ami_hospitalizations

PAA 2021 Slide Deck

Transcript

1. Redlining, Segregation and Acute Myocardial Infarction Hospitalization in St. Louis, Missouri Christopher Prener, Ph.D. Assistant Professor of Sociology Saint Louis University
2. AGENDA 1. Background 2. Data & Methods 3. Redlining’s Legacy in St. Louis 4. AMI Hospitalizations in St. Louis 5. Redlining, Segregation, and AMI Hospitalizations 6. Final Thoughts
3. 1 BACKGROUND
4. 1. BACKGROUND FUNDAMENTAL CAUSE THEORY segregation (Williams & Collins 2001 and Sewell 2016) structural racism (Gee & Ford 2011) systematic racism (Phelan & Link 2015) racial capitalism (Pirtle 2020)
5. 1. BACKGROUND KEY TERMS CHD = Coronary Heart Disease Reduced blood flow to the heart, commonly caused by the the build-up of plaque in the heart’s arteries (known as atherosclerosis). AMI = Acute Myocardial Infarction A consequence of CHD where decreased or stopped blood flow causes damage to the heart muscle.
6. 1. BACKGROUND CHD + AMI disparities for African Americans in risk factors (Graham 2015) higher rates of CHD morbidity (Singh et al. 2015) higher rates of CHD mortality (Nasir et al. 2007) higher rates of AMI (Chi et al. 2020)
7. 1. BACKGROUND FOCUSING ON SOUTHERN CITIES Urban sociology has focused on a relatively small number of cities, and we often view them as a research site rather than an institution. We need to broaden literatures into the literal and figurative American South and produce deeper literatures on specific cities.
8. ST. LOUIS: LABORATORY FOR RACISM DEED COVENANTS EXCLUSIONARY ZONING “SLUM” CLEARANCE REDLINING Clockwise from Upper-left: Erenow; Google; Wikipedia; Missouri Bar Association
9. REDLINING A - “Best” B - “Still Desirable” C - “Definitely Declining” D - “Hazardous”
10. REDLINING A - “Best” B - “Still Desirable” C - “Definitely Declining” D - “Hazardous” “In St. Louis, the white middle class suburb of Ladue was colored green because…it had ’not a single foreigner or negro.’” (Rothstein 2017) Rothstein, Richard. 2017. The Color of Law. New York, NY: W.W. Norton & Co.
11. REDLINING A - “Best” B - “Still Desirable” C - “Definitely Declining” D - “Hazardous” Rothstein, Richard. 2017. The Color of Law. New York, NY: W.W. Norton & Co. “Lincoln Terrace was colored red because ‘it had little or no value today…due to the colored element now controlling the district’” (Rothstein 2017)
12. 2 DATA & METHODS
13. DATA & METHODS 2011-15 AMI Hospitalization Rates (MOPHIMS) 1940 HOLC Boundaries (Gordon 2009) 2011-15 ACS Demographic Data Data Sources Descriptive Mapping Analytical Steps Analysis Platform Local Spatial Autocorrelation (LISA) Correlations OLS Models* * No Residual Autocorrelation
14. 2. DATA & METHODS MEASURING SEGREGATION The Index of Concentration at the Extremes (ICE) provides a sub-county measure of segregation that produces scores per feature from -1 (total segregation of the marginalized group) to 1 (total segregation of the privileged group). Formula: ICEi = (Ai - Pi )/Ti Where: Ai = Privileged [white] Pi = Marginalized [Black] Ti = Total Population Massey, Douglas. 2001. “The prodigal paradigm returns: ecology comes back to sociology.” Pp. 41-48 in Does It Take a Village? Community Effects on Children, Adolescents, and Families, edited by A. Booth and A. Crouter. Mahwah, NJ: Lawrence Erlbaum Associates. Krieger, Nancy, et al. 2017. "Measures of local segregation for monitoring health inequities by local health departments." American Journal of Public Health 107(6): 903-906.
15. 3 REDLINING’S LEGACY IN ST. LOUIS
16. QUESTION 1 Are redlining and contemporary segregation linked? I hypothesize that there is a link between redlining and contemporary segregation, with redlined neighborhoods experiencing higher rates of isolation for African Americans. ?
19. No Population
