On the 27th of January, 2004, Don Brash addressed the Orewa Rotary Club on the subject of race relations, and what he perceived to be the preferential treatment of Maori in health and education.1 According to the New Zealand Herald newspaper, ‘The Orewa Speech’ has had a ‘seismic impact on the political landscape’ in New Zealand.2

The National Party’s meteoric rise in the polls3 suggests that their new leader has indeed struck a chord with the New Zealand public.

This viewpoint article seeks to evaluate, from the perspective of health, whether specifically addressing the needs of Maori is justified—or whether (as Brash suggests) such an approach cannot be justified. One way to determine the legitimacy of a targeted strategy, is to prove that Maori have a need in excess of other societal groups. With regard to need, mortality and life expectancy data are arguably the most frequently used measures of health status. This is due, in part, to their accessibility; certification of death being a legal requirement in most industrialised countries.4 In New Zealand, the most recent life expectancy figures (2000–2002) show that non-Maori live on average 8.5 years longer than Maori.5 Specifically, life expectancy at birth for females of Maori ethnicity is 73.2 years compared with 81.9 years for their non-Maori counterparts. For men, life expectancy at birth is 69.0 for Maori and 77.2 years for non-Maori.

Until recently, meaningful comparisons of ethnic-specific mortality rates and the study of time trends in Maori and Pacific mortality rates have been hampered by changing definitions of ethnicity and a numerator-denominator bias. 6 Mortality rates are calculated using two primary sources: death registration forms (the numerator) and census records (the denominator). Between 1981 and 1995 funeral directors completing deaths registration forms used a biological definition of ethnicity (half or more blood), while the census collected self-identified ethnicity data. This resulted in an undercounting of Maori and Pacific deaths and the underestimation of Maori and Pacific mortality rates.

Ajwani et al (2003) address this bias in their comprehensive description of trends in ethnic-specific mortality rates in New Zealand from 1980 to1999.6 Their results are disturbing. There has been little, if any, decline in Maori and Pacific mortality rates over the last two decades, despite a steady decline in non-Maori non-Pacific rates. Hence, the gaps in life expectancy between these groups increased markedly over the 1980s and 1990s.

The major contribution to the disparity has been the differential chronic disease mortality in the middle to older age groups—most notably from ischaemic heart disease, diabetes, and cancer. For example, for the prioritised series (those persons identifying Maori ethnicity as their primary, but not sole, ethnic group) age-standardised cancer mortality rates were 2.0 (male) and 2.1 (female) times greater for Maori than for non-Maori non-Pacific people during 1996–1999.

Cardiovascular mortality rates have decreased over the last two decades, but to a lesser extent amongst Maori. Consequently, by 1996–99, cardiovascular mortality rates were 3.0 times higher for Maori males compared to non-Maori non-Pacific males, and 4.2 times higher for Maori females compared to non-Maori, non-Pacific females.6 Carr et al (2002) found similar disparities in a study of heart failure outcomes.7 In a retrospective analysis, they compared outcomes for Maori and non-Maori using mortality and hospital admission data from 1988 to 1998. Mortality from heart failure was more than 8.8 times higher among Maori men aged 45–64 years when compared to non-Maori.

To understand what is contributing to these disparities in life expectancy and mortality rates, it is necessary to consider some of the determinants of health status. It has long been recognised, for example, that poor socioeconomic status is associated with poor health outcomes. In fact, the link was established as early as 1840, when reliable data first became available.8

More recently, it has been observed that about 80% of the 80 most-common causes of death occur more frequently in lower socioeconomic groups.9 Ajwani et al (2003) also cite numerous data from both the United Kingdom and United States that confirm that widening health inequalities are strongly associated with widening socioeconomic inequalities.6 Similar gradients are evident within the New Zealand population.

