1. There are four main ways in which researchers have tried to understand why there are racial disparities in mortality, and in health more broadly: the racial-genetic model, the health behavior model, the socioeconomic model, and the psychosocial stress model (Dressler et al also covers the structural-constructivist model, but we'll skip it for simplicity). Dressler et al (2005) have looked at how these models are defined, and whether these models fit the evidence for racial health disparities. Let's look at the racial-genetic model first. "Especially with respect to high blood pressure, there have been appeals by some researchers to racial-genetic factors to account for these disparities (Boyle 1970). For blood pressure, these appeals have been based on both the differences in hypertension prevalence between blacks and whites, and the gradient of blood pressure and hypertension prevalence in relation to skin color within the African American community (Harburg et al. 1978, Keil et al. 1977, Klag et al. 1991). Although poor birth outcomes are less often explicitly attributed to racial-genetic differences, some researchers have suggested as such (Wilcox & Russell 1990, and see critique by David 2001). In current literature it can be dif?cult to ?nd overt attribution of disease risk to a racial-genetic component, perhaps because of widespread knowledge of the critique of race as a biological construct (Cooper 1984, Montagu 1962, Kittles & Weiss 2003). More often than not, as noted above, differences associated with race or ethnicity are simply reported. "Prior to technological advances in molecular biology, evidence had accumulated that a racial-genetic explanation was untenable for broad population differences in blood pressure. Literature reviews showed that prevalence rates were extremely variable across populations in Africa and people of African descent in the New World. Later, focused research (eliminating alternative measurement hypotheses for these differences) demonstrated an east-west gradient in hypertension prevalence: West African samples had the lowest prevalence (16%), West Indian populations had an intermediate prevalence (26%), and African American populations had the highest prevalence (33%) (Cooper et al. 1997). With respect to birth weight, David & Collins (1997) found that the rate of low birth weight infants of African-born black women in Chicago (3.6%) was closer to that of U.S.born white women (2.4%) than to U.S.-born black women (7.5%). Similarly, Kleinman and associates (1991), using a national data set, found that the risk of neonatal mortality was 22% lower for foreign-born compared with U.S.-born black women, whereas there was no difference in risk for whites on the basis of birth country. More recently, Acevedo-Garcia et al. (2005) have shown that lower educated foreign-born black women do not differ in birth outcomes from highly educated U.S.-born black women (less well-educated U.S.-born black women had the highest risk of low birth weight). These data at least suggest that birth weight and blood pressure are subject to substantial environmental in?uence.... "With the advent of technology for identifying genetic variants, the importance of a racial-genetic component in blood pressure has become even less tenable because the search for variant gene structures that contribute to blood pressure has not been particularly successful (Crews & Williams 1999, Harrap 2003, Oparil et al. 2003); and those candidate genes that do appear to be associated with blood pressure are not differentially distributed across conventional racial groups (Cooper et al. 1999, Daniel & Rotimi 2003), nor do they differ between African Americans and ?rst-generation African immigrants (Bouzekri et al. 2004, Carlos Poston et al. 2001). "The racial-genetic model has not disappeared, however. Belief in its importance in the explanation of blood pressure disparities continues in the form of Grim's "slavery hypothesis" (Grim & Robinson 1996). This hypothesis posits that a salt-sparing genetic variant was selected for in Africa (a kind of "thrifty genotype") owing to chronic salt shortages. Then, enslaved Africans were subjected to extreme conditions of sodium deprivation in the Middle Passage and under conditions of slavery, leading to high mortality rates, and this salt-sparing genetic variant in New World African-descent populations was further selected for. According to the slavery hypothesis, owing to the higher prevalence of this racial-genetic trait, African Americans and other black populations in the western hemisphere retain more sodium when it is plentiful in the diet, resulting in high blood pressure (see below on the salt intake hypothesis for hypertension). "The slavery hypothesis is a controversial idea to account for racial and ethnic health disparities and has many critics (Curtin 1992, Jackson 1991). What is striking, however, is its wide acceptance based on virtually no empirical evidence. In a paper full of insight into how hypotheses diffuse, Kaufman & Hall (2002) demonstrate the level of credibility the hypothesis has received in both the professional and popular literature, even being incorporated into recommendations on how high blood pressure should be treated (Brownley et al. 1999), despite the lack of research. As Kaufman & Hall note, "The intellectual resilience of the Slavery Hypothesis may be attributable to several of its ideologic components[....]perhaps the most in?uential of these in the modern era is the beguiling allure of a simplistic genetic determinism" (2002, p. 116). The slavery hypothesis may owe its persistence to its reinforcement of folk models of race. As Kittles & Weiss (2003, p. 34) point out, even specialists in genetics routinely confuse technical and folk uses of the term race. The slavery hypothesis may appear to be true simply because it is consistent with, and in turn reinforces, a Western European and American cultural construction of race as a biologic entity (see also Braun 2002, Sankar et al. 2004)." (Dressler et al 2005: 235-236) What is the evidence for and against the racial-genetic model? To your mind, why is this model pervasive today? 