View Article PDF

Life expectancy is one of the most widely used measures in demographic and health analysis, and in New Zealand is among the highest of any country.1 Equity in health outcomes has long been a goal in New Zealand and is measured mainly in terms of the reduction or elimination of health inequalities between Māori and Pacific. Life expectancy differentials are a frequently cited inequity2 and are persistent within the Māori and Pacific populations when compared to non-Māori and non-Pacific populations.3 In 2012–14 life expectancy differentials were 6.8 years in Māori females and 7.3 years in Māori males. Within the Pacific population, the differential was 5.2 years in females and 5.8 years in males.3 While the life expectancy differentials in Māori have closed since c. 1900,4 a widening occurred in the 1980s and 1990s.5 This widening has partly been attributed to the economic structural reforms that occurred during this period. These reforms had a larger impact on Māori than other ethnic groups.6

The underlying factors contributing to ethnic health inequities, particularly life expectancy, are multifaceted and complex. This is particularly so when the experiences and pathways leading to inequity are likely unique for both the Māori and Pacific populations. Many of the drivers are related to the unequal distribution of the determinants of health. For example, the unequal distribution and access to resources such as income, education and employment, factors which often compound one another.7 These factors also pattern exposures to other risk factors such as tobacco use, poor nutrition, overcrowded and poor-quality housing, and drug and alcohol use.8 This inequitable distribution, particularly for Māori, is rooted in structural hierarchies that are associated with colonisation.9 This is not unique to New Zealand; in countries with a colonial past, indigenous populations frequently will have poorer health, even once socioeconomic position is factored in.7 Furthermore, access to the determinants of health and the associated ethnic inequities in health outcomes can arise from processes related to racism, as one’s social environment and where people are placed in the social hierarchy can shape health.10 The contribution of racism, particularly institutional racism, to health inequities is increasingly being recognised in New Zealand. It not only drives ethnic inequities in poverty, but also compounds the already significant negative effects of poverty.11–13

Quantifying, understanding and monitoring health inequities are important if they are to be addressed. Avoidable mortality is one measure that can be used as an indicator of the contribution of healthcare along with health and social policy to population health and indeed health equity. It is based on the concept that premature deaths from certain conditions should be rare and ideally not occur in the presence of timely, effective and equitable healthcare or other appropriate interventions.14 Avoidable mortality is recognised as a useful component of a suite of indicators to identify areas of potential focus for equity in the health system.15

The concept of avoidable mortality includes two overarching categories: amenable mortality and preventable mortality.16 Amenable mortality includes causes of death that could have potentially been avoided by means of access to high-quality and timely medical interventions. Preventable mortality is broader and includes deaths which could have been avoided through addressing the wider upstream determinants of health, such as individual-level health risk factors, socioeconomic status and environmental factors. The concept of avoidable mortality classifies the causes of death into one of the aforementioned categories, that is conditions that are ‘amenable only’, ‘preventable only’ or both ‘preventable and amenable’. The list of conditions that define avoidable mortality is based on expert review on causes that potentially could have been avoided given knowledge of casual pathways, the determinants of health and therapeutic technologies at the time of death.

Improving and ensuring equitable health outcomes for New Zealanders has long been a goal of the health system. This goal has been reinforced by the revised New Zealand Health Strategy which places equity in health outcomes at the forefront of much of the work of the health system.17 A change in government has also placed a renewed emphasis on equity in health outcomes and an ‘Achieving Equity in Health Outcomes’ work programme has recently been established by the Ministry of Health.18 A key component of this work programme is to build understanding of health equity through data, analytics and insights. In addition, in 2016 the New Zealand Ministry of Health released a framework for the health sector that aimed to improve the health outcomes of the New Zealand population. This framework, known as the System Level Measures framework, focuses on primary care within district health boards working in collaboration with other health system partners using specific quality improvement measures and has a strong focus on equity.19 Reducing the rates of amenable mortality is one of six measures within this framework.

Rates of avoidable mortality have previously been assessed in the New Zealand context16 and an assessment of potential gains in life expectancy from avoidable causes of death has been made.20 Published papers on the decomposition of life expectancy differentials in New Zealand are rare. However, life expectancy decomposition by cause between males and females in New Zealand has previously been undertaken.21 However, this study did not decompose the differential by avoidable causes of death nor decompose the ethnic specific differentials. To our knowledge, no publications have assessed the ethnic specific life expectancy differentials in New Zealand associated with potentially avoidable causes of death.

Analysis of high-quality health data, particularly mortality data allows us to deepen our understanding and provide insight into health inequities in New Zealand, and is a component of the Ministry of Health equity work programme. To enhance the understanding of the underlying causes of death contributing to the ethnic specific inequalities in life expectancy at birth, we evaluated the contribution of avoidable mortality, using a recently updated definition, to all deaths registered in the period 2013 to 2015. We then decomposed the differentials in life expectancy by avoidable causes of death within each of the Māori and Pacific populations when compared to the non-Māori/non-Pacific population. Specifically, the study aimed to identify the proportion of all Māori and Pacific deaths that are potentially avoidable and to decompose the life expectancy differentials by causes of death that are either preventable, amenable or both.

Method

Non-identifiable data relating to all registered deaths in New Zealand were obtained from the New Zealand Mortality Collection held by the New Zealand Ministry of Health for the period 2013 to 2015.22 Mortality data with an International Statistical Classification of Diseases Tenth Edition (Australian Modification) (ICD-10-AM) coded underlying cause of death is often two years delayed to allow coronial processes to be completed; as such 2015 was the most recent year available. New Zealand mortality data allows for multiple coding of ethnicity; this was prioritised to Māori, Pacific and non–Māori/non–Pacific as per standard protocols for health research.23 Prioritised ethnic-specific population estimates for New Zealand corresponding to the same period as the mortality data were obtained from Statistics New Zealand.

Previously published condition lists of avoidable mortality specific to the New Zealand context were identified from the literature, however given their age and advances in medical treatment and technology, these were considered out of date for use in this analysis as many are over a decade old.24 A review of the literature revealed a recently updated list published by the Office of National Statistics in the UK.25 This list was revised in 2016 and was based on previously published definitions developed for use in Australia and New Zealand. Despite the list being revised for use in the UK, it was deemed appropriate for application in the New Zealand context given that casual pathways and risk factors for the preventable conditions were unlikely to be different in New Zealand. However, minor adjustments were made to the list of conditions considered amenable based on recent amendments to the definition of ‘amenable mortality’ published by the New Zealand Ministry of Health26 and removal of some causes which in New Zealand were considered rare, for example protozoan infections. The underlying cause of death for each registered death occurring in those aged 0 to 74 years were categorised as either amenable, preventable or both (Table 1). A small number of exceptions are made to the age cut-off including deaths from injuries and self-harm, which included deaths at all ages, and acute lymphoblastic leukaemia, which only included deaths in the 0 to 44 age group. Those deaths categorised as still births were excluded.

Table 1: Groups of causes of avoidable mortality.

1Groups used in cause-specific decomposition.
2Includes all deaths in those aged 0–74 unless stated otherwise.
3Amenable deaths have been aligned with definition published by the Ministry of Health in July 2016.

Life expectancy at birth using data for 2013–15 (three year aggregated) was calculated using standard life table techniques, with 90 and older being the final age group. The impact of avoidable mortality on differentials in life expectancy was estimated by decomposing the difference in life expectancy using the method developed by Arriaga.27 This method allows differentials in life expectancy to be decomposed into age group and cause-of-death-specific contributions. Each sums to the total differential in life expectancy between groups.

Life tables for Māori males, Māori females, Pacific males, Pacific females, non-Māori/non-Pacific males and non-Māori/non-Pacific females were developed. The sex-specific non-Māori/non-Pacific tables served as the comparator group to compare the Māori and Pacific sex-specific tables. The contribution of avoidable causes of mortality and of the effects by age on life expectancy differentials in both Māori and Pacific compared with non–Māori/non–Pacific was analysed.

This study was outside the scope of ethics review due to the use of non-identifiable and encrypted data.

Results

Avoidable deaths

Between 2013 and 2015 there were 92,196 deaths registered in New Zealand. Of these, 25,210 (27.3% of all deaths and 68.2% of deaths in the 0 to 74 age group) were potentially avoidable (Figure 1). Of those deaths identified as potentially avoidable, 6,624 (7.2% all deaths) were preventable only, 4,056 (4.4% all deaths) amenable only and 14,530 (15.8% all deaths) both preventable and amenable. Within Māori, there were 9,717 deaths of which 5,152 (53.0% all deaths and 74.5% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in Māori 1,451 (14.9% all deaths) were preventable only, 937 (9.6% all deaths) amenable only and 2,764 (28.4% all deaths) both preventable and amenable. Within Pacific, there were 3,720 deaths of which 1,761 (47.3% all deaths and 71.3% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in Pacific, 355 (9.5% all deaths) were only preventable, 441 (11.6% all deaths) only amenable and 965 (25.9% all deaths) both preventable and amenable. Within non-Māori/non-Pacific, there were 78,759 deaths of which 18,297 (23.2% all deaths and 68.0% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in non-Māori/non-Pacific, 4,818 (6.1% all deaths) were only preventable, 2,678 (3.4% all deaths) only amenable and 10,801 (13.7% all deaths) both preventable and amenable. The proportion of all potentially avoidable deaths was twice as high in both the Māori and Pacific populations compared with the non- Māori /non-Pacific population.

Figure 1: Percentage of deaths that were potentially avoidable by ethnicity and total, 2013–2015.

c


Life expectancy at birth

For all deaths registered in 2013–15, estimates of life expectancy were: 73.5 years in Māori males, 77.3 years in Māori females, 75.0 years in Pacific males and 78.3 years in Pacific females. Life expectancy in the comparator groups was 80.9 years in non-Māori/non-Pacific males and 84.3 years in non-Māori/non-Pacific females. Overall these results equated to differentials in life expectancy of 7.4 years in Māori males, 7.0 years in Māori females, 5.9 years in Pacific males, and 6.0 years in Pacific females.

Decomposition by avoidable category and age

Within both Māori and Pacific females, higher mortality rates at a younger age from conditions considered both preventable and amenable made the greatest contribution to the life expectancy differentials, contributing 2.1 years and 1.7 years respectively (Table 2). Within Māori females, this was followed by conditions that are only preventable (1.3 years). However, within Pacific females, preventable conditions contributed only 0.3 years. Conditions only amenable contributed 1.0 years in Māori females and 1.3 years in Pacific females. Within age groups, mortality in the 50–74 year-old group contributed the most to the life expectancy differentials, contributing 4.1 years for Māori females and 3.3 years for Pacific females.

Table 2: Decomposition of the life expectancy differential in females by avoidable category and age.

1Sum of columns along with non-avoidable and avoidable may not equal the exact life expectancy differential due to rounding.

Similar to females, conditions both preventable and amenable were the greatest contributors to the life expectancy differentials in Māori and Pacific males, contributing 2.7 years and 1.7 years respectively (Table 3). Conditions considered preventable only were the second greatest contributor in Māori males and third greatest contributor in Pacific males; however, the number of years contributed within this group was higher for Māori males (1.6 years) compared with Pacific males (0.8 years). Within Pacific males, conditions considered amenable only were the second greatest contributor to the differential. Within age groups, higher mortality rates in the 50–74 year-old group contributed the most to life expectancy differentials, contributing 4.1 years for Māori males and 3.1 years for Pacific males. Even within the youngest age groups (<50 years), there were large contributions to life expectancy differentials for Māori (2.2 years) Pacific (1.5 years) males.

Table 3: Decomposition of the life expectancy differential in males by avoidable category and age.

1Sum of columns along with non-avoidable and avoidable may not equal the exact life expectancy differential due to rounding.

Decomposition by avoidable cause

Within Māori females, trachea, bronchus and lung cancers accounted for 0.9 years of the life expectancy differential, followed by coronary disease (0.6 years), with diabetes and chronic obstructive pulmonary disease each contributing 0.4 years (Figure 2). Within Māori males, coronary disease, trachea, bronchus and lung cancers and diabetes accounted for 1.2 years, 0.8 years and 0.5 years respectively (Figure 3). In Māori males, avoidable injuries had a pronounced contribution to the differential. Injuries contributed 1.0 year to the differential, with suicide and land transport injuries contributing over half of the differential associated with avoidable injuries. A large proportion of the differential caused by injuries was a result of higher mortality rates in the 0–29 year-old age group.

Figure 2: Decomposition of the life expectancy differential by the leading avoidable causes and age—Māori females.

c

Figure 3: Decomposition of the life expectancy differential by the leading avoidable causes and age—Māori males.

c

Within Pacific females, diabetes accounted for 0.6 years of the life expectancy differential, followed by coronary disease (0.5 years), cerebrovascular disease (0.4 years), and uterine and breast cancers (0.3 years each) (Figure 4). The life expectancy differential associated with cancers of the trachea, bronchus and lung was 0.1 years in Pacific females and 0.9 years in Māori females. Within Pacific males, coronary disease accounted for one full year of the life expectancy differential, followed by diabetes (0.7 years) and trachea, bronchus and lung cancers along with cerebrovascular disease (each contributing 0.4 years) (Figure 5). In contrast to Māori males, avoidable injuries were not a significant contributor to the life expectancy differential for Pacific males. However, assessment by age group identified a small contribution as a result of higher mortality rates for suicide in the 0–29 age group; however, this was offset by lower suicide rates in the older Pacific age groups (data not shown). Within the Pacific males and females, the summed contribution of ‘other avoidable causes’ resulted in a net reduction in life expectancy differentials. The largest contributor to this was lower mortality rates from melanoma where mortality rates are higher in the non-Māori/non-Pacific populations. Lower mortality rates from melanoma also had a net reduction on Māori life expectancy differentials.

