customer and patient services

General administration

0.0 16.3 0.0 0.0 0.0 0.0 16.3

29.0 0.0 0.0 0.0 0.0 0.0 29.0

0.0 10.9 16.3 0.0 0.0 0.0 27.1

0.0 0.0 21.7 0.0 0.0 0.0 21.7

0.0 0.0 3.6 0.0 0.0 0.0 3.6

0.0 0.0 14.5 0.0 0.0 0.0 14.5

0.0 0.0 0.0 12.7 0.0 0.0 12.7

4.3 13.0 0.0 4.3 0.0 0.0 21.7

7.2 0.0 0.0 0.0 0.0 0.0 7.2

0.0 0.0 0.0 0.0 5.4 0.0 5.4

0.0 0.0 0.0 0.0 7.2 7.2 14.5

40 (23%)

45 (25%)

50 (28%)

20 (9%)

20 (10%)

5 (4%)

180 (100%)

Claims

Sales and marketing

Underwriting/ actuarial/pricing

Membership and billing

Clinician services and credentialing

Customer service

IT¹

Utilization manage- ment/quality review

Case management

Medical director

Other healthcare services²

Total (percent of total)

21Chapter 2. Analytical framework

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

manage, and collect patient service rev enue. The latter includes administrative spending on two types of nurses: nursing administration, or the managerial layer of nurses who are fully administratively focused and handle nonpatient facing tasks such as staffing and budgeting, and case and disease management, which includes nurses who spend 30 to 40 percent of their time on administra tive tasks such as communication and coordination of patient status across both registration and discharge.

Hospitals spent roughly $20 billion, or 9 percent of total administrative spend ing, on industryspecific operational functions such as medical records management and quality reporting. Finally, hospitals spent about $10 billion, or 5 percent of total administrative spending, on cus tomer and patient services such as call centers. A long tail of smaller expenses were represented by the “other” catego ry, including spending on charity, religious, and spiritual activities (for ex ample, chaplains), and public relations.

Physician groups Similar to hospitals, about 44 percent of the $205 billion that nonemployed phy sician groups spent on administra tive activities in 2019 was for industry agnostic corporate functions ($90 bil lion). These functions include general administration, administrative supplies and services, and nonclinical IT (Ex hibit 2.5). The next largest functional focus area was the financial transactions ecosystem ($50 billion, or about 24 percent of total administrative spend ing), which includes claims and billing, utilization management, and nonclinical IT, such as provider portals. The third

tive spending. The former comprises call centers and the associated IT to support them; the latter includes resources with in case management and those allocat ed to medical directors that enable pri vate payers to manage care for patients with complex needs.

Hospitals Of the $250 billion that hospitals (de fined for our purposes as hospital sys tems and employed physician groups) spent on administrative functions in 2019, the largest functional focus area (approximately $115 billion, or 46 percent of total administrative spending) was industryagnostic corporate functions (Exhibit 2.4). This spending is driven by the large back office operations that hospitals rely on to facilitate care pro vision, as well as by the administrative work that hospitals provide for em ployed physician groups.11 Specific functions include general administra tion, accounting, and nonclinical IT.

Hospitals’ sales and marketing function also resides within industryagnostic corporate functions. In our experience, while spending in this area is growing, the actual level is too low to be a major factor in overall administrative spending. For example, the most sophisticated hospitals adopting digital tools to attract patients are spending no more than 1 to 2 percent of their total administrative budgets on sales and marketing.

The next two largest functional focus areas were the financial transactions ecosystem ($40 billion) and adminis trative clinical support functions ($40 billion). The former includes the “cost to collect” or revenue cycle management function that allows hospitals to identify,

Industry-agnostic corporate functions represented about 45 percent of administrative spending for both hospitals and physician groups.

22 Chapter 2. Analytical framework

McKinsey & Company McKinsey Center for US Health System Reform

patients’ questions, typically about payments and appointment sched uling. The two remaining, smaller functional focus areas were industry

largest functional focus area was cus tomer and patient services ($25 billion, or 13 percent), which includes spending on medical receptionists that answer

Exhibit 2.4

Hospitals: Representative operating activity pro t-and-loss statements

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.4 of 9.2 (total 22 exhibits)

Note: Figures may not sum to 100%, because of rounding. ¹ Includes sales spending (no more than 1–2% of total hospital administrative spending). ² Includes admitting, medical sta�’s administrative activities, and outpatient registration. ³ Comprised of miscellaneous charity and wellness spending (for example, chaplain services, governing board, employee health). Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Functional focus area, $ billion

Financial transactions ecosystem

Industry- agnostic corporate functions

Total administrative spending

Total administrative spending, $ billion, 2019

Industry- speci c operational functions

Administrative clinical support functions Other

Customer and patient services

Hospital admininistration¹

0.0 75.1 0.0 0.0 0.0 0.0 75.1

0.0 14.3 0.0 0.0 0.0 0.0 14.3

0.0 13.1 0.0 0.0 0.0 0.0 13.1

0.0 1.2 0.0 0.0 0.0 0.0 1.2

19.1 0.0 0.0 0.0 0.0 0.0 19.1

11.9 0.0 0.0 0.0 0.0 0.0 11.9

0.0 0.0 0.0 0.0 0.0 9.5 9.5

0.0 0.0 0.0 11.9 0.0 0.0 11.9

0.0 9.5 0.0 0.0 0.0 0.0 9.5

0.0 0.0 0.0 0.0 15.5 0.0 15.5

0.0 0.0 0.0 0.0 21.5 0.0 21.5

0.0 0.0 0.0 0.0 3.6 0.0 3.6

10.7 0.0 0.0 0.0 0.0 0.0 10.7

0.0 0.0 22.6 0.0 0.0 0.0 22.6

0.0 0.0 0.0 0.0 0.0 8.3 8.3

40 (17%)

115 (46%)

20 (9%)

10 (5%)

40 (16%)

20 (7%)

250 (100%)

Other admininistration

General accounting

Other �scal

Patient accounting

Claims submission, credit, and collection

Public relations

Medical receptionists

Personnel

Nursing administration

Admitting and registration²

In-service education

Utilization management

Medical records

Other healthcare services³

Total (percent of total)

23Chapter 2. Analytical framework

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

Summary of analytical framework For the US healthcare system, these five functional focus areas represented about 94 percent of total administrative spend ing, though this varies by stakeholder group (Exhibit 2.6).

For private payers, these five functional focus areas comprised 96 percent of the $180 billion in total administrative spending. The largest (28 percent) was

specific corporate functions ($20 billion, or 9 percent) and administrative clinical support functions ($10 billion, or 5 per cent). The former includes functions such as medical records management and quality reporting; the latter includes resources to support the administrative infrastructure of coordination and com munication of patient care through case management and medical secretaries and transcribers.

Exhibit 2.5

Physician groups: Representative operating activity pro�t-and-loss statements

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.5 of 9.2 (total 22 exhibits)

Note: Figures may not sum to 100%, because of rounding. ¹ Hospital-affiliated and independent physician groups; employed physician groups included in hospitals. Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Functional focus area, $ billion

Financial transactions ecosystem

Industry- agnostic corporate functions

Total administrative spending¹

Total administrative spending, $ billion, 2019

Industry- speci�c operational functions

Administrative clinical support functions Other

Customer and patient services

General administration

0.0 25.5 0.0 0.0 0.0 0.0 25.5

0.0 3.1 0.0 0.0 0.0 0.0 3.1

0.0 11.6 0.0 0.0 0.0 0.0 11.6

0.0 21.6 0.0 0.0 0.0 0.0 21.6

30.1 0.0 0.0 0.0 0.0 0.0 30.1

0.0 0.0 0.0 26.3 0.0 0.0 26.3

4.6 0.0 0.0 0.0 4.6 0.0 9.3

6.3 19.0 6.3 0.0 0.0 0.0 31.7

8.9 0.0 0.0 0.0 0.0 0.0 8.9

0.0 0.0 0.0 0.0 5.4 0.0 5.4

0.0 0.0 11.6 0.0 0.0 0.0 11.6

0.0 9.4 0.0 0.0 0.0 10.2 28.6

50 (24%)

90 (44%)

20 (9%)

25 (13%)

10 (5%)

10 (6%)

205 (100%)

Other admini- strative support

Contracted services

Administrative sup- plies and services

Claims/ billing o�ce

Medical receptionists

Managed care administration

Utilization management

Medical secretaries/ transcribers

Medical records

Other healthcare services

Total¹ (percent of total)

24 Chapter 2. Analytical framework

McKinsey & Company McKinsey Center for US Health System Reform

in industry agnostic corporate func tions, followed by the financial trans actions ecosystem (17 percent), admin istrative clinical support functions (16 percent), industryspecific operational functions (9 percent), and customer and patient services (5 percent). This high proportion of hospital spending on industry agnostic corporate functions was also seen in phy sician groups. A fragmented provider market and the resulting lack of scale could account for this finding.12

For physician groups, these functional focus areas comprised about 94 per cent of the $205 billion in total ad ministrative spending. The largest (44 percent) was in industry agnostic corporate functions, followed by the fin ancial transactions ecosystem (24 percent), customer and patient services (13 percent), industry specific operational functions (9 percent), and administrative clinical support func tions (5 percent). Physician groups spent more than hos pitals on the

industry specific operational functions, followed by industryagnostic corporate functions (25 percent), the financial trans actions ecosystem (23 percent), ad ministrative clinical support functions (10 percent), and customer and patient services (9 percent). Unlike hospitals and physician groups, private payers spent almost 50 percent less on industry agnostic corporate functions as a per centage of total administrative spend ing, owing to higher rates of automation and digitalization. Private payers also had a significantly larger portion of admini strative spending allocated to industry specific operational functions such as underwriting or brokerspecific sales, as these are strategic capabilities that allow them to differentiate them selves in the market (for example, through highly tailored pricing or data driven negotiations with brokers).

For hospitals, these functional focus areas comprised about 93 percent of the $250 billion in total administrative spending. The largest (46 percent) was

Exhibit 2.6

Breakdown of administrative spending by functional focus area

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 2.6 and Exhibit B of 9.2 (total 22 exhibits)

$ billion, 2019

Financial transactions ecosystem

Industry-agnostic corporate functions

Industry-speci�c operational functions

Customer and patient services

Administrative clinical support functions Other

$950 billion $180 billion $250 billion $205 billion

Total¹ Private payers Hospitals Physician groups²

21%

14%

11%

39%

23%

28%

10%

25%

17%

16%

46%

24%

44%

13%

25Chapter 2. Analytical framework

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

investments in auto mation technology to reduce manual work, which can include eligibility checks and pre authorizations or providing selfservice scheduling tools for patients.

financial trans actions ecosystem and customer and patient services as a percentage of total administrative spending. This could reflect the inabili ty of many physician groups to afford

1 James Kahn et al., “The cost of health insurance administration in California: Estimates for insurers, physicians, and hospitals,” Health Affairs, 2005, Volume 24, Number 6, pp, 1629–39, healthaffairs.org.

2 Erin Michael, “A third of US health care spending stems from administrative costs,” Healio, January 6, 2020, healio.com. 3 Joshua Gottlieb and Mark Shepard, “How large a burden are administrative costs in health care,” EconoFact, September 6, 2018, econofact.org. 4 Carlos Angrisano, Diana Farrell, Bob Kocher, Martha Laboissiere, and Sara Parker, “Accounting for the cost of health care in the United

States,” January 1, 2007, McKinsey.com. 5 Diana Farrell, Eric Jensen, Bob Kocher, Nick Lovegrove, Fareed Melhem, Lenny Mendonca, and Beth Parish, “Accounting for the cost of US

health care: A new look at why Americans spend more,” December 1, 2008, McKinsey.com. 6 Jesse Bradford, David Knott, Edward Levine, and Rodney Zemmel, “Accounting for the cost of US healthcare: Prereform trends and the

impact of the recession,” December 2011, McKinsey.com. 7 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. 8 We assume the net cost of health insurance is fully administrative spending, which is defined by the NHEA as “the difference between

calendaryear (CY)incurred premiums earned and benefits paid for private health insurance. This includes administrative spending, and, in some cases, additions to reserves, rate credits and dividends, premium taxes, and plan profits or losses.” Sourced from National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. The managed care portion of the net cost of health insurance was calculated as 60 percent based on the fact that managed care lives across Medicaid and Medicare made up 60 percent of the total Medicaid and Medicaid lives in 2019, based on data from the Kaiser Family Foundation.

9 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov; CMS NHEA, 2019. The managed care portion of the net cost of health insurance was calculated as 60 percent based on the fact that managed care lives across Medicaid and Medicare made up 60 percent of total Medicaid and Medicaid lives in 2019, based on data from the Kaiser Family Foundation.

10 We did not do deep dives into the other two stakeholder groups (public payers and other sites of care) in an effort to focus on the stakeholder groups with the highest administrative spending. However, we do apply adjusted and conservative savings estimates to aggregate the overall opportunity for the US healthcare system. This is important because all stakeholder groups, including Medicare, have a substantial opportunity to simplify administrative functions and could also benefit from the seismic interventions discussed in chapter 8.

11 There may also be IT spending specific to certain functions in other functional focus areas. 12 As of 2018, the top five health systems together account for only about 13 percent of annual hospital admissions. Sourced from Neha Patel,

Lisa Foo, and Saum Sutaria, “The Silent Shapers of Health Care,” September 2018, McKinsey.com.

26 Chapter 2. Analytical framework

McKinsey & Company McKinsey Center for US Health System Reform

The financial transactions ecosystem comprises the movement of all payments, claims, and billing throughout the healthcare ecosystem among payers, hospitals, physician groups, and customers. Representing $200 billion in annual administrative spending (or 21 percent of total administrative spending), it includes functions such as claims processing and prior authorization (PA) for payers and major portions of revenue cycle management for providers. Salient challenges include high levels of product customization that give rise to payment complexity, highly manual processes, lack of coordination between stakeholder groups on data definitions and payment processes, and non-standardized sys- tem-level workflows such as claims payments and underlying medical documen tation. “Within” interventions (such as the simplification of products and their associated rules and the automation of rules-based tasks), along with “between” interventions (such as building a unified claims tracking system and aligning jointly on PA criteria), could deliver $40 billion in annual savings.

CHAPTER 3

Financial transactions ecosystem

27Chapter 3. Financial transactions ecosystem

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

Medi caid revenue to billing problems, compared with 5 percent for Medicare and 3 percent for commercial.1 To identify opportunities within this ecosystem, we separated it into two components:

— Claims processing: The process of submitting, assessing, and adjudicating claims through the US healthcare eco- system to determine payments to pro- viders from payers, which are based on pre-negotiated rates, and from patients through copays

— Prior authorization (PA): The process by which payers determine the medical necessity of specific procedures before allowing physicians to provide the service

This chapter focuses on interventions within the structure of the current US healthcare system. This allowed us to identify opportu- nities in the near term without a substantial change. Other research examines systems such as a single-payer model, though recent literature has found that administrative spending savings from moving to a single- payer model may not be substantially greater than the potential savings in the current US healthcare system construct.2

3.1 CLAIMS PROCESSING

Definition and sizing Claims processing in this report refers to the flow of claims starting after a medical service has been provided. As such, it would be after the PA portion of the finan- cial trans actions ecosystem. It begins when a claim is submitted by the provider after they provide a service and ends when the payment is either rejected by the payer or accepted, at which point payment is made to the provider (Exhibit 3.1).

Of the $165 billion spent on claims pro- cessing, our key stakeholder groups rep- resented $105 billion—physician groups ($40 billion), private payers ($35 billion), and hospitals ($30 billion). For physician groups and hospitals, these represent the

The financial transactions ecosystem involves the processes through which payers and providers manage payments; it starts with the determination of medically necessary services and includes providers’ submission of claims for those services, as well as payment for those services by pay- ers based on pre-negotiated rates between the various parties.

This ecosystem is complex for a number of reasons. First, the United States is a multi- hospital, multi-physician group, and multi- payer system—meaning that to complete the claims and payments process, payers and providers must communicate with a multitude of organizations, many of which have different definitions of medical ne- cessity and services rendered, business rules, and data requirements. Second, the US healthcare system largely operates in a fee-for-service paradigm. Payers and providers are on the op posite sides of a push-and-pull system that engenders a meaningful number of checks and valida- tions across the claims journey, including PA, submission, adjudication, and appeals.

