Population Research and Policy Review 20: 441–455, 2001. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.

441

A reconceptualization of the role of biology in contributing to race/ethnic disparities in health outcomes REANNE FRANK Population Research Center, Department of Sociology, University of Texas at Austin

Abstract. An increasing amount of attention is being given to the inappropriate use of contributions from the biological sciences in research on race/ethnic disparities in health. This discussion will take a first step towards productive interdisciplinary collaboration with the biological sciences by recognizing the methodological and ethical problems underlying current research on race/ethnic disparities. An alternative model is then presented that specifies the role of biology in contributing to race/ethnic disparities in health and mortality without reinforcing a biological conceptualization of race. Keywords: biology, genetics, health, inequality, race “You may group humans into a small number of races if you want to, but you are denied biology as a support for it.”1

Introduction Over this past century, great advances in the public health of racial/ethnic minority populations have been offset by the troubling persistence, and in many cases exacerbation, of many race/ethnic disparities in health and mortality. In the United Sates, for example, African Americans have higher mortality rates than non-Hispanic whites for virtually every major disease (Cooper 1993; Minino and Smith 2001) and in the area of infant health, infant mortality rates between African American and non-Hispanic white infants have actually widened in the last fifty years (Hummer 1993). Exacerbating the existence of these race/ethnic disparities has been the general failure of health researchers to adequately understand and explain their persistence. In part, this failure has been attributed to the methodology underlying research on race/ethnic differences. Attention has been increasingly given to the problems that data constraints, a focus on individual level proximate risk factors, and measurement error pose for successfully accounting for race/ethnic disparities (Kaufman and Cooper 1999; Kaufman et al. 1997; Morgenstern 1997; Hummer 1993). Kaufman and Cooper (2001) have most recently argued that the methodological approaches underlying a large portion of etiologic

442

REANNE FRANK

research on race/ethnic differences in health are inappropriate and their continued usage often works to obscure the nature of race/ethnic disparities in health. The present discussion will add to this methodological critique by complementing it with one located on a conceptual level. It will be argued here that the failure to account for race/ethnic differentials across many health outcomes is not only an artifact of methodological error in and of itself, but also contributing to a lack of methodological rigor is a weakness in the conceptual framework underlying health research. At the core of this inadequate conceptual framework lies the biological concept of race, which encompasses the idea that people who belong to a particular racial group also share a genetic heritage that makes them genetically different from members of other racial groups (Williams 1997; Witzig 1996; Williams et al. 1994; Cooper 1984). Following a brief review of the origin of a biological concept of race, work from biological anthropology will be used to demonstrate the reasons that race is not a biologically sound category. The implications of this fact for health research will be addressed in a discussion of two racially differentiated health outcomes at opposite ends of the life course: infant birth weight and adult onset non-insulin dependent diabetes. Finally, a new conceptual model will be presented that challenges the role given to biology in the majority of current research examining race/ethnic disparities in health.

A biological concept of race The origin of the biological concept of race and the ways in which it has been incorporated into minds of health researchers in the United States involves a rich and complex history. In present times, our conceptualization of race takes racial genetic differences as a starting point. That is, different racial groups are understood to be genetically distinct. However, human variation was not always understood as the discrete phenomenon that it is today. According to C. Loring Brace (forthcoming), our present conceptualization of race emerged from the peculiar situation developed during the settlement of the new world. Prior to then, human variation was understood to be a relatively continuous phenomenon with one population group grading into the next. European colonization involved the bringing together of people from disparate parts of the world to live side by side, erasing any appreciation of the ways in which human population groups actually blend one into the other. This situation was filtered through the intellectual legacy of the Enlightenment, which placed a premium on dividing the world into discrete categorical entities. As such, races were conceived of as categorically dif-