Figure 4: Decomposition of the life expectancy differential by the leading avoidable causes and age—Pacific females.

c

Figure 5: Decomposition of the life expectancy differential by the leading avoidable causes and age—Pacific males.  

c

Discussion

This analysis has identified that potentially avoidable causes of death make up over half of all Māori deaths and nearly half of all Pacific deaths. In addition, they make a prominent contribution to life expectancy differentials in the Māori and Pacific population groups when compared with the non- Māori/non-Pacific group. Higher mortality rates at a younger age from conditions both preventable and amenable are the greatest potentially avoidable contributor to the differential within both populations. For Māori, coronary disease and cancers of the trachea, bronchus and lung in both males and females and avoidable injuries in males were the leading cause specific contributors. A finding specific to Māori males was the contribution of avoidable injuries, such as suicide and land transport accidents. Suicide was the fourth leading avoidable cause of the differential in Māori males. The finding relating to injuries was not replicated within Pacific males. Within both Pacific males and females, coronary disease, diabetes and stroke were the greatest cause-specific contributors to the differentials. Within Pacific females, the smaller contribution from conditions considered only preventable is predominantly due to lower mortality rates from cancers of the trachea, bronchus and lung compared with Māori females. This difference could possibly be explained by the lower smoking rates in Pacific females compared with Māori females.28 A finding unique to Pacific females was the large contribution of uterine cancer which was the leading avoidable cancer contributing to the differential. Higher rates of uterine cancer among Pacific females have been noted elsewhere with identified causal risk factors being diabetes and obesity.29

Pathways to inequity

Various exposures and experiences contribute to poorer health and inequities in avoidable mortality across different population groups. Three main pathways have been suggested that contribute to ethnic inequalities in health outcomes.30 These pathways align well with the concept of avoidable mortality. The pathways also recognise the role of racism in determining these mechanisms leading to inequality. Firstly, differential access and exposure to the upstream determinants of health such as inequitable access to education, employment and income along with other factors such as health literacy. These factors contribute to disparities by fostering illness, delaying care seeking and discouraging good adherence to treatment. This differential access is patterned in other risk factor exposure, for example, Māori and Pacific populations have higher rates of smoking and are more likely to live in more socioeconomically disadvantaged areas compared to non-Māori/non-Pacific people.31 These are risk factors for many conditions classified as preventable such as many avoidable cancers and other chronic conditions.

Intergenerational transmission of the social and economic determinants of health also presents a significant challenge for Māori and Pacific populations, whereby children from low socioeconomic backgrounds have a higher likelihood of remaining in that position through inequitable educational outcomes, restricted employment opportunities and lower incomes.32,33 For Māori, variation in behaviours deemed to be risky among males34,35 is further likely a contributing factor in the differential of Māori life expectancy. This is particularly evident in the contribution of avoidable injuries to the differential of 1.0 year in Māori males. The contribution of suicide to sex-specific differentials in life expectancy in New Zealand has previously been reported.21 However, this is the first time that ethnic-specific differences in New Zealand males have been reported.

Differential access to healthcare and the quality of care received are the second and third pathways that contribute to health inequities. These pathways can be considered downstream factors contributing to life expectancy differentials from conditions considered amenable. The health system literature describes how Māori will often experience longer and slower pathways through the health system.36–38 Furthermore, there is evidence that Māori and Pacific populations have poorer cancer survival compared with other ethnic groups39 with studies showing that Māori patients receive poorer quality treatment for potentially avoidable conditions such as lung and colorectal cancer.40 In addition, Māori and Pacific populations are more likely to report unmet need when engaging with health services.41

Racism at the personal and systems level is increasingly being recognised as a determinant of health in New Zealand, with many of the aforementioned pathways to inequities being underpinned by racism and its multiple expressions. Evidence shows that experience of racism can impact negatively on health and that Māori are more likely to experience interpersonal racism than non-Māori.12,13,42

The role of health services

Hospitals and health services have a crucial role to play as facilitators in reducing health inequities and the burden associated with avoidable mortality43 as the health system in its own right can be considered a determinant of health. The contribution of amenable mortality to life expectancy differentials highlights the need for health services to be further reoriented to ensure timely and equitable access for all population groups. This reorientation also involves ensuring that the health workforce reflects, and is responsive to, the population it serves. Health services, particularly hospitals hold a significant share of power, resources and influence within the health and community structure.43 As such, healthcare services are well positioned to lead the way in work to overcome barriers to achieving health equity, ensuring an equity in all policies approach and leading the intersectoral action required to improve equity in health outcomes. Primary healthcare services are particularly well placed to take action on the determinants of health and be strong campaigners for system change. While many of the social and economic determinants of health are outside the direct influence of medical professionals, they can be powerful advocates for the social and structural change required to reduce health inequities. However, this may only be realised through improved supportive policy and creating greater, systematic links with communities. In addition, health services, intersectoral agencies and in particular the workforce, need to recognise, name and understand the systems, attitudes and actions leading to institutional racism and discrimination. The impact of racism on health equity needs to be acknowledged, openly discussed and be part of medical education, as well as imbedded in health and institutional policy.

The high proportion of avoidable chronic disease, cancer and injury indicates a further need to strengthen policies targeting the respective risk and protective factors of these conditions. These policies need to focus on improving the social and economic determinants of health (such as income, education and housing) so that the opportunities to engage in health-promoting behaviours are equitably distributed across the population. There is also scope for medical treatment to achieve considerable health gain through secondary preventive services (such as diabetes management, cardiovascular risk and cancer screening). However, this requires the cultural safety of health services being addressed. While there have been declines in all-cause mortality rates over time in the Māori and Pacific population,44 there are still substantial mortality gaps that cannot be ignored if we are to close the differentials in life expectancy. This is particularly so for lung cancer which is the largest driver of cancer inequalities.45 This is likely a reflection of the marked ethnic gradient in tobacco smoking in New Zealand28 and highlights the potential for further advances in tobacco control.

Strengths and limitations

This study has several strengths. It benefitted from access to complete mortality data for all registered deaths in New Zealand and robust population data that allowed a complete population-based analysis of avoidable mortality and its contribution to life expectancy differentials. Furthermore, the use of well-established decomposition methods that decompose life expectancy by cause and age-group, allows for a richer analysis of the impact of avoidable mortality on life expectancy differentials experienced across the Māori and Pacific populations. This in itself could be useful insight for decisions on the specific targeting of upstream or structural interventions along with framing social policies that are likely to have the greatest impact in terms of reducing health inequities. In addition, the study made a distinction between mortality that could be tackled by the health system (amenable mortality) and that needing policy interventions and intersectoral action (preventable mortality) and provides a new dimension to understanding life expectancy differentials.

The findings presented in this paper should be interpreted in light of limitations common to ecologic analyses. We used aggregated data that prevent inference to individuals. Numerator/denominator bias may also occur whenever counts are collected using different methods. Ethnicity details on mortality data are sourced from the official record of death held by the Births, Deaths and Marriages.22 The ethnicity data for the official record of death is collected by funeral directors in consultation with a member of the deceased family. Where ethnicity is not available on the official record of death, National Health Index information will be used to supplement for this. On the other hand, ethnicity data for modelled population estimates are based on self-identification as part of the national Census. The level of possible undercount specific to our study is difficult to quantify, however, previous studies have identified undercounts of Māori deaths and overcounts of Pacific deaths in mortality data compared to census data.46 Finally, while a particular condition can be considered to be avoidable, this does not mean that every death from this condition could be prevented. With analysis of avoidable mortality, the precise nature of each death, the extent of disease progression at diagnosis, the existence of other medical conditions or the underlying exposure to risk factors is not taken into account.

It is important to acknowledge the contribution of conditions classified in this analysis as ‘non-avoidable’, particularly deaths among those older than 74 years of age. Many of the conditions within the avoidable mortality definition also occur in the older age groups. This does not mean that these conditions cannot be effectively treated at older ages. The primary reason for the 74-year-old cut off within the avoidable mortality definition is that the prevalence of (multiple) co-morbidity rises rapidly beyond this age. As such, it becomes increasingly difficult to assign a single cause to deaths in the older age groups. The quality of diagnosis and coding also tends to be lower in the very old. Including deaths beyond age 75 would likely impact the validity of the avoidable mortality indicator if a higher upper age limit were applied.

Contribution and past research

To the best of our knowledge, this is the first study to estimate the magnitude of avoidable mortality on life expectancy differentials in Māori and Pacific populations in New Zealand. This study details the potentially avoidable causes of death contributing to the frequently cited ethnic differentials in life expectancy and provides a new dimension to understanding this inequity. Overall rates of potentially avoidable mortality have previously been assessed in New Zealand.16 Tobias and Jackson were able to show in 1996–97 that around 70% of all deaths in the 0–74 age group were considered potentially avoidable and rates in the Māori and Pacific populations were over twice as high as those in the European population. A previous assessment of potential gains in life expectancy from avoidable causes of death has also been made in the Māori and non-Māori population.20 Malcolm identified that in 1985–87, if all potentially avoidable deaths in that period did not occur, Māori males would have gained 2.6 years of life, Māori females 2.7 years of life, non-Māori males 2.3 years of life and non-Māori females 1.7 years of life. However, the analysis by Malcolm did not account for competing mortality, thus the true gains would have been smaller than those presented.

The impact of avoidable mortality on life expectancy has been analysed for other European countries with similar decomposition methods.47,48 However, the causes of death used to define avoidable mortality differ between studies and many have assessed the longitudinal changes in life expectancy as opposed to the absolute difference between population groups at a given point in time. Comparisons are difficult as the detailed lists of causes of avoidable death and ages are not the same between studies. This problem of comparability could be solved with the establishment of a common list that would allow comparisons between studies and across countries.

Conclusion

Avoidable mortality is a large proportion of all Māori and Pacific deaths in New Zealand and is a prominent contributor to the life expectancy differentials in these groups. This study provides detail of the potentially avoidable causes of death contributing to these inequitable life expectancy differentials. It further reinforces the need for policymakers to enhance prevention activities and improve and ensure equitable healthcare delivery. This will require establishing a clear equity focus in all polices across all health and social services. Furthermore, there are clear benefits of looking beyond individual factors and recognising the role of healthcare services in improving health equity, in particular the role of racism. It is important to note that not all approaches used to improve health outcomes for the majority of the population will result in a reduction in inequities, in fact some methods inadvertently increase inequities. To address these inequalities, policymakers should focus on prevention activities and improve healthcare delivery with a clear equity focus, and also support wider improvements in educational achievement and socioeconomic position for the Māori and Pacific populations. Implementing the systems changes that are required to achieve equity will likely improve health for all.

Summary

Abstract

Aim

To determine the contribution of avoidable causes of death to the life expectancy differentials in both Mori and Pacific compared with non-Mori/non-Pacific ethnic groups in New Zealand.

Method

Death registration data and population data for New Zealand between 2013-15 was used to calculate life expectancy. A recent definition of avoidable mortality was used to identify potentially avoidable deaths. Life expectancy decomposition was undertaken to identify the contribution of avoidable causes of death to the life expectancy differential in the Mori and Pacific populations.

Results

Nearly half of all deaths in Pacific (47.3%) and over half in Mori (53.0%) can be attributed to potentially avoidable causes of death, compared with less than one quarter (23.2%) in the non-Mori/non-Pacific population. Conditions both preventable and amenable contribute the greatest to the life expectancy differentials within both ethnic groups, when compared with non-Mori/non-Pacific. Cancers of the trachea, bronchus and lung are significant avoidable causes contributing to the life expectancy differentials in both male and female Mori, contributing 0.8 years and 0.9 years respectively. Avoidable injuries including suicide contribute 1.0 year to the differential in Mori males. Coronary disease, diabetes and cerebrovascular disease are the largest contributors to the differential in both Pacific males and females.

Conclusion

Avoidable causes of death are large contributors to the life expectancy differentials in Mori and Pacific populations. The findings provide further evidence of the need to address the determinants of health and ensure equitable access to health services to reduce the impact of avoidable mortality on inequalities in life expectancy. It also highlights the importance of looking beyond individual factors and recognising the role of healthcare services and the social determinants in improving health equity.

Author Information

Michael Walsh, Epidemiologist, Planning Funding and Outcomes, Waitemata District Health Board, Auckland; Corina Grey, Public Health Physician, Planning Funding and Outcomes, Waitemata District Health Board, Auckland.

Acknowledgements

The authors would like to acknowledge Karen Bartholomew for her helpful contribution. The opinions expressed in this article are those of the authors and not of Waitemata District Health Board.

Correspondence

Michael Walsh, Planning Funding and Outcomes, Waitemata District Health Board, Level 1, 15 Shea Terrace, Takapuna, Auckland.

