Ethnic Classification in the New Zealand Health Care System

Abstract

The ethnic or “racial” classification of Maori and non-Maori is a pivotal feature of New Zealand’s health system and affects government policy and professional practice within the context of Treaty of Waitangi “partnership” politics. Although intended to empower Maori, ethnic categorization can have unintended and negative consequences by ignoring the causality of material forces in social phenomena. The authors begin by showing how the use of ethnic categories in health policy is justified by the Treaty of Waitangi partnership policies. This provides the context for the argument made in the manuscript that an understanding of the social experience of ethnicity within the complex interaction of sociocultural factors such as socioeconomic location and lifestyle is more useful than using the political construct of ethnic categories in explaining the persistence of low health status for a section of the Maori population.

I. INTRODUCTION

In New Zealand, the emergence of Maori retribalization in the 1980s within the wider context of the Maori cultural revival, biculturalism, and the Treaty of Waitangi historical grievance settlements has produced policies and practices in the health sector that classify the Maori section of the population as a discrete ethnic or racial category (McCreanor and Nairn, 2002, U272). This article argues that the ethnic categorization process can be understood as a political construct in Treaty politics. Those politics are based on a post-1987 interpretation of the Treaty as an ongoing political “partnership” between the incorporated tribes and the government (referred to in the Treaty political discourse as the “Crown”) (Te Puni Kokiri Ministry of Maori Development, 2001). According to the Ministry’s interpretation, Maori low health status (Ministry of Health, 2013) is caused by an unequal power relationship between Maori and non-Maori that has its origins in the British colonization of New Zealand dating to the 1840 Treaty (Bishop and Berryman, 2006). The solution to the colonially imposed inequality and the resulting disadvantage is believed to be found in varying degrees of tribal self-determination. This political solution includes the provision of separate health care policies and practices by tribal and other Maori groups, the most recent of which is the Whanau Ora policy, a major government health and social initiative resulting from the Maori Party’s 2008–11 coalition agreement with the governing National Party.

In contrast, the viewpoint presented in this article is that it is in the materiality of ethnic status—in how people who identify ethnically actually carry out their daily lives and are treated by others—that a more useful explanation of the causes of Maori health disadvantage is to be found. Its purpose is to argue against the political construct of ethnicity that classifies all Maori as belonging to one category with a corresponding low health status used to justify separate tribal health provisions, and in support of an understanding of ethnic identification as one of several interdependent social factors implicated in the causes of disadvantage.

In order to secure the status of the tribes as political, even constitutional, “partners” with the “Crown” (Durie, 2009), Treaty politics requires a political accommodation between Maori and non-Maori so that partners are distinguished as separate entities that can then be brought together in a partnership accommodation (Rata, 2011). Ethnic categorization in the health sector contributes to this distinguishing purpose. To unravel the role of ethnic identification as merely one of a number of social causes in Maori disadvantage—and one that affects groups within the Maori population in different ways—is to disturb the justification for the role ethnic categorization plays in that political construct of partnership. Given the importance of ethnic categorization to Treaty partnership politics, section II discusses the political strategies of categorization and the problems resulting from the approach. This is followed in section III with a discussion of the Maori Mental Health Model, Te Whare Tapa Wha (Durie, 2011, 24–36) that illustrates ethnic categorization policies that were developed as a tribal solution to the ongoing low health status of Maori in comparison to New Zealanders of European heritage (Ministry of Health, 2013).

Although classifying Maori as an ethnic/race category is necessary for Treaty partnership politics, this approach represents an ambiguous strategy. Intended to empower Maori and to enact a political accommodation between the tribes and the government, ethnic categorization can have unintended and negative consequences. First, the practice of classifying a population according to race or ethnicity categories may actually enforce disadvantages and perpetuate disparities by reinforcing stereotypes and condoning inequalities (Ellison et al., 1997, 1485). This may lead to racial discrimination, which is one of the social factors in low health status. Second, ethnic/race classification may lead to the conceptual, methodological, and philosophical discrepancies related to claims about health status as discussed in section IV. Third, such classification ignores the causality of material forces in social phenomena, the subject of section V.