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

443

ferent and biologically distinct sub-groups which represented differences in worth. While the majority of present-day health researchers no longer accept the belief that these categorical differences connote differences in worth, the firm belief in the existence of biologically distinct racial groups is still alive and with us today. As observed by Cooper (1984), “the scientific concept of race . . . implies that a package of different genes, all of which ideally should be known, exists between groups” (714). According to Goodman (2000) the way race is conceptualized in health research is best labeled as “scientific racialism” which involves extending a biological deterministic approach to understanding variation (i.e., differences between human groups arise from inherited inborn distinctions), to include the belief that racial differences in health are caused by genetic variation among races. Most often, in health research, this approach takes the form of attributing race/ethnic differentials in a particular health outcome to a set of pre-determined racial genetic conditions after controlling for several environmental or social factors fails to explain the gap. Methodologically, this strategy is fraught with difficulty as it is prone to what Kaufman and Cooper (2001: 294) call “misleading inference”. Adjustment for intermediate variables rarely provides valid estimates of direct effects, and additionally, the inappropriate measurement of intermediate variables may also result in spurious direct effects. Furthermore, the use of race as a causal effect is problematic in that positing a counterfactual race/ethnic state, as is required in an etiologic model, is highly implausible given that race/ethnicity is an unalterable state of lifelong identity. On a conceptual level, this strategy is equally problematic in that the ‘scientific racialism’ approach to understanding racial differences in health has grown out of a European folk taxonomy and lacks a scientific basis. The misunderstanding that there is a genetic basis to race has very real consequences for research design and interpretation of findings. The first step, then, towards reconceptualizing the role of biology in contributing to race/ethnic differences involves clarifying the nature of human biological variation and the role of race therein.

A biological concept of race debunked The essential problem underlying a biological concept of race is that it runs afoul of the available scientific evidence that shows that racial categories are incapable of appropriately capturing biological distinctiveness because human biological variation in not racially patterned (Goodman 2000; Templeton 1999; Williams 1997; Witzig 1996; Marks 1995; Cooper 1984; Williams et al. 1994; Brace et al. 1993; Brace and Hunt 1990). While

444

REANNE FRANK

there do exist biological differences between people and between populations, these differences are not racially patterned. Instead, human biological variation is produced by three different processes (Marks 1995). The first is natural selection, which involves genetic response to local conditions. The changes it produces are adaptive and come about as a result of evolutionary responses to selective forces. They are not restrictive to a particular gene pool, for the selective force that brought them about could have operated in the same intensity on a different gene pool. These clines, or geographic gradients of particular traits, are useful for deconstructing a biological concept of race because they demonstrate that adaptive human variation exists because of the existence of selective forces that maintain the variation, and not because a person belongs to a particular racial group, which is the rationale that supports the present conceptualization of race (Brace et al. 1993; Brace and Hunt 1990; Brace 1964). In the schema of clines, a person from Southeast Asia could exhibit one trait and a person from Africa could exhibit that same trait, as long as the selective force maintaining the trait was present in the same intensity in both places for a sufficient amount of time (Brace 1996). And not surprisingly this is exactly the case with regard to several well-known and supposedly ‘racial’ traits. Perhaps the most well-publicized in health research concerns the presence of hemoglobin S in certain populations, which causes sickle cell anemia in homozygous individuals. First identified as an African disease, it was later correctly identified as a clinally patterned disease by Livingstone (Livingstone 1958). He concluded that the presence of hemoglobin S in certain populations corresponds to the selective force of malaria and is not correlated to specific races. The simple point to be made here is that a large part of human variation exists as a result of evolutionary responses to the forces of selection and a category as ill-defined as race is not capable of capturing the vast majority of human biological variation. As Livingstone (1964) concluded: “(t)here are no races, there are only clines”. Genetic drift is another process that functions to differentiate gene pools and the differences it creates will be reflective of trivial traits that are perpetuated by heredity. Genetic drift explains what Brace (1996) deems ‘cluster variation’ and it is the process most often used to rationalize perceived differences between racial groups. Cluster traits are useful for determining regional relationships because they are trivial traits that are not connected to particular selective forces but instead are perpetuated by heredity. These are biological traits that do associate with each other and that are characteristic of geographically circumscribed populations. However, it must be emphasized that clusters are not akin to races in that one cannot understand the majority of biological variation through them, which is exactly what a classification based on ‘races’ tries to do. Instead clusters are useful for determining regional