Additionally, each touchpoint during a patient visit (for example, inpatient ad- mission, laboratory tests) can result in a separate claim in the billing system; even when a claim is billed at the episode level, each service must be billed as a separate line item. Furthermore, the same service, such as a laboratory test, may be proces- sed differently based on the site of care. Finally, each claim comprises multiple parts depending on a patient’s in surance (for example, copay, deductible). This complexity not only must be managed by payers through their claims processing systems but also by hospitals and physi- cian groups as they track revenue for each pa tient service from initial appointment to final payment across various systems.

The outcome is a non-standardized set of processes that are time-consuming, man ual, and possibly prone to error. Re- cent research estimates that because of complexity, phy sicians lose 17 percent of

28 Chapter 3. Financial transactions ecosystem

McKinsey & Company McKinsey Center for US Health System Reform

“Within” opportunity areas: In our ex- perience, payers, hospitals, and physician groups struggle with incomplete and inac- curate provider and customer data within their own systems. This can result in claims errors that must be manually adjudicated, leading to unnecessary administrative spending. These stakeholder groups also struggle with com plicated and inefficient internal processes that prevent the timely and accurate processing of claims. These challenges present a few specific within opportunities, including:

— Complexity of products and associated rules: A rise in the number and speciali- zation of payer products, largely spurred by market competition, leads to an ever- increasing number of rules (for example, eligibility checks) that must be applied to corresponding claims in the adjudication process. Some of these are due to legacy

major com ponents of the “cost to collect” function and include areas such as claims submission and processing, patient ac- counting, and credit and collections. For private payers, this com prises the entire claims management team, including ex- aminers that review claims, specialists that manage high- complexity cases, and grievance and appeals teams (see chapter 2 for more detail on these estimates).

Key opportunity areas We identified the major pain points that have plagued payers, hospitals, and physician groups in the end-to-end claims submission and processing journey (Exhibit 3.2). This analysis revealed two themes: the presence of data inac curacies and process inefficien- cies within organizations, and a lack of con- sistent data definitions and an integrated claims process between organizations.

Exhibit 3.1

Current claims processing journey (illustrative)

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 3.1 of 9.2 (total 22 exhibits)

Source: 2020 CAQH Index; Peter Orszag and Rahul Rekhi, “Real-time adjudication for health insurance claims,” 1% Steps for Health Care Reform, February 2021, onepercentsteps.com; McKinsey Payer Operations Domain

1 Claims creation

After a member receives a service, the provider �lls

out a detailed claims forms, typically electronically

2 Claims submission Provider sends the

form to payers; 96% of this happens electronically today

3A Claims auto-adjudication Claim comes into payer's

inbound systems; typically more than 80% of claims

are auto-adjudicated

3B Claims manual adjudication For the roughly 20% of claims

that have to be manually adjudicated, they are resolved

by claims processors or experienced claims specialists;

claims that are rejected are sent back to providers

4 Payment

Claims tool determines if the claim is going to be paid, and reimbursement

is initiated

5 Claims tracking

Providers and members can track claim statements online; statements can be received via online portals or through mail; either provider or member may

potentially dispute the claim

6 Audit, grievances, and appeals

Claims excellence and adjust- ments team audits claims on a monthly or quarterly basis

to reduce errors and improve accuracy; team also handles

grievances and appeals

29Chapter 3. Financial transactions ecosystem

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

— Poor data management and coordination: Incomplete or inaccurate claims submis- sion can lead to the failure of the auto- adjudication process, resulting in higher- cost manual edits. A common example is when payers record inconsistent data about the same customer across various contracts, including medical, dental, and pharmacy benefits. Each contract may have different information about that same customer. It may also ask for the same information in different ways, making it difficult to determine whether this is the same customer. Reasons for

building upon legacy, but part of this is by design, as stakeholder groups attempt to better manage medical costs. This has been complicated by the growth of value- based payment models that have poten- tially separate rules. Thus, the greater the number and the more complex the business rules, the more likely any claim will need manual review. Additionally, the ability to “match” required documentation (for example, medical records, eligi bility verification) can become more difficult when the rules themselves are complex, such as for dual eligible members.

Exhibit 3.2

Pain points along current claims processing journey (illustrative)

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 3.2 of 9.2 (total 22 exhibits)

Source: 2020 CAQH Index; Peter Orszag and Rahul Rekhi, "Real-time adjudication for health insurance claims," 1% Steps for Health Care Reform, February 2021, onepercentsteps.com; McKinsey Payer Operations Domain

Pain points

Claims creation After a member receives

a service, the provider �lls out a detailed claims forms,

typically electronically

Claims submission Provider sends the

form to payers; 96% of this happens electronically today

Claims auto-adjudication Claim comes into payer’s

inbound systems; typically more than 80% of claims

are auto-adjudicated

Claims manual adjudication For the roughly 20% of claims

that have to be manually adjudicated, they are resolved

by claims processors or experienced claims specialists;

claims that are rejected are sent back to providers

4 Payment

Claims tool determines if the claim is going to be paid, and reimbursement

is initiated

Claims tracking Providers and members can track claim statements online; statements can be received via online portals or through mail; either provider or member may

potentially dispute the claim

Audit, grievances, and appeals Claims excellence and adjust- ments team audits claims on a monthly or quarterly basis

to reduce errors and improve accuracy; team also handles

grievances and appeals

Complexity of products and associated rulesA

Poor data management and coordinationB

Complex claims tracking processC

Unclear Explanation of Bene�ts for membersD

A C F A B E FA B E

A E GGC D

Complexity of payer-provider contractsE

Lack of consistent data de�nitions and supporting informationF

Unclear and ine�cient claims payment tracking and recovery processG

30 Chapter 3. Financial transactions ecosystem

McKinsey & Company McKinsey Center for US Health System Reform

providers, as well as disruptive for the member. The No Surprises Billing Act has begun to address these issues by requiring payers to provide advance cost estimates to members, called “advanced Explanation of Benefits.”4

“Between” opportunity areas: There is a consistent gap in alignment and collaboration between payers and providers throughout the end-to-end claims workflow. The issues range from upstream complexity of payer- provider contracts, to differing data defini- tions and supporting documentation used by payers and providers, to misalignment on how to track and pay claims. This friction cre- ates rework and requires repeat interactions between payers and providers to process a claim accurately. These issues present a few specific between opportuni ties, including:

— Complexity of payer-provider contracts: There is a proliferation of hyper-customized insurance products in the US market due to customers’ demands for tailored bene- fits. This creates a need for contracting between payers and providers to support each customized product. For example, if an employer wants a certain set of doctors to be in-network for their employees, the payer must negotiate appropriate rates with those providers. This process will need to be repeated for every customized benefits package. That means that the same payer-provider combination may need to negotiate as many combinations of rates as there are customers for the payers. The administrative burden of negotiating, recording, and managing these contracts, especially making sure the claim is processed at the right reim- bursement rate for the member, can be quite high for all involved.

this include different forms used to record customer data or lack of communication between the various teams within the same payer that manage medical, pharmacy, dental, and other ancillary benefits. There are parallels to this situation in hospitals and physician groups as well. Such within inconsistencies can prove costly in terms of the additional labor hours needed to triangulate across different sources of data.

— Complex claims tracking process: Cur- rently, more than 95 percent of claims are submitted by providers versus members who submit directly to payers; only 72 percent of claims status inquiries are electronic.3 Lack of transparency on payments can lead these providers to invest time telephoning payer call cen- ters to investigate claims status. Payers reci procally have to invest in call centers to answer those questions.

— Unclear Explanation of Benefits for members: Explanation of Benefits doc- uments are intended to provide payers’ members with an understanding of the fees for the services they received and whether the payer or the member owes payment. Jargon-heavy language and unintuitive presentation can make it un- clear to a member whether documents received are explanations of what has already been paid or whether there is an additional balance owed. This issue be- comes even more acute in high-deductible health plans. Furthermore, information from the provider may not be consistent with what is provided by the payer, causing further confusion for the member. The admini strative burden of resolving the confusion through grievance and appeals teams can be costly to both payers and

There is a consistent gap in alignment and collaboration between payers and providers throughout the end-to-end claims workflow.

31Chapter 3. Financial transactions ecosystem

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

For example, if one organization wants to reduce manual errors, they may choose to in- vest in the automation of downstream claims processes such as adjudication; another may choose the up-front simplification of products to reduce downstream complexity:

— Simplify products offered: For a payer, interventions that simplify the number of products could have two benefits: reducing the complexity of business rules—including the number of contracts between the same payer-provider pair—that must be applied to claims and increasing the pro- portion of claims that can be auto- adjudicated. Simplifying products, how- ever, must be balanced against the wish to provide customized services to custo mers and to design products nuanced for various risk pools. While this inter vention can be done by individual payers, we also explore a “seismic” intervention at an industry- wide level in chapter 8.

— Streamline claims submission and commu- nication process: Interventions that payers can consider adopting include: creating a simplified provider platform for claims submission, supplementing claims with provider-friendly explanations and simple next steps, and providing push notifica- tions to update providers on claims status.

— Automate adjudication: As noted in Ex- hibit 3.1, only about 20 percent of claims need manual adjudication, but they drive a sizeable savings opportunity. Payers can conduct single- and multi variable analyses of claims data to determine the most common causes about why a claim failed to process automatically. They can also use self-learning algorithms that filter claims in real time upon entering into their systems, using core elements of the data (for example, age, gender, diagnosis-related group codes, primary and secondary diagnoses) to identify cases that could have issues such as fraud. Providers can use automation to ensure a claim is as complete and accurate as possible before submitting it. They can do this by validat- ing that a medical service is, in fact, cov-

— Lack of consistent data definitions and supporting information: There is often a lack of clarity and alignment on how providers should fill out and substantiate claims forms, especially given differing rules and expectations from various pay- ers. This can lead to a burdensome back and forth between payers and providers to ensure they record the right information before a claim can be processed. A com- mon example is missing medical records that are needed to validate the medical necessity of the service (examined as part of the PA process in chapter 3.2). Another example is when providers fail to complete claims forms because they may not under- stand which fields are required; the most frequently missed information on claims forms includes date of accident, date of medical emergency, and date of onset.5

— Unclear and inefficient claims payment tracking and recovery process: Payers, hospitals, and physician groups must often manage a fragmented, manual process of tracking claims and payments between each other. For hospitals, this is called the “cost to collect.” In our experience, hospi- tals spend about 2 to 3 percent of their annual revenue ensuring they receive the right payments and have well-timed cash flows coming in from payers. Payers simi- larly dedicate resources to their payment integrity function to check for fraud and abuse. Research shows that the health- care industry could reduce $20 billion to $30 billion in fraud and abuse annually.6

Known interventions “Within” interventions (5 to 10 percent): We catalogued known interventions that could save 5 to 10 percent of administrative spending on claims processing solely through within interventions. These are interventions that can be controlled and implemented by individual organizations. While these inter- ventions roughly match one-to-one with the opportunity areas, the combination of inter- ventions will depend on the pain points that an individual organization is most acutely facing.

32 Chapter 3. Financial transactions ecosystem

McKinsey & Company McKinsey Center for US Health System Reform

is the Fast Healthcare Information Resource (FHIR), which is creating an internet-based approach to searching and exchanging healthcare information like search engines do with data in other industries.9

— Improve data management and coordi- nation: Interventions here require a con- certed effort by payers and providers to substantially upgrade their capabilities and technologies with regard to data. In our experience, one of the highest return- on- investment interventions includes maximizing the potential for digitalization of claims intake through provider portal improvements. These portals, generally built by third parties, allow providers to log into accounts for different payers and conduct activities such as searching for members’ eligibility, submitting claims, and tracking claims status. Improving the ease of use of portals, including ensuring connectivity to electronic health records so that data do not have to be reentered, can further reduce hospital and physician group administrative spending. Other interventions include: standardizing lan- guage, definitions, and the terminology used in contracts for products and by claims teams, employing natural-language pro- cessing to scrub contracts and streamline interpretation when a non-standard claim comes in, and facilitating claims processing by reducing the number of clearinghouses. (See chapter 8 for a seismic intervention regarding a centralized, automated claims clearinghouse that reduces the need for individual payers, hospitals, and physician groups to launch their own initiatives.)

— Improve coordination and clarity on claims-related communications: Interven- tions could include collaboration between the claims and revenue cycle management

ered under a patient’s health benefits or checking that the total claim amount aligns with the total allowed amount under the patient’s current policy.7

— Clarify Explanation of Benefits: Interven- tions by payers may include simplifying member-facing interfaces, such as offer- ing a digital payment summary that allows members to immediately receive an expla- nation of their financial responsibility. They can also provide the Explanation of Bene- fits and next steps in easy-to-understand language through communication medi- ums tailored to member preferences—for example, text versus phone calls versus emails—and by using cleaner visuals to explain the flow of payments.

“Between” interventions (15 to 20 percent): We also catalogued known interventions that could save 15 to 20 percent of administrative spending on claims processing solely through between interventions. These are interven- tions that require agreement and collabora- tion between organizations but not broader, industry-wide change. While these interven- tions roughly match one-to-one with the opportunity areas, the combination of inter- ventions will depend on the most acute pain points that the payer-provider pair is facing. For example, one payer-provider pair may face issues around claims intake and could invest in provider portals to standardize that process; another set may use portals to standardize the claims tracking process instead. Further, several initiatives are un- derway that provide foundational support to these interventions. One example is the US Core Data for Interoperability (USCDI) pro- gram, which aims to create a patient-centered healthcare data set that specifies standards- based health data classes and elements to improve interoperability.8 Another initiative

Provider portals, which are tools that help payers and providers with data management, can support easier tracking and recovery of claims.

33Chapter 3. Financial transactions ecosystem

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

3.2 PRIOR AUTHORIZATION

Definition and sizing PA is a specific function with in the medical management operations of payers and revenue cycle management functions of hospitals and physician groups. In the market-based US healthcare system, PA has come about as a check and balance between these stakeholder groups. Its primary goal is to assess the medical necessity and coverage of healthcare services and procedures according to established criteria or guidelines under the provisions of payer programs to pre- vent excess and unnecessary utilization. PA also could flag if newer, better treat- ments are available, improving the quality of care (for example, in evolving specialties such as oncology where the standards of care are being refined). PA plays an impor- tant role in the US healthcare system, so there will always be necessary administra- tive spending on this function.10

PA affects a small subset of procedures; for example, more than 90 percent of commercial enrollees are in plans that limit PA to less than 25 percent of medical ser- vices.11 But PA remains a labor-intensive process. It starts when a physician deter- mines that a patient needs a service, such as surgery, and contacts the patient’s payer to as certain if that particular service requires PA. The subsequent steps include payers checking on whether the procedure is medically necessary, providers attaching relevant documentation, and back-and- forth conversations to adjudicate the result (Exhibit 3.3).

Research has also shown that there are in direct costs of PA, such as physician burnout and employee turnover, perhaps resulting from the amount of paperwork providers are required to do.12-14 While we acknowledge this, we are focused on the direct, administrative spending associated with PA in this report.

teams of payers and providers to clarify language on their Explanations of Bene- fits, building a unified system that allows members to track the status of a claim, and simplifying member-facing interfaces.

— Streamline claims payment tracking and recovery process: Provider portals, men- tioned above as tools that help payers and providers with data management, can also support easier tracking and recovery of claims. As claims are being processed, providers can use portals to track claims and identify any trending errors or prob- lem areas that need attention. In addition, the portal could provide notifications regarding current claims status and ex- pected payment data. Teams at the payer and provider could further extend this idea into an end-to-end solution for the providers’ appeals process, including auto-validation and automatic notifica- tions to providers on the status. While this could be a step forward, more admin- istrative savings might come from direct interoperability between payers’ and providers’ systems to submit and process claims. This could lessen the burden on providers to log into multiple systems and submit each claim individually.

— Align incentives between payer and pro- vider through risk-sharing models: In this intervention, payers and providers align on two areas: (1) the metrics they could moni- tor on quality of care and resulting patient outcomes, and (2) the most critical pieces of data that they would need to share to inform those metrics. This alignment could reduce the administrative burden on both sides of the relationship. Nonetheless, if each relationship has a differing set of metrics, it could increase complexity overall for the US healthcare system. Investments in data and analytics to support these new models may also be necessary. We acknowl- edge that doing this at an industry-wide level could have more impact (see chapter 8 for further detail), but given the current US healthcare system, one-by- one agree- ments are the more likely path.