In contrast to the ethnic categorization approach, we argue that a more fruitful way to explain the persistence of low health status for a section of the Maori population is to focus on ethnicity as a social identity that intersects with other social factors. In a seminal paper published by Simon Chapple in 2000, this subgroup was identified as “sole Maori with low literacy, poor education, and living in geographical concentrations that have socio-economic problems, not the Maori group as a whole. There are probably also sub-cultural associations with benefit dependence, sole parenthood, early natality, drug and alcohol abuse, physical violence, and illegal cash-cropping” (Chapple, 2000, 101). Chapple recommended that “the policy issue may need to be viewed primarily at a sub-cultural and socio-economic level rather than the coarse ethno-cultural level of Maori/non-Maori binaries” (Chapple, 2000, 115). Despite this recommendation, government policy continues to be characterized by the categorization of the Maori population in ethno-cultural terms. Although advocates such as Mason Durie (2011) and Chris Cunningham (2000) claim that separate health care for Maori will ameliorate Maori low health status, we argue that it is more likely that the current policy of separate Maori health care will obfuscate the complex links between low socioeconomic status, poverty, marginalized lifestyles, low literacy, and subjective experiences of racial discrimination. This obfuscation may well serve to entrench the health problems for the socially disadvantaged section of the Maori population described by Chapple.

It is important to stress that our argument does not mean that ethnicity is not a contributing factor in Maori health status. It is in the materiality of ethnic status, that is, how people who identify ethnically carry out their daily lives and are treated by others, that a more useful explanation of the causes of Maori health disadvantage is to be found. Therefore, the focus on socioeconomic and other material factors does not exclude ethnicity. Ethnic identification does indeed carry real social meaning, influencing social relations and leading to material consequences. Groups within the Maori population experience their lives in ways where ethnicity matters either through self-identification or as identified by others. A marginalized group’s shared experience of disadvantage is constructed within a complex web of social relations and is reflected at an individual level in the way a person orients him or herself to the world. The relative powerlessness of that position in relation to dominant classes contributes to the complex way in which ethnicity is implicated in low health status. That powerlessness is the experience of some Maori in New Zealand society.

II. JUSTIFYING THE MAORI ETHNIC/RACE CATEGORY IN HEALTH POLICY

The orthodox approach is that ethnic classification in public institutions like health will improve Maori health and the place of Maori in society. According to “He Korowai Oranga, The Maori Health Strategy” (New Zealand Ministry of Health, 2002), both Maori and the Government have aspirations for Maori health and critical roles in achieving the desired whanau (family and extended family) outcomes. “The new health structures encourage Maori to determine their own aspirations and priorities for health and disability, and provide mechanisms for ensuring these are taken into account in the planning and delivery of services. DHBs [District Health Boards] will involve Maori in their decision-making and service delivery and will support effective Maori involvement” (New Zealand Ministry of Health, 2002).

Developing policy based on the concept of an ethnic/race collective is as problematic as the essentialist ethnic identity itself. But because the tribal elite’s claim to economic resources and political authority rests on its legitimacy as the representative of a homogeneous Maori collective, Treaty partnership politics emphasizes that collective. For example, in the 2010 Iwi Infrastructure and Investment Report, tribal leaders claimed that “future demographic realities” point to a “browning” of New Zealand. It is entirely probable there will be “more Maori and Pasifika children in our schools than Pakeha well before the middle of this century” (Iwi Infrastructure and Investment, 2010, ii).

Although the raw statistics state that the numbers of young Maori are increasing at a rapid rate, this may or may not be the case. There is a range of possible ethnic identifications available, such identification itself being a changeable social behavior rather than a fixed ethnic category. For example, according to the latest census in 2006, just over half (52.8%) of all people in the Maori ethnic group identified Maori as their only ethnicity, with 42% stating that they identified with European ethnic groups (Statistics New Zealand, 2011). It is quite possible that many New Zealanders identified with several ethnic groups, including Maori, prior to 2006, but the prioritization principle used by the Department of Statistics until that year meant that those who included Maori as one of their ethnic identities were automatically assigned to the Maori category. This would increase the numbers of Maori but the actual situation is more complex. Many New Zealand families have Maori and non-Maori members (including growing numbers of people from non-European countries and Pacific Island nations). Indeed, the extent of ethnic mixing in New Zealand results in “the majority of Maori ethnic children growing up today have a non-Maori parent” (Chapple, 2000, 105), with “half of Maori living as part of a couple have non-Maori partners” (Callister, 2004, 9).