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

445

relationships, a particularly useful skill for those attempting to understand the human evolutionary record. As Brace and Hunt note, “until recently the whole matter of population relationships that used to be subsumed under the rubric of “racial classification” has been left relatively untouched . . . This [cluster] approach provides a means of dealing with human biological variation that can be used to group and compare human populations without the attendant danger of creating anything comparable to racial typology” (Brace and Hunt 1990: 343–345). However, if schemas based only on clusters are used to understand human biological variation, then the vast majority of the variation that is patterned clinally will be overlooked. This is what led Brace et al. (1993) to conclude that, “neither clines nor clusters alone suffice to deal with the biological nature of a widely distributed population. Both must be used” (Brace et al. 1993: 1). The final process is gene flow that acts to homogenize neighboring populations (Marks 1995). According to Keita and Kittle (1997), “racialized thinking” in its classical form requires the explanation of certain kinds of variation as necessarily the result of gene flow between entities conceptualized as having different traits. When traits that are predefined as belonging to different groups are found in combination, admixture is incorrectly understood to be the only explanation. Together these processes can result in differences in gene pools, but not one of these processes alone can fully explain the variation of the human species. Williams (1997) succinctly summarizes the entire process: “genetic composition of human population groups is not static but evolves over time due to natural selection, adaptations to the environment, mutations, genetic exchanges between different populations, and randomly changing frequencies of generic characteristics from one generation to another” (324). Steven J. Gould (1981) once noted that, “[i]f it is to have any enduring value, sound debunking must do more than replace one social prejudice with another. It must use more adequate biology to drive out fallacious ideas” (322). This section has made clear the fact that human biological variation in not racially patterned. As noted by Camara Phyllis Jones (2001: 299), “There is no denying that there is genetic variability on the planet. However, the pie slicer that we call race does not capture that genetic variability”. The following section will now use “more adequate biology” to demonstrate the ways in which racial disparities in health can be understood without succumbing to overly simplistic attributions to racial genetic variation.

446

REANNE FRANK

Illustrations Infant birth weight and adult onset diabetes (NIDDM) are two continuously distributed health outcomes (i.e., the definition of a disadvantaged outcome is arbitrary) that demonstrate sharp race/ethnic differentials in their distribution. Additionally, in both cases, the search for the explanations behind these differences has proven elusive. In the case of infant birth weight, the strong association between birth weight and infant mortality and the race/ethnic disparities therein has led to intense scrutiny concerning the reasons behind these differences. Classically, differences in perinatal outcomes have been explained either socially or biologically (Wise 1993). David and Collins (1996) summarize the two conflicting explanations behind disparities in the incidence of prematurity: “. . . there are two theoretical possibilities to explain the differences between races in the risk of prematurity and early death: either there is a set of social and environmental risk factors that differ in quantity between races but which exert qualitatively similar effects on blacks and whites, or, if such factors can’t be found, disparities must result from genetic, biological differences between races” (207). And, indeed, it has become increasingly common, amidst heightened frustration with the failure of statistical models to explain race/ethnic disparities in birth weight, to claim that different racial groups are genetically different in size and therefore produce different size babies (see Wilcox and Russell 1989; van de Oore and Rowe 2000; 2001). In addition to the numerous technical limitations hampering the usage of race in etiologic research, one of the most deleterious consequences of attributing unexplained variation to genetic effects is that it absolves the researcher of further investigation into more complex social/environmental influences (Collins and David 1997; Kaufman et al. 1997; Oliver and Shapiro 1995). Researchers become all too willing to accept their measures as sufficient and unbiased when they really remain poorly measured. Perhaps an even more egregious effect of the adoption of a scientific racialism approach is that it functions to preclude exploration into the more mutable biological origins of the race/ethnic differences (Frank 2001). One of the more promising aspects concerning increasing contributions from the biological sciences involves the opportunity to model these biological processes as a scientific alternative to using race as a proxy for genetic variation (Zuberi 2001). Kaufman and Cooper (2001: 306) note that, “. . . for hypotheses about genetic factors and disease, identifying the products of genes and describing a physiologic mechanism by which genetic variation affects the relevant organ system are ultimately more likely to provide an informed perspective on the question than any collection of epidemiologic analyses with various unrealistic and uninterpretable adjustments and inferential extrapolations”. In the case of