Correspondence Email

michael.walsh@waitematadhb.govt.nz

Competing Interests

Nil.

  1. Global Health Observatory (GHO) data - Life expectancy. 2018. (Accessed 25/6/2018, at http://www.who.int/gho/mortality_burden_disease/life_tables/situation_trends/en/)
  2. Marriott L, Sim D. Indicators of Inequality for Māori and Pacific People2014.
  3. New Zealand Period Life Tables: 2012–14. 2015. (Accessed 24/08/2018, at http://archive.stats.govt.nz/~/media/Statistics/Browse%20for%20stats/NZLifeTables/HOTP12-14/NZLifeTables12-14HOTP.pdf)
  4. Woodward A, Blakely T. The Healthy Country? A History of Life and Death in New Zealand. Auckland: University of Auckland Press; 2014.
  5. Tobias M, Blakely T, Matheson D, Rasanathan K, Atkinson J. Changing trends in indigenous inequalities in mortality: lessons from New Zealand. Int J Epidemiol 2009; 38:1711–22.
  6. Blakely T, Tobias M, Atkinson J. Inequalities in mortality during and after restructuring of the New Zealand economy: repeated cohort studies. BMJ 2008; 336:371–5.
  7. Howden-Chapman P, Blakely T, Blaiklock AJ, Kiro C. Closing the health gap. N Z Med J 2000; 113:301–2.
  8. Howden-Chapman P, M T. Social Inequalities in Health: New Zealand 1999. Wellington: Ministry of Health and Wellington School of Medicine; 2000.
  9. Paradies Y. Colonisation, racism and indigenous health. Journal of Population Research 2016; 33:83–96.
  10. Krieger N. A glossary for social epidemiology. J Epidemiol Community Health 2001; 55:693–700.
  11. Harris R, Tobias M, Jeffreys M, Waldegrave K, Karlsen S, Nazroo J. Racism and health: the relationship between experience of racial discrimination and health in New Zealand. Soc Sci Med 2006; 63:1428–41.
  12. Tobias M, Harris R. Racism and health in New Zealand. Ethn Health 2010; 15:318–20.
  13. Harris RB, Stanley J, Cormack DM. Racism and health in New Zealand: Prevalence over time and associations between recent experience of racism and health and wellbeing measures using national survey data. PLoS One 2018; 13:e0196476.
  14. Tobias M, Yeh LC. How much does health care contribute to health gain and to health inequality? Trends in amenable mortality in New Zealand 1981–2004. Aust N Z J Public Health 2009; 33:70–8.
  15. Rutstein DD, Berenberg W, Chalmers TC, Child CG, 3rd, Fishman AP, Perrin EB. Measuring the quality of medical care. A clinical method. N Engl J Med 1976; 294:582–8.
  16. Tobias M, Jackson G. Avoidable mortality in New Zealand, 1981–97. Aust N Z J Public Health 2001; 25:12–20.
  17. Ministry of Health. New Zealand Health Strategy: Future direction. 2016.
  18. Achieving equity. 2018. (Accessed 24/08/2018, at http://www.health.govt.nz/about-ministry/what-we-do/work-programme-2018/achieving-equity)
  19. System Level Measures Framework. 2016. (Accessed 25/08/2018, at http://www.health.govt.nz/new-zealand-health-system/system-level-measures-framework)
  20. Malcolm M. Avoidable mortality and life expectancy in New Zealand. J Epidemiol Community Health 1994; 48:211.
  21. Sandiford P. Gender inequality in New Zealand life expectancy: decomposition by age and cause. N Z Med J 2009; 122:10–7.
  22. Mortality Collection - Data Dictionary. 2017. (Accessed 25/08/2018, at http://www.health.govt.nz/system/files/documents/publications/mortality_data_dictionary_v1.6_final.pdf)
  23. HISO 10001:2017 Ethnicity Data Protocols,. 2017. (Accessed 24/08/2018, at http://www.health.govt.nz/system/files/documents/publications/hiso-10001-2017-ethnicity-data-protocols.pdf)
  24. Page A, Tobias M, Glover J, Wright C, Hetzel D, Fisher E. Australian and New Zealand atlas of avoidable mortality. Adelaide: University of Adelaide; 2006.
  25. Olatunde O, Windsor-Shellard B, Campbell A. Revised Definition of Avoidable Mortality and New Definition for Children and Young People2016.
  26. Defining Amenable Mortality. 2016. (Accessed 24/08/2018, at http://nsfl.health.govt.nz/system/files/.../defining_amenable_mortality_sept2016.docx)
  27. Arriaga EE. Measuring and explaining the change in life expectancies. Demography 1984; 21:83–96.
  28. Regional Data Explorer 2014–17: New Zealand Health Survey [Data File]. 2018. (Accessed 24/10/2018, at http://minhealthnz.shinyapps.io/nz-health-survey-2014-17-regional-update)
  29. Meredith I, Sarfati D, Ikeda T, Atkinson J, Blakely T. High rates of endometrial cancer among Pacific women in New Zealand: the role of diabetes, physical inactivity, and obesity. Cancer Causes Control 2012; 23:875–85.
  30. Jones CP. Invited commentary: “race,” racism, and the practice of epidemiology. Am J Epidemiol 2001; 154:299–304; discussion 5–6.
  31. Ministry of Health. Health Loss in New Zealand 1990–2013: A report from the New Zealand Burden of Diseases, Injuries and Risk Factors Study. Wellington: Ministry of Health; 2016.
  32. Kawharu M. Aotearoa: shine or shame? A critical examination of the Sustainable Development Goals and the question of poverty and young Māori in New Zealand. Journal of Global Ethics 2015; 11:43–50.
  33. Thayer ZM, Kuzawa CW. Ethnic discrimination predicts poor self-rated health and cortisol in pregnancy: insights from New Zealand. Soc Sci Med 2015; 128:36–42.
  34. Turner C, McClure R. Age and gender differences in risk-taking behaviour as an explanation for high incidence of motor vehicle crashes as a driver in young males. Inj Control Saf Promot 2003; 10:123–30.
  35. Morrongiello BA, Rennie H. Why do boys engage in more risk taking than girls? The role of attributions, beliefs, and risk appraisals. J Pediatr Psychol 1998; 23:33–43.
  36. Makowharemahihi C, Lawton BA, Cram F, Ngata T, Brown S, Robson B. Initiation of maternity care for young Māori women under 20 years of age. N Z Med J 2014; 127:52–61.
  37. Robson B. Hauora: Māori Standards of Health IV. A study of the years 2000–2005. Wellington: Te Ropu Rangahau Hauora a Eru Pomare; 2007.
  38. Bramley D, Riddell T, Crengle S, et al. A call to action on Māori cardiovascular health. N Z Med J 2004; 117:U957.
  39. Jeffreys M, Stevanovic V, Tobias M, et al. Ethnic inequalities in cancer survival in New Zealand: linkage study. Am J Public Health 2005; 95:834–7.
  40. Hill S, Sarfati D, Robson B, Blakely T. Indigenous inequalities in cancer: what role for health care? ANZ J Surg 2013; 83:36–41.
  41. Matheson A, Ellison-Loschmann L. Addressing the complex challenge of unmet need: a moral and equity imperative? N Z Med J 2017; 130:6–8.
  42. Harris RB, Cormack DM, Stanley J. The relationship between socially-assigned ethnicity, health and experience of racial discrimination for Māori: analysis of the 2006/07 New Zealand Health Survey. BMC Public Health 2013; 13:844.
  43. Matheson A, Bourke C, Verhoeven A, et al. Lowering hospital walls to achieve health equity. BMJ 2018; 362:k3597.
  44. Disney G, Teng A, Atkinson J, Wilson N, Blakely T. Changing ethnic inequalities in mortality in New Zealand over 30 years: linked cohort studies with 68.9 million person-years of follow-up. Popul Health Metr 2017; 15:15.
  45. Teng AM, Atkinson J, Disney G, Wilson N, Blakely T. Changing socioeconomic inequalities in cancer incidence and mortality: Cohort study with 54 million person-years follow-up 1981–2011. Int J Cancer 2017; 140:1306–16.
  46. Boyd M, Blakely T, Atkinson J. Ethnic counts on mortality, New Zealand Cancer Registry and census data: 2006–2011. N Z Med J 2016; 129:22–39.
  47. Nolte E, Scholz R, Shkolnikov V, McKee M. The contribution of medical care to changing life expectancy in Germany and Poland. Soc Sci Med 2002; 55:1905–21.
  48. Gispert R, Serra I, Bares MA, Puig X, Puigdefabregas A, Freitas A. The impact of avoidable mortality on life expectancy at birth in Spain: changes between three periods, from 1987 to 2001. J Epidemiol Community Health 2008; 62:783–9.

For the PDF of this article,
contact nzmj@nzma.org.nz

View Article PDF

Life expectancy is one of the most widely used measures in demographic and health analysis, and in New Zealand is among the highest of any country.1 Equity in health outcomes has long been a goal in New Zealand and is measured mainly in terms of the reduction or elimination of health inequalities between Māori and Pacific. Life expectancy differentials are a frequently cited inequity2 and are persistent within the Māori and Pacific populations when compared to non-Māori and non-Pacific populations.3 In 2012–14 life expectancy differentials were 6.8 years in Māori females and 7.3 years in Māori males. Within the Pacific population, the differential was 5.2 years in females and 5.8 years in males.3 While the life expectancy differentials in Māori have closed since c. 1900,4 a widening occurred in the 1980s and 1990s.5 This widening has partly been attributed to the economic structural reforms that occurred during this period. These reforms had a larger impact on Māori than other ethnic groups.6

The underlying factors contributing to ethnic health inequities, particularly life expectancy, are multifaceted and complex. This is particularly so when the experiences and pathways leading to inequity are likely unique for both the Māori and Pacific populations. Many of the drivers are related to the unequal distribution of the determinants of health. For example, the unequal distribution and access to resources such as income, education and employment, factors which often compound one another.7 These factors also pattern exposures to other risk factors such as tobacco use, poor nutrition, overcrowded and poor-quality housing, and drug and alcohol use.8 This inequitable distribution, particularly for Māori, is rooted in structural hierarchies that are associated with colonisation.9 This is not unique to New Zealand; in countries with a colonial past, indigenous populations frequently will have poorer health, even once socioeconomic position is factored in.7 Furthermore, access to the determinants of health and the associated ethnic inequities in health outcomes can arise from processes related to racism, as one’s social environment and where people are placed in the social hierarchy can shape health.10 The contribution of racism, particularly institutional racism, to health inequities is increasingly being recognised in New Zealand. It not only drives ethnic inequities in poverty, but also compounds the already significant negative effects of poverty.11–13

Quantifying, understanding and monitoring health inequities are important if they are to be addressed. Avoidable mortality is one measure that can be used as an indicator of the contribution of healthcare along with health and social policy to population health and indeed health equity. It is based on the concept that premature deaths from certain conditions should be rare and ideally not occur in the presence of timely, effective and equitable healthcare or other appropriate interventions.14 Avoidable mortality is recognised as a useful component of a suite of indicators to identify areas of potential focus for equity in the health system.15

The concept of avoidable mortality includes two overarching categories: amenable mortality and preventable mortality.16 Amenable mortality includes causes of death that could have potentially been avoided by means of access to high-quality and timely medical interventions. Preventable mortality is broader and includes deaths which could have been avoided through addressing the wider upstream determinants of health, such as individual-level health risk factors, socioeconomic status and environmental factors. The concept of avoidable mortality classifies the causes of death into one of the aforementioned categories, that is conditions that are ‘amenable only’, ‘preventable only’ or both ‘preventable and amenable’. The list of conditions that define avoidable mortality is based on expert review on causes that potentially could have been avoided given knowledge of casual pathways, the determinants of health and therapeutic technologies at the time of death.

Improving and ensuring equitable health outcomes for New Zealanders has long been a goal of the health system. This goal has been reinforced by the revised New Zealand Health Strategy which places equity in health outcomes at the forefront of much of the work of the health system.17 A change in government has also placed a renewed emphasis on equity in health outcomes and an ‘Achieving Equity in Health Outcomes’ work programme has recently been established by the Ministry of Health.18 A key component of this work programme is to build understanding of health equity through data, analytics and insights. In addition, in 2016 the New Zealand Ministry of Health released a framework for the health sector that aimed to improve the health outcomes of the New Zealand population. This framework, known as the System Level Measures framework, focuses on primary care within district health boards working in collaboration with other health system partners using specific quality improvement measures and has a strong focus on equity.19 Reducing the rates of amenable mortality is one of six measures within this framework.

Rates of avoidable mortality have previously been assessed in the New Zealand context16 and an assessment of potential gains in life expectancy from avoidable causes of death has been made.20 Published papers on the decomposition of life expectancy differentials in New Zealand are rare. However, life expectancy decomposition by cause between males and females in New Zealand has previously been undertaken.21 However, this study did not decompose the differential by avoidable causes of death nor decompose the ethnic specific differentials. To our knowledge, no publications have assessed the ethnic specific life expectancy differentials in New Zealand associated with potentially avoidable causes of death.