34 Chapter 3. Financial transactions ecosystem

McKinsey & Company McKinsey Center for US Health System Reform

this include the complexity of medical policies, lack of infrastructure supporting electronic submission of supporting clinical documentation, limited vendor options, number of web portals, and varying state laws.16

— Lack of standardization: The standard for sharing supporting medical docu- ments for PA has not yet been estab- lished, leading to additional complexity in submitting PA requests.17

— Effectiveness of PAs: Given the high burden that PAs place on providers, one salient pain point involves ensuring that they are as effective as possible. This could include applying discretion when using PAs (for example, in targeted high-cost areas where treatments are still evolving) and where the possibility of denials being overturned is low.

Of the $35 billion spent on PA, our key stakeholder groups represented $30 billion—private payers ($10 billion), hos- pitals ($10 billion), and physician groups ($10 billion). While we focus primarily on PA for medical services, there is potential for pharmacy benefit mana gers and pharma- cies to reduce administrative spending in this area as well.

Key opportunity areas There are three major pain points in the current PA journey (Exhibit 3.4).

— Manual process: PA is one of the least automated transactions in healthcare. About 21 percent of medical services PAs are electronic, compared to approxi- mately 85 percent of eligibility and bene- fit veri fications and 96 percent of claims sub missions (Exhibit 3.5).15 Reasons for

Exhibit 3.3

Current prior authorization journey (illustrative)

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 3.3 of 9.2 (total 22 exhibits)

¹ MD, medical doctor; RN, registered nurse. Source: McKinsey Payer Operations Domain

1 Patient diagnosis Provider decides

service, lab test, or drug prescription

needed for patient

2 Provider checks for requirements Sta� reviews prior

authorization requirements for

specic payer

3 Provider retrieves

payer-speci�c form Provider sta�

contacts payer and retrieves form

4 Submission

Provider submits form and attaches

medical documents

6 Case creation

Payer creates a new case and populates with

relevant biographic, demographic, clinical,

and administrative data

7 Clinical review

At payer, rst-level review for eligibility coverage, level of care, and medical necessity by clinician, typically RN¹; MD¹ reviews for nal determination

when RN recommends denial

8 Case determination

and noti�cation Payer makes nal determination and

communicates approvals or denials to providers

and members

5 Intake validation and case triage

Payer receives clinical information, veries accuracy, and routes

to appropriate reviewing party

9 Case review and appeal

In case of rejection, provider reviews

reason for rejection and potentially

resubmits form/ attachment

35Chapter 3. Financial transactions ecosystem

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

product lifecycle (for example, a branded drug losing its patent), codified PAs may not be revisited or reset to acknowledge this change. In our experience, there are a number of PAs that continue to be moni- tored but are no longer clinically relevant. This leads to unnecessary administrative spending for payers. Annual review and removal of PAs that are never denied can be quick wins that reduce the administra- tive burden of this process.

— Prescreen PA using digital support: Pro- cessing PAs for a payer is largely manual, using skilled labor such as doctors and nurses. In our experience, introducing a digital prescreening process based on clinical pathways could reduce the

Known interventions “Within” interventions (0 to 5 percent): The nature of PA requires payers and pro- viders to collaborate to determine the value- added of various services. As a result, the savings that a single organization working alone could capture may be limited. Still, there are opportunities for individual organi- zations to reduce manual work by prioritizing which PAs are actually needed and triaging the ones that have the highest impact:

— Sunset old PAs: Many PAs are introduced to assess new treatments in areas where medical policies have not yet been formal- ized, requiring careful evaluation to justify the frequently high costs. However, as those treatments go through a normal

Exhibit 3.4

Pain points along current prior authorization journey (illustrative)

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 3.4 of 9.2 (total 22 exhibits)

¹ MD, medical doctor; RN, registered nurse. Source: McKinsey Payer Operations Domain

1 Patient diagnosis Provider decides

service, lab test, or drug prescription

needed for patient

Provider checks for requirements Sta� reviews prior

authorization requirements for

specic payer

Provider retrieves payer-speci�c form

Provider sta� contacts payer and

retrieves form

Submission Provider submits

form and attaches medical documents

Case creation Payer creates a new

case and populates with relevant biographic,

demographic, clinical, and administrative data

Clinical review At payer, rst-level review for eligibility

coverage, level of care, and medical necessity by clinician, typically RN¹; MD¹ reviews for nal determination

when RN recommends denial

Case determination and noti�cation Payer makes nal determination and

communicates approvals or denials to providers

and members

Intake validation and case triage

Payer receives clinical information, veries accuracy, and routes

to appropriate reviewing party

Case review and appeal

In case of rejection, provider reviews

reason for rejection and potentially

resubmits form/ attachment

Pain points

Manual processA Lack of standardizationB E�ectiveness of prior authorization¹C

A B

A B A B A B C

B A B A

36 Chapter 3. Financial transactions ecosystem

McKinsey & Company McKinsey Center for US Health System Reform

“Between” interventions (10 to 20 percent): Organizations could see 10 to 20 percent savings on their PA spending through two pri- mary between interventions that help address pain points involving manual work and that allow them to be more effective in applying PAs to targeted procedures and services:

— Align jointly on PA criteria: Many times, friction between payers and providers is created from differing perspectives on criteria for PA. Some interventions in- clude: prealigning on medical necessity criteria for procedures and services where third- party standards are not available, or identifying PAs where deni- als are frequent and jointly addressing the root causes (for example, when a PA consistently has missing documentation, update what documents are required).21

— Conduct targeted “gold carding”: Gold carding is an arrangement between a payer and select providers to eliminate or reduce PA requirements for specific services, which simplifies administration for both sides. To operationalize this intervention, advanced analytics can help tailor gold-carding approaches to specific specialties and regions. For example, gold carding based on the pro- viders’ adherence to evidence-based clinical guidelines can help lower admini- strative spending while keeping clinical outcomes constant. Our experience shows that gold carding can save 5 to 10

amount of manual labor by separating PAs that are more common, whether for denial or acceptance. Furthermore, the data from the prescreening can provide payers with the ability to identify PAs to sunset or alter.

— Increase proportion of automated PAs: In our experience, interventions to auto- mate PAs throughout the journey could lead to a 5 to 10 percent reduction in total PA spending for payers and providers. Currently, most PAs are handled manu- ally; only about 21 percent are automat- ed.18 A recent pilot run by Council for Affordable Quality Healthcare (CAQH), a healthcare non- profit alliance, and the Cleveland Clinic employed new PA oper- ating rules, excluding attachments. The approach led to staff-time savings of 80 percent, or 12 minutes of processing time per PA.19 A 2021 study showed that 71 percent of experienced providers who implemented electronic PAs reported faster time- to-patient care; they reduc- ed the time between submitting a PA request and receiving a decision from the payer by 69 percent.20 Automation- based interventions such as these are still emerging. They require a set of enablers, including: robust infrastructure to support submission of medical docu- mentation, adoption of automation tech- nology by providers, investment by pay- ers, change in workflow processes, and integration of health information.

Exhibit 3.5

Electronic adoption levels across various claims work�ow transactions

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 3.5 of 9.2 (total 22 exhibits)

¹ See chapter 8 for further exploration of adopting a centralized, automated claims clearinghouse. Source: 2020 CAQH Index

Eligibility and bene�t veri�cation

Prior authorization

Claim submission

Coordination of bene�ts/

crossover claim

Claim status inquiry

Claim payment¹

Remittance advice

84 21 96 89 72 74 57

Percent of transactions that are electronic, 2020

37Chapter 3. Financial transactions ecosystem

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

improve quality. Also, payers would likely need a high share of a practice’s patients to realize the full savings. Another con- cern is that larger providers could be ad- vantaged over smaller providers to meet the necessary conditions for gold card- ing, given greater resources.23 A subop- timal outcome would be if the adminis- trative savings from gold carding were outweighed by significantly increased medical spending. However, in our expe- rience, there are enough data on provider behavior that could help to implement gold carding in targeted ways and still achieve net-positive savings for the US healthcare system.

percent of total PA spending for payers and providers. There is some momentum in US healthcare to expand gold carding. For example, in June 2021, the Texas House of Representatives passed a bill allowing phy sicians to earn “gold card exemptions” based on their track re- cord.22 Because gold carding is still an emerging inter vention, there are several considerations for implementing it successfully. In our experience, there is growing acknowledgment that gold carding will be most successful when providers are in two- sided risk arrange- ments and thus jointly incentivized with payers to both manage spending and

1 Abe Dunn et al., A denial a day keeps the doctor away, National Bureau of Economic Research (NBER), Working paper number 29010, July 2021, nber.org.

2 Arnold Milstein et al., “Reducing administrative costs in US health care: Assessing single payer and its alternatives,” Health Services Research, 2021, Volume 56, Number 4, pp. 615–25, doi.org/10.1111/1475-6773.13649.

3 CAQH Index, “2020 CAQH Index: Closing the gap: The industry continues to improve, but opportunities for automation remain,” Council for Affordable Quality Healthcare, 2021, caqh.org.

4 Stacey Hughes, “No surprises act – Good faith estimates and advanced explanation of benefits,” American Hospital Association, June 2, 2021, aha.org.

5 Michael Sculley, “10 common medical billing mistakes that cause claim denials – Part 1,” Practice Management Newsletter, PracticeSuite, updated January 20, 2021, practicesuite.com.

6 Mahi Rayasam, Justin Tran, Manuk Garg, and Prashanth Reddy, “Using machine learning to unlock value across the healthcare value chain,” August 2, 2018, McKinsey.com.

7 Shubham Singhal, Penelope Dash, Tobias Schneider, Sameer Chowdhary, and Himanshu Aggarwal, “For better healthcare claims management, think digital first,” June 14, 2019, McKinsey.com.

8 United States Core Data for Interoperability (USCDI), The Office of the National Coordinator for Health Information Technology, 2021, healthit.gov.

9 “What Is HL7® FHIR®?,” The Office of the National Coordinator for Health Information Technology, April 2021, healthit.gov. 10 During the COVID-19 pandemic, a number of changes were made to PA practices. For example, national payers suspended PA requirements

for certain COVID-19-related procedures, and some states suspended PA for a broad category of treatments such as cardiology or advanced imaging. However, the impact and duration of these changes have not yet been determined and any evolution may have an impact on our findings.

11 “America’s Health Insurance Plans (AHIP) survey: Prior authorization grounded in clinical evidence and selectively used,” America’s Health Insurance Plans, June 9, 2020, ahip.org.

12 Jacqueline LaPointe, “Prior authorization burden still high despite COVID struggles,” Revcycle Intelligence, April 13, 2021, revcycleintelligence.com.

13 John Epling et al. “Practice characteristics and prior authorization costs: secondary analysis of data collected by SALT-Net in 9 central New York primary care practices,” BMC Health Services Research, March 2014, Volume 14, Number 109, bmchealthservres.biomedcentral.com.

14 Nancy Lin, Harvey Bichkoff, and Michael Hassett, “Increasing burden of prior authorizations in the delivery of oncology care in the United States,” Journal of Oncology Practice, 2018, Volume 14, Number 9, pp. 525–8, ascopubs.org.

15 CAQH Index, “2020 CAQH Index: Closing the gap: The industry continues to improve, but opportunities for automation remain,” Council for Affordable Quality Healthcare, 2021, caqh.org.

16 CAQH Core, “Moving forward: Building momentum for end-to-end automation of the prior authorization process,” Council for Affordable Quality Healthcare, 2019, caqh.org.

17 CAQH Core, “Moving forward: Building momentum for end-to-end automation of the prior authorization process,” Council for Affordable Quality Healthcare, 2019, caqh.org.

18 CAQH Index, “2020 CAQH Index: Closing the gap: The industry continues to improve, but opportunities for automation remain,” Council for Affordable Quality Healthcare, 2021, caqh.org.

19 Tim Kaja, April Todd, and Susan Turney, “Standards subcommittee meeting: Hearing on request for the National Committee on vital and health statistics (NCVHS) review of CAQH CORE operating rules for federal adoption,” Council for Affordable Quality Healthcare, August 2020, ncvhs.hhs.gov.

20 “New analysis shows benefits of electronic prior authorization for patients and providers,” America’s Health Insurance Plans, March 24, 2021, ahip.org.

21 Wendy Warring and Lauren Bedel, “Streamlining prior authorization: Final report & recommendations,” Network for Excellence in Health Innovation, September 2021, nehi-us.org.

22 “Governor approves bill to reduce prior authorization hassles,” Texas Medical Association, last updated June 30, 2021, texmed.org. 23 Wendy Warring and Lauren Bedel, “Streamlining prior authorization: Final report & recommendations,” Network for Excellence in Health

Innovation, September 2021, nehi-us.org.

38 Chapter 3. Financial transactions ecosystem

McKinsey & Company McKinsey Center for US Health System Reform

Industry-agnostic corporate functions comprise back-office, non-clinical tasks that are present in all industries, such as human resources and finance. In healthcare specifically, they are the largest of the five functional focus areas, with $375 billion in annual administrative spending (or 39 percent of total administrative spending). Low levels of efficiency and effectiveness are salient problems that manifest themselves in a high proportion of automatable tasks that are currently manual and prone to errors (for example, creation of financial reports or creating a single source of truth about vendor spending across finance, procurement, and compli- ance teams). Interventions to address these problems include traditional operational excellence levers, such as demand management, lean process redesign, and automation. There are also next-generation interventions to build “functions of the future” by using new technologies and capabilities such as analytics and cloud computing to support faster decision making and extract sharper insights (for example, using data analytics in human resources to improve career development and retention). All of these are “within” interventions and could deliver $95 billion in annual savings.

CHAPTER 4

Industry-agnostic corporate functions

39Chapter 4: Industry-agnostic corporate functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

($45 billion). For both hospitals and phy- sician groups, this represented roughly 45 percent of their total administrative spending, driven by corporate functions such as HR, finance, accounting, and IT. Industry- agnostic corporate functions accounted for about 25 percent of private payers’ total administrative spending, much of which comprises similar back- office functions (see chapter 2 for more detail on these estimates).

Key opportunity areas Low efficiency and effectiveness are a challenge to healthcare organizations’ industry- agnostic corporate functions. Healthcare has been slow to adopt traditional levers to address adminis- trative spending in corporate functions, such as demand management and lean process redesign, and it has also lagged behind other industries in automation.5,6 For example, in finance, auditing ex- pense reports or vacation time is a manual and inefficient process in many healthcare organizations. In other in- dustries, organizations have been able to implement algorithms to cross-check expense reports against travel and personnel data, or systems to compare declared vacation days with badge swipes and computer-usage data. Some organizations have also redesigned employee activities and organizational structures to take advantage of worker capacity freed up by automation.7

The automation potential of technol- ogies such as robotic process auto- mation, machine learning, smart work- flows, or natural language processing is substantial. Across industries in HR functions, there is about a 22 percent automation potential, with some sub- functions, such as time collection, at- tendance, and record keeping, showing more than 30 percent potential. In IT, there is about a 22 percent automation potential, with all subfunctions having 20 to 30 percent potential.8,9

In all industries, there are certain back-office administrative functions that enable an organization to operate. These are generally referred to as corporate or operational functions. For healthcare organizations, many of these functions are in dustry- agnostic; another group represents industry- specific operations. In this chapter, we will focus on industry- agnostic corporate functions, such as finance and human resources (HR).

New technologies such as automation and digitalization have transformed these corporate functions in other industries, delivering the next wave of productivity and savings while also improving customer experience. For example, in banking, new technologies could create an additional $1 trillion of value.1 Further, 60 percent of financial services leaders stated their organizations have used at least one new technology, including robotic process automation (36 percent), virtual assistants (32 percent), and machine learning tech- niques (25 percent).2 These advances also highlight two other important hurdles that organizations are overcoming: how to more effectively manage handoffs between person and machine and how to put off or skip typical process re design or reengineering in favor of automation.

To date, this evolution has largely not been realized in healthcare. Recent research highlights a substantial oppor- tunity for automation among payers and providers.3,4 Capturing these savings will depend on an organization’s ability to scale and coordinate automation across the enterprise, while also find- ing ways to re skill the workforce into higher-productivity roles.

Definition and sizing Of the $375 billion spent on industry- agnostic corporate functions, our key stakeholder groups represented $250 billion—hospitals ($115 billion), physician groups ($90 billion), and private payers

40 Chapter 4: Industry-agnostic corporate functions

McKinsey & Company McKinsey Center for US Health System Reform

demand management to remove low-value activities, smart sourcing that procures resources most effi- ciently, lean process improvement through redesign and workload bal- ance, organizational and governance changes that align the performance of teams with the organization’s vision, consolidation of support services to create scale, and automation to im- prove the quality of service and the precision of outcomes.