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

447

birth weight, this would involve the ways in which physiological pathways for growth are impacted by environmental and social influences. Further investigation into more mutable biological effects would involve such concepts as plasticity, which refers to the ability of many organisms to change their biology or behavior to respond to changes in the environment. The concept of plasticity, in its inception, was used to explain the effects of malnutrition on the human body and was most commonly used in comparative studies between migrant groups and their country of origin populations. Lasker (1995) showed that the mechanism for plasticity in human biology is to be found in those processes that regulate amount of growth and rate of development of the body as a whole. The substantive contribution of the concept of plasticity is that it identifies the ways in which physiologic pathways for growth are changed in direct response to environmental and social influences. Pritchard (1995) notes that humans are produced mainly by the regulative scheme and that “organisms that begin their lives as regulative embryos are founded in plasticity, subject to the molding forces that impinge on them from the outside world or arise as a result of morphogenesis” (22). As such, the concept of plasticity can be utilized to understand the ways in which social and environmental effects on the mother can directly affect the growth processes of the infant in utero. These social and environmental indices have been shown to vary systematically by race/ethnic group and are much more likely to contribute to between group differences in birth weight through interpregnancy physiologic influences on the uterine environment. The strength of a concept like plasticity is that it recognizes, at a biological level, the effect of environmental and social factors on the physiological processes that subsequently determine the growth of both the mother and the fetus, without reverting to a scientific racialism approach that invokes unspecified racial genetic differences. Diabetes At the other end of the life cycle, non-insulin dependent diabetes (NIDDM) represents another health outcome that demonstrates unaccounted-for racial differentials in its distribution. In the middle of what many have deemed an epidemic of diabetes, intense scrutiny has been given to understanding the source of the race/ethnic differentials characterizing this disease. While the search for the genes responsible for adult onset diabetes has proven to be elusive, the explanations invoked to explain these race/ethnic differences have nevertheless increasingly leaned toward a biological deterministic approach in which racial genetic effects are held responsible. Most well-known is the “thrifty gene hypothesis” first formulated by Neel (1962) in which particular racial groups are understood to have developed a diabetic genotype that was

448

REANNE FRANK

efficient in the utilization of food but that in recent times makes its owners susceptible to diabetes. As a result, it is argued that particular race/ethnic groups, most notably indigenous populations, have disproportionately high rates of diabetes. However, in an examination of the prevalence of diabetes among Aboriginal populations in Australia, McDermott (1998) argues that a monopoly of attention to genetic causation has come at the expense of alternative hypotheses that suggest that race/ethnic differentials in diabetes prevalence may be due to inherited or metabolically adapted, and therefore changeable, susceptibility, to NIDDM. The author points to two areas of research that suggest that the inheritability of diabetes may be due to physiological processes in utero. One possibility, first proposed by Freinkel (1980), holds that exposure to hyperglycemia during pregnancy may result in increased susceptibility to obesity and NIDDM later in life. A second possibility, put forth by Barker (1992), suggests that poor nutrition and growth in utero or early infancy may function to increase risk of diabetes. In both cases, attention is given to the physiological effects arising from social and environmental influences at the early stages of the life cycle. As such, they are able to capture the ways in which physiologic pathways for growth are changed in response to such factors as access to better nutrition. In this sense, the explanation for the high rates of diabetes found among the Aboriginal populations in Australia may have very little to do with genetics but instead may have more to do with social, environmental and biological influences. McDermott concludes that diabetes in Aborigines has been simplistically defined as a problem of race and genes in which race has become a “biological entity and an independent risk factor, reified over and over again in repeated studies of disease which take no account of socioeconomic status, history or culture” (1193). This uncritical acceptance of ‘racial genetic effects’ has resulted in a lack of attention given to alternative possibilities, specifically consideration of social determinants of disease. As in the case of birth weight, poorly constructed measures of socioeconomic status also play a role in supporting a scientific racialism approach to understanding differentials in diabetes prevalence. In an extensive analysis of the problems plaguing current measurements of socioeconomic position, Kaufman et al. (1997) point to an analysis of differential NIDDM prevalence between blacks and whites (see Cowie et al. 1993) in which the authors concluded that racial genetic effects were the cause of the differential prevalence after controlling for several confounders failed to explain the gap. However, Kaufman et al. illustrate that incommensurate indicators, i.e., standard SES measures are not commensurate between blacks and whites, biased the results