Analysis of high-quality health data, particularly mortality data allows us to deepen our understanding and provide insight into health inequities in New Zealand, and is a component of the Ministry of Health equity work programme. To enhance the understanding of the underlying causes of death contributing to the ethnic specific inequalities in life expectancy at birth, we evaluated the contribution of avoidable mortality, using a recently updated definition, to all deaths registered in the period 2013 to 2015. We then decomposed the differentials in life expectancy by avoidable causes of death within each of the Māori and Pacific populations when compared to the non-Māori/non-Pacific population. Specifically, the study aimed to identify the proportion of all Māori and Pacific deaths that are potentially avoidable and to decompose the life expectancy differentials by causes of death that are either preventable, amenable or both.

Method

Non-identifiable data relating to all registered deaths in New Zealand were obtained from the New Zealand Mortality Collection held by the New Zealand Ministry of Health for the period 2013 to 2015.22 Mortality data with an International Statistical Classification of Diseases Tenth Edition (Australian Modification) (ICD-10-AM) coded underlying cause of death is often two years delayed to allow coronial processes to be completed; as such 2015 was the most recent year available. New Zealand mortality data allows for multiple coding of ethnicity; this was prioritised to Māori, Pacific and non–Māori/non–Pacific as per standard protocols for health research.23 Prioritised ethnic-specific population estimates for New Zealand corresponding to the same period as the mortality data were obtained from Statistics New Zealand.

Previously published condition lists of avoidable mortality specific to the New Zealand context were identified from the literature, however given their age and advances in medical treatment and technology, these were considered out of date for use in this analysis as many are over a decade old.24 A review of the literature revealed a recently updated list published by the Office of National Statistics in the UK.25 This list was revised in 2016 and was based on previously published definitions developed for use in Australia and New Zealand. Despite the list being revised for use in the UK, it was deemed appropriate for application in the New Zealand context given that casual pathways and risk factors for the preventable conditions were unlikely to be different in New Zealand. However, minor adjustments were made to the list of conditions considered amenable based on recent amendments to the definition of ‘amenable mortality’ published by the New Zealand Ministry of Health26 and removal of some causes which in New Zealand were considered rare, for example protozoan infections. The underlying cause of death for each registered death occurring in those aged 0 to 74 years were categorised as either amenable, preventable or both (Table 1). A small number of exceptions are made to the age cut-off including deaths from injuries and self-harm, which included deaths at all ages, and acute lymphoblastic leukaemia, which only included deaths in the 0 to 44 age group. Those deaths categorised as still births were excluded.

Table 1: Groups of causes of avoidable mortality.

1Groups used in cause-specific decomposition.
2Includes all deaths in those aged 0–74 unless stated otherwise.
3Amenable deaths have been aligned with definition published by the Ministry of Health in July 2016.

Life expectancy at birth using data for 2013–15 (three year aggregated) was calculated using standard life table techniques, with 90 and older being the final age group. The impact of avoidable mortality on differentials in life expectancy was estimated by decomposing the difference in life expectancy using the method developed by Arriaga.27 This method allows differentials in life expectancy to be decomposed into age group and cause-of-death-specific contributions. Each sums to the total differential in life expectancy between groups.

Life tables for Māori males, Māori females, Pacific males, Pacific females, non-Māori/non-Pacific males and non-Māori/non-Pacific females were developed. The sex-specific non-Māori/non-Pacific tables served as the comparator group to compare the Māori and Pacific sex-specific tables. The contribution of avoidable causes of mortality and of the effects by age on life expectancy differentials in both Māori and Pacific compared with non–Māori/non–Pacific was analysed.

This study was outside the scope of ethics review due to the use of non-identifiable and encrypted data.

Results

Avoidable deaths

Between 2013 and 2015 there were 92,196 deaths registered in New Zealand. Of these, 25,210 (27.3% of all deaths and 68.2% of deaths in the 0 to 74 age group) were potentially avoidable (Figure 1). Of those deaths identified as potentially avoidable, 6,624 (7.2% all deaths) were preventable only, 4,056 (4.4% all deaths) amenable only and 14,530 (15.8% all deaths) both preventable and amenable. Within Māori, there were 9,717 deaths of which 5,152 (53.0% all deaths and 74.5% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in Māori 1,451 (14.9% all deaths) were preventable only, 937 (9.6% all deaths) amenable only and 2,764 (28.4% all deaths) both preventable and amenable. Within Pacific, there were 3,720 deaths of which 1,761 (47.3% all deaths and 71.3% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in Pacific, 355 (9.5% all deaths) were only preventable, 441 (11.6% all deaths) only amenable and 965 (25.9% all deaths) both preventable and amenable. Within non-Māori/non-Pacific, there were 78,759 deaths of which 18,297 (23.2% all deaths and 68.0% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in non-Māori/non-Pacific, 4,818 (6.1% all deaths) were only preventable, 2,678 (3.4% all deaths) only amenable and 10,801 (13.7% all deaths) both preventable and amenable. The proportion of all potentially avoidable deaths was twice as high in both the Māori and Pacific populations compared with the non- Māori /non-Pacific population.

Figure 1: Percentage of deaths that were potentially avoidable by ethnicity and total, 2013–2015.

c


Life expectancy at birth

For all deaths registered in 2013–15, estimates of life expectancy were: 73.5 years in Māori males, 77.3 years in Māori females, 75.0 years in Pacific males and 78.3 years in Pacific females. Life expectancy in the comparator groups was 80.9 years in non-Māori/non-Pacific males and 84.3 years in non-Māori/non-Pacific females. Overall these results equated to differentials in life expectancy of 7.4 years in Māori males, 7.0 years in Māori females, 5.9 years in Pacific males, and 6.0 years in Pacific females.

Decomposition by avoidable category and age

Within both Māori and Pacific females, higher mortality rates at a younger age from conditions considered both preventable and amenable made the greatest contribution to the life expectancy differentials, contributing 2.1 years and 1.7 years respectively (Table 2). Within Māori females, this was followed by conditions that are only preventable (1.3 years). However, within Pacific females, preventable conditions contributed only 0.3 years. Conditions only amenable contributed 1.0 years in Māori females and 1.3 years in Pacific females. Within age groups, mortality in the 50–74 year-old group contributed the most to the life expectancy differentials, contributing 4.1 years for Māori females and 3.3 years for Pacific females.

Table 2: Decomposition of the life expectancy differential in females by avoidable category and age.

1Sum of columns along with non-avoidable and avoidable may not equal the exact life expectancy differential due to rounding.

Similar to females, conditions both preventable and amenable were the greatest contributors to the life expectancy differentials in Māori and Pacific males, contributing 2.7 years and 1.7 years respectively (Table 3). Conditions considered preventable only were the second greatest contributor in Māori males and third greatest contributor in Pacific males; however, the number of years contributed within this group was higher for Māori males (1.6 years) compared with Pacific males (0.8 years). Within Pacific males, conditions considered amenable only were the second greatest contributor to the differential. Within age groups, higher mortality rates in the 50–74 year-old group contributed the most to life expectancy differentials, contributing 4.1 years for Māori males and 3.1 years for Pacific males. Even within the youngest age groups (<50 years), there were large contributions to life expectancy differentials for Māori (2.2 years) Pacific (1.5 years) males.

Table 3: Decomposition of the life expectancy differential in males by avoidable category and age.

1Sum of columns along with non-avoidable and avoidable may not equal the exact life expectancy differential due to rounding.

Decomposition by avoidable cause

Within Māori females, trachea, bronchus and lung cancers accounted for 0.9 years of the life expectancy differential, followed by coronary disease (0.6 years), with diabetes and chronic obstructive pulmonary disease each contributing 0.4 years (Figure 2). Within Māori males, coronary disease, trachea, bronchus and lung cancers and diabetes accounted for 1.2 years, 0.8 years and 0.5 years respectively (Figure 3). In Māori males, avoidable injuries had a pronounced contribution to the differential. Injuries contributed 1.0 year to the differential, with suicide and land transport injuries contributing over half of the differential associated with avoidable injuries. A large proportion of the differential caused by injuries was a result of higher mortality rates in the 0–29 year-old age group.

Figure 2: Decomposition of the life expectancy differential by the leading avoidable causes and age—Māori females.

c

Figure 3: Decomposition of the life expectancy differential by the leading avoidable causes and age—Māori males.

c

Within Pacific females, diabetes accounted for 0.6 years of the life expectancy differential, followed by coronary disease (0.5 years), cerebrovascular disease (0.4 years), and uterine and breast cancers (0.3 years each) (Figure 4). The life expectancy differential associated with cancers of the trachea, bronchus and lung was 0.1 years in Pacific females and 0.9 years in Māori females. Within Pacific males, coronary disease accounted for one full year of the life expectancy differential, followed by diabetes (0.7 years) and trachea, bronchus and lung cancers along with cerebrovascular disease (each contributing 0.4 years) (Figure 5). In contrast to Māori males, avoidable injuries were not a significant contributor to the life expectancy differential for Pacific males. However, assessment by age group identified a small contribution as a result of higher mortality rates for suicide in the 0–29 age group; however, this was offset by lower suicide rates in the older Pacific age groups (data not shown). Within the Pacific males and females, the summed contribution of ‘other avoidable causes’ resulted in a net reduction in life expectancy differentials. The largest contributor to this was lower mortality rates from melanoma where mortality rates are higher in the non-Māori/non-Pacific populations. Lower mortality rates from melanoma also had a net reduction on Māori life expectancy differentials.

Figure 4: Decomposition of the life expectancy differential by the leading avoidable causes and age—Pacific females.

c

Figure 5: Decomposition of the life expectancy differential by the leading avoidable causes and age—Pacific males.  

c

Discussion

This analysis has identified that potentially avoidable causes of death make up over half of all Māori deaths and nearly half of all Pacific deaths. In addition, they make a prominent contribution to life expectancy differentials in the Māori and Pacific population groups when compared with the non- Māori/non-Pacific group. Higher mortality rates at a younger age from conditions both preventable and amenable are the greatest potentially avoidable contributor to the differential within both populations. For Māori, coronary disease and cancers of the trachea, bronchus and lung in both males and females and avoidable injuries in males were the leading cause specific contributors. A finding specific to Māori males was the contribution of avoidable injuries, such as suicide and land transport accidents. Suicide was the fourth leading avoidable cause of the differential in Māori males. The finding relating to injuries was not replicated within Pacific males. Within both Pacific males and females, coronary disease, diabetes and stroke were the greatest cause-specific contributors to the differentials. Within Pacific females, the smaller contribution from conditions considered only preventable is predominantly due to lower mortality rates from cancers of the trachea, bronchus and lung compared with Māori females. This difference could possibly be explained by the lower smoking rates in Pacific females compared with Māori females.28 A finding unique to Pacific females was the large contribution of uterine cancer which was the leading avoidable cancer contributing to the differential. Higher rates of uterine cancer among Pacific females have been noted elsewhere with identified causal risk factors being diabetes and obesity.29

Pathways to inequity

Various exposures and experiences contribute to poorer health and inequities in avoidable mortality across different population groups. Three main pathways have been suggested that contribute to ethnic inequalities in health outcomes.30 These pathways align well with the concept of avoidable mortality. The pathways also recognise the role of racism in determining these mechanisms leading to inequality. Firstly, differential access and exposure to the upstream determinants of health such as inequitable access to education, employment and income along with other factors such as health literacy. These factors contribute to disparities by fostering illness, delaying care seeking and discouraging good adherence to treatment. This differential access is patterned in other risk factor exposure, for example, Māori and Pacific populations have higher rates of smoking and are more likely to live in more socioeconomically disadvantaged areas compared to non-Māori/non-Pacific people.31 These are risk factors for many conditions classified as preventable such as many avoidable cancers and other chronic conditions.

Intergenerational transmission of the social and economic determinants of health also presents a significant challenge for Māori and Pacific populations, whereby children from low socioeconomic backgrounds have a higher likelihood of remaining in that position through inequitable educational outcomes, restricted employment opportunities and lower incomes.32,33 For Māori, variation in behaviours deemed to be risky among males34,35 is further likely a contributing factor in the differential of Māori life expectancy. This is particularly evident in the contribution of avoidable injuries to the differential of 1.0 year in Māori males. The contribution of suicide to sex-specific differentials in life expectancy in New Zealand has previously been reported.21 However, this is the first time that ethnic-specific differences in New Zealand males have been reported.

