Income Inequality and Health in Washington State

Transcription

1 Slide 1 of 31 Income Inequality and Health in Washington State Donald L. Patrick, Jesse J. Plascak, Shirley A.A. Beresford University of Washington Autumn Quarterly Meeting Biobehavioral Cancer Prevention and Training Program October 31 st, 2014 ** Funded in part by National Cancer Institute Biobehavioral Cancer Prevention and Control Training Grant (R25CA092408) and P50CA148143

2 Slide 2 of 31 Outline 1. Background 2. Gini Index and top 1% 3. Associations between SES Inequality and health 4. Conclusions

3 Slide 3 of 31

4 Slide 4 of 31 Why the interest in income inequality? Thomas Piketty and Capital in the 21 st Century Public increasingly concerned about wealth gap Public confidence in economy waning, ego opportunity viewed as finite or zero sum game The extent of and continuing increase in inequality in the United States greatly concerns me...i think it is appropriate to ask if this trend is compatible with values rooted in our nation's history, among them the high value Americans have traditionally placed on equality of opportunity Janet Yellen, Federal Reserve Chairwoman

5 Slide 5 of 20 Is income inequality bad for population health?

6 Slide 6 of 31 Previous Literature Higher income inequality associated with all cause mortality, self rated health, depressive symptoms, hypertension, smoking, BMI, disability status Sensitive to geographic scale and lag effects Strongest evidence of association at state level Evidence of 15+ year time lag

7 Slide 7 of 31 A Theory of SES, SES Inequality and Health Health Income ($1,000) If $10,000 is transferred from individual with income=$80,000 and health=95 to individual with income=$20,000 and health=50, Population health gain of 15 (health income=$70,000+ health income=$30,000; (94+66) (95+50)=15) Adapted from Subramanian and Kawachi, Epidemiol Rev 2004

8 Slide 8 of 31 Theory of SES, SES Inequality and Health Socioeconomic Inequality Individual socioeconomic status Health

9 Slide 9 of 31 Measuring Income Inequality and Health Jesse Plascak

10 Slide 10 of 31 Gini Index and the Lorenz Curve The average of the absolute differences between all pairs of scores, divided by twice the mean Gini = A/(A+B) Cumulative % of Income A B Cumulative % of Individuals Top 1%

11 Slide 11 of 31 Gini Index Shortcomings Viable data U.S. Census: household, gross income (but has whole distribution) IRS: left truncated of (but IRS can account for redistributive tax systems [net income]) More sensitive to change in middle incomes Non decomposable inequality in every subgroup does not necessarily lead to Gini

12 Slide 12 of 31

13 Slide 13 of 31

14 Slide 14 of 31 Trends in Income Inequality and Health Shirley Beresford

15 50 45 Income Inequality (Top 1%) in the U.S., Washington, North Dakota, Alaska, New York, and Delaware, Slide 15 of 31 % OF PRE TAX INCOME HELD BY TOP 1% YEAR U.S. WA ND (Lowest) AK (2nd lowest) NY (2nd highest) DE (Highest)

16 LOG INFANT MORTALITY RATE (PER 1,000) 80 Slide 16 of 31 Top 1% and Infant Mortality Rate by Race, , United States % OF PRE TAX INCOME HELD BY TOP 1% YEAR Infant Mortality Rate All Races Infant Mortality Rate non White Top 1% Infant Mortality Rate White Infant Mortality Rate Black

17 INFANT MORTALITY EVENT RATE RATIO Slide 17 of 20 Gini index and Infant Mortality Rate Ratio Disparity by Race, , United States GINI INDEX YEAR Rate Ratio Black vs White Rate Ratio non White vs White Gini

18 GINI INDEX OF PRE TAX INCOME Income Inequality , Pre tax Washington State Slide 18 of % OF PRE TAX INCOME HELD BY TOP 1% YEAR Gini % of Income Held by Top 1%

19 70 Top 1% and Infant Mortality Rate , Washington State Slide 19 of INFANT MORTALITY RATE (PER 1,000) % OF PRE TAX INCOME HELD BY TOP 1% YEAR Infant Mortality Rate (per 1,000) Top 1%

20 70 Gini Index and Infant Mortality Rate , Washington State Slide 20 of INFANT MORTALITY RATE (PER 1,000) GINI INDEX YEAR Infant Mortality Rate (per 1,000) Gini

21 100 Slide 21 of 31 Gini Index and Infant Mortality Rate (Log) by Race, , Washington State 0.65 LOG INFANT MORTALITY RATE (PER 1,000) GINI INDEX YEAR Infant Mortality Rate All Race Infant Mortality Rate Black Infant Mortality Rate American Indian Infant Mortality Rate White Infant Mortality Rate Asian Gini

22 Gini index and % Poor/Fair Health, Washington State Slide 22 of % POOR/FAIR HEALTH YEAR % Poor/Fair Health Gini GINI INDEX

23 Slide 23 of 31 % < 100% FPL by County,

24 Slide 24 of 31 County Level Washington State Health Disparity Indicators by SES % Poor/Fair Health (N=39) Low Birth Weight (<2500 g) (N=36) Teenage Birth Rate (N=39) Corr Coefficient Corr Coefficient Corr Coefficient (P value) (P value) (P value) % < 100% FPL 0.38 (0.015) 0.29 (0.082) 0.38 (0.017) Median Household Income 0.45 (0.004) 0.10 (0.567) 0.27 (0.101) % < High School Diploma 0.74 (<.001) 0.48 (0.003) 0.93 (<.001)

25 Slide 25 of 31 Associations with Poor/Fair Health, Household Income Washington State, < $15, OR (95% CI) P value $15,000 $25, ( ) < $25,001 $35, ( ) < $35,001 $50, ( ) < $50, ( ) < 0.001

26 Slide 26 of 31 Associations with Poor/Fair Health, Washington State, OR (95% CI) P value Education Attainment < High School Diploma 1.00 High School Diploma 0.42 ( ) < Some College / 0.30 ( ) < Associate s Degree Bachelor s Degree 0.14 ( ) < 0.001

27 Slide 27 of 31 Conclusions and Discussion Donald Patrick

28 Slide 28 of 31 Conclusions Internationally, higher income inequality higher infant mortality In the US, income inequality has steadily risen over the last 40 years In the US, income inequality is not consistently associated with infant mortality disparities by race in different time epochs In WASHINGTON STATE, income inequality not consistently associated health indicators in multilevel analyses HOWEVER< lower SES indicators (poverty, median income, and education) are associated with poor/fair health at both a county and individual level Multilevel studies of income inequality are difficult or impractical yet Income inequality is important to our notions of equity and justice

29 Slide 29 of 31 Policy Implications Interventions at policy and individual level may be best directed toward reducing poverty and increasing education for all Distributive Justness: John Rawls Each person as an equal right to the most extensive basic liberty compatible with a similar liberty for others Social and economic inequalities are to be arranged so they are to the greatest benefit to the least advantaged members of society

30 Slide 30 of 31 Data Sources Internal Revenue Service household income inequality by year Mark Frank of Sam Houston State University (State data) Immanuel Saez of University of California, Berkeley (U.S. data) U.S. Census Bureau American Community Survey County level % < 100% federal poverty level, median household income, % < high school diploma, income Gini Behavioral Risk Factor Surveillance System Individual level % poor/fair health, household income, education attainment, age, sex National Center for Health Statistics Infant mortality rate by race and year (U.S.) Low birth rate, teenage birth rate (Washington State) Washington State Department of Health, Center for Health Statistics Infant mortality rate by race and year The World Bank (international data) Infant mortality and income Gini

31 Thank you! Slide 31 of 31