2. Next is the health-behavior model, which is a model that starts with the idea that racial disparities in health are related to the different behaviors performed by individuals that fall along racial lines for cultural or financial reasons - consuming a high salt or fat intake, or low physical activity, for instance. The idea, then, is that if we could change the health behaviors of those races with worse outcomes, we could "fix" the disparity. Let's see if this idea holds water once we look at the literature. "Body composition is clearly associated with higher blood pressure (Sowers et al. 2002), but differences in body composition do not explain health disparities. National Center for Health Statistics (NCHS) (Cent. Dis. Control Prev. 2004c) data show that in 1999-2000, 67.3% of white American men would be considered overweight (a body mass index >25.0), compared with 60.3% of African American men. This discrepancy therefore could not account for the differences in hypertension prevalence between black and white men. Fifty-seven percent of white women are overweight compared with 77.7% of black women. Although this seemingly could account for prevalence differences between black and white women, it apparently does not. Bell and associates (2004) recently analyzed data from the NHANES (National Health and Nutrition Examination Survey) for women, and black women remained twice as likely to have high blood pressure after controlling for obesity. Furthermore, virtually every study of blood pressure routinely uses a measure of body composition (e.g., the body mass index) as a control variable, with little effect on black-white disparities. "With respect to birth weight, heavier women tend to have heavier babies (Inst. Med. 1990), so the higher prevalence of overweight among black women cannot explain the higher prevalence of low birth weight. "Physical activity levels affect both weight and overall risk of disease, and there is evidence of differences among racial and ethnic groups in levels of physical activity. Nationally, 34.4% of white men and 38.3% of white women report being physically inactive compared with 45.1% of black men and 55.1% of black women (Schoenborn et al. 2004, p. 42). Bell et al. (2004) found that controlling for reported levels of physical activity made no difference in the differential risk of hypertension between black and white women. With respect to low birth weight, strenuous occupational activity (such as standing for long periods) does not alter differences in low birth weight between black and white women (Homer et al. 1990, Teitelman et al. 1990). "Smoking is a risk factor that has been directly implicated in low birth weight (McCormick et al. 1990) but only tangentially, if at all, in association with blood pressure (Janzon et al. 2004). Again, smoking is not a factor likely to account for health disparities because there are virtually no differences in rates of smoking between black and white men (27.1% versus 25.2%) or between black and white women (19.5% versus 22.2%) (Schoenborn et al. 2004, p. 21). "Alcohol intake is discouraged during pregnancy because it contributes to low birth weight, whereas when considering blood pressure there appears to be a J-shaped relationship between alcohol consumption and blood pressure; nondrinkers and mild drinkers (=3 drinks/day) had comparably low blood pressures, and heavy drinkers had higher blood pressures (Estruch et al. 2004). According to NCHS survey data, 70.8% of white men versus 55.8% of black men report being drinkers; corresponding ?gures for white versus black women are 60.4% versus 39.4%. Rates of heavy drinking, de?ned as 14 or more drinks per week, are very similar across groups, ranging from 2.2% among African American women to 5.6% among white men (Schoenborn et al. 2004, p. 7)." (Dressler et al 2005:236-238) Do health behaviors affect health? Do you think they are a likely source of health disparities ? Why or why not? What are the most compelling examples to support your argument? 3. The final model for consideration is the psychosocial stress model, which suggests that the stresses of racism are the main source of racial health disparities. This model is biocultural because it proposes a clear, physiological mechanism between the lived experience of racism and the chronic activation of the stress response. But there are a few approaches within this model worth talking about. "The ?rst approach can be best exempli?ed by the social epidemiologists Krieger (1999, 2003) and Williams (Williams &Collins 1995, Wyatt et al. 2003). In this approach, there is a clear distinction made between institutional racism and perceived racism, the former referring to the system of structured inequality that places black Americans lower on all indicators of economic well-being, and the latter referring to the conscious perception of discriminatory acts and practices and the distress associated with that perception. Institutional racism results in the limited access of racial and ethnic