— Build for “functions of the future”: Healthcare is starting from a deficit in terms of automation in industry- agnostic corporate functions. This also creates an opportunity to jump ahead and bring industry-agnostic corporate functions into the next generation. For example, a tra ditional function such as HR could be reima- gined to address the strategic needs of the business and the changing nature of the workforce. This could mean building an analytics capability within HR to mine data to hire, devel-

Known interventions “Within” interventions (20 to 30 per- cent): To address these key opportunity areas, we catalogued known interven- tions. These could reduce administrative spending on industry- agnostic corporate functions by 20 to 30 percent, using solely within interventions. Individual organizations can achieve much of these savings, but some potential gains can be unavailable to smaller organizations because of limited economies of scale. However, the more recent entrance of vendors that offer these capabilities to payers and providers has created partnership op portunities that decrease the necessary scale. As a result of this partnership model, more healthcare organizations are now able to pursue the interventions below, regardless of their size:

— Promote operational excellence using traditional levers: For corporate functions, traditional levers to capture value that have been proven in other industries (Exhibit 4.1) include:

Exhibit 4.1

Traditional levers for operational excellence in corporate functions

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 4.1 of 9.2 (total 22 exhibits)

Source: McKinsey Payer Operations Domain

• Non-labor cost reduction • Improved decision quality • Improved service outcomes

• Support infrastructure consolidation • Labor centralization for economy of skills • Service factories (eg, call centers, COEs)

• Strategy and vision alignment • Organization structure optimization • Governance e�ectiveness • Functional performance management

• Price versus usage transparency • Self-service promotion • Low value-added activity elimination

• Labor arbitrage opportunities • Outsourced services optimization • Strategic purchasing practices

• Process �ows improvement • Workload balance and management • Task e�ciency improvement

Example initiatives by lever

4 Organization

and governance optimization

5 Support service

consolidation

6 IT and

automation

1 Manage demand

Traditional levers

2 Smart

sourcing

3 Improve process

(lean)

41Chapter 4: Industry-agnostic corporate functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

could allow smaller or ga nizations to implement these interventions with- out substantial capital investment in new technology.

“Between” interventions (not appli- cable): Given the internal nature of industry- agnostic corporate functions, the majority of savings are within orga- nizations, not between different stake- holder groups.

op, and retain the best employees.10 Skills-focused sourcing is another opportunity, where HR professionals are supplied with the analytical tools and relevant data necessary to identi- fy temporary labor for their organiza- tion’s changing needs. This may be a critical time saver and source of effi- ciency for providers that increasingly rely on agency staffing. Finally, the growth of cloud-based technology

1 Suparna Biswas, Brant Carson, Violet Chung, Shwaitang Singh, and Renny Thomas, “AI-bank of the future: Can banks meet the AI challenge?,” September 19, 2020, McKinsey.com.

2 Suparna Biswas, Brant Carson, Violet Chung, Shwaitang Singh, and Renny Thomas, “AI-bank of the future: Can banks meet the AI challenge?,” September 19, 2020, McKinsey.com.

3 James Manyika, Michael Chui, Mehdi Miremadi, Jacques Bughin, Katy George, Paul Willmott, and Martin Dewhurst, “A future that works: Automation, employment, and productivity,” January 2017, McKinsey.com.

4 Brandon Carrus, Sameer Chowdhary, and Rob Whiteman, “Making healthcare more affordable through scalable automation,” September 16, 2020, McKinsey.com.

5 James Manyika, Michael Chui, Mehdi Miremadi, Jacques Bughin, Katy George, Paul Willmott, and Martin Dewhurst, “A future that works: Automation, employment, and productivity,” January 2017, McKinsey.com.

6 Brandon Carrus, Sameer Chowdhary, and Rob Whiteman, “Making healthcare more affordable through scalable automation,” September 16, 2020, McKinsey.com.

7 Frank Plaschke, Ishaan Seth, and Rob Whiteman, “Bots, algorithms, and the future of the finance function,” January 9, 2018, McKinsey.com. 8 Alexander Edlich, Fanny Ip, and Rob Whiteman, “How bots, algorithms, and artificial intelligence are reshaping the future of corporate support

functions,” November 15, 2018, McKinsey.com. 9 When deploying these levers, especially automation, it is important to guard against algorithmic bias that could negatively affect equity or

access for vulnerable populations. 10 Asmus Komm, Florian Pollner, Bill Schaninger, and Surbhi Sikka, “The new possible: How HR can help build the organization of the future,”

March 12, 2021, McKinsey.com.

42 Chapter 4: Industry-agnostic corporate functions

McKinsey & Company McKinsey Center for US Health System Reform

Industry-specific operational functions are back-office, non-clinical tasks that are generally only found in healthcare, such as clinician credentialing and management of medical records, or that have health- care- specific nuances, such as underwriting or broker-based sales. They are the third largest of the five functional focus areas, with $135 billion in annual administrative spending (or 14 percent of total admini strative spending). Similarly to industry-agnostic corporate functions, they suffer from low levels of efficiency and effectiveness, for example labor- intensive, customized processes to validate and credential pro viders. That is the case because these healthcare-specific functions have not received as much attention regarding automation or adoption of new tech nologies. Interventions include using predictive analytics to build “smart services,” for example ones that use self- learning algorithms to guide payers’ members to provider options that best fit them. Other interventions include making foundational data investments—for instance, building data lakes that link disparate data sources, such as claims, customers, and contracts. All of these are “within” interventions and could deliver $30 billion in annual savings.

CHAPTER 5

Industry-specific operational functions

43Chapter 5: Industry-specific operational functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

functions represented 10 percent of total administrative spending and manifest themselves in functions such as medical records, non-clinical provider-specific IT, and quality management (see chapter 2 for more detail on these estimates).

Key opportunity areas An example of an industry-specific op- erational function with high opportunity for automation is enrollment and billing. In our experience, this function has an automation potential of approximately 25 percent. The function is the first step for payers to enter new customers into their systems, but it still relies heavily on manu- al data entry. For example, Small Groups (employers with less than 50 employees), Individual, and Medicare Advantage members are typically enrolled and con- figured through paper processes, which then require manual entry. Furthermore, paper- and fax-based enrollment pro- cesses often have no tracking, confirm- ation, or status updates. So, communi- cations between the payer and the custo mer are routinely delayed, leading to holdups in setting up coverage. This can also make coordination more difficult between teams within a payer, for exam- ple, a contracts team that needs informa- tion about the members being enrolled to finalize documentation.

Known interventions “Within” interventions (20 to 30 per- cent): To address these key opportunity areas, we catalogued known interven- tions that could save 20 to 30 percent of administrative spending on industry- specific operational functions using solely within interventions. Much of these sav- ings can be achieved by individual organi- zations, either through direct investments in technologies or partnerships with at- scale vendors. The interventions listed below include both traditional levers, which apply as much to industry-specific operational functions as they do to

As discussed in chapter 4, healthcare organizations are supported by a set of administrative back-office functions that enable them to operate. These are generally referred to as corporate and operational functions. Although many of these functions are industry- agnostic in healthcare, others focus on healthcare- specific tasks. In this chapter, we will focus on industry-specific operational functions, such as enrollment and billing, quality reporting, underwriting and actu- arial, and clinician credentialing.

In most cases, healthcare organizations suffer from heavy manual work that re- sults in low reliability and inaccurate out- puts. Take underwriting, which is one of the most strategic operational capabili- ties for payers. This unit determines the premium rates for different customers based on customer-specific factors, such as size, mix, and utilization; the strength of the payer’s network with providers, as well as its ability to route members to specific services; and macro economic factors, such as inflation. Because much of the data collection and analysis for underwriting is still manual, payers have been unable to employ new technologies such as pre dictive analytics, which gen- erates pricing tailored to each customer (Exhibit 5.1). This represents a problem in both efficiency and effectiveness that results in unnecessary admini strative spending.

Definition and sizing Of the $135 billion spent on industry- specific operational functions, our key stakeholder groups represented $90 billion—private payers ($50 billion), hos pitals ($20 billion), and physician groups ($20 billion). For payers, this represented about 28 percent of their total administrative spending, driven by areas such as underwriting and pricing, broker-based sales, membership and billing, and clinician cre dentialing. For hospitals and physician groups, these

44 Chapter 5: Industry-specific operational functions

McKinsey & Company McKinsey Center for US Health System Reform

supporting them with chatbots to answer frequently asked questions as they weigh their options, deploying self-service tools (for example, guided portals) for follow-ups, and using analytics-based agent coaching tools for personalized support. In addition to direct member-facing interventions, smart back-office interventions include using optical-character- recognition platforms to rapidly digitalize forms and cut down processing time, automating data extraction from digital forms utilizing natural language processing or robotic process automation tools, and embedding workflow tools with reporting capabilities to provide status updates. Post-enrollment, interventions include enabling digital and voice assis-

industry- agnostic corporate functions (see chapter 4), and newer interventions, such as smart services, that have become avail able to healthcare organizations with the introduction of new technologies.

— Promote operational excellence using traditional levers: As laid out in chapter 4, the levers that can improve industry-agnostic corporate functions can be applied for many industry- specific operational functions.

— Build smart services: Interventions that use predictive analytics to aid decision making can have an impact n administrative settings. In the mem- ber engagement process, predictive analytics can “smarten” navigation by guiding members to best-fit options,

Exhibit 5.1

Evolution of underwriting as a “smart” function (illustrative)

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 5.1 of 9.2 (total 22 exhibits)

Source: McKinsey Commercial Payer Analytics Domain

1 Data

gathering

2 Rating/ quoting

3 Rate release to sales and negotiation

4 Post-sale

implementation

5 Performance

tracking

Common pain points

Underwriting example

• Multiple database platforms to gather data • Manual scrubbing

• Fragmented rating engine (eg, over 5 rating agencies) • Highly manual work

• Lack of standardized work�ows • Lack of clear goals or incentives for underwriting

• Accountability issues when collaborating with billing and other teams • Underwriters burdened by administrative tasks

• Lack of centralized solution • Lack of consistency • Lack tracking of critical metrics

• Standardized data intake forms • Automated data-gathering process using optical-character-recognition technology and robotic process automation

• Automated processes where underwriters manually log, summarize, and transfer data across systems • Eliminated learning curves between specialized underwriters who perform speci�c tasks

• Provided connected experience for agents/ underwriters, sales associates, administrative supports, and billing to work on the same process with role-relevant views, rules exposed within/called by electronic processes, and automated reports • Developed capacity planning capability against forecasted case volume

• Improved visibility of underwriting data between teams through a common dashboard functionality • Integrated with a knowledge library to provide guidance on resolving edits and processing accounts/cases

Example “smart” solutions

45Chapter 5: Industry-specific operational functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

“speak to each other” across systems; and data-driven insights, or integrated data that aid decision making. As shown earlier in this chapter, payers’ underwriting functions may be too manual and complex from start to fin- ish (Exhibit 5.1). In our experience, one payer addressed this issue by investing in a cloud-based data lake that houses all customer, contracts, and claims data and links them through common data dictionaries and standardized identifiers (for example, using “ABC” instead of “ABC Rx” for the name of a prescription drug). This investment resulted in needing less time to model personalized agreements with cus- tomers and lower error rates, leading to a 40 percent reduction in the time it took to formalize a customer deal.

“Between” interventions (not appli cable): Given the internal nature of industry- specific operational functions, the majority of savings are within organizations and not between different stake holder groups.

tants to onboard and educate members about benefits and developing multiple formats for identification cards (for ex- ample, physical, digital, and biometrics). In our experience, one Medi care Advan- tage payer was able to move from less than 25 percent paper- based enroll- ment to 90 percent digital enrollment, reducing the total processing time for an application from 25 minutes to five minutes. To do this, the payer built a digital enrollment platform that used techniques such as auto- population of known fields to eliminate manual work (for example, name of city once zip code was filled in), video guides for members to self-enroll, and multichannel access to the platform, including voice- enabled intake from cell phones.

— Empower a function through founda- tional data investments: Healthcare organizations may lack: foundational data liquidity, or data that are stored in similar formats and easily accessi- ble; interoperability, or data that can

46 Chapter 5: Industry-specific operational functions

McKinsey & Company McKinsey Center for US Health System Reform

Customer and patient services comprise tasks that address questions and concerns expressed by payers’ customers (members, hospitals, and physician groups) and providers’ customers (patients). These services are increasingly moving to digital channels, such as self-service websites and applications, but much of this work in healthcare still occurs through phone calls. This is a smaller area of spending, accounting for $80 billion in annual administrative spending (or 9 percent of total administrative spending). Salient pain points include a high volume of customers that have questions about billing or services rendered, as well as process complexities such as multiple routings to answer a question. “Within” interventions could create savings by reducing volume upstream through issue resolution, reducing call times downstream by using artificial in- telligence that can address problems before a live agent is needed, or outsourcing to specialized vendors. “Between” interventions rely on payers, hospitals, and physician groups to collaborate by sharing unified communications (for example, a list of in-network physicians or clearer Explanation of Benefits) on integrated customer-facing platforms. These interventions could deliver $20 billion in annual savings.

CHAPTER 6

Customer and patient services

47Chapter 6: Customer and patient services

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

the digital divide, where the elderly or economically disadvantaged may not be able to access or use digital channels, raising questions about equity for all customers.

This customer service revolution is already taking place in industries such as telecommunications and retail. Certain leading healthcare organizations are also integrating many of these same new technologies and capabilities, as well as the same talent, into their customer and patient services. COVID-19 accelerated the transition to digital channels as pay- ers, hospitals, and phy sician groups were compelled to move away from face- to- face interactions and build remote com- munication capabilities, whether digital or via phone.3 Yet, in our experience, these kinds of innovations remain the exception in healthcare because of underinvestment; the decentralized and subscale nature of operations across payers, hospitals, and physician groups; and lack of prioritization.

Customer and patient services also have an impact on customer experience, which can lead to customer attraction and retention. We do not focus on that in this report but acknowledge these second-order effects.4

Definition and sizing Spending on customer and patient services across industries varies by the complexity of the business and the per- centage of call center volume that is out- sourced. From our research, businesses in complex and regulated industries such as healthcare, banking, and insurance spend 2 to 3 percent of total revenue on call centers, whereas businesses in less complex industries such as retail and tele- communications spend 1 percent or less of total revenue. This lower percentage is likely due to the simpler, more straightfor- ward questions that are being asked and answered, and the ability to effectively outsource a substantial portion of the call

Customer and patient services reflect activities and processes that let custom- ers ask questions and express concerns. These have largely depended on call centers, but there is increasing reliance on digital and self-service applications. In healthcare, the definition of “customer” is complex; for payers, services need to be provided to members and providers, and for providers, services need to be provided to patients.

The type of calls and touchpoints in the healthcare industry differ in complexity and duration. For example, most patients call providers with questions about basic logistics, such as appointment scheduling or whether they accept a new in surance coverage, or with simple medical questions about issues such as interpretation of lab results or medication side effects. Similarly, for payers, a large number of questions are related to tactical logistics such as benefits verification or billing. However, there are patients with complex needs that are more likely to require greater engagement, including speaking to a human being. So, payers and providers can be expected to spend a certain baseline amount on cus- tomer and patient services.

Across industries, customer service is undergoing disruption. Customer expec- tations are changing and reshaping what “good” means. Expectations are rising: first-contact resolution and knowledge- able representatives are now seen as the hallmark of good service.1 Customers familiar with the ease of transacting in other industries, such as in software or banking, are now seeking better service in healthcare. In addition, there are size- able shifts in consumer behavior as multi- ple kinds of service emerge: 50 percent of customers now use three to five channels or endpoints for service and information across industries.2 Finally, there is still value in live interactions despite the avail- ability of other channels. As healthcare organi zations try to respond to these market changes, they should also address

48 Chapter 6: Customer and patient services

McKinsey & Company McKinsey Center for US Health System Reform

groups (13 percent of total admini- strative spending), followed by private payers (9 percent), and hospitals (5 percent). For private payers, this includes spending on call centers that respond to members and providers, which are often staffed by call center represen tatives. For hospitals and physician groups, call center staff includes medical receptionists who help answer patient’s questions about their care (see chapter 2 for more detail on these estimates).

center volume. Organizations in more complex industries generally keep 20 to 40 percent of call center volume in-house.

