SOCIAL DETERMINANTS OF HEALTH INEQUALITIES: MEASUREMENT &

INTERVENTION

Angela G. Reyes, MPHFounding and Executive Director

Detroit Hispanic Development Corporation

Amy J. Schulz, PhD, MPHProfessor, Department of Health Behavior and Health Education

University of Michigan School of Public Health

Presentation for Council of Michigan FoundationsSeptember 19, 2016

Objectives• Describe social determinants of health equity• Consider implications of social determinants of health for interventions to promote health & health equity

• Discuss four brief case examples of interventions that address social determinants of health, and evaluation/measurement of effects.

2

SOCIAL DETERMINANTS OF HEALTH

3

Social determinants of health

•Social & economic & physical conditions under which people are born, live, work, learn & age, & which determine their health

•These conditions determine the availability of resources that are necessary to maintain health.

5Introduction

6Introduction

7

HEALTH DISPARITIES VS.HEALTH INEQUITIES

8

“HEALTH DISPARITIES”

“BROADLY DEFINED AS POPULATION-SPECIFIC DIFFERENCES IN HEALTH INDICATORS”

“most dictionaries define disparity as: inequality; difference in age, rank, condition, or excellence.”

9Introduction

Carter-Pokras, O. and Baquet, C. “What is a Health Disparity?” PHR. Vol 117. September–October 2002. pp. 425-434.

Health Inequities are inequalities that are related to differences in health status or medical treatment that are unfair to disadvantaged people and

that are avoidable.

Source: Braveman and Tarimo, Soc Sci and Med:54:1621-1635 (2002). Image from “Unnatural Causes: When the Bough Breaks”.

10Introduction

11

4% 4%

8%

13%12%11%

27%

24%

0%

5%

10%

15%

20%

25%

30%

White Asian or PacificIslander

Black Hispanic

High School Dropouts Below the Poverty Level

An Enduring Relationship Exists Between Race and Income/Educational Levels

Sources: US Census Bureau, Statistical Abstract of the United States: 2014; US Department of Education, National Center for Education Statistics. 2014. The Condition of Education 2014.

There is also an enduring relationship between various demographic and social

factors and health

12

Self-Reported Health and Activity Limitation by Level of Education, 2011

13

35%

19%

14%

7%

31%

25% 25%

19%

0%

10%

20%

30%

40%

Less than High School High School or Equivalent Some post-High School College Graduate

Fair/Poor Health Activity Limitations (all causes)

Source: Behavioral Risk Factor Surveillance System, Prevalence and Trends Data, 2011. Accessed Apr. 19, 2015 at: http://apps.nccd.cdc.gov/brfss/page.asp?yr=2011&state=UB&cat=CH#CH.

http://apps.nccd.cdc.gov/brfss/page.asp?yr=2011&state=UB&cat=CH#CH

What do social determinants of health have to do with health inequities?

15

Health inequities occur when there are systematic differences in the distribution of social and economic resources – the social determinants of health – across communities or groups of people.

Differences in the distribution of these social determinants of health are largely responsible for health inequities.

Health and the Built Environment

16

The design of neighborhoods impacts residents’ health

Health and the Physical Environment

17

18

• Definition of Allostatic Load: “A measure of the cumulative physiological burden exacted on the body through attempts to adapt to life's demands.”

• Sources of stress include: Economic insecurity Job insecurity Lack of social support Inadequate child care Low-control jobs Racism Sexism Discrimination Unsafe neighborhoods Elements of the built environment

Allostatic Load: Stress and Health

Connection between Stress and Health• Neighborhood poverty higher stress poorer health

• People who live in disadvantaged neighborhoods are more likely to suffer heart attacks than people in middle-class neighborhoods

• People in neighborhoods with many abandoned buildings have higher rates of early death from cancer and diabetes

• Higher allostatic load is associated with significantly increased risk for 7-year mortality, declines in cognitive and physical functioning, increased risk for cardiovascular disease and metabolic disorders

19

Sources: Teresa E. Seeman, Bruce S. McEwen, John W. Rowe, and Burton H. Singer. Allostatic load as a marker of cumulative biological risk: MacArthur studies of successful aging. PNAS 2001 98: 4770-4775; Helen Epstein, New York Times, Ghetto Miasma: Enough to Make you Sick? 10/12/2003.; Denise Grady, New York Times, “Profiles in Science: Charting Her Own Course, 4/8/2013, http://www.nytimes.com/2013/04/09/science/elizabeth-blackburn-molecular-biologist-charts-her-own-course.html?pagewanted=1&_r=0&smid=tw-share

• Innovative research on telomeres• Short telomeres are linked to heart disease, diabetes, cancer – and

chronic stress• Ways to protect telomeres include through diet, exercise – and easing

emotional stress

http://www.nytimes.com/2013/04/09/science/elizabeth-blackburn-molecular-biologist-charts-her-own-course.html?pagewanted=1&_r=0&smid=tw-share

IMPLICATIONS FOR INTERVENING TO REDUCE HEALTH INEQUITIES

20

Social Determinants of Health Frameworks…• …open new possibilities for interventions to promote health• Interventions that mitigate the impact of social, economic or physical environmental conditions on people’s lives & health

• Interventions that directly address the social, economic and physical environmental conditions that affect health

21

Core Aspects of Effective Solutions• Place-based solutions.