A wide range of lifestyle and socioeconomic class differences contribute to how people identify ethnically. The Maori middle class is growing. According to the 2013 census data the percentage of Maori professionals has increased from 14.0% in 2006 to 16.4% in 2013 with the numbers in managerial positions also increasing from 11.6% to 13.1% over the same period. There is a decrease in the numbers of Maori in laboring jobs, down from 21.2% in 2006 to 19.4% in 2013 (Statistics New Zealand, 2013). The statistics confirm the pattern of Maori middle class growth identified by Coleman, Dixon, and Mare (2006) as occurring from the early 1960s. The pattern is significant because socioeconomic class location intersects in complex ways with ethnic/race identification. Those in higher socioeconomic categories are more likely to identify as Maori and other, rather than Maori-only (Callister, 2003, 109–40). Those who do identify as sole Maori tend to be amongst the disadvantaged and the “most disadvantaged are less likely to identify with a tribe” (Gill et al., 2010, 19).

III. EXAMPLES OF A SEPARATE MAORI HEALTH APPROACH

In this section, we examine two examples of separate Maori health provision justified by the construction of an essentialist ethnic category, a strategy that we located in Treaty partnership politics. The first example is the Maori mental health, Te Whare Tapa Wha (literally, the four cornerstones of Maori health) developed by Mason Durie during the 1980s (Durie, 1985, 483–86) and implemented into the health system during the 1990s (Ministry of Health, 2009). Its endorsement of “a Maori world view . . . and an ecological approach to health that accorded with indigenous philosophies” (Durie, 2011, 30) was seen to compensate for what was regarded as a lack of a spiritual dimension in the Western-based secular health services. This involves “the inclusion of the wairua (spirit), the role of the whanau (family) and the balance of the hinengaro (mind) as being as important as the physical manifestations of illness” in addition to taha tinana or physical well-being (Maori Mental Health Model, 2010). The second example is a Treaty research model in health designed to show researchers a practical means to acknowledge the principles of the Treaty of Waitangi (Wyeth et al., 2010, 303–16). Both examples demonstrate how ethnic categorization is justified in terms of the perceived essentialist character of Maori identity rather than in terms of the needs of subgroups whose social experiences—such as the ones described by Chapple (2000 cited above)—are more likely to lead to health disorders.

Since the 1980s and 1990s, the reported figures of mental health disorders prevalent among New Zealanders have been characterized by marked disparities between Maori and non-Maori ethnic groups; this is particularly so for schizophrenia and bipolar disorder (Baxter, 2008). However, it is male Maori adults under the age of 40 who are most likely to be clients of mental health services (Ministry of Health, 2006), rather than the Maori category as a whole. In addition, the growing disparity in this subgroup of anxiety and eating problems along with the more traditional “mood, alcohol and drug disorders” aggravates this predicament further (Ministry of Health, 2006). Although disparities for those affected negatively are always of concern, they are particularly poignant in the domain of public health because, at the extreme, they represent differential access to the benefits and rights of our society, reduced ability to contribute to future generations, and premature death (Te Ropu Rangahau Hauora a Eru Pomare, 2000, 1–6).

The official discourse of a distinctive Maori model of health exerts a significant influence on health professionals, educators, and researchers to accept a distinction between Maori and non-Maori people. This is based on the belief that Maori have a different paradigm of health beliefs and practices from other groups (Cunningham, 2000, 62–69; Durie, 2006, 1138–43). The belief that cosmological or spiritual forces may cause illness and psychological distress is contrasted with the Western biomedical model (Durie, 1999, 5–12). The purported differences in world and health views are not only deemed irreconcilable but also not amenable to exegesis, since, for the proponents of these ideas, “indigenous knowledge cannot be verified by scientific criteria” (Durie, 2006, 1138–43). Yet a comparative study conducted to investigate lay perceptions of Maori and non-Maori participants using major depressive disorder as a paradigm observed no evidence of different views between the two groups (Marie, Forsyth, and Miles, 2004, 225–52). These findings challenge the claim (Durie, 2011, 24–36) that Maori and non-Maori individuals have a different understanding and familiarity with mental illness (Marie and Miles, 2007, 34–41).