Differential access to healthcare and the quality of care received are the second and third pathways that contribute to health inequities. These pathways can be considered downstream factors contributing to life expectancy differentials from conditions considered amenable. The health system literature describes how Māori will often experience longer and slower pathways through the health system.36–38 Furthermore, there is evidence that Māori and Pacific populations have poorer cancer survival compared with other ethnic groups39 with studies showing that Māori patients receive poorer quality treatment for potentially avoidable conditions such as lung and colorectal cancer.40 In addition, Māori and Pacific populations are more likely to report unmet need when engaging with health services.41

Racism at the personal and systems level is increasingly being recognised as a determinant of health in New Zealand, with many of the aforementioned pathways to inequities being underpinned by racism and its multiple expressions. Evidence shows that experience of racism can impact negatively on health and that Māori are more likely to experience interpersonal racism than non-Māori.12,13,42

The role of health services

Hospitals and health services have a crucial role to play as facilitators in reducing health inequities and the burden associated with avoidable mortality43 as the health system in its own right can be considered a determinant of health. The contribution of amenable mortality to life expectancy differentials highlights the need for health services to be further reoriented to ensure timely and equitable access for all population groups. This reorientation also involves ensuring that the health workforce reflects, and is responsive to, the population it serves. Health services, particularly hospitals hold a significant share of power, resources and influence within the health and community structure.43 As such, healthcare services are well positioned to lead the way in work to overcome barriers to achieving health equity, ensuring an equity in all policies approach and leading the intersectoral action required to improve equity in health outcomes. Primary healthcare services are particularly well placed to take action on the determinants of health and be strong campaigners for system change. While many of the social and economic determinants of health are outside the direct influence of medical professionals, they can be powerful advocates for the social and structural change required to reduce health inequities. However, this may only be realised through improved supportive policy and creating greater, systematic links with communities. In addition, health services, intersectoral agencies and in particular the workforce, need to recognise, name and understand the systems, attitudes and actions leading to institutional racism and discrimination. The impact of racism on health equity needs to be acknowledged, openly discussed and be part of medical education, as well as imbedded in health and institutional policy.

The high proportion of avoidable chronic disease, cancer and injury indicates a further need to strengthen policies targeting the respective risk and protective factors of these conditions. These policies need to focus on improving the social and economic determinants of health (such as income, education and housing) so that the opportunities to engage in health-promoting behaviours are equitably distributed across the population. There is also scope for medical treatment to achieve considerable health gain through secondary preventive services (such as diabetes management, cardiovascular risk and cancer screening). However, this requires the cultural safety of health services being addressed. While there have been declines in all-cause mortality rates over time in the Māori and Pacific population,44 there are still substantial mortality gaps that cannot be ignored if we are to close the differentials in life expectancy. This is particularly so for lung cancer which is the largest driver of cancer inequalities.45 This is likely a reflection of the marked ethnic gradient in tobacco smoking in New Zealand28 and highlights the potential for further advances in tobacco control.

Strengths and limitations

This study has several strengths. It benefitted from access to complete mortality data for all registered deaths in New Zealand and robust population data that allowed a complete population-based analysis of avoidable mortality and its contribution to life expectancy differentials. Furthermore, the use of well-established decomposition methods that decompose life expectancy by cause and age-group, allows for a richer analysis of the impact of avoidable mortality on life expectancy differentials experienced across the Māori and Pacific populations. This in itself could be useful insight for decisions on the specific targeting of upstream or structural interventions along with framing social policies that are likely to have the greatest impact in terms of reducing health inequities. In addition, the study made a distinction between mortality that could be tackled by the health system (amenable mortality) and that needing policy interventions and intersectoral action (preventable mortality) and provides a new dimension to understanding life expectancy differentials.

The findings presented in this paper should be interpreted in light of limitations common to ecologic analyses. We used aggregated data that prevent inference to individuals. Numerator/denominator bias may also occur whenever counts are collected using different methods. Ethnicity details on mortality data are sourced from the official record of death held by the Births, Deaths and Marriages.22 The ethnicity data for the official record of death is collected by funeral directors in consultation with a member of the deceased family. Where ethnicity is not available on the official record of death, National Health Index information will be used to supplement for this. On the other hand, ethnicity data for modelled population estimates are based on self-identification as part of the national Census. The level of possible undercount specific to our study is difficult to quantify, however, previous studies have identified undercounts of Māori deaths and overcounts of Pacific deaths in mortality data compared to census data.46 Finally, while a particular condition can be considered to be avoidable, this does not mean that every death from this condition could be prevented. With analysis of avoidable mortality, the precise nature of each death, the extent of disease progression at diagnosis, the existence of other medical conditions or the underlying exposure to risk factors is not taken into account.

It is important to acknowledge the contribution of conditions classified in this analysis as ‘non-avoidable’, particularly deaths among those older than 74 years of age. Many of the conditions within the avoidable mortality definition also occur in the older age groups. This does not mean that these conditions cannot be effectively treated at older ages. The primary reason for the 74-year-old cut off within the avoidable mortality definition is that the prevalence of (multiple) co-morbidity rises rapidly beyond this age. As such, it becomes increasingly difficult to assign a single cause to deaths in the older age groups. The quality of diagnosis and coding also tends to be lower in the very old. Including deaths beyond age 75 would likely impact the validity of the avoidable mortality indicator if a higher upper age limit were applied.

Contribution and past research

To the best of our knowledge, this is the first study to estimate the magnitude of avoidable mortality on life expectancy differentials in Māori and Pacific populations in New Zealand. This study details the potentially avoidable causes of death contributing to the frequently cited ethnic differentials in life expectancy and provides a new dimension to understanding this inequity. Overall rates of potentially avoidable mortality have previously been assessed in New Zealand.16 Tobias and Jackson were able to show in 1996–97 that around 70% of all deaths in the 0–74 age group were considered potentially avoidable and rates in the Māori and Pacific populations were over twice as high as those in the European population. A previous assessment of potential gains in life expectancy from avoidable causes of death has also been made in the Māori and non-Māori population.20 Malcolm identified that in 1985–87, if all potentially avoidable deaths in that period did not occur, Māori males would have gained 2.6 years of life, Māori females 2.7 years of life, non-Māori males 2.3 years of life and non-Māori females 1.7 years of life. However, the analysis by Malcolm did not account for competing mortality, thus the true gains would have been smaller than those presented.

The impact of avoidable mortality on life expectancy has been analysed for other European countries with similar decomposition methods.47,48 However, the causes of death used to define avoidable mortality differ between studies and many have assessed the longitudinal changes in life expectancy as opposed to the absolute difference between population groups at a given point in time. Comparisons are difficult as the detailed lists of causes of avoidable death and ages are not the same between studies. This problem of comparability could be solved with the establishment of a common list that would allow comparisons between studies and across countries.

Conclusion

Avoidable mortality is a large proportion of all Māori and Pacific deaths in New Zealand and is a prominent contributor to the life expectancy differentials in these groups. This study provides detail of the potentially avoidable causes of death contributing to these inequitable life expectancy differentials. It further reinforces the need for policymakers to enhance prevention activities and improve and ensure equitable healthcare delivery. This will require establishing a clear equity focus in all polices across all health and social services. Furthermore, there are clear benefits of looking beyond individual factors and recognising the role of healthcare services in improving health equity, in particular the role of racism. It is important to note that not all approaches used to improve health outcomes for the majority of the population will result in a reduction in inequities, in fact some methods inadvertently increase inequities. To address these inequalities, policymakers should focus on prevention activities and improve healthcare delivery with a clear equity focus, and also support wider improvements in educational achievement and socioeconomic position for the Māori and Pacific populations. Implementing the systems changes that are required to achieve equity will likely improve health for all.

Summary

Abstract

Aim

To determine the contribution of avoidable causes of death to the life expectancy differentials in both Mori and Pacific compared with non-Mori/non-Pacific ethnic groups in New Zealand.

Method

Death registration data and population data for New Zealand between 2013-15 was used to calculate life expectancy. A recent definition of avoidable mortality was used to identify potentially avoidable deaths. Life expectancy decomposition was undertaken to identify the contribution of avoidable causes of death to the life expectancy differential in the Mori and Pacific populations.

Results

Nearly half of all deaths in Pacific (47.3%) and over half in Mori (53.0%) can be attributed to potentially avoidable causes of death, compared with less than one quarter (23.2%) in the non-Mori/non-Pacific population. Conditions both preventable and amenable contribute the greatest to the life expectancy differentials within both ethnic groups, when compared with non-Mori/non-Pacific. Cancers of the trachea, bronchus and lung are significant avoidable causes contributing to the life expectancy differentials in both male and female Mori, contributing 0.8 years and 0.9 years respectively. Avoidable injuries including suicide contribute 1.0 year to the differential in Mori males. Coronary disease, diabetes and cerebrovascular disease are the largest contributors to the differential in both Pacific males and females.

Conclusion

Avoidable causes of death are large contributors to the life expectancy differentials in Mori and Pacific populations. The findings provide further evidence of the need to address the determinants of health and ensure equitable access to health services to reduce the impact of avoidable mortality on inequalities in life expectancy. It also highlights the importance of looking beyond individual factors and recognising the role of healthcare services and the social determinants in improving health equity.

Author Information

Michael Walsh, Epidemiologist, Planning Funding and Outcomes, Waitemata District Health Board, Auckland; Corina Grey, Public Health Physician, Planning Funding and Outcomes, Waitemata District Health Board, Auckland.

Acknowledgements

The authors would like to acknowledge Karen Bartholomew for her helpful contribution. The opinions expressed in this article are those of the authors and not of Waitemata District Health Board.

Correspondence

Michael Walsh, Planning Funding and Outcomes, Waitemata District Health Board, Level 1, 15 Shea Terrace, Takapuna, Auckland.

Correspondence Email

michael.walsh@waitematadhb.govt.nz

Competing Interests

Nil.

  1. Global Health Observatory (GHO) data - Life expectancy. 2018. (Accessed 25/6/2018, at http://www.who.int/gho/mortality_burden_disease/life_tables/situation_trends/en/)
  2. Marriott L, Sim D. Indicators of Inequality for Māori and Pacific People2014.
  3. New Zealand Period Life Tables: 2012–14. 2015. (Accessed 24/08/2018, at http://archive.stats.govt.nz/~/media/Statistics/Browse%20for%20stats/NZLifeTables/HOTP12-14/NZLifeTables12-14HOTP.pdf)
  4. Woodward A, Blakely T. The Healthy Country? A History of Life and Death in New Zealand. Auckland: University of Auckland Press; 2014.
  5. Tobias M, Blakely T, Matheson D, Rasanathan K, Atkinson J. Changing trends in indigenous inequalities in mortality: lessons from New Zealand. Int J Epidemiol 2009; 38:1711–22.
  6. Blakely T, Tobias M, Atkinson J. Inequalities in mortality during and after restructuring of the New Zealand economy: repeated cohort studies. BMJ 2008; 336:371–5.
  7. Howden-Chapman P, Blakely T, Blaiklock AJ, Kiro C. Closing the health gap. N Z Med J 2000; 113:301–2.
  8. Howden-Chapman P, M T. Social Inequalities in Health: New Zealand 1999. Wellington: Ministry of Health and Wellington School of Medicine; 2000.
  9. Paradies Y. Colonisation, racism and indigenous health. Journal of Population Research 2016; 33:83–96.
  10. Krieger N. A glossary for social epidemiology. J Epidemiol Community Health 2001; 55:693–700.
  11. Harris R, Tobias M, Jeffreys M, Waldegrave K, Karlsen S, Nazroo J. Racism and health: the relationship between experience of racial discrimination and health in New Zealand. Soc Sci Med 2006; 63:1428–41.
  12. Tobias M, Harris R. Racism and health in New Zealand. Ethn Health 2010; 15:318–20.
  13. Harris RB, Stanley J, Cormack DM. Racism and health in New Zealand: Prevalence over time and associations between recent experience of racism and health and wellbeing measures using national survey data. PLoS One 2018; 13:e0196476.
  14. Tobias M, Yeh LC. How much does health care contribute to health gain and to health inequality? Trends in amenable mortality in New Zealand 1981–2004. Aust N Z J Public Health 2009; 33:70–8.
  15. Rutstein DD, Berenberg W, Chalmers TC, Child CG, 3rd, Fishman AP, Perrin EB. Measuring the quality of medical care. A clinical method. N Engl J Med 1976; 294:582–8.
  16. Tobias M, Jackson G. Avoidable mortality in New Zealand, 1981–97. Aust N Z J Public Health 2001; 25:12–20.
  17. Ministry of Health. New Zealand Health Strategy: Future direction. 2016.
  18. Achieving equity. 2018. (Accessed 24/08/2018, at http://www.health.govt.nz/about-ministry/what-we-do/work-programme-2018/achieving-equity)
  19. System Level Measures Framework. 2016. (Accessed 25/08/2018, at http://www.health.govt.nz/new-zealand-health-system/system-level-measures-framework)
  20. Malcolm M. Avoidable mortality and life expectancy in New Zealand. J Epidemiol Community Health 1994; 48:211.
  21. Sandiford P. Gender inequality in New Zealand life expectancy: decomposition by age and cause. N Z Med J 2009; 122:10–7.
  22. Mortality Collection - Data Dictionary. 2017. (Accessed 25/08/2018, at http://www.health.govt.nz/system/files/documents/publications/mortality_data_dictionary_v1.6_final.pdf)
  23. HISO 10001:2017 Ethnicity Data Protocols,. 2017. (Accessed 24/08/2018, at http://www.health.govt.nz/system/files/documents/publications/hiso-10001-2017-ethnicity-data-protocols.pdf)
  24. Page A, Tobias M, Glover J, Wright C, Hetzel D, Fisher E. Australian and New Zealand atlas of avoidable mortality. Adelaide: University of Adelaide; 2006.
  25. Olatunde O, Windsor-Shellard B, Campbell A. Revised Definition of Avoidable Mortality and New Definition for Children and Young People2016.
  26. Defining Amenable Mortality. 2016. (Accessed 24/08/2018, at http://nsfl.health.govt.nz/system/files/.../defining_amenable_mortality_sept2016.docx)
  27. Arriaga EE. Measuring and explaining the change in life expectancies. Demography 1984; 21:83–96.
  28. Regional Data Explorer 2014–17: New Zealand Health Survey [Data File]. 2018. (Accessed 24/10/2018, at http://minhealthnz.shinyapps.io/nz-health-survey-2014-17-regional-update)
  29. Meredith I, Sarfati D, Ikeda T, Atkinson J, Blakely T. High rates of endometrial cancer among Pacific women in New Zealand: the role of diabetes, physical inactivity, and obesity. Cancer Causes Control 2012; 23:875–85.
  30. Jones CP. Invited commentary: “race,” racism, and the practice of epidemiology. Am J Epidemiol 2001; 154:299–304; discussion 5–6.
  31. Ministry of Health. Health Loss in New Zealand 1990–2013: A report from the New Zealand Burden of Diseases, Injuries and Risk Factors Study. Wellington: Ministry of Health; 2016.
  32. Kawharu M. Aotearoa: shine or shame? A critical examination of the Sustainable Development Goals and the question of poverty and young Māori in New Zealand. Journal of Global Ethics 2015; 11:43–50.
  33. Thayer ZM, Kuzawa CW. Ethnic discrimination predicts poor self-rated health and cortisol in pregnancy: insights from New Zealand. Soc Sci Med 2015; 128:36–42.
  34. Turner C, McClure R. Age and gender differences in risk-taking behaviour as an explanation for high incidence of motor vehicle crashes as a driver in young males. Inj Control Saf Promot 2003; 10:123–30.
  35. Morrongiello BA, Rennie H. Why do boys engage in more risk taking than girls? The role of attributions, beliefs, and risk appraisals. J Pediatr Psychol 1998; 23:33–43.
  36. Makowharemahihi C, Lawton BA, Cram F, Ngata T, Brown S, Robson B. Initiation of maternity care for young Māori women under 20 years of age. N Z Med J 2014; 127:52–61.
  37. Robson B. Hauora: Māori Standards of Health IV. A study of the years 2000–2005. Wellington: Te Ropu Rangahau Hauora a Eru Pomare; 2007.
  38. Bramley D, Riddell T, Crengle S, et al. A call to action on Māori cardiovascular health. N Z Med J 2004; 117:U957.
  39. Jeffreys M, Stevanovic V, Tobias M, et al. Ethnic inequalities in cancer survival in New Zealand: linkage study. Am J Public Health 2005; 95:834–7.
  40. Hill S, Sarfati D, Robson B, Blakely T. Indigenous inequalities in cancer: what role for health care? ANZ J Surg 2013; 83:36–41.
  41. Matheson A, Ellison-Loschmann L. Addressing the complex challenge of unmet need: a moral and equity imperative? N Z Med J 2017; 130:6–8.
  42. Harris RB, Cormack DM, Stanley J. The relationship between socially-assigned ethnicity, health and experience of racial discrimination for Māori: analysis of the 2006/07 New Zealand Health Survey. BMC Public Health 2013; 13:844.
  43. Matheson A, Bourke C, Verhoeven A, et al. Lowering hospital walls to achieve health equity. BMJ 2018; 362:k3597.
  44. Disney G, Teng A, Atkinson J, Wilson N, Blakely T. Changing ethnic inequalities in mortality in New Zealand over 30 years: linked cohort studies with 68.9 million person-years of follow-up. Popul Health Metr 2017; 15:15.
  45. Teng AM, Atkinson J, Disney G, Wilson N, Blakely T. Changing socioeconomic inequalities in cancer incidence and mortality: Cohort study with 54 million person-years follow-up 1981–2011. Int J Cancer 2017; 140:1306–16.
  46. Boyd M, Blakely T, Atkinson J. Ethnic counts on mortality, New Zealand Cancer Registry and census data: 2006–2011. N Z Med J 2016; 129:22–39.
  47. Nolte E, Scholz R, Shkolnikov V, McKee M. The contribution of medical care to changing life expectancy in Germany and Poland. Soc Sci Med 2002; 55:1905–21.
  48. Gispert R, Serra I, Bares MA, Puig X, Puigdefabregas A, Freitas A. The impact of avoidable mortality on life expectancy at birth in Spain: changes between three periods, from 1987 to 2001. J Epidemiol Community Health 2008; 62:783–9.