• Assess community to identify the unique ways its environment impacts health outcomes. • Meaningful place-based solutions are holistic, focus on prevention, and engage community

members and partners from multiple sectors.

• Intentional focus on race, nationality, ethnicity, and culture. • Race affects where and how we all live, work and play. • Attention must be placed on addressing racial equity.

• Communication strategies. • Explain and amplify the problem• Highlight inequities with supporting data• Offer solutions.

• Policy and systems change.• Critical elements in sustaining health equity efforts and maintaining a culture of health.

22

HEALTH INEQUALITIES: CASE STUDY 1

INTERVENTIONS THAT RECOGNIZE SOCIAL DETERMINANTS OF HEALTH

23

Healthy Environments Partnership

Chandler Park Conservancy | Detroit Health Department | Detroit Hispanic Development Corporation | | Eastside Community Network | Friends of Parkside | Henry Ford Health System | Institute for Population Health |

University of Michigan School of Public Health | Community Members At-Large

A community-based participatory research partnership working together since 2000 to understand and promote heart health in Detroit. We examine aspects of the social & physical environment that contribute to racial & socioeconomic inequities in cardiovascular disease (CVD), and develop, implement & evaluate interventions to address them.

Age adjusted cardiovascular mortality rates and median household income

0100200300400500600700

Median Household income (in100's)Heart Disease Mortality Rate(per 100,000)

Community Planning Process:Building placed-based solutionsChallenges

• “There is no equipment – youth play basketball in the street”

• Local recreation centers closed• Places that are not clean• “immigrants don’t want to walk

outside – they feel vulnerable to the border patrol”

• “the wooded areas are dangerous –why take the chance?”

• Traffic – cars driving up and down the streets fast”

Facilitating Factors

• Outdoor community events• Dancing/fun• Activities for youth & families• Trails, parks & facilities that are safe

& easy to get to• More people out walking – more

likely to use the spaces• Support for walking• Organizations that support walking

and activity friendly spaces

CATCH Multilevel Intervention:Pathways to Heart Health

• Promote Walking• Promote Community Leadership & Sustainability

• Promote Activity Friendly Neighborhoods

Walk Your Heart to Health• Walking Group Aims:Promote heart healthy behaviors via

walkingProvide opportunities for other heart

health activity (e.g., food demos)Offer social support for heart healthy

activities

• Evaluation: Pre & Post Surveys (e.g., health indicators, attitudes, social support)

Pedometers – monitor stepsParticipant observationAttendance recordsSession summary sheets

What We Learned1. WALKING GROUPS INCREASE PHYSICAL ACTIVITY

Mean Number of Daily Steps Walked by WYHH Participants

4,729

5,800 5,796 5,751 5,711

6,993 6,956 6,893 6,839

9,899 10,097 10,161 10,221

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

11000

12000

Baseline 8 Weeks 16 Weeks 24 Weeks 32 Weeks

Steps on days participants did not walk with the group

Overall mean steps

Steps on days participants walked with the group

What We Learned

30.0

35.0

40.0

45.0

50.0

Baseline 8 Weeks 32 weeks

HBP

prev

alen

ce

(%)

2. WYHH WALKING GROUPS REDUCED CVD RISK FACTORS

Adjusted High Blood Pressure Prevalence Estimates for WYHH Participants with an Average Increase of 4000 Steps per Day

Chart2

Average StepsBaseline8 Weeks32 weeks47.40886302690741942.19978123325356742.199781233253567

HBP prevalence (%)

unadjusted prevalences

Unadjusted HBP prevalence ( Among the n=603 that have consistent particiaption and have complted all three instruments, final n=368)

Hypertensive

Baseline8 weeks32 weeksTotal

%44.639.748.4

Interpretation for HBP

Table 5b. Does Physical activity measures by number od steps reduce healtj outcomes? ( GEE models Using N=603)

(baseline 8 week and 32 weeks)

