can information technology transform health care? potential health benefits, savings and costs...

Can Information TechnologyTransform Health Care?

Potential Health Benefits, Savings and Costs

Richard Hillestad, Ph.D.

Presented at The HIT Symposium, Massachusetts Institute of Technology, July 17-20, 2006

A6387b-2

Background

• A two year RAND Health study completed in Spring 05

• Results appear in 2 articles in September 05 Health Affairs and 4 RAND reports

• Funded by the private sector -- Cerner Corp., General Electric, Hewlett-Packard, Johnson & Johnson, and Xerox

• 14 member steering group headed by Dr. David Lawrence, former CEO of Kaiser

A6387b-3

The Problem in Context• U.S. health care is one of the largest and most

inefficient information enterprises because it still operates mostly with paper records

A6387b-4

• U.S. health care is one of the largest and most inefficient information enterprises because it still operates mostly with paper records

The Problem in Context

• Despite health spending of $1.7 trillion nationally and projected to grow to over $4 trillion in 10 years, it doesn’t provide the best care

– recommended care is provided only about 55% of the time

– and, by a number of measures, health in the U.S. is worse than OECD averages

A6387b-5

• U.S. health care is one of the largest and most inefficient information enterprises because it still operates mostly with paper records

The Problem in Context

• Despite health spending of $1.7 trillion nationally and projected to grow to over $4 trillion in 10 years, it doesn’t provide the best care

– recommended care is provided only about 55% of the time

– and, by a number of measures, health in the U.S. is worse than OECD averages

• How much could Electronic Medical Record Systems (EMR-S) help?

A6387b-6

What Is an Electronic Medical Record System?

• EMR -- replaces the paper medical record

• EMR-S adds functions:– Clinical decision support– Patient tracking and reminders– Personal health records– Computerized physician order entry– Regional health information networks

A6387b-7

Key Findings

• Efficiency savings enabled by EMR-S could reach ~$80B/year at 90% adoption

• Costs to achieve that in 15 years average ~$8B/yr

• Safety benefits include avoiding 2.2 million adverse drug events

• Health benefits from prevention and management of chronic diseases alone could be 400,000 deaths avoided and 40M added workdays

• The market is not leading to this result because of important barriers and disincentives

• Therefore, there is a clear role for government action

A6387b-8

EMR-S Now in Only 20-25% of Hospitals and10-15% of Physicians’ Offices

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1990 1995 2000 2005 2010 2015 2020 2025 2030

Inpatient EMR-S

Outpatient EMR-S

Inpatient CPOE

Fraction ofhospitals or officesadopting

A6387b-9

Problem Is To Estimate Impact atFull Adoption

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1990 1995 2000 2005 2010 2015 2020 2025 2030

Adoptionperiod15 yrs

“Full” adoption

Complex,networkedtechnology

Inpatient EMR-S

Outpatient EMR-S

Inpatient CPOE

Fraction ofhospitals or officesadopting

A6387b-10

The RAND Study of EMR-S• RAND study developed computer simulation models

to estimate potential benefits and costs, assuming– Widespread adoption (90%)– Interoperability (across providers)– Related health care process changes, for

example:• Restructured hospital and physician office

workflow• Increased preventative interventions• Team care for chronic disease

• Extrapolates limited published evidence of EMR-S benefits

A6387b-11

• Reduced waste, e.g., reduced duplication of tests

• Improved/changed processes, e.g., improved workflow and patient flow

• Fewer resources, e.g., reduced administration of paper records, better antibiotics usage

• Lower cost substitutions, e.g., generic drug utilization

Efficiency Savings Enabled by EMR-S

A6387b-12

Efficiency Savings in the Inpatient and Outpatient Settings

Outpatient$20B/yrInpatient

$60B/yr

• Length of stay• Nursing administrative time• Medical records administration• Lab and radiology utilization

• Drug utilization• Lab and radiology utilization• Chart administration• Efficient patient scheduling

• • •

• • •

$80B/yr at 90% Adoption

A6387b-13

It Will Take Some Time to RealizeSuch Savings

Savings/yr$B Fraction

adopting

Evidenceand scalinguncertainty

Adoptioncurve

$80B/yr

2004 2006 2008 2010 2012 2014 2016 2018

1.0

0.8

0.6

0.4

0.2

80

60

40

20

0 0

A6387b-14

• Costs include EMR-S software license, hardware and its maintenance

• As well as planning, training and implementation

• And reduced revenue or increased provider costs during implementation

Costs of EMR-S

A6387b-15

We Estimated the Cost of Adoption over Time by Simulating Adoption with Current Costs

Ambulatory EHR-S costs/yr Inpatient EHR-S costs/yr

$B

2.0

1.5

1.0

0.5

Cumulative cost: $97.4BOne-time cost: $30.4BMean yearly cost: $6.5B

Fractionadopting

2004 2006 2008 2010 2012 2014 2016 2018

1.0

0.8

0.6

0.4

0.2

0

$B

0

2

4

6

8

10

Cumulative cost: $17.2BOne-time cost: $7.3BMean yearly cost: $1.1B

Fractionadopting

2004 2006 2008 2010 2012 2014 2016 2018

1.0

0.8

0.6

0.4

0.2

00

A6387b-16

Although EMR-S Implementation Costs Are Substantial . . .