The acceptance of a fundamental Maori/non-Maori distinction has a number of consequences. For example, some drug and alcohol treatment programs and regional mental health units identify individuals as Maori—based on self-identification, appearance, and/or surname—and proceed to offer treatment within the Maori mental health framework on those premises. Individuals may need to specifically request the alternative “mainstream” option (Carlos Zubaran, personal communication). In addition, a common belief within the New Zealand mental health community is that patients from a Maori cultural background develop more side effects to conventional antipsychotic medication in comparison with the general population, particularly with respect to movement disorders (Mahmoud, Johnson, and Tawhai, 2008, S164). Consequently, there is a stigma attached to the use of these neuroleptics in this population, which may negatively influence the attitudes and reactions of Maori patients to their recommended treatment. This demonstrates at an individual level the effects of justifying separate health care according to perceived categorical differences, such as the difference between Western and indigenous worldviews. The individual may reject the neuroleptics because they are considered to be “Western” and therefore not appropriate medication for someone who identifies as Maori, rather than for reasons based on his own medical needs, which may arise from social factors.

IV. CONCEPTUAL AND METHODOLOGICAL PROBLEMS

Ethnic discourses contain confused concepts of (1) “race” and (2) “ethnicity” that lead to methodological problems. The “racial” explanation of ethnic belonging, which mixes biology and culture in the idea of genealogical descent and draws on beliefs in the spiritual transmission of culture, is confused with the discourse of genetics. For example, according to a Maori leader, “If you are born a Maori, then you have to accept the consequences of that biological fact, and the culture that comes with it” (Mead, 1997). In contrast to the link between biological descent and culture, the term “ethnicity” is used to emphasize the cultural component of an identity located in a group with historical heritage. The New Zealand government’s definition of ethnicity refers only to ahistorical sociocultural factors, such as language, and also allows for cultural self-identification, so that individuals who do not have Maori ancestry may claim to be Maori. The term “race” is not included in the government definition (Statistics New Zealand, 2005). This is an attempt to avoid the association with the biologically determined notion of “race” with its historically unacceptable connotations.

However, locating social identity in ancestry—as is the case if an individual wishes to identify with a tribe (despite the government definition)—means that “ethnicity,” with its cultural component, takes on the unchangeable quality attributed to “race.” This contrasts with the use of “ethnicity” to refer to beliefs and practices that are culturally located, which therefore enables “ethnicity” to be distanced from the meaning of “race” with its biological “blood” connotations. Yet the problem is that the contemporary ethnic group interprets “ethnicity” as a biological category. The quotation from Mead above makes that clear. To complicate matters, the resurgence of tribal politics where tribal membership is based on biological antecedents reinforces the biological meaning.

Maori leaders have found a way through this conceptual confusion by describing Maori experiences in terms of an individual’s “spiritual” inheritance as an indigenous person (Rata, 2011). To that end, poor Maori mental health status is seen as the result of being Maori; that is, to be Maori today is to experience life as an indigenous—but colonized and “damaged”—individual within a decimated social structure, viz. the tribe.

According to Durie (2003, 510–11), the “climate of material and spiritual oppression” faced by members of indigenous populations worldwide produces an “increased susceptibility to disease and injury.” Although genetic descent is the criterion for tribal membership, the concept of genetic transmission in indigenous discourse tends to be metaphysical rather than “racial,” with the genetic connection between generations expressed in “spiritual” terms. The transmitted oppression is believed to operate at the psychosocial level as the intergenerational “spirit” (mauri or wairua) carries the memory of oppression from one generation to the next to cause mental or spiritual illness in succeeding generations. This explanation of ongoing disadvantage experienced by the Maori category as a whole avoids the biological determinism—with its racial overtones—associated with tribal membership. However, the explanation does not solve the conceptual difficulties that are aggravated by the confusion between Statistics New Zealand’s (2005) definition of ethnicity as a self-referential social identity, on the one hand, and policies in education and health, including such as scholarship criteria and researcher’s status, on the other hand. These require a person claiming Maori identity to show evidence of Maori ancestry (increasingly glossed as “background”).

