Cooper Health Care And The Affluence Poverty Nexus Detroit

Health Care Health Care and the and the

Affluence Poverty NexusAffluence Poverty Nexus

Richard A. Cooper, M.DRichard A. Cooper, M.D..Leonard Davis Institute of Health EconomicsLeonard Davis Institute of Health Economics

University of PennsylvaniaUniversity of Pennsylvania

WCMS Foundation Francis P. Rhoades, MD Memorial Lecture

March 26, 2010

Geographic Variation in Health CareGeographic Variation in Health CareDARTMOUTH ATLASDARTMOUTH ATLASPeter Peter OrszagOrszag, 2007, 2007

Three Myths of Geography and Poverty

1. Hospital Referral Regions: Variation in health care Variation in health care utilization among hospital referral regions (HRRs) utilization among hospital referral regions (HRRs) is due to the overuse of supplyis due to the overuse of supply--sensitive services. sensitive services.

2. 2. Academic Medical CentersAcademic Medical Centers: Variation in physician : Variation in physician inputs among academic medical centers is a sign of inputs among academic medical centers is a sign of waste and inefficiency.waste and inefficiency.

3. 3. HRR QuintilesHRR Quintiles: If the entire US could achieve : If the entire US could achieve spending equivalent to the lowestspending equivalent to the lowest--spending region, spending region, 30% of health care spending could be saved.30% of health care spending could be saved.

“Regional differences in poverty and incomeexplain almost none of the observed variation.”

Skinner and Fisher 2009

Geographic variation in health care is principally the result of geographic differences in poverty.

Payment changes made according to geographic norms will harm to low-income patients and the providers who care for them.

The Inconvenient TruthThe Inconvenient Truth**************************************************

Regional Poverty

Poverty, 20000 - 20%

20 - 40%40 - 60%60 - 80%80 - 100%

Urban Poverty

PhiladelphiaPhiladelphiaIncome = 118% of US Average

The Bruton CenterThe University of Texas at Dallas

BaltimoreBaltimoreIncome = 114% of US Average

DetroitDetroitIncome = 96% of US Average

““Unexplained geographic variation is due to Unexplained geographic variation is due to the overuse of supplythe overuse of supply--sensitive specialty services.sensitive specialty services.””

Myth #1Myth #1

Milwaukee HRRWisconsin

MilwaukeeMilwaukee

0

100

200

300

400

500

600

day/1000_1864Days per 1,000

MilwaukeeMilwaukee

Hospital Days in Wisconsin HRRsHospital Days in Wisconsin HRRs

Wisconsin HRRsWisconsin HRRs

HospitalDaysper

1,000

30% excess utilization

Milwaukee HRRMilwaukee HRRPer Capita Income = 108% of US Average

The Bruton CenterThe UT at Dallas

Milwaukee is the third most

segregated city in the nation

R2 = 0.65

0

250

500

750

1,000

$- $10,000 $20,000 $30,000 $40,000 $50,000

Per Capita Income

Day

s pe

r 1,0

00Milwaukee

Hospital Days vs. Per Capita Income

4-fold

Power

Poor

Rich

ZIP Codes -

Ages 18-64

“Poverty Corridor”42% of total population92% of Black population74% of Latino population33% of income

MilwaukeeMilwaukee’’s s ““Poverty CorridorPoverty Corridor””

0

100

200

300

400

500

600

day/1000_1864Days per 1,000

MilwaukeeMilwaukee

Hospital Utilization in Wisconsin HRRsHospital Utilization in Wisconsin HRRs

Wisconsin HRRsWisconsin HRRs

HospitalDaysper

1,000

Poverty CorridorPoverty Corridor

Milwaukee minus Milwaukee minus ““Poverty CorridorPoverty Corridor””

“Preventable”

Hospital Admissions Milwaukee

0

2

4

6

8

Diabetes Asthma COPD CHF

Ratio of Poorest

toWealthiest

Zones

6-fold

1999

Los Angeles HRR

Los Angeles

The Bruton CenterThe UT at Dallas

Los Angeles CountyLos Angeles County7.5 million adults

Average Income = 108% of US Average

R2 = 0.61

0

400

800

1,200

$- $50,000 $100,000 $150,000 $200,000 $250,000Mean Household Income

Day

s P

er 1

,000

Los Angeles Hospital Days Per Capita vs. Household Income

4-fold

Poor

Rich

ZIP Codes -

Ages 45-64

Poverty Zone 1.8 million adults (25%)

Poverty Zone25%

Watts

Poverty Core

Poverty Core375,000 (5%)

5%

0

400

800

1,200

$- $50,000 $100,000 $150,000 $200,000 $250,000Mean Household Income

Day

s P

er 1

,000

Los Angeles Hospital Days vs. Household Income

ZIP Codes -

Ages 45-64

Household Income >$100,0001.4 million

(18%)

0%

50%

100%

150%

200%

All Ages .

