03. Quality Measurement and Report: Implications for Disparities

www.CenterForUrbanHealth.org

Using Quality Measurement and Reporting to Confront

Disparities

Yiscah Bracha, M.S.Research Director

Center for Urban Health at HCMCMinneapolis Medical Research Foundation


Today’s presentation will discuss:

• Goals & presumed mechanisms of QM&R• Reason to use QM&R to address disparities

Locus of Minnesota’s problems in population health

Demographic changes in the state

• Ways to use QM&R to address disparities Disparities-relevant measures Disparities-relevant reports

Stratified measures Structure reports to favor providers who do most with

least

• Conclusions



• Goals & presumed mechanisms of QM&R • Reason to use QM&R to address disparities

Locus of Minnesota’s problems in population health

Demographic changes in the state



least

• Conclusions


Goal of Quality Measurement & Reporting

• Improve population health by• Improving the quality of medical care

delivered to the population• Assumed mechanisms:

Individual patients choose providers of highest reported quality

Providers improve quality in order to earn: Increased market share Improved public image Bonus payments from health plans


When does care most matter to health?

• Improved quality of medical care makes the most difference to health among those: Who are acutely ill With complex chronic disease With lifestyles and exposures that place

them at high risk for ill health


Who is most sick and at risk in Minnesota?

• Racial and ethnic minorities• Persons of low SES


Premature mortality in MN by race:


Child health indicators in MN by race

02468

1012141618

Inadequateprenatal

care

Lowbirthweight

babies

Infantmortality

rate

Rat

es

White

Asian

Hispanic

African American

Native American

*Source: Minnesota Department of Health, Spring 2006


Indicators of SES by Minnesota race

0

10

20

30

40

50

60

70

80

Poverty EmployerInsurance

Medicaid

Pe

rce

nt

of

Min

ne

so

tan

s

White

Black

Hispanic

Other

46 14 23 12 6 16 11 8 39 4 n/a 6

Numbers inside bars represent Minnesota’s rank among states.Employer-sponsored insurance and Medicaid for non-elderly.


Ways to improve MN’s overall health:

• Improve quality of medical care for majority population, which already is healthiest in the nation?

-OR-

• Improve quality of medical care for minority populations, which have some of the lowest health indicators in the nation?



• Goals & presumed mechanisms of QM&R • Reason to use QM&R to address disparities

Locus of Minnesota’s quality problem Demographic changes in the state



least

• Conclusions


Growth in MN non-white population:

Source: Minnesota State Demographic Center, August 2006


Change in MN youth population:



Projected changes in MN population:



Conclusions:

• Our state is rapidly diversifying• Much more diversity expected in the

future• Reasons to target resources to disparities:

Justice: Gaps are indefensible Efficiency: Direct resources to places where

there is most room to improve Sustainability: As the state grows more

diverse, the minority in poor health may become the majority







least

• Conclusions


How QM&R could address disparities:

1. Help low-income patients use reports 2. Develop disparities-relevant measures

3. Develop disparities-relevant reportsa. Stratify reports to reveal disparitiesb. Structure reports to reward providers who

i. Do the best with the most challenging patientsii. Do the best with the most limited resources


Helping ptts use reports (?)

• Empirical Q: Do patients switch providers on the basis of quality reports? Research: Few patients consult reports. Workers switch health plans on the basis of

cost, not reported quality

• Normative Q: Should patients switch providers on the basis of quality reports? Many say no. Switching disrupts continuity,

which is necessary for quality



Help low-income patients better use reports

1. Develop disparities-relevant measures

2. Develop disparities-relevant reportsa. Stratify reports to reveal disparitiesb. Structure reports to reward providers who

i. Do the best with the most challenging patientsii. Do the best with the most limited resources


Goal of Quality Measurement & Reporting

• Improve population health by• Improving the quality of medical care

delivered to the population• Assumed mechanism:

Individual patients choose providers of highest reported quality

Providers improve quality in order to earn: Increased market share Improved public image Bonus payments from health plans