Along with the conceptual confusion in the terms used for ethnic categorization, the quality of data used to support ethnic classification in public policy also contains problems. Treaty categorization politics requires a defined “partner,” but the “non-Maori” category lacks any specific characterization because it does not represent any ethnic group in particular (Marie, Forsyth, and Miles 2004, 225–52). In addition, intra-ethnic variation may be more salient than the inter-ethnic differences of the Maori/non-Maori dichotomy (Broughton, 1984, 290–91). That “all Maori have some degree of non-Maori ancestry” (Butterworth and Mako, 1989, 1) means there is a range of possible identifications available to the Maori population. In addition, as a result of international migration (currently almost a quarter of New Zealand’s population is born overseas) and intermarriage in New Zealand between people from different countries, significant methodological challenges arise for categorizing people into static ethnic groups (Callister et al., 2007, 299–320). To further complicate the ethnic categorization process, a significant proportion of people born in New Zealand, including Maori, live abroad (Bedford et al., 2004, 131–41).

The classification of humans according to categories based on “race” and “ethnicity” is of questionable if not unsustainable scientific validity. It does not support the political construction of such categories, even if such categorization is advocated on the basis of a political accommodation. Current genetic data reject the notion that purportedly different “human races” constitute genetically distinct human populations. Studies developed during the last decade provide evidence that there is no connection between self-identified “race” or ethnicity and frequency of particular genetic variants (Marshall, 1998, 654–55; Angier, 2000, F1). Low level of genetic variability and of structuring of the human species is incompatible with the existence of “race” as a biological entity (Pena, 2005, 321–46). Furthermore, differences in human skin color are fallacious indicators of biological differences among populations (Parra, Kittles, and Shriver, 2004, S54–60). “Race,” whether imposed or self-identified, is considered a weak surrogate for genetic and nongenetic factors associated with health status (Royal and Dunston, 2004, S5–7).

Apart from the application of a “racial” taxonomy and the use of “ethnicity” as a surrogate for “race” (Zubaran, 2009, 43–52), there are other fundamental difficulties in using these concepts as variables in public health surveillance. These limitations include differences in terminology, data collection procedures, perceptions of group identity, and changing demographics of population subgroups. A further and more fundamental challenge to biomedical research is to resist reifying the dogmas of “race” while advancing further in the genetic and molecular correlates of disease (Durie, 2006, 1138–43).

A series of studies has highlighted inconsistencies in the way that “race” and, more recently, “ethnicity” have been defined in New Zealand for the purpose of health care and policy making (Thomas, 2001, 86–88; Blakely, Robson, and Woodward, 2002, 149–50; Smartt et al., 2002, 167–69). The definitions of “race” and “ethnicity” have been criticized in the context of ischemic heart disease (Priest and Jackson, 1994, 3–4), diabetes (Swan, Lillis, and Simmons, 2006, U2103), breast cancer (Curtis, Wright, and Wall, 2005, U1298), and pediatric community care (Bramley and Latimer, 2007, U2779). The authors of the latter publication concluded that the poor quality of data based on ethnicity was “an ongoing issue” in the New Zealand health sector (Bramley and Latimer, 2007, U2779). In fact, continuous investigation over more than two decades in New Zealand has consistently revealed inconsistencies in the way ethnicity data are coded in hospital records (Swan, Lillis, and Simmons, 2006, U2103). In addition, the validity of Maori mortality statistics has been challenged due to discrepancies in ethnic classification (Graham et al., 1989, 124–26).