Total cost (15 years)

Hospitals 97

Physician offices 17

Connectivity 6

Total $120B

Costs

A6387b-17

. . . Costs Are Modest Compared to Potential Savings, Even During Implementation

Total savings(15 years)

Hospitals 470

Physician offices 160

Total $630B

Efficiency Savings

Total cost (15 years)

Hospitals 97

Physician offices 17

Connectivity 6

Total $120B

Costs

A6387b-18

Safety Benefits of EMR-S

• Reduced errors from handwriting

• Allergy warnings

• Warnings of drug-drug interactions

• Dosage monitoring

A6387b-19

EMR-S with Computerized Physician Order Entry Can Increase Safety -- Medicare Share ~40% --

$3.1B

65+

0–6465+

65+

0–640–64

$0.9B

Thousandsof

events

A6387b-20

Health Benefits Enabled by EMR-S

• Improved compliance with prevention activities

• Better management and prevention of chronic diseases

• Coordination of care across providers

• Patient involvement in care and healthy life style

A6387b-21

EMR-S Can Promote Prevention with Guidelines, Reminders, and Outreach

Targetpopulation

% Populationnot now

compliant

Cost/yr for100%

compliance

Health benefitswith 100%

compliance

Breast cancerscreening

Women40 andolder

30% $1.5B

50K cancersdetected early,

4K fewer deaths/yr

Colorectalcancerscreening

50 andolder

66% $4.0B 23.5K fewer deaths

Influenzavaccination

65 andolder

37% $0.2B7.5K fewerdeaths/yr

Pneumococcalvaccination

65 andolder

47% –$0.1B21K fewerdeaths/yr

A6387b-22

Chronic Disease Management Is a High Leverage Application of EMR-S

• The chronically ill absorb about 75% of national health expenditure

• Chronic disease management requires– Community support and team care– Coordination and communication across

providers– Patient monitoring and involvement

• Regional demonstration projects with EMR-S often focus on chronic disease management

A6387b-23

Targetdiagnoses

Analysis Process for Disease ManagementICD9 and

CCC codes

Clinical judgment

MEPS

Partitionfile

MEPS subjectswithout target

diagnoses

MEPS subjectswith targetdiagnoses

Summarize by sex, age, eth., event type.Re-weight without file to match with pop.

Disease pop - Utilization* - Expenditures* - Outcomes

Control pop - Utilization - Expenditures - Outcomes

* Differentiate between events associated vs. unassociated with a target diagnosis

Disease managementliterature

Clinical judgment

% change inutilization byseverity - IP stays - ED visits - Oth Amb - Rx use

Selectparticipation

rates

Calculate system-widebenefits & costs

System-wide benefits & costs(includes utilization & cost ofdisease management program) - Utilization - Expenditure - Outcome

Activities &resourcesper patientby severity - MD hrs - RN hrs - Oth hrs

A6387b-24

Disease Management Attempts to Reduce Acute Episodes

Reduced ER visits and hospital stays

0

2

4

6

8

10

12

14

Inpatient stays ER visits

% reduction

Upper Bound:Assumes 100%participation in management ofemphysema, asthma, CHF, and diabetes.

A6387b-25

EMR-S Enabled Prevention and Disease Management Can Reduce Mortality and the

Economic Impact of Chronic Illnesses

Participation Rates

Disease Management

Lifestyle Change

80%

50%

50%

20%

Days Affected (millions)

School days lost

Work days lost

Total days in bed

-11

-39

-270

-7

-21

-160

Mortality (thousands of deaths) -400 -250

Results for emphysema, asthma, CHF and diabetes

A6387b-26

Other Industries Health Care Industry

Champion Firm No

Integrated System Disaggregate System

Standards Low Implementation

High IT Investment Low EMR-S Investment

Market Forces Market Failure

Consumer Involvement No Consumer Involvement

Barriers to Adoption of EMR-S in Health Care

A6387b-27

The Most Significant Barrier: Physicians and Hospitals Do Not See Most

Savings from EMR-S Investments

-40

-30

-20

-10

0

10

20

30

40

Social savingsPhysicianrevenue

Hospitalrevenue

Pharmarevenue

$B

EMR-S Purchasers

Revenue and Savings From Chronic Disease Management

A6387b-28

The Government Should Intervene Now

• The market is not working well – Providers have little incentive or capability to

institute standards-based, interconnected EMR systems

– Current adoption process may lead to a 2-tiered health care system and inhibit future change