minorities to resources, both in the sense of limited access to high-paying jobs or educational opportunities and in the sense of limited access to resources that would support the attainment of better health status (e.g., living in neighborhoods with markets that stock fresh fruits and vegetables, neighborhoods in which it is safe to walk for exercise). The concept of institutional racism has mainly offered a framework for the interpretation of racial and ethnic health disparities that is an alternative to other (e.g., racial-genetic) models, providing what Krieger (1999, p. 310) calls an "indirect" approach to the study of discrimination and health. "Perceived racism, by comparison, is measured directly by self-reports of respondents about their experiences of discriminatory acts, both in institutional settings (e.g., on the job) and in mundane social interactions (Krieger 1990, Krieger & Sidney 1996). The empirical record for measures of perceived discrimination is mixed. In a recent review, Williams and associates (2003) report eleven studies that examine the association of perceived discrimination and blood pressure. Of these studies, three ?nd a direct association, three ?nd no association, and ?ve ?nd associations that pertain only to particular subgroups (e.g., gender or occupational groups; see also Brondolo et al. 2003). A recent study reports a direct association of perceived discrimination and blood pressure, although data were collected from a convenience sample (Din-Dzietham et al. 2004). Two studies in the review by Williams et al. (2003) examined perceived discrimination and low birth weight; one found no association and one found a conditional association. Amore recent study found that controlling for self-reported discrimination reduced by half the risk for black women of reporting having had a low birth weight baby (Mustillo et al. 2004; see also Collins et al. 2004). "The second approach to the study of psychosocial stress and health disparities employs a more general understanding of the term stress as negative affect (depression, anxiety) experienced by individuals, which in turn can be associated with deleterious health outcomes. This approach has been taken in the incorporation of psychosocial data into large national studies, such as CARDIA (Cardiovascular Disease in Young Adults) and the various waves of the NHANES (see Williams 1999 for a useful discussion of national data sets). Jonas et al. (1997) and Jonas & Lando (2000) looked at overall negative affect as a prospective predictor of incident hypertension in two different follow-up waves of the NHANES, ?nding that those who report negative affect are at a higher risk for developing hypertension and that this association is greater for African Americans. Davidson et al. (2000) found a similar pattern of results using the CARDIA data. Finally, using a subset of the CARDIA data, Knox and colleagues (2002) found that young African Americans who were more reactive to stressful stimuli in the laboratory in turn had higher ambulatory blood pressures three years later. "The third approach to the study of psychosocial stress and health outcomes is best represented by the early work of Harburg and associates (1973), and the subsequent work of James on the John Henryism hypothesis (James et al. 1983). These researchers adapted general models of the stress process to the speci?c ethnographic realities of the African American community. For example, Harburg et al. (1973) argued that persons, black or white, living in high "socioecologic stress" areas (characterized by low SES and high rates of social instability as measured by crime) were at a higher risk for stressful experiences on a daily basis, increasing the likelihood of high blood pressure. For African Americans, and especially darker-skinned black men, there was the added insult of racist interactions (with police or other representatives of the white power establishment). These racist interactions were in turn likely to provoke hostility on the part of the black participant in the interaction, who may then suppress that hostility to avoid negative repercussions. The model thus predicted that darker-skinned black men who lived in high stress areas and suppressed hostility would have the highest blood pressures. Research results have been generally consistent with these predictions, although the strength of the anger expression and suppression effect has been found to be modest (Schum et al. 2003).... "Finally, studies by Dressler (1990, 1991a) are relevant here. Like James's studies of John Henryism, Dressler eschewed the attempt to account for racial or ethnic differences in disease risk, focusing instead on factors within the African American community. On the basis of ethnographic observations, he adapted the concept of status incongruity, arguing that individuals' struggles to achieve a middle-class lifestyle in the face of limited economic resources would be a potent stressor. At the same time, traditional features of social organization in the black community, especially reliance on the extended family for social support, would moderate that stressor. He found that the interaction of status incongruence and social support was associated with blood pressure within a Southern black community; however, the interaction of kin support and status incongruence was signi?cant only for older (>45 years) respondents. For younger respondents, non-kin support buffered the pressor effect of status incongruence." (Dressler et al 2005:238-241) How might this model be seen as an advancement over the other models?