Data from the coronary heart disease register revealed that, when ethnicity information coded by the interviewer during hospital admission was compared with self-defined ethnicity information, the number of those initially classified as Maori increased almost 12%, whereas only 0.5% of those classified as “other” considered themselves to be Maori (Priest and Jackson, 1994, 3–4). The misclassification of ethnic identity was found “particularly problematic” in the group labeled as the “other” ethnic group (Swan, Lillis, and Simmons, 2006, U2103). In recent studies, Maori and non-Maori individuals continue to be classified according to official demographic categories (Wheeler et al., 2011, 30–38) or fall short of providing detailed information on how Maori research participants are distinguished from their non-Maori counterparts (Thomas et al., 2010, 208–16).

Bias can arise when comparing health standards in Maori and non-Maori populations. In a study that reviewed 98 research reports comparing Maori and non-Maori samples from 1980 to 1996, only 19% of the articles reported any information about the criteria used for categorizing ethnicity and only less than 3% mentioned how people of dual or multiple ethnicity were categorized. This led to the conclusion that much of the New Zealand research comparing Maori and non-Maori samples is flawed, and that most articles in that period did not meet minimum expected standards for reporting procedures for categorizing ethnicity (Thomas, 2001, 86–88). Even in a well-developed study using linked census and mortality cohort data sets, changes in definition of ethnicity have been recognized as a possible source of bias in the study (Blakely et al., 2006, 44–52). It is plausible to infer that similar circumstances may occur in other countries with diverse population compositions.

The authors of a study in which ethnic differences in mental health and health style choices of New Zealand “Europeans,” Pacific Islanders, and Maori were compared observed that the priority system of coding ethnicity “gives special priority to Maori,” which results in a faulty system that “loses detail and makes [unwarranted] assumptions” (Goodyear-Smith et al., 2005, U1374). Furthermore, people classified within the group of “unknown ethnicities” belong in fact to a variety of ethnic groups, and their inclusion in the group of “others” is also inappropriate (Bramley and Latimer, 2007, U2779). The use of the term “non-Maori” to those who are not Maori was described as an “extreme” example of inaccurate labeling of human populations (Reinken, Salmond, and de Lacy, 1983, 174–76). Although some researchers have recommended methodological “adjustments” for classification and measurement for public health surveillance purposes in New Zealand (Blakely, Robson, and Woodward, 2002, 149–50), the persisting inaccuracies and methodological obstacles for that are likely to persist.

The excessive emphasis on genetics, the conceptual problems associated with “race” and “ethnicity,” and the major methodological problems associated with the use of these terminologies tend to obfuscate a myriad of other possible explanations for health disparities between subgroups.

V. SOCIAL EXPLANATIONS

Significant evidence supports the claim that socioeconomic differentials are likely to be a fundamental explanation for the observed inequalities in health status between minority and predominant groups (Nazroo, 2003, 277–84). Indeed, the New Zealand Ministry of Health (2009) does recognize that the experience of socioeconomic disadvantage—associated with factors such as low household incomes, limited education, and living in neglected areas—contributes to differences in mental health prevalence between Maori and non-Maori individuals.

According to Nash (2001, 23–36), the placement of Maori within very broad socioeconomic categories is significant in understanding the disparities between Maori and non-Maori. Although his research was undertaken in education, the findings are relevant to health, given that the same disparities exist across the social sectors. Nash argued that the “issue at stake” in understanding social disparity causation was “the significance of within class differences in literate resources in both populations” (italics in the original). Like Marie, Fergusson, and Boden (2008, 183–96), Nash (2001, 23–36) was unable to locate evidence of ethno-cultural factors contributing to Maori disadvantage in education.

The national mental health survey also found that the prevalence of mental disorders was greatest among Maoris with lowest household income and levels of education (Baxter et al., 2006, 914–23). Associations between social deprivation and adverse outcomes in health, such as long-term drug use among the poor and disenfranchised, are relevant to the understanding of the low status of Maori mental health. A study designed to compare the profile of individuals attending alcohol and other drug treatment services in New Zealand revealed that Maori patients consumed more cannabis, were younger, and were less likely to attend follow-up appointments (Adamson et al., 2006, U2284).

The relative youth of the Maori population is a contributing factor to the size of the marginalized group. According to Natalie Jackson (2011), in 2006, the median age for the Maori population was just 23 years, compared with 38 years for the “European/New Zealanders/Other populations” (Jackson, 2011, 2). It has been proposed that the “differences in the timing and speed of demographic transition between the Maori and European-origin populations have led to significant differences in age structure” (Jackson, 2011, 7). Young people are more likely to engage in behaviors that put their health at risk.