Several tools are available for measuring socioeconomic status. The NZDep96 is an area-based measure of socioeconomic deprivation that uses nine variables obtained from 1996 census data to provide a summary deprivation score for each meshblock (a geographical unit containing a median of 90 people) in New Zealand. The score is based on the proportion of people within each meshblock who don’t have access to a telephone or car, are unemployed, in receipt of a means-tested income, live in a low income or single-parent family, have no educational qualifications, live in non-tenured homes, and live in crowded households. The meshblocks are ranked into deciles with one being the least deprived and 10 being the most deprived.10

Although care should be taken in using NZDep96 as the sole indicator of socioeconomic status, the ranking of individuals by the decile assigned to their neighbourhood is, on average, strongly and linearly related to health and other social outcomes.11 An examination of NZDep96 by ethnicity reveals that the Maori ethnic group is over-represented in the most deprived areas. Reid et al (2000) found that more than half (56%) of Maori live in areas represented by the three lowest deciles.12

Indeed, only 3% of Maori are represented in decile one (the most affluent) whilst 26% reside in decile 10 (the most deprived). This pattern is consistent within both categories of the Maori ethnic group: sole Maori (those who give Maori as their only ethnicity) and mixed Maori (those who give Maori as only one of their ethnicities). Other socioeconomic data reveal that Maori households (when compared with non-Maori) earn on average $10,000 less than non-Maori households, are more likely to rely on income support than non-Maori, and are more likely to run out of food because of a lack of money.13

Hence, these data suggest that Maori are over-represented in the lower socioeconomic classes, and that this level of deprivation puts them at increased risk of poor health outcomes. However, there is clear evidence that poor health outcomes in Maori are not simply due to socioeconomic deprivation. For example, Reid et al (2000) combined NZDep96 data from deciles 1–7 (due to the relatively few number of Maori in this category) and deciles 8–9, and compared them with decile 10.12 They found that life expectancy of Maori from birth is consistently less than that of non-Maori at each of these three deprivation points. These authors state that such data provide support for the argument that ‘above and beyond the effects of increasing deprivation in New Zealand, there is an additional and significant health effect of ethnicity’.

Similar patterns are noted by Sporle et al (2002), who used data from 1996–97 to update previous studies of social-class mortality differences in Maori and non-Maori New Zealand men aged 15–64 years.14 They found that within each social class, Maori all-cause mortality was significantly higher than non-Maori, and that these disparities increased with increasing disadvantage. Moreover, the mortality rate of Maori in the highest social class was still higher than that of non-Maori in the lowest social class. The greatest disparities between Maori and non-Maori were found in an analysis of amenable causes of death, that is, deaths that should have been avoidable had the individuals received appropriate medical care.

Sporle and colleagues conclude that the persistently high Maori mortality rates, when controlled for social class, indicate that the poor state of Maori health cannot be explained solely by relative socioeconomic disadvantage. They further note that the high Maori rate of potentially preventable deaths indicates that the health sector is still not meeting the serious health needs of this ethnic group. Indeed, in terms of targeting health resources, this research indicates that Maori ethnicity and high need are synonymous.

In relation to other determinants of health, there is evidence to suggest that Maori are over-represented among those with intermediate risk factors for disease. Provisional results of the 2003/03 New Zealand Health Survey demonstrate that the prevalence of smoking in Maori is more than twice that of non-Maori, and higher than any other ethnic group.15 This data also shows that the prevalence of hypertension (a major risk factor for heart failure) is highest in Maori and that approximately 30% of Maori are obese. These risk factors have a considerable impact on both the rate and outcomes of amenable disease and are likely to contribute to the significant burden of disease suffered by Maori.

Disparities between Maori and non-Maori in the treatment of chronic disease have also been observed. For example, diabetic Maori in one study appeared to have poor glycaemic control and poor knowledge about diabetes.16 The South Auckland Diabetes Study confirmed these findings17 whilst other research demonstrates that Maori patients are more likely to have diabetic complications such as nephropathy and retinopathy.18

Research into cardiovascular disease and outcomes suggests that, despite high hospitalisation rates for cardiac diseases, intervention rates for Maori are much lower.19 Similarly, Tukuitonga and Bindman (2002) state that even though Pacific and Maori peoples have higher rates of coronary artery disease morbidity and mortality, revascularisation rates are lower in both groups.20 Adequate access to care is also raised by Ellison-Loschmann et al (2004) who found that asthma hospitalisation rates are higher in Maori than in non-Maori, despite asthma prevalence being similar in Maori and non-Maori children.21 They conclude that the differences are likely to reflect differences in asthma exacerbation and disease severity due to reduced access to asthma health services.

