Assessing relationships between discrimination and health: emphasis on measurement and methodology
Access StatusFull text of the requested work is not available in DASH at this time ("dark deposit"). For more information on dark deposits, see our FAQ.
Lawrence, Jourdyn Alexis
MetadataShow full item record
CitationLawrence, Jourdyn Alexis. 2021. Assessing relationships between discrimination and health: emphasis on measurement and methodology. Doctoral dissertation, Harvard University Graduate School of Arts and Sciences.
AbstractResearch investigating relationships between discrimination and health has grown over the past two decades, establishing a body of evidence that documents the adverse health outcomes associated with inequitable systems. Efforts to strengthen the available evidence in this area have called for increased attention to measurement, mechanisms, and methodology. Specifically, prior work has called attention to the need for future research to focus on identifying experiences that are salient in regard to impacting health, evaluating hypothesized processes of embodiment, and utilizing methodological approaches to strengthen causal inference. These topics serve as the foundation of the present dissertation.
In Chapter 2, I conduct a systematic meta-analysis to estimate the mean correlation coefficient between self-reported discrimination and molecular biomarkers among studies that have operationalized discrimination using the Everyday Discrimination Scale. This analysis aims to (1) provide an understanding of how experiencing discrimination may become embodied to result in poor health and to (2) maximize cross-study comparability by only assessing those relationships among studies using the same measurement of discrimination. Examining relationships between discrimination and intermediate indicators of physiological wellbeing such as biomarkers allows for the assessment of indicators of inflammation, stress, and accelerated aging that have been associated with adverse mental and physical health outcomes.
Literature examining relationships between discrimination and blood pressure have been inconsistent, with some studies observing positive correlations between measures of discrimination and elevated blood pressure and others finding null or inverse relationships. Differences in the associations between discrimination and blood pressure have also been observed by gender and indicators of socioeconomic status. Contributing to efforts to minimize threats to causal inference that could contribute to the differences in findings, such as measurement error and unmeasured confounding, I implemented instrumental variable (IV) estimation to assess the relationship between racial discrimination and blood pressure in Chapter 3. The analyses demonstrate that alternative methodological approaches, specifically IV, may be useful in accounting for potential measurement error and omitted variable bias. These findings contribute to a body of research that demonstrates the adverse effects of institutional discrimination on health and provide context for intervention.
Defined as accumulated “wear and tear” of physiologic systems due to exposure to chronic stress, allostatic load serves as a useful outcome to assess the system-wide impact of psychosocial stressors like discrimination. In much of the literature, allostatic load is evaluated as a summary index, however, this may obscure specific physiologic responses important to understanding the pathways through which discrimination contributes to adverse health outcomes. Using three measures of discrimination (i.e., everyday, lifetime, and appraised burden of discrimination), I assessed whether each form of discrimination operated distinctly or potentiated associations with other forms of discrimination (e.g., greater frequency of everyday discrimination and appraisal of discrimination as a significant burden) to heighten allostatic load. This investigation adds to the existing literature by identifying the extent to which the relationship between discrimination and allostatic load varied by measure used. Results from this analysis suggest distinct mechanisms through which everyday and major lifetime experiences or appraisals of discrimination contribute to allostatic load to impact mental and physical health outcomes.
Taken together, the findings from these three analyses provide guidance for future research, specifically regarding pathways through which discrimination may adversely impact health, methodological approaches used, and the importance and theoretical implications of how discrimination is measured and utilized.
Citable link to this pagehttps://nrs.harvard.edu/URN-3:HUL.INSTREPOS:37368381
- FAS Theses and Dissertations