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

449

and that correct adjustment would have resulted in findings contrary to the published conclusion.

Towards a reconceptualization Given the confusion surrounding the role of biology in contributing to race/ethnic disparities in health and mortality, as well as the methodological bias resulting from this confusion, conceptual clarity is a necessary prerequisite if any hopes at better understanding the role of different sets of risk factors in contributing to race/ethnic differentials are to be achieved. Figure 1 presents a conceptual model that moves away from a scientific racialism explanation of the role of biology in contributing to race/ethnic disparities in health outcomes. First and foremost, beginning with the biological/genetic contribution, population level human biological variation is not specified as racially patterned, but rather is best understood as consisting of clinal and cluster variation. That is, while the model acknowledges the existence of biological/genetic variation between populations, it also stipulates that a system of racial categorization is not the best way to capture this variation. As shown in the infant birth weight and non-insulin dependent diabetes illustrations, this distinction is critical to moving past scientific racialism frameworks and simplistic causal attributions. Perhaps the most important contribution of genetic heritage is that it confers upon individuals the phenotypic markers that are used to channel individuals into different racial groups. It is in this way that biological markers are used in the social construction of racial groups. Culture/nativity forms an important part of racial identity and this dynamic relationship is depicted with the double arrow running between the two categories. A system of institutional racism and discrimination then channels individuals into differential positions of social advantage based upon an assortment of physical characteristics and heredity. As such, membership in a particular racial group connotes differential access to societal resources. Poverty, educational attainment, employment, income and general living conditions are all racially patterned indices that contribute to differences in health outcomes. This racial classification system results in dramatic differences in life chances based on race in that it directs the distribution of risks and opportunities (Jones 2001). In this sense, the overriding importance of race in contributing to race/ethnic disparities in health is that it functions as an excellent measure of exposure to racism. Yet despite the importance of racism and socioeconomic status in determining racial/ethnic disparities in health outcomes, they remain poorly constructed and measured in current research, thus allowing unexplained variation to be erroneously allocated to

450

REANNE FRANK

Figure 1. Conceptual model depicting the role of particular factors in contributing in race/ethnic disparities in health outcomes.

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

451

‘racial genetic differences’. More attention needs to be given to measuring SES, racial discrimination and contextual variables, collecting more sophisticated data bases and developing more appropriate modeling strategies. In fact, Kaufman and Cooper (2001) argue that measurement of the external effects of race/ethnicity on disparate health outcomes is one of the few areas of epidemiological research that has a high potential to result in valid findings and conclusions. The identification and quantification of the external effects of race in contributing to disparate health outcomes promises to be a complex area of emerging research. For example, using a life cycle approach to understanding the racial gap in chronic health conditions, Hayward et al. (2000) conclude that racial differences in health represent the combinations of different experiences over the life cycle. Given that racism is a root cause of racial disparities in health, the importance of adequately understanding the characteristics and manifestations of racism should be at the top of our collective research agenda. While socioeconomic factors are understood to be important driving forces in determining differential health outcomes, they are not usually understood to directly affect health, but rather they largely operate through a host of proximate or intervening factors such as medical care, social support, levels of stress, nutrition, substance use, exercise, and other health behaviors. Differential social position influences preferences toward major lifestyle behaviors by affecting the ability to control everyday life circumstances (Jones 2001). Social status categories thus result in differential exposure to risk factors and resources that ultimately contribute to differential health outcomes (Williams 1997). The study of the health experience of many Hispanic groups, most notable Mexican Americans, has challenged a strict socioeconomic approach and highlighted the important role of lifestyle factors in determining differential health outcomes (Williams and Collins 1995). The Hispanic population is characterized by health outcomes that are more similar to the non-Hispanic white population than to other population groups with whom they share similar socioeconomic risk profiles. One of the leading explanations behind the more favorable health outcomes of Hispanics, and in particular the foreign born component of the Hispanic population, is that they are able to successfully limit the harmful effects of their disadvantaged socioeconomic position in the U.S. through salutary, culturally-based behavior (Scribner 1996; Weeks and Rumbaut 1991; Scribner and Dwyer 1989). In this sense, socioeconomic factors may not be as important in determining differential health risk in populations that are able to limit the deleterious consequences attached to a disadvantaged social position. In turn, without a strong cultural base many economically disadvantaged individuals may resort to harmful stress reduction behaviors (James 1993). Conceptually,