Mean Steps when walking with the group

ORStErr95%CIp-value

HBP1β0Intercept

β1age1.070.01(1.05,1.09)0.000

β2female0.881.38(0.47,1.67)0.702

β3black3.081.24(2.03,4.68)0.000

β4StepsV00.711.16(0.54,0.94)0.019

β5delta1_StepsV0.811.06(0.72,0.91)0.000

β6delta2_StepsV0.841.31(0.49,1.43)0.521

1: point estimates for Steps V0, delta1_Steps abd delta2_steps are evaluated ar 4000 steps

Interpretation of the parameter estimates int he model:

β4On average, Individuals that walked 4000 steps at baseline had a 29% likelihood of

not being hypertensive

β5One average, Individuals that from baseline to 8 weeks increased by 4000 steps had an

additional 19% likelihood of not being hypertensive

β6One average, Individuals that from 8 to 32 weeks increased by 4000 steps had an

additional 16% likelihood of not being hypertensive

Bar Plots

Walking with the WYHH GroupHBPStepsHBPBaseline8 Weeks32 weeks

EstimateStderrLowerCLUpperCLProbZInterpretation of β4 ( using H BP as an example)AverageAverage Steps47.442.242.2

Intercept-3.6880.545-4.755-2.6210.000Assumptions( A)

age0.0650.0090.0480.0820.000Mean age45

female-0.1250.325-0.7620.5130.702female1

black1.1240.2130.7061.5420.000Black1

Mean_V10-0.0000.000-0.000-0.0000.019Baseline Steps4000

delta1_StepsV1-0.0000.000-0.000-0.0000.000

delta2_StepsV1-0.0000.000-0.0000.0000.521HBP estimate-0.103738412

exp(.)0.901461078

P(HBP=1)0.4740886303

p47.408863026944.62.8088630269

At baseline those that walk more have significantly lower likelihood of having HBP

Interpretation of β5 ( using HBP as an example)

Assumptions( A)

Mean age45

gender1Adjusted HBP prevalence estimates for an average increase of 4000 steps per day for those that consistenly participated in the WYHH program

Race/ethnicity1

Baseline Steps4000

change in steps from baseline to 8 weeks4000Consistent participation: Individual that attended at least one session per week of the WYHH program

HBP estimate-0.3145775236

exp(.)0.730097258

P(HBP=1)0.4219978123

p42.199781233339.72.4997812333

Those that at baseline walked more and that from baseline to 8 weeks continue to walk more have significant reduction in their likelihood of having HBP

Interpretation of β6 ( using HBP as an example)

Assumptions( A)

Mean age45

gender1

Race/ethnicity1

Baseline Steps4000

change in steps from baseline to 8 weeks4000

change in steps from 8 to 32 weeks0

HBP estimate-0.3145775236

exp(.)0.730097258

P(HBP=1)0.4219978123

p42.199781233348.4-6.2002187667

Average StepsBaseline8 Weeks32 weeks47.40886302690741942.19978123325356742.199781233253567

HBP prevalence (%)

Sheet4

What We Learned

“I loved it! The people in the group and the Community Health Promoters, we became family...Everybody in my household walks, I changed my diet & lost weight. The program should never end…”

3. WALKING GROUPS strengthen social relationships

Changing Social & Physical Environments

• WYHH Network of Community Organizations to Support Walking Groups

• Strengthen Social Relationships/Social Capital• Supporting Walking Groups (SWAG)Training • Walking Group Capacity Building Mini-grants• Policy Advocacy Capacity Building Workshops

Next Steps• Entrepreneurial mindset in Detroit - unique opportunity.

• Self-sustaining models that maintain a focus on promoting walking in low resourced neighborhoods with high cardiovascular risks.

• Exploring corporate partnerships.• Foundation support for piloting & investigation phase.

33

HEALTH INEQUALITIES: CASE STUDY 2

INTERVENTIONS THAT REDUCE INEQUITIES IN ENVIRONMENTAL EXPOSURES AS A SOCIAL

DETERMINANT OF HEALTH

34

Community Action to Promote Healthy Environments (CAPHE)

Community ActCAPHE is Funded by the National Institute of Environmental Health Sciences – Grant # RO1ES022616 and by the Erb Family Foundation

36

37

38

Schulz, Mentz, Sampson et al, 2016. Race and the distribution of social and physical environmental risk. In press.