Of the $80 billion spent on customer and patient services, our key stake- holder groups represented $55 billion— physician groups ($25 billion), private payers ($20 billion), and hospitals ($10 billion). Among these stakeholder groups, customer and patient services represented the greatest portion of administrative spending for physician

Exhibit 6.1

Pain points along current customer journey through payer's call center (illustrative)

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 6.1 of 9.2 (total 22 exhibits)

¹ Routing engines distinguish between customers (ie, providers versus members) entering process. ² CSR, customer service representative. CSR can be involved in other processes such as utilization management and case management. Source: McKinsey Payer Operations Domain

Pain points

Customers attempt to solve issues on their own and call customer service if help is required

Escalation to internal/

external specialist

Third-party interpreter

Private service centers

Converted into digital

form

Call

Billing, Explanation of Bene�ts,

etc.Outbound communi-

cation

Customer (member/ provider) outreach Written

Commercial plans

Government business

Government service centers

Issue resolved

Customer is routed to CSR and may also be connected to IVR and self-service channels

Customers reach general representatives who use available customer information to resolve issue (to the extent possible)

Issues are either esca- lated to a higher-level rep (for example, mana- gers) or resolved as is using documentation

Issue identi�cation Routing Issue handling Resolution

Increased interaction volumeA

A Walk-in centers

Digital (web

portal)

DA DB

Know- ledge data- base

C Docu-

mentation, reporting,

and analytics

C CSR²

DB Routing engine¹

Increased process complexityB Low productivityC

Process description

Interactive voice

recognition (IVR)

49Chapter 6: Customer and patient services

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

and spend substantial time gathering basic information (for example, billing history) rather than solving the actual problem.

— Low productivity: Poor layout, ineffi- cient vendor management, and poor agent productivity contribute to low levels of productivity within many call centers. Some of the problems stem from outdated technology, underused staffing, overspecialized agents, and an unoptimized mix of full- and part- time employees. This is especially true for smaller hospitals and physician groups with subscale operations.

Known interventions “Within” interventions (approximately 25 percent): We catalogued known inter- ventions that could save 25 percent of administrative spending on customer and patient services by using solely within interventions. These are interventions that can be controlled and implemented by individual organizations. For example, organizations could save via upstream reduction of volume through issue reso- lution or via downstream reduction of call times through artificial intelligence (AI):

— Reduce transaction volume through proactive issue resolution and inter- face improvements: Interventions to reduce the volume of calls or touch- points that are transactional (for ex- ample, questions that do not require human interaction, such as refreshing passwords or updating billing infor- mation) can take two forms: minimiz- ing the need for interaction by fixing the root cause and making the infor- mation readily available on digital channels. The former can be done by using “SWAT” teams that conduct root-cause ana lysis on calls to identify the most common issues and ensuring they get fixed. This might include simplification of confusing commu - nica tions on Ex planation of Benefits documents if they form a sizeable pro-

Key opportunity areas There is no typical customer service journey due to variations in the com- plexity of service needs (Exhibit 6.1). How ever, our experience shows that all journeys face similar challenges, generally driven by a lack of standard- ized data. This includes insufficient re- cording of customer his tory, incomplete linking of customer interactions across multiple calls and requests, missing documentation, and time-consuming manual work.

As a result, three main issues account for unnecessary administrative spend- ing in customer and patient services:

— Increased interaction volume: Call demand has grown as call centers become the frontline of customer and patient services. Call centers are now the preferred method for customers to receive personalized service, rather than going in person or not asking questions at all. Two other factors are increasing the volume of interactions: growth in the Individual insurance segment and a greater focus on member satisfaction metrics in Medicare Advantage.

— Increased process complexity: Responding to service requests has become more challenging. Organiza- tions have experienced call-routing management issues such as a large number of personalized matters not directed to the appropriate channel or staff. Also, they face service-time variability, including situations where customers with high needs are not immediately connected with the ap- propriate representative. A common example of a complex issue is patient scheduling. For example, rigidity in the system may force 15-minute ap- pointments into 30-minute slots. In another case, the frontline call center staff may have incomplete information about patients’ previous experiences

50 Chapter 6: Customer and patient services

McKinsey & Company McKinsey Center for US Health System Reform

technol ogy provides guidance to the representative based on the speed, volume, and tone of the conversation. For example, it might nudge the representative to speak more slowly if his or her pace is faster than best practice, or it might give an “empathy cue” if it thinks the customer does not feel heard. By deploying this tech- nology at scale, this European organi- zation reduced average handle time by about 14 percent, increased the rate of issue resolution by about 6 percent, and raised employee engage- ment by about 63 percent. Further, customers perceived the representa- tives as more confident and empa- thetic and felt that their issues were being resolved.

— Outsource to highly skilled vendors: This intervention is to outsource call center volume to vendors who special- ize in customer and patient services. However, outsourcing call centers should be evaluated within the context of customer and community prefer- ences (for example, if the existing call center was a local employer) as well as overall considerations of brand.

“Between” interventions (0 to 5 percent): We catalogued known interventions that could save up to 5 percent of administrative spending on customer and patient services by using solely between interventions. This type of intervention requires agree- ment and collaboration between organi- zations but not broader, industry- wide change. Although they are just emerging, these interventions aim to address infor- mation gaps between organizations in customer and patient services.

portion of incoming calls. The latter can be carried out by updating organi- zations’ websites and apps for easier usage, redirecting to self-service options during calls, and introducing tech nology such as predictive interac- tive voice response to resolve issues without an agent. In our experience, one US telecommunications organi- zation reduced transactional call vol- umes by 75 percent and admini stra tive spending by 20 percent by updating its website, making it easier to read bills, and improving user interface and experience functionality through more intuitive design.

In healthcare, research has shown that digitalization of scheduling led to an increase in care continuity between patients and their primary care physicians.5 However, this re- search also found greater adoption by younger, White, commercially insured patients. So, digitalization of this kind could have the unintended effect of widening the digital divide and worsening socioeconomic dis- parities in primary care access if not managed appropriately.

— Improve handle time and issue re solution via AI: This intervention is focused on using AI-based technolo- gies to provide real- time coaching to call center agents. In our experience, a European healthcare organization used a technology that gives real-time guidance to customer representatives based on speech patterns and past observations; it improved customer and patient services and turnaround times. By listening in real time, the

Research has shown that digitalization of scheduling led to an increase in care continuity between patients and their primary care physicians.

51Chapter 6: Customer and patient services

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

platforms to refer to a digital list of in- network specialists and advise patients on whom they should see. Similarly, members could log onto this platform to understand exactly what their out-of-pocket costs might be for a service before receiving it. Organi- zations are beginning to see success with this type of intervention. For example, a regional payer and large health system are collaborating on a platform that streamlines customer communications to provide infor ma- tion such as the network status of physicians or expected out-of-pocket fees. By doing this, they expect to reduce administrative spending while improving customer satisfaction.

— Build strategic payer-provider plat- forms to reduce demand: Patients in the United States generally have to interact with two different types of entities—payers and providers—who often have different information about their care. Common examples include patients not knowing if the specialist their primary care physician refers them to is in network and see- ing high out-of-pocket costs when payers decline to cover a service that their physician had recommended. A payer- provider platform can relieve much of the stress that derives from the amount of time it can take before a patient receives an answer. For ex ample, providers can use such

1 State of Customer Care Survey, 2019. 2 State of Customer Care Survey, 2019. 3 Oleg Bestsennyy, Greg Gilbert, Alex Harris, and Jennifer Rost, “Telehealth: A quarter-trillion-dollar post-COVID-19 reality?,” July 9, 2021,

McKinsey.com. 4 Jenny Cordina, Dan Jamieson, Rohit Kumar, and Monisha Machado-Pereira, “Improving acquisition and retention in Medicare,” March 2016,

McKinsey.com. 5 Ishani Ganguli et al., “Patient and visit characteristics associated with use of direct scheduling in primary care practices,” JAMA Network Open,

August 2020, Volume 3, Number 8, jamanetwork.com.

52 Chapter 6: Customer and patient services

McKinsey & Company McKinsey Center for US Health System Reform

Administrative clinical support functions represent activities that can be customer-facing, including members and patients, and require some clinical expertise, but that are not related to hands-on care. Relevant functions include nursing administration (for example, staffing and scheduling of nurses) and case and disease management (for example, patient admitting and discharge). They account for $105 billion in annual administrative spending (or 11 percent of total administrative spending). Salient pain points include the manual nature of tasks for nurse managers and case managers, such as scheduling nurse shifts or non- automated outreach to patients, and disconnected tools that govern major tasks, such as separate tools for scheduling and overtime management. “Within” interventions such as automation of those manual tasks and building application programming interfaces (APIs) to share data across tools could deliver $15 billion in annual savings.

CHAPTER 7

Administrative clinical support functions

53Chapter 7: Administrative clinical support functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

follow- ups. Case management is also seen as a critical component of revenue cycle management. It provides coordination and communication on ad mitting and discharge planning to ensure patients have the right care within and outside of provider settings. Other parts of revenue cycle management, including claims management and utilization management, are covered in chapter 3.

7.1 NURSING ADMINISTRATION

Definition and sizing Nursing administration encompasses nurse managers (and select other roles) that are fully dedicated to an admini strative role in both inpa- tient and out patient settings. According to the Journal of Nursing Management, “the expecta- tion is that nurse managers are adept at fi nancial management, negotiation, staff recruitment and development, conflict resolution, technologic advancements, and leadership.”1

Beyond nursing administration, hospitals and physician groups also hire nurses as case man- agers (chapter 7.2) or clinically- focused nurses (not covered in this report). The latter includes front line nurses who spend most of their time on direct patient care and some portion on admini- strative activities (Exhibit 7.2).

Of the $35 billion spent on nursing administra- tion, our key stakeholder groups represented $25 billion—hospitals ($20 billion) and physician groups ($5 billion). Private payers also hire nurs- es, but these are often covered under customer and patient ser vices (chapter 6) or other adminis- trative clinical support functions such as case and disease management ( chapter 7.2).

In our experience, hospitals spend about 30 percent of revenue on nurses, of which 5 to 7 percent is spent on nursing administration. In our analysis, the $20 billion hospitals spend on nursing administration would amount to 1.5 percent of total hospital revenue in the United States.2 Physician groups rely less on nursing administration. They often call staffers who per- form administration duties office managers or office admini strators as they may wear multiple

Administrative clinical support functions represent activities that can be customer- facing and require some clinical expertise but are not related to the hands- on care of patients. His torically, payers and providers have not sought savings from this functional focus area due to the clinical association of the work, but they are increasingly seeing oppor tunities for effici encies without com- promising access or quality.

Of the $105 billion spent on administrative clinical support functions, our key stakeholder groups represented $70 billion—hospitals ($40 billion), private payers ($20 billion), and physician groups ($10 billion). Within these stakeholder groups, these functions repre- sented the greatest portion of spending for hos pitals (16 percent of total administrative spending), followed by private payers (10 percent) and physician groups (5 percent).

Within this functional focus area, we identi- fied two specific subareas of administrative spending that predominantly involve nurses (Exhibit 7.1):

— Nursing administration: This is administrative spending on nursing management resources that are 100 percent dedicated to admini- strative activities such as staffing, float pool management, house supervision, patient placement, and transfers. Other elements of clinical administration, including manage- ment of staff such as nursing tech nicians, pharmacists, respiratory therapists, and dieticians, are not covered in this report be- cause they often fall under clinical nursing.

— Case and disease management: This is administrative spending on resources to support coordination for high- need patients and members. For payers, case and disease management encompasses activities such as supporting these members as they navigate their care journey (for example, coor dinating housing support and helping them find pro- viders in their neighborhoods) and managing high- complexity diseases such as diabetes and congestive heart failure. For providers, aspects of disease management may be em bedded in a range of care activities and

54 Chapter 7: Administrative clinical support functions

McKinsey & Company McKinsey Center for US Health System Reform

vacation schedule, and availability of part- time versus full- time nurses. Staffing be- comes especially burdensome given the vola tility in patient flows, which can cause ebbs and flows in the supply and demand equation between patients and nurses.

— Lack of integration across daily- use systems and tools: Nurse managers often deal with an array of tools and systems to make decisions about staffing, budgeting, resource allocation, and quality control. Without a common set of data, definitions, and insights in one place, they have to manually connect information from one system to another. For example, they may manually note information from self- scheduling systems that identify shift availability for nurses and use that to as- sign nurses to patients on another portal.

hats. We found $5 billion of administrative spending in physician groups, which is slightly less than 1 percent of total annual revenue (see chapter 2 for more detail on these estimates).

Key opportunity areas Based on our experience, there are two key opportunity areas in nursing admini stration:

— Manual nature of key tasks for nurse man- agers: Most activities that nurse managers are responsible for, from patient flow man- agement and budgeting to communications and nurse staffing, remain manual and labor- intensive. For example, with staffing, nurse managers have to manage an intricate web of constraints to build schedules for each day, taking into account patient acuity, staffing ratios, each nurse’s travel and

Exhibit 7.1

Four components of care management

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 7.1 of 9.2 (total 22 exhibits)

Source: McKinsey Payer Operations Domain

Example activities Component of care management Providers

For more information, see

Managing referral requirements; work- ing with care teams to ensure appro- priateness of care being delivered

Payers

Running 24-hour nurse hotline; managing prior authorization

Utilization management • Evaluation of the appropriateness, medical necessity, and e�ciency of healthcare services and procedures according to established criteria under the provisions of a payer’s program • Proactive processes such as discharge planning, concurrent planning, precerti�cation, and clinical case appeals

Chapter 3.2: Prior authorization

Developing care plans and coordi- nating discharge planning

Conducting outreaches and screening patients; coordinating social supports (for example, housing)

Case management • Method of managing the provision of healthcare to members and patients with high-cost medical conditions • Goal is to coordinate the care to improve continuity, increase quality of care, and lower spending

Chapter 7.2: Case and disease management

Partially covered by case management and care plans

Sending quarterly mailings to patients identi�ed with diabetes

Disease management • Process of reducing healthcare spending and improving quality of life for individuals by preventing or minimizing the e�ects of a disease, usually a chronic condition, through a system of coordinated healthcare interventions

Chapter 7.2: Case and disease management

N/ARewards for members hitting goals

Not reviewed in this report Wellness • Resources and tools that help individuals understand their overall health status and take an active role in their personal health

N/A

55Chapter 7: Administrative clinical support functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

implemented by individual organizations. In previous chapters, we identified interventions that could result in the need for fewer resources. That may not be the case here. While these known interventions could free up nurse manag- ers’ capacity so they can better focus on their tasks and support their staff through training and mentorship, they may not necessarily reduce the overall number of nurses an organization needs. This is due to both operational constraints (for example, staffing cycles) and regu latory or work- rule constraints such as California’s Title 22 stipulations on nurse- to- patient ratios. However, the intan gible benefits of the following interven- tions that reduce the manual work burden and boost produc tivity through technology may lead to financial savings within a few years through enhanced retention and nurse satisfaction:

— Digitalize manual administrative activities: Nurse managers spend much of their time on manual activities dealing with patient flows, staffing, and communication. There are prov- en uses of technology to add efficiencies in

Although these challenges are not new, a number of barriers have prevented the uptake of known interventions. Hospitals and physician groups that lack substantial scale generally find the cost required to implement new technologies prohi- bitive, further amplified by the long- term nature of realizing the savings. Also, nurses are often trained in a specific unit that has its own stand- ards and procedures for using IT systems. Fur- ther, any administrative interventions that could have second- order impact on front line care givers tend to be pursued with careful consideration for the caregiver’s bandwidth and potential disrup- tion. Finally, certain activities, including critical review, have traditionally been kept manual given their heightened importance in patient care.

Known interventions “Within” interventions (5 percent): We cat- alogued known interventions that could save about 5 percent on nursing administration by using solely within interventions. These are interventions that can be controlled and

Exhibit 7.2

Breakdown of time spent by nursing sta� across administrative and non-administrative activities in hospitals

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 7.2 of 9.2 (total 22 exhibits)

¹ Case management sta� at hospitals are comprised of a 3:1 ratio between nursing sta� and non-nursing sta� (for example, clinical coordinators, discharge planners, and unit secretaries).