452

REANNE FRANK

this dynamic relationship is depicted in Figure 1 with the arrows running between culture/nativity, racism/discrimination and socioeconomic status and health behaviors. As the relationship between culture and health behaviors weakens, with time in the U.S. or through assimilation, then the arrows running between racism/discrimination, socioeconomic status and health behaviors becomes more salient in determining race/ethnic disparities in health outcomes. Finally, the most proximate stage in the model accounts for the role of biological processes in contributing to race/ethic disparities in health outcomes. In this stage the interactions between environmental, social and biological processes are taken into account. It is also here that the most important insights from the biological sciences for understanding race/ethnic disparities are to be found. As Hayward et al. note, ‘Biology’s major role is to define the array of possible physiologic responses to the social, economic and environmental conditions associated with the increased risk of morbidity and mortality. These conditions influence the expression of genetic susceptibility” (913). In this model, race/ethnic disparities are created by different physiologic processes that are preceded by an elaborate schema that involves interactions between both biological and social factors at different stages in the path to a final health outcome. The model is able to integrate the complex of interactions that may intervene between gene products and their ultimate phenotypic expression. Nowhere in this elaborated model does a genetic category appear as an undisaggregated grouping that suggests the existence of predetermined genetic packages that are driven by race. Instead, it is recognized that genetic systems are interactive, with social and environmental forces determining the extent to which a given trait is influenced by genetic factors (Cartmill 1999). This model refutes the use of race as an undisaggregated biological category in health research without totally disregarding either race or biology in contributing to differential health outcomes. By identifying what components of race are related to health and documenting their role in producing particular patterns in health outcomes, we can come a long way in appropriately handling racial differentials without succumbing to the explanatory temptations that have plagued past work.

Conclusion In the area of health research, the way we conceptualize the role of biology in contributing to race/ethnic disparities in health represents an important conceptual distinction that can impinge on study design and influence interpretation of findings. While methodologically the usage of a biological

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

453

concept of race in health research most likely results in misclassification bias (as there is a geographic aspect to race which correlates with all sorts of clinally distributed traits), the conceptual bias resulting from its continued usage promises to be much greater. Comparing populations defined along conventional racial criteria and attributing differentials to racial genetic differences simplifies the nature of human biological variation and even worse, allows for other more consequential contributions to race/ethnic health differentials to be dismissed or ignored. Leaving the basis of these disparities poorly explained has the very tangible consequence of limiting efforts at primary prevention. If researchers do indeed share a commitment to the elimination of racial disparities in health outcomes then their first step is to stop using an outdated conceptual framework to understand these differences. Until this is done, the deleterious scientific and social consequences that have cumulatively resulted and continue to result from the irresponsible use of race in health research should lie on the conscience and mar the academic credibility of each and every researcher who uncritically utilizes a biological concept of race in their analyses, findings or conclusions. Acknowledgements The author gratefully acknowledges the comments of Samuel Echevarría. Notes 1. Marks (1994).