For the PDF of this article,
contact nzmj@nzma.org.nz

View Article PDF

Life expectancy is one of the most widely used measures in demographic and health analysis, and in New Zealand is among the highest of any country.1 Equity in health outcomes has long been a goal in New Zealand and is measured mainly in terms of the reduction or elimination of health inequalities between Māori and Pacific. Life expectancy differentials are a frequently cited inequity2 and are persistent within the Māori and Pacific populations when compared to non-Māori and non-Pacific populations.3 In 2012–14 life expectancy differentials were 6.8 years in Māori females and 7.3 years in Māori males. Within the Pacific population, the differential was 5.2 years in females and 5.8 years in males.3 While the life expectancy differentials in Māori have closed since c. 1900,4 a widening occurred in the 1980s and 1990s.5 This widening has partly been attributed to the economic structural reforms that occurred during this period. These reforms had a larger impact on Māori than other ethnic groups.6

The underlying factors contributing to ethnic health inequities, particularly life expectancy, are multifaceted and complex. This is particularly so when the experiences and pathways leading to inequity are likely unique for both the Māori and Pacific populations. Many of the drivers are related to the unequal distribution of the determinants of health. For example, the unequal distribution and access to resources such as income, education and employment, factors which often compound one another.7 These factors also pattern exposures to other risk factors such as tobacco use, poor nutrition, overcrowded and poor-quality housing, and drug and alcohol use.8 This inequitable distribution, particularly for Māori, is rooted in structural hierarchies that are associated with colonisation.9 This is not unique to New Zealand; in countries with a colonial past, indigenous populations frequently will have poorer health, even once socioeconomic position is factored in.7 Furthermore, access to the determinants of health and the associated ethnic inequities in health outcomes can arise from processes related to racism, as one’s social environment and where people are placed in the social hierarchy can shape health.10 The contribution of racism, particularly institutional racism, to health inequities is increasingly being recognised in New Zealand. It not only drives ethnic inequities in poverty, but also compounds the already significant negative effects of poverty.11–13

Quantifying, understanding and monitoring health inequities are important if they are to be addressed. Avoidable mortality is one measure that can be used as an indicator of the contribution of healthcare along with health and social policy to population health and indeed health equity. It is based on the concept that premature deaths from certain conditions should be rare and ideally not occur in the presence of timely, effective and equitable healthcare or other appropriate interventions.14 Avoidable mortality is recognised as a useful component of a suite of indicators to identify areas of potential focus for equity in the health system.15

The concept of avoidable mortality includes two overarching categories: amenable mortality and preventable mortality.16 Amenable mortality includes causes of death that could have potentially been avoided by means of access to high-quality and timely medical interventions. Preventable mortality is broader and includes deaths which could have been avoided through addressing the wider upstream determinants of health, such as individual-level health risk factors, socioeconomic status and environmental factors. The concept of avoidable mortality classifies the causes of death into one of the aforementioned categories, that is conditions that are ‘amenable only’, ‘preventable only’ or both ‘preventable and amenable’. The list of conditions that define avoidable mortality is based on expert review on causes that potentially could have been avoided given knowledge of casual pathways, the determinants of health and therapeutic technologies at the time of death.

Improving and ensuring equitable health outcomes for New Zealanders has long been a goal of the health system. This goal has been reinforced by the revised New Zealand Health Strategy which places equity in health outcomes at the forefront of much of the work of the health system.17 A change in government has also placed a renewed emphasis on equity in health outcomes and an ‘Achieving Equity in Health Outcomes’ work programme has recently been established by the Ministry of Health.18 A key component of this work programme is to build understanding of health equity through data, analytics and insights. In addition, in 2016 the New Zealand Ministry of Health released a framework for the health sector that aimed to improve the health outcomes of the New Zealand population. This framework, known as the System Level Measures framework, focuses on primary care within district health boards working in collaboration with other health system partners using specific quality improvement measures and has a strong focus on equity.19 Reducing the rates of amenable mortality is one of six measures within this framework.

Rates of avoidable mortality have previously been assessed in the New Zealand context16 and an assessment of potential gains in life expectancy from avoidable causes of death has been made.20 Published papers on the decomposition of life expectancy differentials in New Zealand are rare. However, life expectancy decomposition by cause between males and females in New Zealand has previously been undertaken.21 However, this study did not decompose the differential by avoidable causes of death nor decompose the ethnic specific differentials. To our knowledge, no publications have assessed the ethnic specific life expectancy differentials in New Zealand associated with potentially avoidable causes of death.

Analysis of high-quality health data, particularly mortality data allows us to deepen our understanding and provide insight into health inequities in New Zealand, and is a component of the Ministry of Health equity work programme. To enhance the understanding of the underlying causes of death contributing to the ethnic specific inequalities in life expectancy at birth, we evaluated the contribution of avoidable mortality, using a recently updated definition, to all deaths registered in the period 2013 to 2015. We then decomposed the differentials in life expectancy by avoidable causes of death within each of the Māori and Pacific populations when compared to the non-Māori/non-Pacific population. Specifically, the study aimed to identify the proportion of all Māori and Pacific deaths that are potentially avoidable and to decompose the life expectancy differentials by causes of death that are either preventable, amenable or both.

Method

Non-identifiable data relating to all registered deaths in New Zealand were obtained from the New Zealand Mortality Collection held by the New Zealand Ministry of Health for the period 2013 to 2015.22 Mortality data with an International Statistical Classification of Diseases Tenth Edition (Australian Modification) (ICD-10-AM) coded underlying cause of death is often two years delayed to allow coronial processes to be completed; as such 2015 was the most recent year available. New Zealand mortality data allows for multiple coding of ethnicity; this was prioritised to Māori, Pacific and non–Māori/non–Pacific as per standard protocols for health research.23 Prioritised ethnic-specific population estimates for New Zealand corresponding to the same period as the mortality data were obtained from Statistics New Zealand.

Previously published condition lists of avoidable mortality specific to the New Zealand context were identified from the literature, however given their age and advances in medical treatment and technology, these were considered out of date for use in this analysis as many are over a decade old.24 A review of the literature revealed a recently updated list published by the Office of National Statistics in the UK.25 This list was revised in 2016 and was based on previously published definitions developed for use in Australia and New Zealand. Despite the list being revised for use in the UK, it was deemed appropriate for application in the New Zealand context given that casual pathways and risk factors for the preventable conditions were unlikely to be different in New Zealand. However, minor adjustments were made to the list of conditions considered amenable based on recent amendments to the definition of ‘amenable mortality’ published by the New Zealand Ministry of Health26 and removal of some causes which in New Zealand were considered rare, for example protozoan infections. The underlying cause of death for each registered death occurring in those aged 0 to 74 years were categorised as either amenable, preventable or both (Table 1). A small number of exceptions are made to the age cut-off including deaths from injuries and self-harm, which included deaths at all ages, and acute lymphoblastic leukaemia, which only included deaths in the 0 to 44 age group. Those deaths categorised as still births were excluded.

Table 1: Groups of causes of avoidable mortality.

1Groups used in cause-specific decomposition.
2Includes all deaths in those aged 0–74 unless stated otherwise.
3Amenable deaths have been aligned with definition published by the Ministry of Health in July 2016.

Life expectancy at birth using data for 2013–15 (three year aggregated) was calculated using standard life table techniques, with 90 and older being the final age group. The impact of avoidable mortality on differentials in life expectancy was estimated by decomposing the difference in life expectancy using the method developed by Arriaga.27 This method allows differentials in life expectancy to be decomposed into age group and cause-of-death-specific contributions. Each sums to the total differential in life expectancy between groups.

Life tables for Māori males, Māori females, Pacific males, Pacific females, non-Māori/non-Pacific males and non-Māori/non-Pacific females were developed. The sex-specific non-Māori/non-Pacific tables served as the comparator group to compare the Māori and Pacific sex-specific tables. The contribution of avoidable causes of mortality and of the effects by age on life expectancy differentials in both Māori and Pacific compared with non–Māori/non–Pacific was analysed.

This study was outside the scope of ethics review due to the use of non-identifiable and encrypted data.

Results

Avoidable deaths

Between 2013 and 2015 there were 92,196 deaths registered in New Zealand. Of these, 25,210 (27.3% of all deaths and 68.2% of deaths in the 0 to 74 age group) were potentially avoidable (Figure 1). Of those deaths identified as potentially avoidable, 6,624 (7.2% all deaths) were preventable only, 4,056 (4.4% all deaths) amenable only and 14,530 (15.8% all deaths) both preventable and amenable. Within Māori, there were 9,717 deaths of which 5,152 (53.0% all deaths and 74.5% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in Māori 1,451 (14.9% all deaths) were preventable only, 937 (9.6% all deaths) amenable only and 2,764 (28.4% all deaths) both preventable and amenable. Within Pacific, there were 3,720 deaths of which 1,761 (47.3% all deaths and 71.3% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in Pacific, 355 (9.5% all deaths) were only preventable, 441 (11.6% all deaths) only amenable and 965 (25.9% all deaths) both preventable and amenable. Within non-Māori/non-Pacific, there were 78,759 deaths of which 18,297 (23.2% all deaths and 68.0% of deaths in the 0 to 74 age group) were potentially avoidable. Of the avoidable deaths in non-Māori/non-Pacific, 4,818 (6.1% all deaths) were only preventable, 2,678 (3.4% all deaths) only amenable and 10,801 (13.7% all deaths) both preventable and amenable. The proportion of all potentially avoidable deaths was twice as high in both the Māori and Pacific populations compared with the non- Māori /non-Pacific population.