² Spending on nurses equals ~30% of total hospital revenue, which was estimated to be ~$1.2T in 2019 by National Health Expenditure Accounts of the Centers for Medicare & Medicaid Services.

Source: Centers for Medicare & Medicaid Services; McKinsey analysis

Total spending, $ billion (percent of total nursing sta�)

Percent of time spent on admin- istrative tasks

Total resulting administrative spending, $ billionTypes of nurses

15–25100%15–25 (5%–10%)

Nursing administration: Nurse managers that focus on tasks such as sta�ng, �oat pool management, house supervision, and patient placement and transfers; not responsible for direct hands-on care of patients Covered in chapter 7.1

15–2020%–30%60–70 (15%–20%)

Case managers¹: Nurses that manage coordination of high- need patients and communication tasks such as discharge planning but also have direct patient care responsibilities Covered in chapter 7.2

~30~10%260–270 (~75%)

Clinical nursing: Frontline nurses who spend majority of their time on direct patient care but may have some administrative responsibilities Not covered in report

60–75345–365 (100%)

Total spending on nurses²

56 Chapter 7: Administrative clinical support functions

McKinsey & Company McKinsey Center for US Health System Reform

Of the $70 billion spent on case and disease management, our key stakeholder groups rep- resented $45 billion—hospitals ($20 billion), private payers ($20 billion), and physician groups ($5 billion). Hospitals and physician groups are pre dominantly focused on case management. They typically employ a 3:1 ratio of nurses to non- nursing staff, which includes clinical coordi- nators, discharge planners, and unit secretaries. This staff can be thought of as essential to reve- nue cycle management, as its activities intersect with clinical care and revenue cycle management teams (for example, discharge planning for ap- propriate length of stay, readmission prevention, and care plans for med ical necessity). For private payers, this group includes nurses, social workers, and administrative staff. Their activities comprise conducting outreach and screening patients, assisting in coordi nation of social supports such as trans portation and meals- on- wheels, and providing documentation support (see chapter 2 for more detail on these estimates).

Key opportunity areas Payers, hospitals, and physician groups face similar challenges when it comes to administra- tive spending in case and disease management:

— Manual, repetitive work: Case and disease management are functions that have the most touchpoints with members (for payers) and patients (for providers). Given the “human” nature of the activities, there are many tasks that are done manually. How ever, many of these are often rules- based, repetitive work that could be automated. For example, payers and providers routinely reach out to members and patients for mundane items such as scheduling or public health offerings (for example, flu shots).

— Disconnected tools and systems: Over the last two decades, organi zations have created many point solutions to support case and disease management for payers and provid- ers.4 Such providers could find themselves managing a myriad of vendors within the realm of case management, from vendors focusing on particular aspects of population health management to those that support specific parts of revenue cycle management.

these areas. For example, self- scheduling capabilities or predictive analytics can help better manage patient flow through optimal discharge timing or capa city planning.3 One caveat is that when these technologies are introduced, change management issues should also be considered in order for the interventions to take hold. For example, the new technologies should ideally fit within pre-existing workflows so nurse managers can naturally take them up in the course of business.

— Enable management of larger spans: Typi- cally, nurse managers have large spans of control—the number of people that nurse managers look after—because the full- time equivalents (FTE) are generally homogeneous in terms of skills and expertise. Nonetheless, workforce management tools (for example, float pool visuali zations and overtime pre- vention software) can help nurse managers in two ways. First, they could allow nurse managers to be more effective and efficient in their day- to- day tasks, permitting them to spend more time coaching and developing their teams. Second, they could make it easier to manage a large span. For example, work- force management tools can flag when indi- viduals on their team have worked overtime. This might prompt a nurse manager to check in with indi viduals to assess burnout.

“Between” interventions (not appli cable): Given the internal nature of nursing administra- tion, the majority of savings are within organiza- tions, not between different stakeholder groups.

7.2 CASE AND DISEASE MANAGEMENT

Definition and sizing Care management generally includes four components (Exhibit 7.1)—utilization manage- ment, case management, disease management, and wellness. In this report, we focus on case and disease management, which is defined as the coordination of support for high- need patients and members.

57Chapter 7: Administrative clinical support functions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

for providers could be a portal that, possibly through electronic health records (EHRs), commu nicates care plans to patients and their care givers, potentially reducing the need for patients to call or make in- person visits.

— Improve operational discipline and ensure “top of license” practices: Introducing stand- ard operating pro cedures and metrics such as staffing ratios, case length expectations, and case graduation criteria can enhance the performance of case management re- sources. In addition, in stituting a rig orous performance management culture could lead to better alignment on pri orities and more effective use of time. Some organiza- tions employ dashboards that mea sure key performance indicators (for example, case length, case graduation rates) and weekly performance reviews.

“Between” interventions (not appli cable): In the fee-for-service payment model, while case and disease management functions do require coordination across payers and pro- viders (for example, payers may send their case managers with patients to monitor phy sician visits), there is less emphasis on joint ex ecution. However, collaboration is in- creasing with value-based payment models as payers are delegating case and disease management responsibilities to providers. Further, in our experience, some payers are supporting providers to build the necessary capabilities such as managing social and medical support for a complex patient. Yet, given the low penetration of value-based payment models today, we did not focus on these between interventions in this report.

Known interventions “Within” interventions (20 percent): We catalogued known interventions that could save 20 percent of administrative spending on case and disease management by using solely within interventions. These are interven tions that could be control led and implemented by in di vidual orga ni zations and that may require changing pre-existing, in- organization work- flows such as scheduling or admitting through digitalization, technology upgrades, and per- formance management:

— Digitalize and automate processes: There are several steps that could be automated in the case and disease management work- flow at both payers and providers. Reducing time to look up and confirm member or patient information is one example. Payers could automate tasks such as outbound dialing, case prioritization, and assignment of members to case managers. For provid- ers, ex amples include admitting patients and registering outpatient visits.5

— Integrate suite of tools and solutions: Improving the usability of tools and making them interoperable can reduce the time taken to switch from system to system. In our ex perience, one Medicaid- focused payer implemented an enterprise- wide care man - agement package that captured member engagement activities in one place and inter- grated that information with other systems such as utilization management software. The change led to reduced administrative spending through saved time and enabled the payer to offer personalized and real- time communications to its members. An example

1 Eloise Balasco Cathcart, Miriam Greenspan, and Matthew Quin, “The making of a nurse manager: The role of experiential learning in leadership development,” Journal of Nursing Management, May 2010, Volume 18, Number 4, pp. 440-7, doi:10.1111/j.1365-2834.2010.01082.x.

2 National health expenditure data, Centers for Medicare & Medicaid Services, accessed September 17, 2021, cms.gov. 3 Lisa Massarweh, “Hospital staffing technology: Hazard and opportunity risks,” Nursing Management (Springhouse), 2018, Volume 49,

Number 11, pp. 48-53, lww.com. 4 Shubham Singhal, Basel Kayyali, Rob Levin, and Zachary Greenberg, “The next wave of healthcare innovation: The evolution of ecosystems,”

June 23, 2020, McKinsey.com. 5 As this intervention evolves, there is likely an opportunity for a between intervention where payers and providers jointly agree to digitalize these

processes. We have not seen this in practice at scale and so have not documented it in this report, but we acknowledge this could become another intervention in the future.

58 Chapter 7: Administrative clinical support functions

McKinsey & Company McKinsey Center for US Health System Reform

“Seismic” interventions are different from “within” or “between” inter ventions in that they cannot be achieved by an organization operating on its own or with smallscale collaboration between payers, hospitals, and physician groups. In this chapter, we draw inspiration from other industries to identify a noncomprehensive list of seismic interventions that could accelerate administrative simplification in US healthcare. They take the form of technology platforms (for example, adopting a centralized, automated claims clearinghouse), operational alignment (standardizing medical policies), or payment design (modularizing product design). These interventions address spending in most of the functional focus areas and could deliver about $105 billion in annual savings. Much of that total comes from simplifying the financial transactions ecosystem or streamlining duplicative healthcare processes such as statespecific clinician credentialing and payerspecific medical policy customization.

CHAPTER 8

Seismic interventions

59Chapter 8: Seismic interventions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

Technology platforms Adopt a centralized, automated claims clearinghouse Analogous to full adoption of a banking payment clearinghouse The US healthcare system does not have a sole, centralized clearinghouse through which it manages claims payments. Payers instead may use a number of clearing houses. A provider may send a claim to its contracted clearinghouse, which would then pass the claim to the payer’s clear inghouse. Multiple intermediaries could add complexity and cost to claims proces sing and increase the likelihood of delays and errors that could lead to customer service issues and business disruption.

We see the potential to incentivize more payers and providers to consider avenues such as the Automated Clearing House (ACH) network or processing networks for credit card payments. Focusing on the for mer, this national payment system connects all US bank accounts and facilitates the move ment of money and information; it has already been adopted for use in healthcare payments. Over the past few years, the National Automated Clearing House Asso ciation (NACHA) has sponsored efforts to increase the use of the ACH for electronic funds transfer (also called ACH/EFT) in healthcare. As of 2020, about 74 percent of medical claims were paid through the ACH/EFT, up from 63 percent in 2018.3 By comparison, only 13 percent of dental claims were paid through the ACH/EFT in 2020, leaving the remaining 87 percent to be paid manually through paper checks.

There could be a meaningful return on investment when claims are paid through a system like the ACH. For example, trans ac tion costs for payments through

In previous chapters, we reviewed the opportunity by functional focus area for “within” interventions that indi vidual organizations can control and implement and “between” in terventions that require agreement and collaboration between organiza tions, but we did not examine broader, industry wide change.

This chapter draws inspiration from other industries to examine a third type of intervention that we call “seismic.” These interventions are considered seismic because they cannot be achiev ed by an organization operating on its own or with smallscale collaboration between payers and providers. To understand the potential of these ap proaches for US administrative spend ing, we identified a few examples based on analogs. These interventions are not meant to be a comprehensive list nor offer a pointofview on what is best, but they are intended to show what may be possible in US healthcare.

The chapter explores three broad themes: technology platforms (for exam ple, adopting a centralized, automated claims clearinghouse), operational align ment (for example, standardizing medi cal policies), and payment design (for example, modularizing product design). Our research yielded seven seismic interventions that meet our “known” definition, which means they could be carried out in the next three years if there was agreement and collaboration from all necessary stakeholder groups (Exhibit 8.1). We estimated that these interventions could generate approxi mately $105 billion of savings, or 11 per cent of current administrative spending.1,2

There could be a meaningful return on investment when claims are paid through a system like the ACH. 60 Chapter 8: Seismic interventions

McKinsey & Company McKinsey Center for US Health System Reform

as ex perience in other countries and industries shows.7 Further, grounding a centralized platform with the proper supporting infrastructure, such as data integrity (for example, a single source of truth provider directory) and technology (for example, application programming interfaces with provider portals or claims submission and tracking capa bilities), could enhance its usability and potenti ally increase adoption.

an ACH compared to manual payments are seven times lower for private payers and three times lower for pro viders.4,5 Additionally, ACH fraud rates are the lowest across all payment types.6 Cre ating incentives—both bonuses and penalties—so that all healthcare claims payments go through a governing technology such as the ACH has the potential to materially reduce admini strative spending on claims processing,

Exhibit 8.1

Seismic interventions and associated analogs

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 8.1 of 9.2 (total 22 exhibits)

Example seismic interventions Analog Functional focus area impacted

Adopt a centralized, automated claims clearinghouse: Set up an automated clearinghouse that systematizes payment between payers and providers

Banking payments clearinghouse

Technology platform

Source: McKinsey analysis

Financial transactions ecosystem

Standardize medical policies: Apply uniform medical policies across all payers

Standardizing P&C insurance

Operational alignment

Payment design

Financial transactions ecosystem

Prioritize high-value interoperability use cases: Align stakeholders on high-value interoperability use cases and organize investments to scale necessary technology

Universal Product Codes in retail Sharing platforms for �nancial data

Customer and patient services Administrative clinical support functions

Standardize physician licensure: Eliminate state-speci�c requirements for physicians and allow all who meet federal requirements to practice across state lines

Standardized aviation licenses in the United States

Industry-speci�c operational functions

Streamline quality reporting: Reduce number of quality measures to the highest-value subset and provide support for automated, digital methods of capturing data

Car safety measures and FICO scores

Industry-speci�c operational functions

Adopt globally capitated models for segments of the care delivery system: Eliminate service-based payments from payers to providers by shifting all payments to a “capped” or �xed amount per patient

Evolution from cost- per-impression to pay-per-click for online advertising

Financial transactions ecosystem Customer and patient services

Modularized product design: Shift toward a modularized set of bene�ts packages for customers

Modularized o�erings in enterprise software

Industry-speci�c operational functions Customer and patient services

61Chapter 8: Seismic interventions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

To fulfill their promise, alignment may be beneficial on the most valuable use cases (for example, aggregated patient health records) across stakeholder groups. Once this prioritization is agreed upon, the exist ing interoperability approaches mention ed above could be reevaluated to ensure they are in support of those use cases.

To launch each prioritized use case, a trusted party, such as a wellregarded nonprofit or a public–private consortium, could orchestrate and be seen as the lead er of the effort. This could help patients feel comfortable adopting the innovation. Next, the ap propriate rules to govern these use cases may need to be defined (for example, who has access to a patient record and who is responsible to update this record when new data are available). After this, the right technical partner can be chosen to build the enabl ing infra structure (for example, a universal patient identifier and preference engines that record patients’ choices on who can ac cess their record). This structured pro cess could help ensure that highvalue use cases are built with the right buyin up front and an enabling technology foundation to ensure full value capture of the administrative spending savings.

Operational alignment Standardize physician licensure Analogous to standardization of aviation licensure in the United States that allows pilots to fly across state lines The federated nature of US physician licensure has resulted in a complex set of statespecific rules and requirements on where physicians can practice. Each state and territory has its own board for medi cal and osteopathic licensing, each of which has unique and specific require ments for physician licensure in that jurisdiction.11 For example, while all states require physicians to show proof of grad uation from an accredited medical school, others also ask for specialtyspecific tests and coursework. This structure not

Prioritize highvalue interoperability use cases Analogous to prioritization of a Unified Product Code in retail, which became a foundational use case from which other uses cases, such as tracking of products, were launched; another example is build ing sharing platforms for financial data that act as a consumer’s single source of truth for specific use cases, such as credit cards, bank accounts, and loans It is difficult to build a longitudinal view of a patient’s health record over his or her life due to the fragmented nature of how patient data are stored and shared in the US health care system. Investing in interoperability that is supported by an aggregated, real time, and patientcentered database, while preserving privacy and patient confidenti ality, could reshape how healthcare is organized, delivered, and managed in the United States. Through an admini strative lens, interoperability could lead to reduced spending in several functional focus areas. One example of this includes cutting time spent on medical records matching in prior authorization (PA) by making medical re cords easily accessible and readable; an other example is minimizing the need for call center touchpoints to answer customer queries on claims status by allowing mem bers to have those data at their fingertips.

Interoperability is not a new idea, and there are numerous ongoing efforts on this front. These include: the Health Information Exchange of the Centers for Medicare & Medicaid Services (HIE CMS), an interop erability framework that offers a repository of patient information accessible to relevant providers; the Trusted Exchange Frame work and Common Agreement (TEFCA) to enable the nationwide exchange of electronic healthcare informa tion; the US Core Data for Interoperability (USCDI) effort that is aligning data definitions nationally; and continuing efforts to build a universal patient identifier.8-10 How ever, these technologybased initiatives have often been launched without clarifying the use cases they are meant to address.

62 Chapter 8: Seismic interventions

McKinsey & Company McKinsey Center for US Health System Reform

multiple credentialing processes for the same physician. Further, such standardi zation could reduce constraints for start ups that may provide lowercost vended services to payers and providers.