Day

s pe

r 1,0

00,

% in

ZIP

Cod

es w

ith M

HI >

$100

KHousehold Income >$100KPoverty CorePoverty Zone w/o CoreTotal County

Hospital Days in Los Angeles Per Cent of Days in ZIPs

with Household Income >$100,000

Days per 1,000 in all of LA County

are 36% greater thanin ZIPs >$100K

Days per 1,000 in the Poverty Core are double the rate

of ZIPs >$100K

Hospital Days Among Eight California Counties Adults (18-64)

0

75

150

225

300

Total Adult Income >$100K

Days Per

1,000

LOS ANGELES

SACRAMENTO

SAN FRANCISCO

ALAMEDA

SAN DIEGO

ORANGE

SAN MATEO

MARIN

Variation Among

All Adults

Hospital Days Among Eight California Counties Adults (18-64)

ZIP Code Household Income

0

75

150

225

300

Total Adult Income >$100K

Days Per

1,000

LOS ANGELES

SACRAMENTO

SAN FRANCISCO

ALAMEDA

SAN DIEGO

ORANGE

SAN MATEO

MARIN

Variation Among the

Wealthiest

Hospital Days in California Counties Adults (18-64)

ZIP Code Household Income

0

75

150

225

300

Total Adult Income >$100K

Days Per

1,000

LOS ANGELES

SACRAMENTO

SAN FRANCISCO

ALAMEDA

SAN DIEGO

ORANGE

SAN MATEO

MARIN

34% greater use of hospital days

below $100K income

“Unexplained variation”

is explained by poverty.

Conclusion

Myth #2Myth #2DartmouthDartmouth’’s Quintiless Quintiles

““The 30% SolutionThe 30% Solution””

“If the entire nation could bring its costs down to match the lower-spending regions,

the country could cut perhaps 20 to 30 percent off its health care bill, a tremendous saving.”

New York Times, 2007

Lowest Medicare

Highest Medicare

Medicare SpendingMedicare Spending

BostonChicagoDetroitHoustonLos Angeles McAllenMiamiPhiladelphia PittsburghNewarkNew OrleansNew YorkPensacolaTexakanaWashington

WashingtonOregonIdahoUtahWyomingMontanaNebraskaNorth DakotaSouth DakotaIowaMinnesotaWisconsin...except for their

major cities

The Quintiles Study

Compare With

Total Health Care Spending Total Health Care Spending

Lowest Total

Highest Total

No DifferencesNo Differences

11--year Mortality year Mortality No differences No differences 55--year Mortalityyear Mortality

Lowest better; others the same Lowest better; others the same

Functional status Functional status No differencesNo differencesSatisfactionSatisfaction

No consistent differencesNo consistent differences

Access Access No better (one slightly worse)No better (one slightly worse)**

QualityQuality

No better on most measures, No better on most measures, worse for some preventive careworse for some preventive care

** ““Trouble seeing a doctorTrouble seeing a doctor”” 3.1% vs. 2.5% 3.1% vs. 2.5%

Dartmouth Doubletalk

Outcomes were no different because differences could not be discerned. But since outcomes were no better, spending in “high spending”

“regions”

must have been wasted.

And because this “wasted spending”

could not be explained (by them), it must have been due to an over-supply of specialists providing low value care.

“Waste”“Inefficiency”

“Supply-sensitive”“Value”

The 30% solution is a mirage.The 30% solution is a mirage.

ConclusionConclusion

Academic medical centers vary by more than 3-fold in the quantity of physician services at the end of life.

Given this apparent inefficiency, the supply pipeline is sufficient to meet future needs for physicians through 2020.”

Goodman et al, 2006

Myth #3

BostonChicagoDetroit (2)Houston (2)Los Angeles

Philadelphia (3)PittsburghNewarkNew York (2)Washington

15 Cities with 15 Cities with“Highest Efficiency”

“Lowest Efficiency”

Hospitals HospitalsCincinnatiCincinnatiIndianapolis Indianapolis Salt Lake CitySalt Lake CityAugusta Augusta DartmouthDartmouthMadison WIMadison WIRichmond VARichmond VATemple TXTemple TXRochester NYRochester NYJackson MSJackson MSColumbia MOColumbia MOLexington KYLexington KYOklahoma CityOklahoma CityAtlanta (Grady)Atlanta (Grady)Rochester MN (Mayo)Rochester MN (Mayo)

15 Cities with 15 Cities with“Highest

Efficiency”

“Lowest Efficiency”

Hospitals Hospitals

250,000250,000 PopulationPopulation 1,500,0001,500,000

22%22% Blacks + LatinosBlacks + Latinos 51%51%

8%8% Seniors in povertySeniors in poverty 17%17%

Medicare Spending/Decedent (Last 2 years of life, 2001-2005)

Douglas Wood, MD, Mayo Clinicfrom Dartmouth Atlas, Appendix Table 1.