References Barker, D.J.P. (ed.) (1992), Fetal and Infant Origins of Adult Disease. British Medical Journal. Brace, C.L. (1964), A Nonracial approach towards understanding of human diversity, pp. 103– 152 in A. Montagu (ed.), The Concept of Race, New York: The Free Press. Brace, C.L. (1996), A four-letter word called “Race”, pp. 106–141 in L.T. Reynolds & L. Lieberman (eds.), Race and Other Misadventures: Essays in Honor of Ashley Montagu in His Ninetieth Year, Dix Hills: General Hall, Inc. Brace, C.L. (forthcoming), The concept of race in physical anthropology, in Peter Peregrine (ed.), Physical Anthropology Original Readings on Method and Practice. Prentice Hall. Brace, C.L. & Hunt, K.D. (1990), A nonracial craniofacial perspective on human variation: A(ustralia) to Z(uni), American Journal of Physical Anthropology 82: 341–360. Brace, C.L., Tracer, D.P., Yaroch, L.A., Robb, J., Brandt, K. & Nelson, A.R. (1993), Clines and clusters versus “Race”: A test in ancient Egypt and the case of a death on the Nile, Yearbook of Physical Anthropology 36: 1–31.

454

REANNE FRANK

Cartmill, M. (1999), The status of the race concept in physical anthropology, American Anthropologist 100: 651–660. Collins, J.W. & David, R.J. (1997), Urban violence and African-American pregnancy outcomes: An ecologic study, Ethnicity and Disease 7: 184–190. Cooper, R. (1984), A note on the biological concept of race and its application in epidemiologic research, American Heart Journal 108: 715–723. Cooper R. (1993), Health and social status of blacks in the United States, Annals of Epidemiology 3: 37-144. Crowie, C.C., Harris, M. I., Silverman, R.E., Johnson, E.W. & Rus, K.F. (1993), Effect of multiple risk factors on differences between blacks and whites in the prevalence of non-insulin-dependent diabetes mellitus in the United States, American Journal of Epidemiology 137: 719–732. David, R.J. & Collins Jr., J.W. (1996), Racism and bad pregnancy outcomes among AfricanAmericans: The case against the biological concept of race in perinatal epidemiology, pp. 206–216 in L.T. Reynolds and L. Lieberman (eds.), Race and Other Misadventures: Essays in Honor of Ashley Montagu in His Ninetieth Year, Dix Hills: General Hall, Inc. Frank, R. (2001), The misuse of biology in demographic research on racial/ethnic differences: A reply to Van den Oord and Rowe, Demography 38(4): 563–567. Freinkel, N. (1980), Of pregnancy and progeny, Diabetes 29: 1023–1035. Goodman, A.H. (2000), Why genes don’t count (for racial differences in health), American Journal of Public Health 90: 1699–1702. Gould, S.J. (1981), The Mismeasure of Man. 1st Edition. New York: Norton. Hayward, M.D., Crimmins, E.M., Miles, R.P. & Yang, Y. (2000), The significance of socioeconomic status in explaining the racial gap in chronic health conditions, American Sociological Review 65: 910–930. Hummer, R.A. (1993), Racial differentials in infant mortality in the U.S.: An examination of social and health determinants, Social Forces 72: 529–554. James, S.A. (1993), Racial and ethnic differences in infant mortality and low birth weight: A psychosocial critique, Annals of Epidemiology 3(2): 130–136. Jones, C.P. (2001), Invited commentary: “Race”, racism and the practice of epidemiology, American Journal of Epidemiology 154(4): 299–304. Kaufman J.S., Cooper, R.S. & McGee, D.L. (1997), Socioeconomic status and health in blacks and whites: The problem of residual confounding and the resiliency of Race, Epidemiology 8: 621–628. Kaufman J.S. & Cooper, R.S. (1999), Seeking causal explanations in social epidemiology, American Journal of Epidemiology 150(2): 113–120. Kaufman J.S. & Cooper, R.S. (2001), Commentary: Considerations for use of racial/ethnic classification in etiologic research, American Journal of Epidemiology 154(4): 291–298. Keita, S.O.Y. & Kittles, R.A. (1997), The persistence of racial thinking and the myth of racial divergence, American Anthropologist 99: 534–544. Lasker, G.W. (1995), The study of migrants as a strategy for understanding human biological plasticity, pp. 110–114 in C.G.N. Mascie-Taylor & Barry Bogin (eds.), Human Variability and Plasticity, Cambridge: Cambridge University Press. Livingstone, F.B. (1958), Anthropological implications of sickle cell gene distribution in West Africa, American Anthropologist 60: 533–562. Livingstone, F.B. (1964), On the nonexistence of human races, pp. 46–60 in A. Montagu (ed.), The Concept of Race. New York: the Free Press. Marks, J. (1994), Racial categories and cultural constructs masquerading as biology, Natural History December: 32–35.