Although there is considerable evidence for the existence of significant disparities between Maori and non-Maori in socioeconomic status, distribution of risk factors for disease, and access to healthcare, there is less definitive research identifying the barriers and obstacles to good health outcomes. It has been suggested that the younger age structure of the Maori population, in comparison to the general population, may play a role in the inequalities.22 Failure to account for these structural differences, when designing and delivering services, may help explain why health promotion and education strategies have had less impact on this ethnic group. Cost has been recognised as a barrier to care with regard to diabetes and, given the socioeconomic data previously discussed, it is likely that this obstacle plays a significant role in other areas of Maori health care.22,23

Although caution is required in attributing blame, the Government reforms of the 1980s and early 1990s may also have had a disproportionate effect on Maori. The sharp rise in Maori suicide rates may suggest that Maori suffered considerably during this period.

In 1980, Maori men aged 24–44 years had suicide rates 22% less than non–Maori—but by 1999, the rates were 70% higher than non-Maori.6 More recently, Malcolm (2003) has questioned whether health resources are actually filtering down to where they are needed most and suggests that the inverse care law (ie, those populations in greatest need are those least likely to receive the services they need) remains a dominant feature of New Zealand’s primary care system.24

To summarise, the epidemiological data highlight several important points in terms of healthcare policy and funding. First, Maori have significantly lower life expectancy than non-Maori; the high mortality rates are largely due to the burden of chronic disease. Second, Maori are over-represented in socioeconomic-deprivation statistics; a measure highly correlated with greater morbidity and mortality. Third, and most importantly in terms of ‘The Orewa Speech’, Maori have worse health outcomes independent of socioeconomic measures. Finally, Maori suffer disproportionately from diseases amenable to health service intervention.

Together, this evidence provides a compelling argument for specific initiatives focused on improving Maori health outcomes and reducing disparities. Contrary to the opinions of Dr Brash, current evidence identifies a need for health policies to continue to directly target Maori and further, aim to elucidate the barriers to care that presently exist.

The arguments above cite epidemiological evidence for targeting Maori as an ethnic group. However, there are other grounds, the most obvious of which is the Treaty of Waitangi.

The Treaty, signed in 1840, consists of three articles:

It may be argued that the current inequalities in health status between Maori and non-Maori violate this contract and that the Crown has not provided equal protection and rights to Maori compared with non-Maori. For example, Reid et al (2000) state that the current disparities are a breach of the Treaty that sought to protect Maori from marginalisation as a result of colonisation.12

Durie (1989) notes that the deteriorating health of Maori may actually have been the driving force behind a formal relationship between the British Crown and Maori.25 He states that a British protectorate was proposed in 1837 due to the ‘miserable’ condition of the Maori people, particularly their ‘high mortality rate’ resulting from ‘total European’ impact.

The Treaty of Waitangi represents the New Zealand Government’s contractual obligation to explicitly ensure equitable outcomes for Maori. Thus it offers a legal impetus for addressing the continuing health needs of this specific ethnic group.

The Ministry of Health, amongst other Government bodies, recognise these responsibilities and obligations. In terms of the New Zealand Health Strategy,26 recognition of the Treaty translates to the participation of Maori at all levels, partnership with Maori in service delivery, and the protection and improvement of Maori health status.

In conclusion, epidemiological data provide strong support for specifically addressing Maori health need, whilst the Treaty of Waitangi represents a contractual obligation on behalf of the New Zealand Government to ensure equity of outcome for Maori. Underpinning both the epidemiological and legal arguments are ethical principles. The central tenets of medicine, to reduce suffering, improve and prolong the quality of life, should provide a strong driving force to address these inequalities.

Health policies should be shaped by the four basic principles of medical ethics: beneficence, non-maleficence, justice and autonomy. The moral imperatives that form the very basis of the medical profession should drive us (its practitioners and students) to strongly resist any health policy that attempts to thwart the progress of Maori towards equality in health status.

In essence, with regard to The Orewa Speech, Dr Brash’s claims that Maori have the ‘upper hand’ remain unsubstantiated. Unfortunately, these views may provide yet another threat to Maori health.

Author information: Cindy Towns, PhD Student, Department of Anatomy and Structural Biology; Nathan Watkins, Medical Student; Arapera Salter, Medical Student; Patricia Boyd, Medical Student, Dunedin School of Medicine; Lianne Parkin, Lecturer in Epidemiology, Department of Social and Preventive Medicine, University of Otago, Dunedin.

Correspondence: Cindy Towns, C/- Department of Anatomy and Structural Biology, University of Otago, Dunedin. Fax: 03 479 7254; email: cindy.towns@anatomy.otago.ac.nz