Developmental histories of perceived racial discrimination and diurnal cortisol profiles in adulthood: A 20-year prospective study
Introduction
Racial and ethnic disparities exist across a wide range of adult health conditions (Williams and Collins, 1995, Mensah et al., 2005, Myers, 2009, Williams and Mohammed, 2009). Differing health care access and health behaviors do not appear to fully account for these disparities, leading investigators to propose that race-based social stress such as perceived racial discrimination (PRD) may play a role, by way of its influence on stress biology (Kuzawa and Sweet, 2009, Williams and Mohammed, 2009). The hypothalamic–pituitary–adrenal (HPA) axis and its primary product cortisol are frequently implicated in theoretical models of race-based stress and health (Myers, 2009). Racial/ethnic differences have been found in diurnal cortisol rhythms (Cohen et al., 2006, DeSantis et al., 2007), and perceived discrimination has been associated with altered basal/diurnal levels of cortisol in past research (Kaholokula et al., 2012, Zeiders et al., 2012).
Most existing research on discrimination and cortisol, including the current study, has focused primarily on interpersonal forms of discrimination, including daily hassles and microaggressions related to race/ethnicity (Harrell, 2000, Sue et al., 2007). Past research on discrimination and cortisol has also been cross-sectional, focusing on current or recent PRD, rather than cumulative histories of past PRD exposure. In the current study, we examined the impact of cumulative exposure to PRD assessed prospectively over a 20-year period. We also examined whether PRD experiences measured in adolescence were more strongly related to adult cortisol than PRD experiences measured in early adulthood. We controlled for potential confounds related to cortisol levels and/or discrimination experiences, including health behaviors, socioeconomic variables, and depressive symptoms. We also tested whether effects of histories of PRD on cortisol are stronger for Black adults and for individuals with lower socioeconomic status. Before describing our methods, we give a brief introduction of perceived discrimination and on HPA axis activity, then review past research on associations between them. Methodological issues, such as the importance of considering developmental timing, socioeconomic context, and potential confounds are discussed. Finally, the current project is described.
Perceived discrimination involves an individual perceiving that they are receiving or have received unfair treatment on the basis of membership in a group (Tajfel, 1982, Fishbein, 1996, Brown and Bigler, 2005, Major and Kaiser, 2008). When asked about perceived discrimination across a range of categories (e.g., gender, race, ethnicity, age, religion, physical appearance, sexual orientation), racial and ethnic minorities in the U.S. describe higher levels of discrimination than their White counterparts (Kessler et al., 1999). Due to the fact that racial/ethnic categorization is relatively stable, and as result of historical mistreatment and oppression based on race within U.S. society, discrimination based on race/ethnicity (perceived racial/ethnic discrimination, PRD) may have particularly strong effects on the well-being of racial/ethnic minority individuals, including Black Americans (Branscombe et al., 1999, Harrell, 2000, Feagin et al., 2001).
Basal cortisol levels follow a strong circadian or diurnal rhythm, involving high levels upon waking, a substantial (50–60%) increase in the 30–40 min after waking (the cortisol awakening response or CAR), and a subsequent decline across the day, reaching a nadir around midnight (Pruessner et al., 1997, Adam and Kumari, 2009).
Periodic activation of the HPA axis is considered adaptive and necessary to cope with acute stress, and cortisol levels are particularly responsive to stress of a social-evaluative nature (Dickerson and Kemeny, 2004). Stress-related changes in several aspects of the diurnal cortisol rhythm have been identified. The CAR has been found to increase in the presence of acute daily stressors (Adam et al., 2006, Chida and Steptoe, 2009, Fries et al., 2009). It has been found to be lower, however, in the presence of traumatic stress, particularly when accompanied by post-traumatic stress symptoms or disorders (Rohleder et al., 2004, Wessa et al., 2006). Chronic stress has also been associated with lower waking cortisol levels flatter diurnal cortisol slopes (Adam., 2012, Doane et al., 2013). Flatter cortisol slopes have been linked to higher depression (Doane et al., 2013), fatigue (Bower et al., 2005, Kumari et al., 2009), and cardiovascular disease (Matthews et al., 2006, Kumari et al., 2011). Chronic, and particularly traumatic stress, also predicts overall reductions in cortisol across the day, known as hypocortisolism (Heim et al., 2000, Fries et al., 2005, Miller et al., 2007). Hypocortisolism is associated with the presence of fatigue and pain syndromes, and overactivation of immune and inflammatory systems (Fries et al., 2009).