Mayo Mayo ––

1515thth

Most Most ““EfficientEfficient””

Medicare Spending During Last 2 Years of Life Medicare Spending During Last 2 Years of Life

20012001--20052005

SinaiSinai--Grace Hospital, DetroitGrace Hospital, Detroit 99thth

Least Least ““EfficientEfficient””

AMCAMC

““Occupying a campus of red brick buildings amid Occupying a campus of red brick buildings amid abandoned houses, checkabandoned houses, check--cashing stores and wig cashing stores and wig shops on the cityshops on the city’’s West Side, Sinais West Side, Sinai--Grace is a Grace is a classic urban hospital. It has eight hundred classic urban hospital. It has eight hundred physicians, seven hundred nurses and two physicians, seven hundred nurses and two thousand other medical personnel to care for a thousand other medical personnel to care for a population with the lowest median income of any population with the lowest median income of any city in the country.city in the country.””

AtulAtul

GawandeGawandeThe New YorkerThe New Yorker

0

3

6

9

12

15

18

University of California HospitalsUniversity of California Hospitals

Days

Dartmouth UCLALast 6 Months of Life 6 Months of Severe

CHF .

Unexplained differences

Dartmouth:The volume of care during the last 6 months of life varies among University of California hospitals by 45%.

45%

Dartmouth:The study focused only on patients who died, so we could be sure that all patients were similarly ill.

By definition, the prognosis was identical –

all were dead.

Therefore, variations among hospitals cannot be explained by differences

in the severity of in the severity of

patientpatient’’s illnesses.s illnesses. Dartmouth Atlas OnlineDartmouth Atlas Online

Frequently asked questionFrequently asked question::

But how do you ensure that patients were not more But how do you ensure that patients were not more severely ill at some hospitals than at others?severely ill at some hospitals than at others?

0

3

6

9

12

15

18

Days

Last Six Months Six Months of life with CHF of life with severe CHF

UCLA All patients (dead or not)

adjusted for income and illness

Circulation, Cardiovascular Quality and OutcomesCirculation, Cardiovascular Quality and Outcomes, 2009, 2009

University of California HospitalsUniversity of California Hospitals

DartmouthSimilarly dead;

similarly ill

Remarkablysimilar

Unexplained differences

VVariation is due to variation in patientsariation is due to variation in patients’’ income and burden of disease.income and burden of disease.

ConclusionConclusion

$0

$2,500

$5,000

$7,500

$10,000

<$10,000 $10-15,000

$15-20,000

$20-25,000

$25-50,000

>$50,000

Income Groups

Annual Medicare Spending

Medicare Spending and IncomeMedicare Spending and IncomeNational Medicare Spending by Income GroupsNational Medicare Spending by Income Groups

Sutherland, Fisher, Skinner, 2009, from CMS

$0

$2,500

$5,000

$7,500

$10,000

<$10,000 $10-15,000

$15-20,000

$20-25,000

$25-50,000

>$50,000

Income Groups

Annual Medicare Spending

Patients, Not GeographyPatients, Not GeographyNational Medicare Spending by Income GroupsNational Medicare Spending by Income Groups

34% of Medicare Expenditures

Health Care Reform Has Taken OffHealth Care Reform Has Taken Off

Dorothy to the Wizard: Come back! Come back! Don't leave without me! Come back!

Wizard of Orszag: I can't come back! I don't know how it works! Good-bye folks!

An incentive payment of $400M for providers in the 25% of counties that have the lowest Medicare expenditures

Payments for Payments for ““Efficient CountiesEfficient Counties””

Lowest Spending

Highest Spending

Medicare per Enrollee

Incentive payments of up to 2% for physicians and Incentive payments of up to 2% for physicians and hospitals that attain hospitals that attain ““efficiency standardsefficiency standards””

developed by the Secretary.developed by the Secretary.

Payments for Payments for ““ValueValue””

Advocacy states

Other Low Medicare States

Penalties for Hospital ReadmissionsPenalties of 3% to 5% for hospitals withPenalties of 3% to 5% for hospitals with

““excessexcess””

levels of levels of ““preventablepreventable””

readmissions.readmissions.

Reductions in Disproportionate Share PaymentsReductions in Disproportionate Share Payments

Lowest DSH

Highest DSH

$20B reduction in DSH over 9 years, $10B yearly therafter

“The IOM will recommend strategies for addressing geographic variation by altering payments for physicians and hospitals.”

Institute of Medicine (IOM) Institute of Medicine (IOM) Study of Geographic VariationStudy of Geographic Variation

Geographic variation in health care is principally related to geographic differences in poverty.

Payment changes made according to geographic norms would result in major harm to low-income patients and the providers who care for them.

ConclusionsConclusions**************************************************

ThoTho' a man may be in doubt of what he know,' a man may be in doubt of what he know, very quickly he will fight to prove very quickly he will fight to prove

that what he does not know is so.that what he does not know is so.

King of SiamKing of Siam

PHYSICIANS AND HEALTH CARE REFORM

Commentaries and Controversies

Visit Visit http://http://buzcooper.combuzcooper.com