Standardize medical policies Analogous to standardization of property and casualty insurance through consis tency of offerings—for example, the minimum amount that must be insured before a mortgage is allowed or stan dard ized faultdetermination methodology Medical policies are guidelines written by payers to help providers determine if cer tain medical services are covered. These are generally published in provider facing portals, with the intent of outlining how payers assess the medical necessity and appropriateness of a service. For example, a payer may require a step therapy where a patient must have received a generic version of a drug to lower cholesterol before a more expensive one can be pre scribed. Complexity has grown over the last century, starting in the 1930s when the first set of medical policies were writ ten.19 While the foundational require ments for medical policies are governed by states, each payer refines its own poli cies and continually refreshes them as medical research surfaces new treat ments and insights. Commercial payers, which cover more than 175 million Ameri cans, are not standardized, and thus, keeping pace with them may put substan tial burden on hospitals and physician groups.20,21 (By comparison, Medicare has pursued national standardization for the diagnosis or treatment of an illness or injury.) While any standardization must be undertaken in compliance with competi tion laws, a range of potential approaches

only may create complications for physi cians seeking to practice in other states (precluding costeffective delivery models such as telemedicine), but it may also in crease overall administrative spending at the national level. Standardizing creden tialing has historically faced challenges due to concerns about lower quality care provision and deemphasizing localized standards of care.12 Also, even with standardization, physicians must still adhere to state specific regulations, such as malpractice requirements.13

Proponents of standardization say that physicians already must meet criteria to become eligible to practice in any state— for example, graduating from an American Medical Association–accredited medical school, passing a comprehensive national medical licensing examination sponsored by the Federation of State Medical Boards (FSMB) and the National Board of Medical Examiners (NBME), and meeting stand ards for work history and health status.14,15 In addition, there are already examples of standardized credentialing in the United States. The Nursing Licensure Compact (NLC), agreed to by almost 35 states as of March 2021, allows nurses to have multi state licenses.16 Further, the Interstate Medical Licensing Compact or “Telemedi cine Compact” (IMLCC) facilitates a legal agreement among 29 states to allow phy sicians licensed in one state to provide services through telemedicine in other states.17,18 Thus, administrative spending savings may result from lower credential ing spending for payers (currently $15 billion annually). For providers, standard ized physician licensing would potentially make crossstate care provision simpler, reducing administrative spending on the

The Nursing Licensure Compact, agreed to by almost 35 states as of March 2021, allows nurses to have multi-state licenses.

63Chapter 8: Seismic interventions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

moderni zation (for example, digitizing sources of data).25 Further, physicians now have the flexibility to choose which measures they report: MIPS, for example, lets providers select six measures from a longer list.26 Even with these changes, quality reporting could remain burdensome for pro viders, especially since the shift toward valuebased payment models gen erally requires greater measurement of physician performance. Anticipating this shift, programs such as the Core Quality Measurement Collaborative (CQMC), a multi stakeholder group effort working to develop quality measure sets by specialty and condition, are already prioritizing mea sures related to valuebased payment models.27 Overall, these examples high light how administrative spending could be reduced through largerscale collabo ration between public and private payers on a short list of quality measures and standardized requirements to better enable digital collection of necessary data. For example, the CMS Star ratings program, which targets payers, is a helpful analog for streamlined quality measures within healthcare. It has narrowed a list down to roughly 40 measures that provide a standard method to evaluate the quality of a payer’s Medicare Advantage product.28

Payment design Modularize product design Analogous to tiering of enterprise software packages—for example, modules for financial asset management, reporting, customer service, and business intelligence One feature of the US healthcare system is the substantial variation in insurance product offerings. This variation is often viewed by employers (a payer’s customers) as a competitive advantage in recruiting employees as well as meeting the needs of their workforce. However, not all custo mers place the same value on this choice: research shows that nearly twothirds of midsize employers (those with 50 to 500 employees) would be willing to switch

could be considered, such as standard izing data requirements for PA.

Overall, there may be two scenarios in which hospitals and physician groups could benefit administratively from this intervention. First, they might spend less on PAs, as less nurse and physician time would likely be used in reviewing payer specific medical policies. Second, they could spend less on claims, as there may be fewer followon denials and appeals.

Streamline quality reporting Analogous to standardization of car safety metrics into a fivestar system by the National Highway Traffic Safety Administration; another example is the creation of FICO scores as one highvalue, standardized measure of credit risk Over the past few decades, measuring the quality of care has risen in importance and become a core part of the “triple aim,” a framework developed by the Institute for Healthcare Improvement.22 The Centers for Medicare & Medicaid Services (CMS) originally created programs to assess quality by implementing payment and reporting incentives. These include both payer facing measures (for example, CMS Star ratings) and provider facing measures (for example, the Merit based Incentive Payment System, or MIPS). Today, CMS requires reporting on more than 1,700 quality measures.23 Research has shown that physicians spend 2.6 hours per week— the equivalent of caring for nine patients— reporting on quality measures; staff other than physicians spend 12.5 hours per physician per week on the same tasks, with the largest proportion (6.6 hours) by licensed practical nurses and medical assistants.24 In an effort to simplify qual ity reporting, CMS introduced two different programs: in 2017, “Meaningful Measures” reduced the number of quality measures by 18 percent, and in 2020, “Meaningful Measures 2.0: Moving from Measure Reduction to Modernization,” continued the effort to require only the highestvalue measures and to accelerate

64 Chapter 8: Seismic interventions

McKinsey & Company McKinsey Center for US Health System Reform

models is to pay providers based on value versus on a perservice basis, as well as to ensure payers and providers are collabo ratively taking on risk to support patient outcomes. When in these models, providers are more likely to implement certain initia tives that simplify administration, such as triage call centers (55 percent of providers who participate in these payment models versus 31 percent of providers who remain in feeforservice payment models) and remote patient monitoring (49 percent versus 30 percent).31 However, adopting globally capitated payment models will likely not automatically reduce admini stra tive spending. For example, in our experi ence, providers such as integrated delivery networks using these payment models have captured none or as little as about a third of expected administrative savings. Some reasons for this result may include health care regulations that require additional administrative spending in these models (for example, risk coding in Medicare Advan tage) or the transfer of certain necessary administrative processes to the provider, such as PA for new standards of care.

While the administrative spending savings may be limited, we acknowledge that the pri mary financial benefit for globally capi tat ed payment models is medical cost savings (not in the scope of this report). Still, if some of the interventions discus sed in the previous chapters, such as AIenabled PA, were implemented, the potential for administrative spending savings could increase. Further, though potentially limit ed, the administrative spending savings could fund the provider's transition to these payment models.

Finally, of the seismic interventions in this chapter, this is the least likely to gain traction in the near term. As of June 2020, less than three percent of payments were fully capi tated.32 In addition, smaller providers may lack the ability or resources to invest in the capabilities that capitation might require, such as population health management tools, which could further slow adoption.

payers for a premium reduction of 10 per cent or less.29 This raises the question of how much product variation is truly need ed and if there is room to preserve choice while simplifying the administration that supports this variation.

From an administrative perspective, there are three potential benefits of modularized product design:

1. Enterprise benefits that simplify the suite of contracts between payers and providers, thereby reducing the man ual entry of the customized product design agreed upon with each customer

2. Ability to deploy digital marketplaces that permit rapid updates in pricing, given the limited option set (compared to the current process of rerunning complex models to provide the custom er with an updated price)

3. More intuitive customer engagement tools that nudge customers to select the best option at the moment of the decision given simplified underlying rules engines

A seismic intervention (for example, an ag gregator site that publishes every payer’s products and rates online, such as Kayak) that incentivizes the modularization of product design could deliver substantial administrative spending savings.

Adopt globally capitated payment models for segments of the care delivery system Analogous to moving from cost perimpression to payperclick for online advertising The US healthcare system largely employs a feeforservice payment model. A globally capitated model for certain segments (for example, populations like Medicare or ill nesses like endstage renal disease) could eliminate servicebased payments from payers to providers. Instead, providers may receive a capped or fixed amount per patient for delivering a predetermined set of healthcare services, such as $500 per patient per month.30 The goal of these

65Chapter 8: Seismic interventions

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

1 We acknowledge that these partnerships should be structured and carried out consistently with all applicable laws and regulations governing competition.

2 There is approximately a $50 billion overlap in savings between seismic and within/between interventions identified in earlier chapters. 3 CAQH Index, “2020 CAQH Index: Closing the gap: The industry continues to improve, but opportunities for automation remain,” Council for

Affordable Quality Healthcare, 2021, caqh.org. 4 Transaction costs for payment through ACHs are $0.08 per claim for private payers, compared to $0.57 for manual payments; similarly,

processing payments through ACH costs $1.19 for providers compared to $3.18 for manual payments. 5 CAQH Index, “2020 CAQH Index: Closing the gap: The industry continues to improve, but opportunities for automation remain,” Council for

Affordable Quality Healthcare, 2021, caqh.org. 6 ACH Fraud Report, “As payment card security rises, ACH fraud grows,” Marqeta, 2020, marqeta.com. 7 CAQH Index, “2020 CAQH Index: Closing the gap: The industry continues to improve, but opportunities for automation remain,” Council for

Affordable Quality Healthcare, 2021, caqh.org. 8 “Health Information Exchange,” Centers for Medicare & Medicaid Services, accessed October 15, 2021, medicaid.gov. 9 “Trusted Exchange Framework and Common Agreement,” Office of the National Coordinator for Health Information Technology, last reviewed

October 7, 2021, healthit.gov. 10 “United States Core Data for Interoperability (USCDI),” Office of the National Coordinator for Health Information Technology, accessed

October 15, 2021, healthit.gov. 11 Aaron Young et al., “FSMB census of licensed physicians in the United States, 2018,” Journal of Medical Regulation, 2019, Volume 105, Number

2, pp. 7–23, fsmb.org. 12 Donnie Bell and Mitchell Katz, “Modernize medical licensing, and credentialing, too—lessons from the COVID-19 pandemic,” JAMA Internal

Medicine, January 2021, Volume 181, Number 3, pp. 312–5, jamanetwork.com. 13 Health Affairs Blog, “Doctors without state borders: Practicing across state lines,” blog entry by Bob Kocher, February 18, 2014, healthaffairs.org. 14 “About physician licensure: How physicians gain licenses to practice medicine,” Federation of State Medical Boards, December 11, 2020, fsmb.org. 15 “Who is USMLE?,” United States Medical Licensing Examination, 2021, usmle.org. 16 Kathleen Gaines, “What are nursing compact states?,” Full Beaker, July 7, 2021, nurse.org. 17 “A faster pathway to physician licensure,” Interstate Medical Licensure Compact, 2021, imlcc.org. 18 “About physician licensure: How physicians gain licenses to practice medicine,” Federation of State Medical Boards, December 11, 2020, fsmb.org. 19 George Moseley III, “The U.S. health care nonsystem, 1908-2008,” Virtual Mentor, 2008, Volume 10, Number 5, pp. 324–31, journalofethics.

amaassn.org. 20 We defined commercial payer as those with employerbased coverage or nongroup coverage. 21 Health Insurance Coverage of the Total Population, Kaiser Family Foundation, 2019, kff.org. 22 The triple aim is defined as improving care, improving population health, and reducing costs per capita; see “Triple aim for populations,”

Institute for Healthcare Improvement, 2021, ihi.org. 23 Gail Wilensky, “The need to simplify measuring quality in health care,” Journal of the American Medical Association, June 19, 2018, Volume 319,

Number 23, pp. 2369–70, jamanetwork.com. 24 Lawrence Casalino et al., “US physician practices spend more than $15.4 billion annually to report quality measures,” Health Affairs, March

2016, Volume 35, Number 3, healthaffairs.org. 25 “Meaningful measures 2.0: Moving from measure reduction to modernization,” Centers for Medicare & Medicaid Services, last modified March

31, 2021, cms.gov. 26 To meet data submission requirements and data completeness criteria, physicians must report at least six measures or one specialty measure

set and do so for each one for at least 70 percent of applicable patients. In addition to the sixmeasure requirement, groups of 16 or more eligible clinicians that meet a case minimum of at least 200 cases are subject to the 30day allcause hospital readmission measure. This measure is automatically calculated using claims data and is counted in addition to the quality reporting requirement.

27 “AHIP, CMS, and NQF partner to promote measure alignment and burden reduction,” Core Quality Measure Collaborative, May 12, 2021, qualityforum.org.

28 “Part C and D performance data,” Centers for Medicare & Medicaid Services, last modified September 13, 2021, cms.gov. 29 McKinsey Broker Health Benefits Survey, 2018. 30 We acknowledge that there may still be feeforservicebased transaction costs for billing between the capitated provider and other providers

such as postacute facilities and laboratories. 31 “Premier Inc. survey: Clinically integrated networks in alternative payment models expanded valuebased care capabilities to manage COVID

19 surge,” Premier, May 13, 2020, premierinc.com. 32 Julius Bruch, Adi Kumar, and Christa Moss, “Implications for valuebased payment programs: Weathering COVID-19,” June 2020, McKinsey.com.

66 Chapter 8: Seismic interventions

McKinsey & Company McKinsey Center for US Health System Reform

Why has the US healthcare system not yet captured the $265 billion in annual savings from known interventions identified in this report? The answer is different for “within” and “between” versus “seismic” interven- tions. For the first two types of interventions, the major barriers are at the organization level. These include the need for healthcare organizations to manage labor displacement in an industry that is a driver of US workforce growth and a lack of prioritization from industry leaders on administrative simplification. Organizations that have been successful in addressing these issues prioritize administrative simplification as a strategic initiative, commit to transformational change versus incremental steps, engage the broader partnership ecosystem for the right capabilities and investments, and disproportionally allocate resources to the underlying drivers of productivity, such as technology and talent. Seismic interventions, on the other hand, have not yet been adopted fully because of a lack of motivation to innovate at the organization level and therefore require stakeholders to take on industry-level roles, such as setting top-down guidelines for action (for example, inter operability frameworks); creating public–private part- nerships; and collaborating with third parties, such as foundations, to research facts to galvanize action.

CHAPTER 9

How to catalyze change

67Chapter 9: How to catalyze change

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

critical to deliver the administrative spending savings from within and between interventions.

Make it a strategic priority: The or- ganizations that have demonstrated success at administrative simplification make it clear at the highest level that the efforts are a strategic priority. This transforms the dynamic from a cost-cutting exercise to a major pro- ductivity initiative underpinning other organizational priorities such as growth, agility, innovation, and resilience. Re- search has shown that organizations that prioritize industry-leading produc- tivity programs were twice as likely to be in the top quintile of their peers, as measured by economic profit.1,2

Commit to transformational change: The organizations that conduct these bold moves create programs that are transformations, not one-off initiatives.3 To be successful in these transforma- tional programs requires substantially more effort and planning than for dis- crete projects (for example, completing a full-potential analysis to set ambitious targets, instilling rigorous performance management, focusing on change management, pursuing systematic, multilever change). An analysis of more than 80 large-scale transformations of public companies showed that organi- zations that are comprehensive in this way were correlated with top-quartile total-return-to-shareholder gains.4,5

Engage the broader partnership eco system: Administrative simplifica- tion benefits from partnerships. The rise of healthcare service technology (HST) companies and a deepening entry of technol ogy giants into health- care has created a broader partnership eco system, where newer technologies, such as cloud services, analytics, or member engagement, have been an enabler of growth.6,7 These HST com panies reduce the scale require- ment for a healthcare organization to

In chapters 3 to 7 of this report, we identified $210 billion in savings from “within” and “between” interventions in five functional focus areas. In chapter 8, we identified $105 billion from “seis- mic” interventions. Since many of these seismic interventions can partially replace within and between interventions, we found total savings of $265 billion, or 11 percent of total administrative spending, after we accounted for overlap. This represents where the opportunity is in administrative spending today and what we need to do to get there (see Exhibit 9.1 for approximately 30 within and between interventions).

Yet, it would be natural to ask why the opportunity has not yet been realized and how to bring the interventions to fruition. The purpose of our analytical framework was to provide a roadmap for stakeholder groups to capture the opportunity; in this chapter, we will outline the why but focus on the how for each type of intervention.

Catalyzing "within" and "between" interventions The opportunity and interventions identified in this report are not new. At the same time, however, little has been realized in the US healthcare system in a scaled and sustained way. Based on our firsthand experience, as well as input from many industry experts, some of the main challenges include the need to manage labor displacement in an industry that is a driver of US workforce growth and a lack of prioritization of administrative simplification by health- care leaders.