The claim that “the policy issue may need to be viewed primarily at a sub-cultural and socio-economic level rather than the coarse ethno-cultural level of Maori/non-Maori binaries” (Chapple, 2000, 101) has gained additional support (Collins, 2004, S13–15; Sankar et al., 2004, 2985–89). According to their research, which criticizes the excessive emphasis on genetics as a foremost explanatory element for health disparities, racial or ethnic disparities are considered to derive primarily from differences in a myriad of factors, including culture, diet, socioeconomic status, education, access to health care, discrimination, and additional social determinants (Durie, 1999, 5–12; Marie, Forsyth, and Miles, 2004, 225–52). Additional evidence substantiates the assertion that socioeconomic deprivation is the main factor in explaining different health outcomes in distinct ethnic groups (Chandola, 2001, 1285–96; Bramley et al., 2005, 844–50). In fact, it has been claimed that analyzing ethnic or racial inequalities in health status without a groundwork on socioeconomic factors is naïve (Bhopal, 2002, 156–57) and that attention should be directed to all socioeconomically disadvantaged individuals (Ellison et al., 1997, 1485). Significant evidence from international studies supports the claim that socioeconomic differentials are likely to be a fundamental explanation for the observed inequalities in health status among minority groups (Nazroo, 2003, 277–84).¹

In understanding the reasons for the low health status of various subgroups of minority populations, it is necessary not only to focus on socioeconomic status but also to examine the influence of racism and discrimination. Further, it is necessary to examine the complex relationship that exists between socioeconomic position and experiences of racism and discrimination. These experiences, rather than ethnicity or “race” per se, are among the main vectors through which limited access to health care and unfavorable health outcomes affect members of minority groups. Indeed, ethnicity or “race” as a blanket category is insufficient to explain ethnic or racial inequalities in terms of access to health care (Alegria et al., 2002, 1547–55).

There is also extensive international research evidence indicating that racism plays a major role as a health determinant; the subjective experiences of racism and discrimination, rather than affiliation to a “race” or ethnic category per se, are what is associated with poor health outcomes (Krieger, 2003, 194–99; Paradies, 2006, 888–901; Peters, 2006, 234–40; Barnes et al., 2008, 1241–47; Dominguez, 2008, 360–70). A number of studies have identified the association between self-reported experiences of racial discrimination and poor physical and mental health outcomes for a range of ethnic groups in various countries (Karlsen and Nazroo, 2002, 624–30; Williams, Neighbors, and Jackson, 2003, 200–08; Karlsen and Nazroo, 2004, 1017–18). Personal experiences of racism, harassment, and discrimination are associated with multiple indicators of poorer physical and mental health status (Nazroo, 2003, 277–84). Socially inflicted trauma and internalized oppression are considered among the leading pathways through which racism can harm health (Krieger, 2003, 194–99). In New Zealand, self-reported experiences of discrimination are associated with unfavorable health outcomes (Harris et al., 2006, 1428–41). It is likely that experiences of discrimination such as those described in the studies cited above may explain the “unexplained disparity” (Ministry of Health, 2006) that is claimed to exist between Maori and non-Maori, even when socioeconomic factors are taken into account.

However, disagreement exists among researchers concerning the extent of the direct influences of racism, how racism affects the meaning people give to ethnic identity, and how the interplay between these experiences of racism and socioeconomic status act to affect health status. In fact, racism is experienced more frequently by individuals in marginalized socioeconomic groups rather than by people in the middle and upper middle class (Adamson et al., 2003, 895–904). Discrimination is multidimensional and produces negative emotional states such as anxiety and depression, which in turn alter biological processes or patterns of behavior (James, 2003, 189).