Figure 1: Percentage of deaths that were potentially avoidable by ethnicity and total, 2013–2015.

c


Life expectancy at birth

For all deaths registered in 2013–15, estimates of life expectancy were: 73.5 years in Māori males, 77.3 years in Māori females, 75.0 years in Pacific males and 78.3 years in Pacific females. Life expectancy in the comparator groups was 80.9 years in non-Māori/non-Pacific males and 84.3 years in non-Māori/non-Pacific females. Overall these results equated to differentials in life expectancy of 7.4 years in Māori males, 7.0 years in Māori females, 5.9 years in Pacific males, and 6.0 years in Pacific females.

Decomposition by avoidable category and age

Within both Māori and Pacific females, higher mortality rates at a younger age from conditions considered both preventable and amenable made the greatest contribution to the life expectancy differentials, contributing 2.1 years and 1.7 years respectively (Table 2). Within Māori females, this was followed by conditions that are only preventable (1.3 years). However, within Pacific females, preventable conditions contributed only 0.3 years. Conditions only amenable contributed 1.0 years in Māori females and 1.3 years in Pacific females. Within age groups, mortality in the 50–74 year-old group contributed the most to the life expectancy differentials, contributing 4.1 years for Māori females and 3.3 years for Pacific females.

Table 2: Decomposition of the life expectancy differential in females by avoidable category and age.

1Sum of columns along with non-avoidable and avoidable may not equal the exact life expectancy differential due to rounding.

Similar to females, conditions both preventable and amenable were the greatest contributors to the life expectancy differentials in Māori and Pacific males, contributing 2.7 years and 1.7 years respectively (Table 3). Conditions considered preventable only were the second greatest contributor in Māori males and third greatest contributor in Pacific males; however, the number of years contributed within this group was higher for Māori males (1.6 years) compared with Pacific males (0.8 years). Within Pacific males, conditions considered amenable only were the second greatest contributor to the differential. Within age groups, higher mortality rates in the 50–74 year-old group contributed the most to life expectancy differentials, contributing 4.1 years for Māori males and 3.1 years for Pacific males. Even within the youngest age groups (<50 years), there were large contributions to life expectancy differentials for Māori (2.2 years) Pacific (1.5 years) males.

Table 3: Decomposition of the life expectancy differential in males by avoidable category and age.

1Sum of columns along with non-avoidable and avoidable may not equal the exact life expectancy differential due to rounding.

Decomposition by avoidable cause

Within Māori females, trachea, bronchus and lung cancers accounted for 0.9 years of the life expectancy differential, followed by coronary disease (0.6 years), with diabetes and chronic obstructive pulmonary disease each contributing 0.4 years (Figure 2). Within Māori males, coronary disease, trachea, bronchus and lung cancers and diabetes accounted for 1.2 years, 0.8 years and 0.5 years respectively (Figure 3). In Māori males, avoidable injuries had a pronounced contribution to the differential. Injuries contributed 1.0 year to the differential, with suicide and land transport injuries contributing over half of the differential associated with avoidable injuries. A large proportion of the differential caused by injuries was a result of higher mortality rates in the 0–29 year-old age group.

Figure 2: Decomposition of the life expectancy differential by the leading avoidable causes and age—Māori females.

c

Figure 3: Decomposition of the life expectancy differential by the leading avoidable causes and age—Māori males.

c

Within Pacific females, diabetes accounted for 0.6 years of the life expectancy differential, followed by coronary disease (0.5 years), cerebrovascular disease (0.4 years), and uterine and breast cancers (0.3 years each) (Figure 4). The life expectancy differential associated with cancers of the trachea, bronchus and lung was 0.1 years in Pacific females and 0.9 years in Māori females. Within Pacific males, coronary disease accounted for one full year of the life expectancy differential, followed by diabetes (0.7 years) and trachea, bronchus and lung cancers along with cerebrovascular disease (each contributing 0.4 years) (Figure 5). In contrast to Māori males, avoidable injuries were not a significant contributor to the life expectancy differential for Pacific males. However, assessment by age group identified a small contribution as a result of higher mortality rates for suicide in the 0–29 age group; however, this was offset by lower suicide rates in the older Pacific age groups (data not shown). Within the Pacific males and females, the summed contribution of ‘other avoidable causes’ resulted in a net reduction in life expectancy differentials. The largest contributor to this was lower mortality rates from melanoma where mortality rates are higher in the non-Māori/non-Pacific populations. Lower mortality rates from melanoma also had a net reduction on Māori life expectancy differentials.

Figure 4: Decomposition of the life expectancy differential by the leading avoidable causes and age—Pacific females.

c

Figure 5: Decomposition of the life expectancy differential by the leading avoidable causes and age—Pacific males.  

c

Discussion

This analysis has identified that potentially avoidable causes of death make up over half of all Māori deaths and nearly half of all Pacific deaths. In addition, they make a prominent contribution to life expectancy differentials in the Māori and Pacific population groups when compared with the non- Māori/non-Pacific group. Higher mortality rates at a younger age from conditions both preventable and amenable are the greatest potentially avoidable contributor to the differential within both populations. For Māori, coronary disease and cancers of the trachea, bronchus and lung in both males and females and avoidable injuries in males were the leading cause specific contributors. A finding specific to Māori males was the contribution of avoidable injuries, such as suicide and land transport accidents. Suicide was the fourth leading avoidable cause of the differential in Māori males. The finding relating to injuries was not replicated within Pacific males. Within both Pacific males and females, coronary disease, diabetes and stroke were the greatest cause-specific contributors to the differentials. Within Pacific females, the smaller contribution from conditions considered only preventable is predominantly due to lower mortality rates from cancers of the trachea, bronchus and lung compared with Māori females. This difference could possibly be explained by the lower smoking rates in Pacific females compared with Māori females.28 A finding unique to Pacific females was the large contribution of uterine cancer which was the leading avoidable cancer contributing to the differential. Higher rates of uterine cancer among Pacific females have been noted elsewhere with identified causal risk factors being diabetes and obesity.29

Pathways to inequity

Various exposures and experiences contribute to poorer health and inequities in avoidable mortality across different population groups. Three main pathways have been suggested that contribute to ethnic inequalities in health outcomes.30 These pathways align well with the concept of avoidable mortality. The pathways also recognise the role of racism in determining these mechanisms leading to inequality. Firstly, differential access and exposure to the upstream determinants of health such as inequitable access to education, employment and income along with other factors such as health literacy. These factors contribute to disparities by fostering illness, delaying care seeking and discouraging good adherence to treatment. This differential access is patterned in other risk factor exposure, for example, Māori and Pacific populations have higher rates of smoking and are more likely to live in more socioeconomically disadvantaged areas compared to non-Māori/non-Pacific people.31 These are risk factors for many conditions classified as preventable such as many avoidable cancers and other chronic conditions.

Intergenerational transmission of the social and economic determinants of health also presents a significant challenge for Māori and Pacific populations, whereby children from low socioeconomic backgrounds have a higher likelihood of remaining in that position through inequitable educational outcomes, restricted employment opportunities and lower incomes.32,33 For Māori, variation in behaviours deemed to be risky among males34,35 is further likely a contributing factor in the differential of Māori life expectancy. This is particularly evident in the contribution of avoidable injuries to the differential of 1.0 year in Māori males. The contribution of suicide to sex-specific differentials in life expectancy in New Zealand has previously been reported.21 However, this is the first time that ethnic-specific differences in New Zealand males have been reported.

Differential access to healthcare and the quality of care received are the second and third pathways that contribute to health inequities. These pathways can be considered downstream factors contributing to life expectancy differentials from conditions considered amenable. The health system literature describes how Māori will often experience longer and slower pathways through the health system.36–38 Furthermore, there is evidence that Māori and Pacific populations have poorer cancer survival compared with other ethnic groups39 with studies showing that Māori patients receive poorer quality treatment for potentially avoidable conditions such as lung and colorectal cancer.40 In addition, Māori and Pacific populations are more likely to report unmet need when engaging with health services.41

Racism at the personal and systems level is increasingly being recognised as a determinant of health in New Zealand, with many of the aforementioned pathways to inequities being underpinned by racism and its multiple expressions. Evidence shows that experience of racism can impact negatively on health and that Māori are more likely to experience interpersonal racism than non-Māori.12,13,42

The role of health services

Hospitals and health services have a crucial role to play as facilitators in reducing health inequities and the burden associated with avoidable mortality43 as the health system in its own right can be considered a determinant of health. The contribution of amenable mortality to life expectancy differentials highlights the need for health services to be further reoriented to ensure timely and equitable access for all population groups. This reorientation also involves ensuring that the health workforce reflects, and is responsive to, the population it serves. Health services, particularly hospitals hold a significant share of power, resources and influence within the health and community structure.43 As such, healthcare services are well positioned to lead the way in work to overcome barriers to achieving health equity, ensuring an equity in all policies approach and leading the intersectoral action required to improve equity in health outcomes. Primary healthcare services are particularly well placed to take action on the determinants of health and be strong campaigners for system change. While many of the social and economic determinants of health are outside the direct influence of medical professionals, they can be powerful advocates for the social and structural change required to reduce health inequities. However, this may only be realised through improved supportive policy and creating greater, systematic links with communities. In addition, health services, intersectoral agencies and in particular the workforce, need to recognise, name and understand the systems, attitudes and actions leading to institutional racism and discrimination. The impact of racism on health equity needs to be acknowledged, openly discussed and be part of medical education, as well as imbedded in health and institutional policy.

The high proportion of avoidable chronic disease, cancer and injury indicates a further need to strengthen policies targeting the respective risk and protective factors of these conditions. These policies need to focus on improving the social and economic determinants of health (such as income, education and housing) so that the opportunities to engage in health-promoting behaviours are equitably distributed across the population. There is also scope for medical treatment to achieve considerable health gain through secondary preventive services (such as diabetes management, cardiovascular risk and cancer screening). However, this requires the cultural safety of health services being addressed. While there have been declines in all-cause mortality rates over time in the Māori and Pacific population,44 there are still substantial mortality gaps that cannot be ignored if we are to close the differentials in life expectancy. This is particularly so for lung cancer which is the largest driver of cancer inequalities.45 This is likely a reflection of the marked ethnic gradient in tobacco smoking in New Zealand28 and highlights the potential for further advances in tobacco control.

Strengths and limitations

This study has several strengths. It benefitted from access to complete mortality data for all registered deaths in New Zealand and robust population data that allowed a complete population-based analysis of avoidable mortality and its contribution to life expectancy differentials. Furthermore, the use of well-established decomposition methods that decompose life expectancy by cause and age-group, allows for a richer analysis of the impact of avoidable mortality on life expectancy differentials experienced across the Māori and Pacific populations. This in itself could be useful insight for decisions on the specific targeting of upstream or structural interventions along with framing social policies that are likely to have the greatest impact in terms of reducing health inequities. In addition, the study made a distinction between mortality that could be tackled by the health system (amenable mortality) and that needing policy interventions and intersectoral action (preventable mortality) and provides a new dimension to understanding life expectancy differentials.

The findings presented in this paper should be interpreted in light of limitations common to ecologic analyses. We used aggregated data that prevent inference to individuals. Numerator/denominator bias may also occur whenever counts are collected using different methods. Ethnicity details on mortality data are sourced from the official record of death held by the Births, Deaths and Marriages.22 The ethnicity data for the official record of death is collected by funeral directors in consultation with a member of the deceased family. Where ethnicity is not available on the official record of death, National Health Index information will be used to supplement for this. On the other hand, ethnicity data for modelled population estimates are based on self-identification as part of the national Census. The level of possible undercount specific to our study is difficult to quantify, however, previous studies have identified undercounts of Māori deaths and overcounts of Pacific deaths in mortality data compared to census data.46 Finally, while a particular condition can be considered to be avoidable, this does not mean that every death from this condition could be prevented. With analysis of avoidable mortality, the precise nature of each death, the extent of disease progression at diagnosis, the existence of other medical conditions or the underlying exposure to risk factors is not taken into account.

It is important to acknowledge the contribution of conditions classified in this analysis as ‘non-avoidable’, particularly deaths among those older than 74 years of age. Many of the conditions within the avoidable mortality definition also occur in the older age groups. This does not mean that these conditions cannot be effectively treated at older ages. The primary reason for the 74-year-old cut off within the avoidable mortality definition is that the prevalence of (multiple) co-morbidity rises rapidly beyond this age. As such, it becomes increasingly difficult to assign a single cause to deaths in the older age groups. The quality of diagnosis and coding also tends to be lower in the very old. Including deaths beyond age 75 would likely impact the validity of the avoidable mortality indicator if a higher upper age limit were applied.