20. 3. REDLINING’S LEGACY IN ST. LOUIS CORRELATIONS There is an “arc” over time in the strength of the relationship between increased percent redlined and ICE values (remember -1 is highest segregation for African Americans). Year Pearson’s r with Redlining 1940 -0.141* 1950 -0.235*** 1960 -0.468*** 1970 -0.573*** 1980 -0.565*** 1990 -0.566*** 2000 -0.555*** 2010 -0.475***
21. 3. REDLINING’S LEGACY IN ST. LOUIS CORRELATIONS Increased redlining is moderately associated with increased segregation for Black residents (remember -1 is highest segregation for African Americans). Variable Pearson’s r with Redlining AMI Hospitalization Rate 0.145* ICE, Race (2010) -0.493*** Median Income -0.510*** % in Labor Force -0.209*** % Owner Occupied -0.543*** % Vacant 0.629***
22. 4 AMI HOSPITALIZATIONS IN ST. LOUIS
23. QUESTION 2 Are there geographic patterns in AMI hospitalization in St. Louis? Based on the literature on heart disease and AIM, I hypothesize that clusters are present in parts of St. Louis City and County that HOLC redlined. ?
26. 4. AMI HOSPITALIZATIONS IN ST. LOUIS CORRELATIONS Weak relationship between AMI hospitalization and redlining, every so slightly stronger relationship with segregation (remember -1 is highest segregation for African Americans). Variable Pearson’s r with AMI Hosp % Redlined, 1940 0.145* ICE, Race (2010) -0.354*** Median Income -0.350*** % in Labor Force -0.395*** % Owner Occupied -0.125* % Vacant 0.384***
27. 5 REDLINING, SEGREGATION, AND AMI HOSPITALIZATIONS
28. QUESTION 3 After accounting for contemporary segregation and socioeconomic status, is redlining a salient factor in shaping AMI hospitalization patterns? I hypothesize that redlining does have an independent effect on hospitalizations after controlling for other factors. ?
29. OLS MODELS (1) (2) (3) % Redlined, 1940 0.016* (0.006) -0.004 (0.007) -0.015* (0.007) ICE, Race -2.236*** (0.372) -0.142 (0.482) Median Income -0.0001*** (0.00001) % in Labor Force -0.131*** (0.027) % Owner Occupied 0.083*** (0.015) % Vacant 0.125*** (0.032) Constant 8.830*** (0.306) 9.993*** (0.349) 14.976*** (2.201) AIC 1723.054 1690.667 1648.426 BIC 1734.185 1705.509 1648.426 Observations 302 302 302 Adjusted R2 0.018 0.120 0.316 Residual Std. Error 4.167 (df = 300) 3.943 (df = 299) 3.478 (df = 295) F Statistic 6.424* (df = 1; 300) 21.617*** (df = 2; 299) 24.159*** (df = 6; 295) Note: * - p < 0.05; ** p < 0.01; *** - p < 0.001
30. 6 FINAL THOUGHTS
31. 6. FINAL THOUGHTS DISCUSSION ▸ Question 1 - there is a moderate relationship between redlining and contemporary segregation. ▸ Question 2 - there is concentration of AMI hospitalization rates spatially, and some overlap with areas redlined by HOLC. ▸ Question 3 - redlining’s relationship remains with AMI hospitalization after accounting for other factors, but is weak. ▸ Suggests there may be some residual legacy of historical racism that continues to shape contemporary outcomes, and would be consistent with the clinical literature (reduced treatment despite increased morbidity/ mortality)
32. 6. FINAL THOUGHTS LIMITATIONS ▸ Low overall amount of variation explained - potential for omitted variable bias ▸ The overlaps between redlining and AMI hospitalizations look different between St. Louis City and St. Louis County - disaggregating models is necessary ▸ Detailed, patient-level data is needed to further explore these patterns but significant barriers remain to building such a data set.
33. ST. LOUIS: LABORATORY FOR RACISM DEED COVENANTS EXCLUSIONARY ZONING “SLUM” CLEARANCE REDLINING Clockwise from Upper-left: Erenow; Google; Wikipedia; Missouri Bar Association
34. Slides available via SpeakerDeck speakerdeck.com/chrisprener/ami-hospitalizations Raw data, code available via GitHub github.com/PrenerLab/ami_hospitalizations chris.prener@slu.edu chris-prener.github.io LEARN MORE THANKS FOR COMING! @chrisprener chrisprener.substack.com