The type of discrimination associated with higher mental health needs is relevant to understanding which subgroups in the population are most affected. One study conducted in multiple New Zealand regional hospitals revealed that volunteers who self-identified as Maori or Pacific Islanders had reported more psychotic phenomena when compared to ethnic groups included within the “all other” category. The differences between ethnic groups also varied according to socioeconomic deprivation level (Trauer, Eagar, and Mellsop, 2006, 310–21). In addition, discrimination is not confined to the Maori/non-Maori binary. In the work setting, Asians reported the highest prevalence of discrimination (Harris et al., 2006, 1428–41). Other immigrants, including non-English Europeans of white appearance, may be particularly vulnerable to racial discrimination due to limited proficiency in the native language, undocumented immigrant status, and different cultural and religious backgrounds.

A study of the personal experiences of racial discrimination—such as ethnically motivated physical or verbal attacks and different types of unfair treatment that affected different minority groups in New Zealand, including Asians, Pacific peoples, and Maori—found the highest prevalence of “ever” experiencing any of the forms of racial discrimination (34%) in the Maori group, followed by similar levels among Asian (28%) and Pacific peoples (25%) (Harris et al., 2006, 1428–41). However, this finding allows for different interpretations in that both the method of classification and appearance (e.g., skin color) are implicated in any analysis of discrimination toward Maori. Socioeconomic class does appear to be implicated in links between discrimination and skin color. A Maori person with “white” physical appearance is more likely to identify as both Maori and non-Maori, a group tending to be “socially and economically much better off than all other Maori” (Callister, 2008, 18–25) and therefore less exposed to the full spectrum of racism and discrimination experienced by individuals who record Maori-only ethnicity in official surveys. According to Callister (2008), this group may be those who “look more Maori,” and “if discrimination is common in New Zealand, the Maori-only (or Pacific peoples) group would be more likely to suffer discrimination from police, landlords and healthcare providers” (Callister, 2008, 21).

VI. CONCLUSION

The argument presented in this article is that the Treaty of Waitangi partnership politics unfortunately requires that Maori be regarded as a discrete ethnic/race category. This means that the complex nature of social experience as an ethnicized person—one who regards him or herself as a member of an ethnic group or is treated by others in ethnic terms—is essentialized to the entire ethnic or “race” group in the interests of tribal politics. It also means that the wide variation of intragroup differences within the Maori population, including the sociocultural lifestyles of the marginalized subgroup identified by Chapple (2000), is portrayed as consequences of the historical experiences of the entire ethnic group and the cause of the persistence of disadvantage. It is more likely that sociocultural factors that may or may not have an ethnic dimension are the cause of poor health outcomes.

However, ethnic classification into a Maori/non-Maori binary deflects attention from the specific subgroups within the Maori population that experience discrimination by obfuscating the complex links between discriminatory experiences, low socioeconomic status, and marginalized lifestyles. Although it is recognized that being Maori for some does have social meaning that contributes to poor health, that social meaning is the result of specific material circumstances, which produce subjective experiences for individuals. In other words, it is in the materiality of the ethnic state, not in ethnicity as an essentialized state, that a more useful explanation of the causes of Maori health disadvantages is to be found. Furthermore, racial discrimination in particular is the subjective experience of ethnicity that links ethnic identification to poor health.

The recognition that “socio-economic performance may determine choice of ethnic group for some people” (Chapple, 2000, 101) sensibly addresses the possibility of causative links between class position and ethnic identity. If, as claimed, “ethnicity may itself be caused by disadvantage [with] negative stereotypes contribut[ing] to a social climate where reverse causality is possible” (Chapple, 2000, 101), the ethnic factor may be acknowledged in research and policy, but not as a broad categorization of an essential group. Unfortunately, the commitment by successive New Zealand governments since the 1980s to ethnic categorization as a research and public policy tool and the uncritical adoption of this approach in some quarters may well contribute to the negative stereotypes that produce racism (Ellison et al., 1997, 1485), thereby feeding into the cycles of discrimination for those Maori subgroups.

The quest for alternatives to ameliorate the health standards of Maori people is a timely and much needed endeavor. However, the answer to this demand does not lie in ethnically based health provision. The understanding of all Maori as a discrete ethnic category as well as the development of health and other public policies based on categorical ethnic distinctions should be abandoned in favor of a more tenable approach that recognizes a series of other causal factors in social disadvantage.