Contribution and past research

To the best of our knowledge, this is the first study to estimate the magnitude of avoidable mortality on life expectancy differentials in Māori and Pacific populations in New Zealand. This study details the potentially avoidable causes of death contributing to the frequently cited ethnic differentials in life expectancy and provides a new dimension to understanding this inequity. Overall rates of potentially avoidable mortality have previously been assessed in New Zealand.16 Tobias and Jackson were able to show in 1996–97 that around 70% of all deaths in the 0–74 age group were considered potentially avoidable and rates in the Māori and Pacific populations were over twice as high as those in the European population. A previous assessment of potential gains in life expectancy from avoidable causes of death has also been made in the Māori and non-Māori population.20 Malcolm identified that in 1985–87, if all potentially avoidable deaths in that period did not occur, Māori males would have gained 2.6 years of life, Māori females 2.7 years of life, non-Māori males 2.3 years of life and non-Māori females 1.7 years of life. However, the analysis by Malcolm did not account for competing mortality, thus the true gains would have been smaller than those presented.

The impact of avoidable mortality on life expectancy has been analysed for other European countries with similar decomposition methods.47,48 However, the causes of death used to define avoidable mortality differ between studies and many have assessed the longitudinal changes in life expectancy as opposed to the absolute difference between population groups at a given point in time. Comparisons are difficult as the detailed lists of causes of avoidable death and ages are not the same between studies. This problem of comparability could be solved with the establishment of a common list that would allow comparisons between studies and across countries.

Conclusion

Avoidable mortality is a large proportion of all Māori and Pacific deaths in New Zealand and is a prominent contributor to the life expectancy differentials in these groups. This study provides detail of the potentially avoidable causes of death contributing to these inequitable life expectancy differentials. It further reinforces the need for policymakers to enhance prevention activities and improve and ensure equitable healthcare delivery. This will require establishing a clear equity focus in all polices across all health and social services. Furthermore, there are clear benefits of looking beyond individual factors and recognising the role of healthcare services in improving health equity, in particular the role of racism. It is important to note that not all approaches used to improve health outcomes for the majority of the population will result in a reduction in inequities, in fact some methods inadvertently increase inequities. To address these inequalities, policymakers should focus on prevention activities and improve healthcare delivery with a clear equity focus, and also support wider improvements in educational achievement and socioeconomic position for the Māori and Pacific populations. Implementing the systems changes that are required to achieve equity will likely improve health for all.

Summary

Abstract

Aim

To determine the contribution of avoidable causes of death to the life expectancy differentials in both Mori and Pacific compared with non-Mori/non-Pacific ethnic groups in New Zealand.

Method

Death registration data and population data for New Zealand between 2013-15 was used to calculate life expectancy. A recent definition of avoidable mortality was used to identify potentially avoidable deaths. Life expectancy decomposition was undertaken to identify the contribution of avoidable causes of death to the life expectancy differential in the Mori and Pacific populations.

Results

Nearly half of all deaths in Pacific (47.3%) and over half in Mori (53.0%) can be attributed to potentially avoidable causes of death, compared with less than one quarter (23.2%) in the non-Mori/non-Pacific population. Conditions both preventable and amenable contribute the greatest to the life expectancy differentials within both ethnic groups, when compared with non-Mori/non-Pacific. Cancers of the trachea, bronchus and lung are significant avoidable causes contributing to the life expectancy differentials in both male and female Mori, contributing 0.8 years and 0.9 years respectively. Avoidable injuries including suicide contribute 1.0 year to the differential in Mori males. Coronary disease, diabetes and cerebrovascular disease are the largest contributors to the differential in both Pacific males and females.

Conclusion

Avoidable causes of death are large contributors to the life expectancy differentials in Mori and Pacific populations. The findings provide further evidence of the need to address the determinants of health and ensure equitable access to health services to reduce the impact of avoidable mortality on inequalities in life expectancy. It also highlights the importance of looking beyond individual factors and recognising the role of healthcare services and the social determinants in improving health equity.

Author Information

Michael Walsh, Epidemiologist, Planning Funding and Outcomes, Waitemata District Health Board, Auckland; Corina Grey, Public Health Physician, Planning Funding and Outcomes, Waitemata District Health Board, Auckland.

Acknowledgements

The authors would like to acknowledge Karen Bartholomew for her helpful contribution. The opinions expressed in this article are those of the authors and not of Waitemata District Health Board.

Correspondence

Michael Walsh, Planning Funding and Outcomes, Waitemata District Health Board, Level 1, 15 Shea Terrace, Takapuna, Auckland.

Correspondence Email

michael.walsh@waitematadhb.govt.nz

Competing Interests

Nil.

  1. Global Health Observatory (GHO) data - Life expectancy. 2018. (Accessed 25/6/2018, at http://www.who.int/gho/mortality_burden_disease/life_tables/situation_trends/en/)
  2. Marriott L, Sim D. Indicators of Inequality for Māori and Pacific People2014.
  3. New Zealand Period Life Tables: 2012–14. 2015. (Accessed 24/08/2018, at http://archive.stats.govt.nz/~/media/Statistics/Browse%20for%20stats/NZLifeTables/HOTP12-14/NZLifeTables12-14HOTP.pdf)
  4. Woodward A, Blakely T. The Healthy Country? A History of Life and Death in New Zealand. Auckland: University of Auckland Press; 2014.
  5. Tobias M, Blakely T, Matheson D, Rasanathan K, Atkinson J. Changing trends in indigenous inequalities in mortality: lessons from New Zealand. Int J Epidemiol 2009; 38:1711–22.
  6. Blakely T, Tobias M, Atkinson J. Inequalities in mortality during and after restructuring of the New Zealand economy: repeated cohort studies. BMJ 2008; 336:371–5.
  7. Howden-Chapman P, Blakely T, Blaiklock AJ, Kiro C. Closing the health gap. N Z Med J 2000; 113:301–2.
  8. Howden-Chapman P, M T. Social Inequalities in Health: New Zealand 1999. Wellington: Ministry of Health and Wellington School of Medicine; 2000.
  9. Paradies Y. Colonisation, racism and indigenous health. Journal of Population Research 2016; 33:83–96.
  10. Krieger N. A glossary for social epidemiology. J Epidemiol Community Health 2001; 55:693–700.
  11. Harris R, Tobias M, Jeffreys M, Waldegrave K, Karlsen S, Nazroo J. Racism and health: the relationship between experience of racial discrimination and health in New Zealand. Soc Sci Med 2006; 63:1428–41.
  12. Tobias M, Harris R. Racism and health in New Zealand. Ethn Health 2010; 15:318–20.
  13. Harris RB, Stanley J, Cormack DM. Racism and health in New Zealand: Prevalence over time and associations between recent experience of racism and health and wellbeing measures using national survey data. PLoS One 2018; 13:e0196476.
  14. Tobias M, Yeh LC. How much does health care contribute to health gain and to health inequality? Trends in amenable mortality in New Zealand 1981–2004. Aust N Z J Public Health 2009; 33:70–8.
  15. Rutstein DD, Berenberg W, Chalmers TC, Child CG, 3rd, Fishman AP, Perrin EB. Measuring the quality of medical care. A clinical method. N Engl J Med 1976; 294:582–8.
  16. Tobias M, Jackson G. Avoidable mortality in New Zealand, 1981–97. Aust N Z J Public Health 2001; 25:12–20.
  17. Ministry of Health. New Zealand Health Strategy: Future direction. 2016.
  18. Achieving equity. 2018. (Accessed 24/08/2018, at http://www.health.govt.nz/about-ministry/what-we-do/work-programme-2018/achieving-equity)
  19. System Level Measures Framework. 2016. (Accessed 25/08/2018, at http://www.health.govt.nz/new-zealand-health-system/system-level-measures-framework)
  20. Malcolm M. Avoidable mortality and life expectancy in New Zealand. J Epidemiol Community Health 1994; 48:211.
  21. Sandiford P. Gender inequality in New Zealand life expectancy: decomposition by age and cause. N Z Med J 2009; 122:10–7.
  22. Mortality Collection - Data Dictionary. 2017. (Accessed 25/08/2018, at http://www.health.govt.nz/system/files/documents/publications/mortality_data_dictionary_v1.6_final.pdf)
  23. HISO 10001:2017 Ethnicity Data Protocols,. 2017. (Accessed 24/08/2018, at http://www.health.govt.nz/system/files/documents/publications/hiso-10001-2017-ethnicity-data-protocols.pdf)
  24. Page A, Tobias M, Glover J, Wright C, Hetzel D, Fisher E. Australian and New Zealand atlas of avoidable mortality. Adelaide: University of Adelaide; 2006.
  25. Olatunde O, Windsor-Shellard B, Campbell A. Revised Definition of Avoidable Mortality and New Definition for Children and Young People2016.
  26. Defining Amenable Mortality. 2016. (Accessed 24/08/2018, at http://nsfl.health.govt.nz/system/files/.../defining_amenable_mortality_sept2016.docx)
  27. Arriaga EE. Measuring and explaining the change in life expectancies. Demography 1984; 21:83–96.
  28. Regional Data Explorer 2014–17: New Zealand Health Survey [Data File]. 2018. (Accessed 24/10/2018, at http://minhealthnz.shinyapps.io/nz-health-survey-2014-17-regional-update)
  29. Meredith I, Sarfati D, Ikeda T, Atkinson J, Blakely T. High rates of endometrial cancer among Pacific women in New Zealand: the role of diabetes, physical inactivity, and obesity. Cancer Causes Control 2012; 23:875–85.
  30. Jones CP. Invited commentary: “race,” racism, and the practice of epidemiology. Am J Epidemiol 2001; 154:299–304; discussion 5–6.
  31. Ministry of Health. Health Loss in New Zealand 1990–2013: A report from the New Zealand Burden of Diseases, Injuries and Risk Factors Study. Wellington: Ministry of Health; 2016.
  32. Kawharu M. Aotearoa: shine or shame? A critical examination of the Sustainable Development Goals and the question of poverty and young Māori in New Zealand. Journal of Global Ethics 2015; 11:43–50.
  33. Thayer ZM, Kuzawa CW. Ethnic discrimination predicts poor self-rated health and cortisol in pregnancy: insights from New Zealand. Soc Sci Med 2015; 128:36–42.
  34. Turner C, McClure R. Age and gender differences in risk-taking behaviour as an explanation for high incidence of motor vehicle crashes as a driver in young males. Inj Control Saf Promot 2003; 10:123–30.
  35. Morrongiello BA, Rennie H. Why do boys engage in more risk taking than girls? The role of attributions, beliefs, and risk appraisals. J Pediatr Psychol 1998; 23:33–43.
  36. Makowharemahihi C, Lawton BA, Cram F, Ngata T, Brown S, Robson B. Initiation of maternity care for young Māori women under 20 years of age. N Z Med J 2014; 127:52–61.
  37. Robson B. Hauora: Māori Standards of Health IV. A study of the years 2000–2005. Wellington: Te Ropu Rangahau Hauora a Eru Pomare; 2007.
  38. Bramley D, Riddell T, Crengle S, et al. A call to action on Māori cardiovascular health. N Z Med J 2004; 117:U957.
  39. Jeffreys M, Stevanovic V, Tobias M, et al. Ethnic inequalities in cancer survival in New Zealand: linkage study. Am J Public Health 2005; 95:834–7.
  40. Hill S, Sarfati D, Robson B, Blakely T. Indigenous inequalities in cancer: what role for health care? ANZ J Surg 2013; 83:36–41.
  41. Matheson A, Ellison-Loschmann L. Addressing the complex challenge of unmet need: a moral and equity imperative? N Z Med J 2017; 130:6–8.
  42. Harris RB, Cormack DM, Stanley J. The relationship between socially-assigned ethnicity, health and experience of racial discrimination for Māori: analysis of the 2006/07 New Zealand Health Survey. BMC Public Health 2013; 13:844.
  43. Matheson A, Bourke C, Verhoeven A, et al. Lowering hospital walls to achieve health equity. BMJ 2018; 362:k3597.
  44. Disney G, Teng A, Atkinson J, Wilson N, Blakely T. Changing ethnic inequalities in mortality in New Zealand over 30 years: linked cohort studies with 68.9 million person-years of follow-up. Popul Health Metr 2017; 15:15.
  45. Teng AM, Atkinson J, Disney G, Wilson N, Blakely T. Changing socioeconomic inequalities in cancer incidence and mortality: Cohort study with 54 million person-years follow-up 1981–2011. Int J Cancer 2017; 140:1306–16.
  46. Boyd M, Blakely T, Atkinson J. Ethnic counts on mortality, New Zealand Cancer Registry and census data: 2006–2011. N Z Med J 2016; 129:22–39.
  47. Nolte E, Scholz R, Shkolnikov V, McKee M. The contribution of medical care to changing life expectancy in Germany and Poland. Soc Sci Med 2002; 55:1905–21.
  48. Gispert R, Serra I, Bares MA, Puig X, Puigdefabregas A, Freitas A. The impact of avoidable mortality on life expectancy at birth in Spain: changes between three periods, from 1987 to 2001. J Epidemiol Community Health 2008; 62:783–9.

Contact diana@nzma.org.nz
for the PDF of this article

Subscriber Content

The full contents of this pages only available to subscribers.

LOGINSUBSCRIBE