Reviewing a number of high-performing healthcare organizations, we identified four common principles to overcome these challenges. While these principles are not necessarily new, they are often overlooked, resulting in failed attempts. Ensuring these are in place up front and committing to them as an organization is

68 Chapter 9: How to catalyze change

McKinsey & Company McKinsey Center for US Health System Reform

one-time investment, generally 0.7 to 1.0 times the annual run-rate savings. The total investment can be substantial and may require commitment and financial prudence. It has multiple components. Part of it will be capital, as organizations invest in core technology or large-scale partnerships to pursue scale benefits (for example, investing in enterprise- level core administration platforms that allow seamless data processing across

simplify administrative functions. Capability-driven partnerships have become another avenue for two or more organizations to overcome the economies- of-scale hurdle on specific strategic capabilities (for example, joint ventures between payers on claims adjudication platforms).8

Allocate resources disproportionally: In our experience, capturing admini- strative savings requires an up-front,

Exhibit 9.1

Identi�ed “within” and “between” interventions across functional focus areas

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 9.1 of 9.2 (total 22 exhibits)

Functional focus area “Within” interventions “Between” interventions

Financial transactions ecosystem

• Simplify products o�ered • Streamline claims submission and communication process • Automate adjudication • Clarify Explanation of Bene�ts • Sunset old prior authorizations • Prescreen prior authorizations using digital support • Increase proportion of automated prior authorizations

• Improve data management and coordination • Improve coordination and clarity on claims-related communications • Streamline claims payment tracking and recovery process • Align incentives between payer and provider through risk-sharing models • Align jointly on prior authorization criteria • Conduct targeted “gold carding”

• Promote operational excellence using traditional levers • Build for “functions of the future”

Industry-agnostic corporate functions

• Promote operational excellence using traditional levers • Build smart services • Empower a function through foundational data investments

Industry-speci�c operational functions

• Reduce transaction volume through proactive issue resolution and interface improvements • Improve handle time and issue resolution via arti�cial intelligence • Outsource to highly skilled vendors

• Build strategic payer-provider platforms to reduce demand

Customer and patient services

• Digitalize manual nursing management administrative activities • Enable management of larger spans • Digitalize and automate case and disease management processes • Integrate suite of tools and solutions • Improve operational discipline and ensure “top of license” practices

Administrative clinical support functions

Source: McKinsey analysis

69Chapter 9: How to catalyze change

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

2. Ability to sunset outdated infrastruc- ture: Latitude and market mechanisms to remove outdated technology and business models

Motivation to innovate:

3. Existence of feedback loop: Feedback loops between customers and organi- zations to motivate investment into and adoption of the right innovations

4. Existence of incentives for product or service improvement: Financial or non-financial motivation to improve performance of the innovation

5. Existence of budget constraints for end users: Budgetary constraints on the end user of innovation to force right prioritization

We reviewed each of the seismic inter- ventions against these conditions (Ex- hibit 9.2). In general, implementation of seismic interventions is not held back by the ability to innovate. In our example seismic interventions, there were clear pockets of innovation and experimenta- tion across all (for example, standardiza- tion of nursing licensure or creation of a unified Medicare medical policy), though with varying degrees of completion. The more common hurdle was the motivation to innovate. In general, the financial pressure to act is not apparent at the organization level (for example, limited incentive to use a centralized claims clearing house). Therefore, given the outsize impact these seismic interven- tions could create, stakeholder groups across the healthcare spectrum may need to play specific role at an industry level to create the right motivations:

— Government could set a framework: Federal and state governments could consider setting the guardrails within which healthcare organizations could operate in areas where natural mar- ket competition has not led to optimal behaviors. The interoperability rules that the Centers for Medicare & Med- icaid (CMS) have set are an example;

claims, care management, enrollment, billing). Another component is the need to attract varying kinds of talent. Change may depend on giving the right people the space to try new approach- es. Many of these inter ventions could benefit from skills in advanc ed analytics and new technologies (for example, machine learning, cloud computing, natural language processing). In addition, people who are able to trans- late between technical and business teams and support a unified goal will be needed. Finally, opera tional leaders in the core business that under stand and adopt the digital and analytical capa- bilities could be helpful. This new talent pool can be built both through hiring and through reskilling the current workforce. Given the amount of needed talent, organizations will likely need to do both.

Catalyzing “seismic” interventions The non-exhaustive set of potential seis- mic interventions we outlined in chapter 8 relies on collaboration across public and private stakeholder groups, including individual consumers. The fact that this collaboration is likely needed leads to an important observation: these seismic interventions may require industry-level action since organization-level actions alone are not proving effective. By apply- ing a framework on how to enable dis- ruptive innovation in these types of situ- ations, we could identify the underlying gap for each of the proposed seismic interventions, including the ideal stake- holder group (private or public sector) to carry out the right mechanism (for ex- ample, technology platform, operational alignment, payment design).9 We started by examining the five conditions for innovation. They fall into two categories:

Ability to innovate:

1. Ability to experiment: Existence of experimental infrastructure to test and validate new innovations

70 Chapter 9: How to catalyze change

McKinsey & Company McKinsey Center for US Health System Reform

Exhibit 9.2

Seismic interventions have not yet occurred because organizations lack the motivation to innovate

Web 2021 Administrative simpli�cation: How to save a quarter-trillion dollars in US healthcare Exhibit 9.2 of 9.2 (total 22 exhibits)

Five conditions for breakthrough innovation

Technology platforms

Adopt centralized automated claims

clearinghouse

Ability to experiment Existence of thoughtful methodology, approach, and funding for “testing” the idea

Prioritize high- value interoper- ability use cases

Standardize physician licensure

Standardize medical policies

Streamline quality

reporting Modularize

bene�ts

Adopt globally-capitated

payment models

Operational alignment Payment design

Condition already met Path forward to meet condition already underway Potential action needed to meet condition

Existence of incentives for product or service improvement Motivation to improve performance (�nancial or non-�nancial)

Existence of budget constraints for end users Budgetary restrictions that will force prioritiza- tion of the right actions

Existence of feedback loop Building strong feedback loops between stake- holder groups to motivate investment into and adoption of the right innovations

Ability to sunset outdated infrastructure Capacity to freely remove outdated technology and business models

Met: Automated clearinghouses have already been shown to work in banking, and 70%+ of healthcare payments already are submitted through the National Automated Clearing House Asso- ciation (NACHA)

Met: Multiple tech- nology initiatives have been launched at the US healthcare system level to tackle this (eg, FHIR, interoper- ability rules)

Met: Nursing licensure compact, which is signed by almost 35 states and allows nurses to work across state lines, has already proven this model can be used in the United States

Met: Medicare is able to in�uence the market in a way that promotes standardized policies

Met: Well- recognized problem; CMS has already launched multiple initiatives to simplify quality reporting

Met: Bene�ts of standardized o�erings in both Individual market and Medicare Advantage are already visible

Met: Providers already operating in a value-based payment paradigm

Potential action needed: Create �nancial incentives for organizations that opt out of ACH

Path forward: Use case alignment will allow organizations to capture a greater ROI on technology investments as well as allow vendors to better hone their products/service o�erings

Potential action needed: Create a commission with repre- sentation from states and physicians that monitors impact of standardized licensure and recommends changes if needed

Potential action needed: Payers will have to �nd other ways to di�erentiate than medical policies (eg, product design, pricing)

Potential action needed: Payers and providers are better able to di�erentiate performance in VBC settings if there is a standardized, focused list of metrics to reach

Met: Payers will be pushed to develop simpler and more compelling digital tools as part of sales process with customers

Met: Revenue in these models is predeter- mined on a per person basis, creating incen- tives to provide better products/services for payers to increase member attraction and retention

Potential action needed: Financial incentives for organizations that opt into ACH could create competitive advantages

Met: Payers and providers are already spending substantially on individual technol- ogy projects related to interoperability; alignment on use cases could remove unnecessary spending

Potential action needed: States may need o�set from reduction in licensing revenues (eg, quanti- fying bene�ts of in- creased physician access within their areas)

Potential action needed: Currently no budget incentive for a payer to standardize medical policies; payers would need �nancial incentives to follow national standards

Met: Physicians already spend substan- tial time and operating expenses to meet quality reporting requirements

Path forward: Ongoing price pres- sures in Commercial/ fully-insured segments are likely to create more impetus for �nding ways to reduce payer expenses

Met: Global capitated models do provide incentives for organi- zations to improve productivity, including administrative spend- ing, given shift in pro�t economics

Met: Feedback loops between payer and customers (members and providers) already exist, and channel will continue to exist

Potential action needed: Convene a group of stakeholders to discuss and align on high-value, system-level use cases that help prioritize interoperability-related technology initiatives

Potential action needed: Launch a process that solicits states' inputs on what the federal licensure should look like

Met: CMS already engages with provider stakeholders (eg, AMA, AHIP) to solicit feed- back on which quality metrics to keep or remove

Potential action needed: Though there is momentum to shift to VBC, there are con- straints (eg, lack of op- portunity for providers, payers lacking data on delegated risk); need to incorporate these issues as the interven- tion is launched

Path forward: Long-tail of local/ regional claims pro- cessing warehouses may need to integrate their work�ow into single ACH

Path forward: Agreement on which handful of existing initiatives will align with high-value use cases (eg, patient- centered health- care database) and which to sunset

Path forward: Ability to remove state- speci�c licensures in nursing compact already proven

Met: Medical policies are already regularly updated as new inno- vations and standards of care emerge

Met: Providers can focus on narrower set of prioritized metrics; payers will potentially have to replace quality metrics in contract arrangements with providers

Path forward: Review all contracts with a given provider and determine the “least common denominator” version that could satisfy many custom- ers; integrate new, simpler bene�t design into core functions (eg, underwriting, sales, marketing)

Potential action needed: While there is a shift to VBC models generally, most of this is still built on top of FFS foundation with only 3% fully capitated; broader action would be needed to fully shift infrastructure (eg, contracts, payment platforms) to global capitated

Is there motivation to innovate?

Is there ability to innovate?

Seismic interventions

Path forward: Customers have suggested willingness to switch carriers for reduced premiums

Potential action needed: Create an ongoing, diverse commission to review and update medical policies

Source: Lawrence Casalino et al., “US physician practices spend more than $15.4 billion annually to report quality measures,” Health A�airs, March 2016, Volume 35, Number 3, healtha�airs.org; Andis Robeznieks, "AMA to CMS: Work to simplify Quality Payment Program regulations," American Medical Association, August 21, 2017, ama-assn.org; Gail Wilensky, “The need to simplify measuring quality in health care,” Journal of the American Medical Association, June 19, 2018, Volume 319, Number 23, pp. 2369–70, jamanetwork.com

71Chapter 9: How to catalyze change

Administrative simplification: How to save a quarter-trillion dollars in US healthcare

— Third parties can provide objective fact gathering and analyses: Given the benefits for collaboration across stakeholder groups, neutral and objective third parties such as foun- dations or bipartisan groups can be the arbitrators of facts. Whether it is benchmarking to highlight healthcare organiza tions that are leading in ad- ministrative simplification or funding small-scale experiments, such as a randomized control trial of a small group of private payers or hospitals to test what conditions may be neces- sary for administrative savings, the publication of these organization- level data could galvanize action among other healthcare organizations.

While the goal of addressing unnecessary administrative spending in US healthcare has seemed elusive, this report aims to create a roadmap that: (1) identifies a concrete set of known interventions ( within, between, and seismic) against a set of functional focus areas, (2) breaks down how these savings could be achieved within the US healthcare sys- tem, and (3) offers a framework for what roles different stakeholder groups could play to deliver the opportunity.

it has defined both what it expects to see (for example, payers and provid- ers to build application programming interfaces (APIs) to readily share data) and timelines for realization.10 But the agency has left it up to indi- vidual organizations to determine how to get there.

— Investors can prove ideas with pilots: In healthcare, there has been a recent boom in investment from institutional investors (private equity, venture capital, and hedge funds).11 This type of funding allows for rapid testing, scaling, and evolution of innovation in healthcare in a way that the public sector could not replicate. Creating public–private partnerships could in- fuse the benefits of private investing into seismic interventions that may be be driven by the public sector. For ex- ample, instead of each state building a health information exchange (HIE) by itself, a public– private partnership in a pilot state could prove the model and then disseminate the technology platform to other states for rapid scale- up. This may lessen the overall public sector budget burden while also in- centivizing private sector investment.

1 Martin Hirt, “How to create a real hockey stick strategy,” February 25, 2018, McKinsey.com. 2 Chris Bradley, Martin Hirt, and Sven Smit, “Strategy to beat the odds,” February 2018, McKinsey.com. 3 “The path to true transformation,” October 28, 2020, McKinsey.com. 4 “The path to true transformation,” October 28, 2020, McKinsey.com. 5 Kevin Laczkowski, Tao Tan, and Matthias Winter, “The numbers behind successful transformations,” October 2019, McKinsey.com. 6 Shubham Singhal, Basel Kayyali, Rob Levin, and Zachary Greenberg, “The next wave of healthcare innovation: The evolution of ecosystems,”

June 23, 2020, McKinsey.com. 7 Greg Gilbert, Jay Krishnan, and Drew Ungerman, “What payers and providers can learn from successful cloud transformations in other

industries,” June 28, 2021, McKinsey.com. 8 We acknowledge that these partnerships should be structured and carried out consistently with all applicable laws and regulations governing

competition. 9 Nikhil Sahni, Maxwell Wessel, and Clayton Christensen, “Unleashing breakthrough innovation in government,” Stanford Social Innovation

Review, Summer 2013, ssir.org. 10 Calder Lynch, “Implementation of the CMS Interoperability and Patient Access Final Rule and Compliance with the ONC 21st Century Cures Act

Final Rule,” Centers for Medicare & Medicaid Services, August 14, 2020, medicaid.gov. 11 “Venture investments in digital health during the first half of 2021 have already surpassed funding raised in all of 2020 and is the largest

amount raised in a single year since 2010,” from Heather Landi, “Digital health dollars hit $15B high driven by telehealth investment in 2021,” Fierce Healthcare, July 19, 2021, fiercehealthcare.com.

72 Chapter 9: How to catalyze change

McKinsey & Company McKinsey Center for US Health System Reform

Published by McKinsey & Company, 150 West Jefferson,

Suite 1600, Detroit, Michigan 48226.

Administrative simplification: How to save a quarter-

trillion dollars in US healthcare meets the Forest Steward-

ship Council (FSC) chain of custody standards.

The paper used in Administrative simplification: How to

save a quarter-trillion dollars in US healthcare is certified

as being produced in an environ mentally responsible, socially

beneficial, and economically viable way.

Printed in the United States of America.

Nikhil R. Sahni ([email protected]) is a partner in the Boston office and a fellow in the Economics Department at Harvard University.

Prakriti Mishra ([email protected]) is an associate partner in the Washington, DC, office.

Brandon Carrus ([email protected]) is a senior partner in the Cleveland office.

David M. Cutler is the Otto Eckstein professor of applied economics at Harvard University.

Administrative simplification: How to save

a quarter-trillion dollars in US healthcare is

published by the McKinsey Center for US Health

System Reform, part of McKinsey’s Healthcare

Systems and Services Practice. The McKinsey

Center for US Health System Reform is led by

Stephanie Carlton, David Knott, Edward Levine,

and Nikhil Sahni.

McKinsey & Company

Healthcare Systems and Services Practice

150 West Jefferson, Suite 1600

Detroit, Michigan 48226

This publication is not intended to be used as the basis for trading in the shares of any company or for undertaking any other complex

or significant financial transaction without consulting appropriate professional advisers. No part of this publication may be copied or

redistributed in any form without the prior written consent of McKinsey & Company.

Productivity imperative leaders: David Knott, Edward Levine

Global healthcare practice leader: Shubham Singhal

North America healthcare practice leader: Drew Ungerman

Analysts: Gary Chia, Julius Ewungkem, Nicolas Garcia, Brooke Istvan,

Neda Bassir Kazeruni, Crosbie Marine, Chrissy Meder, Garam Noh,

Rahi Punjabi, Mara Reichle

Editor: Allan Gold, Elizabeth Newman, Mark Staples

Assistant editor: Lyris Autran

Art director: Ginny Hull Hartline

Copy editor: Susan Schwartz, Sarah Smith

External relations: Sharmeen Alam

Lead reach and relevance partner: Scott Blackburn

Practice managers: Julia Barclay, Cameron Kennedy, Heather Megosh

Knowledge manager: Eija Erasmus

McKinsey Center for US Health System Reform

Administrative simplification: How to save a quarter-trillion dollars in US healthcare Perspectives on the Productivity Imperative in US Healthcare Delivery

M cK

insey on H ealthcare: A

dm inistrative sim

plification: H ow

to save a quarter-trillion dollars in U S

healthcare

October 2021

McKinsey Center for US Health System Reform

McKinsey Healthcare Systems & Services Practice

October 2021

Cover image: We Are Aphra Ltd

McKinsey.com

@McKinsey