• The government is the largest employer and health care payer (and has considerable leverage on the industry)

• EMR-S enabled changes could moderate unsustainable health care cost growth and improve quality

A6387b-29

Key Government Actions

• Promote standards and EMR-S certification

• Implementation support

• Promote interoperability and regional connectivity

• Medicare leadership with incentives– Pay for use of EMR-S– Pay for quality measured by EMR-S

A6387b-30

Per Encounter Pay-for-Use Incentive

With incentive

Without incentive

Value of incentive: $16.2 BCost of incentive: $2 BPer-encounter payment: $1.5Duration: 3 yearsDemand elasticity: -.5Adoption period: 15 years

Fractionadopting

1.0

0.8

0.6

0.4

0.2

02004 2006 2008 2010 2012 2014 2016 2018

A6387b-31

Can Information Technology Transform Health Care?

• Yes, but --– not without much wider adoption– not without standards and interoperability– not without associated process and health care

system changes– not without measurement of quality and

efficiency

• And, probably not without government intervention

A6387b-33

Dissemination of These Findings• Publications for multiple audiences

– 2 peer reviewed articles in Health Affairs– 4 RAND monographs– RAND Research Brief and Congressional Newsletter

• Widespread media coverage with press releases by both Health Affairs and RAND

• Briefings for congressional and committee staff– Alliance for Health Reform (Frist-Rockefeller Group) briefing (300+ attendees, 60

Congressional)– RAND Congressional briefing cosponsored by 21st Century Health Care Caucus (Reps.

Murphy and Kennedy), 26 attendees

• Meetings with key committee staff– Senate Finance Committee– Senate LHHS Appropriations Subcommittee– House Energy & Commerce Health Subcommittee– House Ways & Means Subcommittee– House LHHS Appropriations Subcommittee– House Armed Service Committee

A6387b-34

Dissemination (2)• Meetings with Senators and staff

– Senator Max Baucus (D-MT)– Senator Pat Roberts’ Staff (R-KS)– Senator Michael Enzi (R-WY)– Senator Orrin Hatch’s Staff (R-UT)

• Workshop on “Economic Impact of EHR Adoption Gap” with David Brailer, National Coordinator, Health Information Technology, HHS, at RAND, Santa Monica

• Briefings at private sector activities– Hewlett-Packard Worldwide Health Symposium, Las Vegas– Cerner Health Care Leadership Forum, Orlando– Xerox Health Care Forum, Rochester (in December)

• Briefings for other interest groups– Institute for Behavioral Health Informatics

A6387b-35

U.S. Health Expenditures Per Capita Are the Highest Among OECD Countries

Source: OECD Health Data 2004, 1st Edition, Table 9

Note: The presented countries represent the range of expenditures for OECD countries. Due to space limitations, all OECD countries are not presented, however the average was calculated from 29 countries. Turkey’s data was not available.

$0

$1,000

$2,000

$3,000

$4,000

$5,000

$6,000

Un

ited

Sta

tes

Sw

itze

rlan

d

No

rway

Can

ad

a

Ger

ma

ny

Fra

nce

Au

stra

lia

Sw

ed

en

Ital

y

Ave

rag

e

Jap

an

Un

ited

Kin

gd

om

Sp

ain

Hu

ng

ary

Ko

rea

Po

lan

d

Mex

ico

2001 healthexpenditures

per Capita(US$ PPP)

A6387b-36

U.S. Life Expectancy IsSlightly Below the OECD Average

60

65

70

75

80

85

90

Icel

and

Jap

an

Sw

ed

en

Can

ad

a

Au

stra

lia

Ital

y

Un

ited

Kin

gd

om

Ger

ma

ny

Sp

ain

Fra

nce

Fin

lan

d

Ave

rag

e

Un

ited

Sta

tes

Mex

ico

Po

lan

d

Hu

ng

ary

Tu

rkey

Years

Males at birth (2001) Females at birth (2001)

Source: OECD Health Data 2004, 1st Edition, Table 1

Note: The presented countries represent the range of life expectancies for OECD countries. Due to space limitations, all OECD countries are not presented, however the average was calculated from all OECD countries.