Implications for interventions• Interventions to mitigate adverse health effects of air pollutants

• Air filters in homes and schools to clean pollutants from the air• Land use policies that forbid siting homes or schools within 150 meters of freeways• Direct resources to communities experiencing greatest cumulative risk (e.g., community

benefits agreements, California policy for distributing $$ to communities with highest cumulative risk)

• Interventions that reduce exposure to air pollution• Reducing emissions from point sources (e.g., smokestacks on industrial facilities)• Reducing emission from mobile sources (e.g., retrofitting diesel truck engines)

40

Implications for measurement• Mitigation efforts

• Measure beneficial effects on health• Measure reductions in health inequities

• Reductions in air pollutants• Measure reductions in air pollutants, with particular attention to areas with high cumulative risk

• Measure reductions in health inequities

41

HEALTH INEQUALITIES: CASE STUDY 3

DIRECTLYADDRESSING SOCIAL DETERMINANTS OF HEALTH

42

Changing the Context: Addressing Intermediate Predictors of Cardiovascular Risk

• Public Works Project: Municipal investment in built & social environment in subset of neighborhoods serviced by Gondola

• Sample• Neighborhoods in Medellín serviced by Gondola that received public

works intervention• Comparable neighborhoods serviced by Gondola that did not receive

public works intervention• Study Design

• Pre-post comparison of intervention and comparable control neighborhoods

• Outcomes: Intervention vs. Control Neighborhoods• 66% decline in homicide rate• 75% decrease in reports of violence

Cerda et al., 2012

Health Effects of Interventions to Promote EquityCivil Rights Act (1964) & Voting Rights Act (1965)

• Sample– Non-Latino blacks & non-Latino whites

• Study Design: Compared national mortality data for:– 1955-1964 (before Civil Rights Act)– 1965-1974 (After Civil Rights Act)– Measures

• Racial & regional differences in sex-specific age-adjusted mortality due to heart disease, cerebrovascular disease, and cancer

• Findings– Significant decline in stroke & heart disease mortality

rate for non-Latino black women relative to non-Latino white women

– Health gains not seen for non-Latino black men

Kaplan, G. A., et al. (2008).

SUMMARY

45

Social Determinants of Health• Characteristics of the contexts in which we live, work, and play.• Inequalities in SDOH largely responsible for health inequities.• Implications for interventions:

• Reduce/mitigate adverse effects of unequal contexts (e.g., supporting physical activity even when environments are less conducive)

• Directly address the SDOH (e.g., infrastructure change, policy change to promote equity)• Multilevel: Simultaneously reduce adverse effects AND address the contexts themselves

(e.g., walking group intervention + complete streets legislation

46

Implications for assessing change• Assess process as well as impact

• Does the intervention process reinforce or challenge underlying inequalities? e.g., exclude those most affected from being part of the solution?

• Were efforts made to modify policy? [not just whether the policy actually changed]

• Timeline• Addressing social determinants of health may require longer funding

periods and measurement to capture change

47

John F. Kennedy, message to Congress, February 28, 1963

• "The Negro baby born in America today ... has about one-half as much chance of completing high school as a white baby born in the same place on the same day-one-third as much chance of completing college-one third as much chance of becoming a professional man-twice as much chance of becoming unemployed ... a life expectancy which is seven years less-and the prospects of earning only half as much."1

48

Social Determinants of Health Inequalities: Measurement & InterventionObjectivesSocial determinants of healthSocial determinants of healthSlide Number 5Slide Number 6Slide Number 7Health Disparities vs.�Health Inequities“Health Disparities”��“broadly defined as population-specific differences in health indicators” ��“most dictionaries define disparity as: inequality; difference in age, rank, condition, or excellence.”Health Inequities �are inequalities that are related to differences in health status or medical treatment that are unfair to disadvantaged people and that are avoidable.Slide Number 11There is also an enduring relationship between various demographic and social factors and healthSelf-Reported Health and Activity Limitation by Level of Education, 2011Slide Number 14What do social determinants of health have to do with health inequities?Health and the Built EnvironmentHealth and the Physical EnvironmentSlide Number 18Connection between Stress and HealthImplications for intervening to reduce health inequitiesSocial Determinants of Health Frameworks…Core Aspects of Effective SolutionsHealth Inequalities: caSe study 1��Interventions that recognize social determinants of healthHealthy Environments PartnershipAge adjusted cardiovascular mortality rates and median household incomeCommunity Planning Process:�Building placed-based solutionsCATCH Multilevel Intervention:�Pathways to Heart HealthWalk Your Heart to HealthWhat We LearnedWhat We LearnedWhat We LearnedChanging Social & Physical EnvironmentsNext StepsHealth Inequalities: caSe study 2��Interventions that reduce inequities in environmental exposures as a social determinant of healthCommunity Action to Promote Healthy Environments (CAPHE)Slide Number 36Slide Number 37Slide Number 38Slide Number 39Implications for interventionsImplications for measurementHealth Inequalities: caSe study 3��DIRECTly Addressing social determinants of healthChanging the Context: Addressing Intermediate Predictors of Cardiovascular RiskSlide Number 44SummarySocial Determinants of HealthImplications for assessing changeJohn F. Kennedy, message to Congress, February 28, 1963