2. Develop disparities-relevant reportsa. Stratify reports to reveal disparitiesb. Structure reports to reward providers

whoi. Do the best with the most challenging

patientsii. Do the best with the most limited resources


1. Measures relevant to disparities

• Diversity measures: % patients served proportionate to

demographics in community % health care workers with demographics

proportionate to those in community

• Access measures: Cancelled appointment rates Availability of transportation and child care % patients served who are uninsured or MA


Other measures relevant to disparities:

• Patient-centeredness. Develop indicators of good care specific to: Multiple chronic conditions Gender and age Patient stated preferences for

aggressive vs. conservative medical therapy








2a. Stratify reports

• For all measures, show outcomes within strata such as: Race/ethnicity Estimate of SES (from census data) Number of co-morbidities

• For all strata, show % patients served within stratum








2b. Structure of reports

• All structure decisions favor some at the expense of others; thus choice of structure reflects normative values.

• Two critical dimensions of structure: Use raw outcomes vs. outcomes

adjusted by patient characteristics Display attainment of absolute threshold

vs. attainment of improvement


Who is favored by what structure decision?

Outcome measure is:

Reward Based on Achieving

Absolute target Improvement

Unadjusted

High resource providersHigh resource patients

Low resource providers High resource patients

Adjusted or stratified:

High resource providersAny kind of patient

Low resource providersAny kind of patient

Observed to Expected

High resource providersLow resource patients

Low resource providersLow resource patients


Decisions now favor:

Outcome measure is:



Unadjusted










Decisions could favor:

Outcome measure is:



Unadjusted















least

• Conclusions


Questions, answers & implications:

Question Answer Policy implicationsDo existing measures assess equity or equality in quality?

NoDevelop & use new measures relevant to disparities.

Does patient race & SES affect MNCM outcomes measures?

Very likely. Known that low SES worse outcomes

Stratify reports by SESUse SES to risk-adjust or calculate observed-to-expected outcomes

Which non-medical agents affect MNCM outcome measures?

Family, patient, community, public policies

Reimbursement higher when contributions from non-medical agents are low


The Bad News:

• Minnesota has a disparities problem• If not addressed, this problem will:

Challenge our commitment to equality Waste health improvement resources by not

directing them to the places they can do the most good

Undermine the future vitality of the state, as low-income, minority populations continue to grow

• Quality measurement & reporting methods Currently do not address the problem May exacerbate it


The good news:

• Quality measurement & reporting framework is state-of-the-art: Excellent cooperation among health

plans Strong support from business and state Willingness to address the disparities

issue

• We can utilize the existing framework to address disparities


This problem is solvable!Let’s start.

Extra slides


Chronic Care Model

from E.H. Wagner 1998. What will it take to improve care for chronic illness? Effective Clinical Practice. 1(1):2-4


Visioning a new reimbursement structure:

• Based on episodes of care

• Fosters collaboration and mutual accountability among all responsible actors: Schools and community based social agencies Municipalities & counties (e.g. public health

impact of development decisions) State (e.g. MA eligibility & reimbursement

policies)


How healthy is Minnesota?

• Minnesota has ranked as one of the top two healthiest states since1990*

• According to United Health Foundation, our strengths include: Low uninsurance rate Low CVD death rate Low premature death rate Low infant mortality rate

* Source: United Health Foundation’s America’s Health Rankings.


MN Health Strengths by Race*

*Source: Minnesota Department of Health, Spring 2006.


MN population growth rates by race:


Changes in MN demography by county


Some answers to empirical Qs:

• Effect of patient characteristics on measures of diabetes quality: Low SES patients have higher rates of smoking,

higher BP, higher chol, higher HbA1c. Reductions in HbA1c less likely in patients with

multiple chronic conditions, have diabetes of longer duration, youngest & oldest, racial minorities, low SES.

Risk-adjusting provider report card by patient SES can eliminate apparent outliers

• Strength of this knowledge claim: Very good.

03. Quality Measurement and Report: Implications for Disparities

Health & Medicine

disparities structure

basis of quality reports

measures structure reports

ptts use reports

disparities yiscah bracha

health plans

ill health

urban health