Past research has consistently found flatter diurnal cortisol rhythms in African Americans as compared to Whites (Cohen et al., 2006, DeSantis et al., 2007, DeSantis et al., 2015, Skinner et al., 2011, Martin et al., 2012). Higher PRD has been proposed as a potential mediator between race/ethnicity and flatter diurnal cortisol slopes (DeSantis et al., 2007). There are relatively few empirical studies of associations between PRD and diurnal cortisol rhythms. Some studies have examined major life events related to racism, although most have focused primarily on everyday interpersonal discrimination (daily race-related hassles and microaggressions).
One study of young adults (Skinner et al., 2011) assessed retrospective lifetime racism and racial daily hassles and found that discrimination predicted flatter diurnal cortisol slopes among both Black and White youth. Another study found associations between everyday PRD and flatter diurnal cortisol slopes among racial/ethnic minority young adults, but not racial/ethnic majority group members (Zeiders et al., 2014). This latter result is in line with prior evidence showing that racial/ethnic minorities are more sensitive to the effects of stress on the cortisol diurnal rhythm (DeSantis et al., 2015). In pregnant women, Suglia et al. (2010) found cumulative stress, including measures of major and everyday PRD, predicted lower morning cortisol and flatter waking to bedtime cortisol slopes for Black, but not Hispanic women. By contrast, one study of adults found that everyday discrimination predicted flatter diurnal cortisol slopes in White adults, but steeper diurnal cortisol slopes in Black adults (Fuller-Rowell et al., 2012b). Another study of preadolescents did not find significant associations between everyday discrimination and cortisol diurnal rhythms (Martin et al., 2012).
Less research has focused on PRD and average cortisol levels, however a study of Hawaiians found perceived racism to be associated with an overall lowering of cortisol levels across the day among Native Hawaiians (Kaholokula et al., 2012). Thus, with a few exceptions, existing research suggests that PRD is associated with a flattening of the diurnal cortisol rhythm and potentially an overall lowering of the diurnal cortisol curve across the day, with some evidence that effects are stronger for racial/ethnic minorities.
Prior empirical research on discrimination and cortisol has focused on recent discrimination, rather than taking into account histories of exposure. As a result, past work has not been able to assess effects of chronic PRD exposure, or the relative impacts of PRD exposure at different developmental stages. Life-span developmental theories suggests that experiences may have cumulative impacts on biology over time; events occurring during times of rapid developmental transition are, however, likely to have larger effects, becoming “built-in” to an individual’s changing biology or psychology (Halfon and Hochstein, 2002). The biological “embedding” of experiences during infancy and early childhood have received considerable attention (Hertzman, 1999, Shonkoff et al., 2009, Miller and Chen, 2013, Nelson, 2013). More recently, adolescence has been recognized as an additional sensitive period, in part due to notable changes in brain and neuroendocrine development during this time period (Spear, 2000, Chambers et al., 2003, Dahl, 2004).
Adolescence is also a key period in the development of identity (Kroger, 2006), and for racial/ethnic minority youth, developing a racial/ethnic identity becomes salient (French et al., 2006, Umaña-Taylor et al., 2014). The presence of a strong racial/ethnic identity, conferred in part through racial/ethnic socialization by parents, has been found to be protective against the negative impact of discrimination (Neblett et al., 2008, Neblett et al., 2012). For adolescents, however, racial/ethnic socialization and racial/ethnic identity development are still in progress (Boykin and Toms, 1985). Adolescents may also be less ready to employ important coping skills, ranging from support seeking to emotion regulation (Brondolo et al., 2009). PRD experiences may thus have stronger effects in adolescence as compared to adulthood, due to immature buffering mechanisms and coping resources. Adolescent PRD may also affect adult outcomes by becoming “built in” to developing biological and psychological systems (for example, by impacting self-esteem). As a result, we hypothesize that PRD experiences during adolescence will be a stronger predictor of adult HPA axis functioning than PRD experiences in young adulthood, particularly for Black Americans.
Racial discrimination and racial disparities in stress hormones are not experienced in a vacuum; they occur in particular neighborhood, family and socioeconomic contexts (Brondolo, 2015). The socioeconomic context in which perceived discrimination occurs may be particularly important in several ways (Harrell, 2000, Meyers, 2009). First, given that Blacks are more likely to live in lower income families, it is important to ensure that effects attributed to perceived racial discrimination for Blacks are not accounted for by the economic conditions of the family, and co-occurring stressors (Meyers, 2009). Second, the additional stress associated with living in a low-income environment may exacerbate the negative effects of perceived discrimination, multiplying the negative impacts of PRD. By contrast, individuals living in higher SES environments may be exposed to more or differing kinds of racial discrimination, due to exposure to more multiracial contexts (Harrell, 2000). Research should therefore both covary the effects of SES and consider interactions between SES and PRD when considering its relationships with stress biology and health outcomes (Harrell, 2000, Myers, 2009).