THE ROLE OF BIOLOGY IN CONTRIBUTING TO RACE/ETHNIC DISPARITIES

455

Marks, J. (1995), Human Biodiversity: Genes, Race, and History. New York: Aldine de Gruyter. McDermott R. (1998), Ethnics, epidemiology and the thrifty gene: Biological determinism as a health hazard, Social Science and Medicine 47: 1189–1195. Minino A.M. & Smith, B.L. (2001), Deaths: Preliminary data for 2000. National Vital Statistics Reports. Vol. 49 no. 12. Hyattsville, Maryland: National Center for Health Statistics. Morgenstern, H. (1997), Defining and explaining race effects, Epidemiology 8: 609–610. Neel, J.V. (1962), Diabetes mellitus: A “thrifty" genotype rendered detrimental by “progress?” American Journal of Human Genetics 14: 353–362. Oliver, M.L. & Shapiro, T.M. (1995), Black Wealth/White Wealth: A New Perspective on Racial Inequality. New York: Routledge. Pritchard, D.J. (1995), Plasticity in early development, pp. 18–45 in C.G.N. Mascie-Taylor & Barry Bogin (eds.), Human Variability and Plasticity, Cambridge: Cambridge University Press. Scribner, R. (1996), Paradox as Paradigm-the health outcomes of Mexican Americans, American Journal of Public Health 86(3): 303–305. Scribner, R. & Dwyer, J.H. (1989), Acculturation and low birth weight among Latinos in the Hispanic HANES, American Journal of Public Health 79(9): 1263–1267. Templeton, A.R. (1999), Human races: A genetic and evolutionary perspective, American Anthropologist 100: 632–650. Van den Oord, E.J.C.G. & Rowe, D.C. (2000), Racial differences in birth health risk: A quantitative genetic approach, Demograph 37(3): 285–298. Van den Oord, E.J.C.G. & Rowe, D.C. (2001), A step in another direction: Looking for maternal genetic and environmental effects on racial differences in birth Weight, Demography 38(4): 573–576. Weeks, J.R. & Rumbaut, R.G. (1991), Infant mortality among ethnic immigrant groups, Social Science and Medicine 33(3): 327–334. Wilcox, A.J. & Russel, I.T. (1989), Birthweight and perinatal mortality: I. On the frequency of birthweight, International Journal of Epidemiology 12: 314–318. Williams, D.R. (1997), Race and health: Basic questions, emerging directions, Annuals of Epidemiology 7: 322–333. Williams, D.R., Lavizzo-Mourey, R. & Warren, R.C. (1994), The concept of race and health status in America, Public Health Reports 109: 26–41. Williams, D.R. & Collins, C. (1995), U.S. socioeconomic and racial differences in health: Patterns and explanations, Annual Review of Sociology 349–386. Wise, P.H. (1993), Confronting racial disparities in infant mortality: Reconciling science and politics, American Journal of Preventative Medicine (supplement) 7-16. Witzig, R. (1996), The medicalization of race: Scientific legitimization of a flawed social construct, Annals of Internal Medicine 125: 675–679. Zuberi, T. (2001), One step back in understanding racial differences in birth Weight, Demography 38(4): 569–571. Address for correspondence: Reanne Frank, University of Texas at Austin, Population Research Center, 1800 MAI, Austin, TX, 78712 E-mail: [email protected]