We examined relations between histories of PRD reported from adolescence through young adulthood and diurnal cortisol rhythms measured in adulthood. We accomplished this by adding diurnal cortisol measures to the Maryland Adolescent Development in Context Study (MADICS) (Eccles et al., 1997, Wong et al., 2003, Eccles et al., 2006). MADICS is a longitudinal study in which PRD was measured over a 20-year period from early adolescence through approximately age 32. In addition to examining average/cumulative measures of PRD over a 20-year period, we examined whether PRD in two different time periods – adolescence and young adulthood – differentially predicted adult diurnal cortisol profiles. We also examined whether associations between perceived discrimination and adult stress biology differed for self-identified Blacks and Whites, and whether effects were moderated by socioeconomic status. Finally, we examined whether racial disparities existed in diurnal cortisol rhythms, and whether histories of racial discrimination from adolescence through adulthood helped to explain these disparities.
Section snippets
Study overview
Participants and data were drawn from the MADICS Study, a longitudinal study of 1482 adolescents (n = 879 Black, 49% women) from Prince Georges County, Maryland (Eccles et al., 1997, Eccles et al., 2006, Wong et al., 2003). Participants were recruited in 7th grade, at age 12, and followed for 20 years, through approximately age 32. There were eight waves of data collection across the follow-up period, including assessments in the 7th grade (Waves (W) 1 and 2), 8th grade (W3), 11th grade (W4), 1
Descriptive information
Descriptive information on levels of cortisol, PRD, covariates, and contextual variables for the full sample and for Black and White participants are presented in Table 1. Independent sample t-tests revealed that Blacks had marginally lower waking cortisol (0.236 vs. 0.274 μg/dL) and significantly higher bedtime cortisol (0.085 vs. 0.050 μg/dL) than Whites. Racial differences were also apparent for time of waking; Black participants reported later waking times than White participants. Racial
Discussion
Our results suggest that greater developmental histories of PRD are related to a pattern of flatter diurnal cortisol slopes, lower waking cortisol levels, and lower average cortisol across the waking day. The pattern of flatter slopes with high PRD was present for both Blacks and Whites, whereas the lowering of waking and average cortisol levels was specific to Blacks reporting high PRD. These results are in accord with past studies that have found higher PRD to be associated with flatter
Contributors
EKA conceived of and designed the study, oversaw the acquisition of data and data analysis, and took the lead on the first draft and revisions of the paper. JAH ran the analysis, helped write the results section, prepared the tables and graphs and commented on multiple drafts of this paper. KHZ wrote part of the introduction, contributed to data analysis and commented on and edited drafts of this paper. JAR contributed to study design and commented on drafts of this paper. ECR oversaw data
Funding
The funding sources listed in the acknowledgements had no influence of study design, collection or analysis or interpretation of the data; they did not influence the writing of the report or the decision to submit the article for publication.
Conflicts of interest
None.
Acknowledgements
The first three waves of the study were funded by the MacArthur Network on Successful Adolescent Development in High Risk Settings (Chair: R. Jessor). Waves 4, 5 were funded by NICHD Grant #R01HD33437 to JSE and Arnold J. Sameroff. Wave 6 was funded by Spencer Foundation Grant MG #200000275 to Tabbye Chavous and JSE. Waves 7, 8 were funded by NICHD Grant #R01HD048970 to JSE and SCP and the Wave 8 biomarker data were funded by NIA Grant #RC2AG03678001 to JSE, SCP, EKA, JR, MK, and Wendy Berry
References (87)
Transactions among adolescent trait and state emotion and diurnal and momentary cortisol activity in naturalistic settings
Psychoneuroendocrinology
(2006)- et al.
Adverse adolescent relationship histories and young adult health: cumulative effects of loneliness, low parental support, relationship instability, intimate partner violence, and loss
J. Adolesc. Health
(2011) - et al.
Prospective prediction of major depressive disorder from cortisol awakening responses in adolescence
Psychoneuroendocrinology
(2010) - et al.
Assessing salivary cortisol in large-scale, epidemiological research
Psychoneuroendocrinology
(2009) - et al.
Diurnal cortisol rhythm and fatigue in breast cancer survivors
Psychoneuroendocrinology
(2005) - et al.
Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis
Biol. Psychol.
(2009) - et al.
Racial/ethnic differences in cortisol diurnal rhythms in a community sample of adolescents
J. Adolesc. Health
(2007) - et al.
Synthesis of a cortisol-biotin conjugate and evaluation as a tracer in an immunoassay for salivary cortisol measurement
J. Steriod Biochem.
(1992) - et al.
Ethnicity as a social context for the development of African–American adolescents
J Sch. Psychol.
(2006) - et al.
Free cortisol awakening responses are influenced by awakening time
Psychoneuroendocrinology
(2004)