June Atkinson

Changing trends in indigenous inequalities in mortality: lessons from New Zealand

BACKGROUND We describe trends from 1951 to 2006 in inequalities in mortality between the indigenous (Māori) and non-indigenous (non-Māori, mainly European-descended) populations of New Zealand. We relate these trends to the historical context in which they occurred, including major structural adjustment of the economy from the mid 1980s to the mid 1990s, followed by a retreat from neoliberal social and economic policies from the late 1990s onwards. This was accompanied by economic recovery and the introduction of health reforms, including a reorientation of the health system towards primary health care. METHODS Abridged period lifetables for Māori and non-Māori from 1951 to 2006 were constructed using standard demographic methods. Absolute [standardized rate difference (SRD)] and relative [standardized rate ratio (SRR)] mortality inequalities for Māori compared with European/Other ethnic groups (aged 1-74 years) were measured using the New Zealand Census-Mortality Study (an ongoing data linkage study that links mortality to census records) from 1981-84 to 2001-04. The SRDs were decomposed into their contributions from major causes of death. Poisson regression modelling was used to estimate the extent of socio-economic mediation of the ethnic mortality inequality over time. RESULTS Life expectancy gaps and relative inequalities in mortality rates (aged 1-74 years) widened and then narrowed again, in tandem with the trends in social inequalities (allowing for a short lag). Among females, the contribution of cardiovascular disease to absolute mortality inequalities steadily decreased, but was partly offset by an increasing contribution from cancer. Among males, the contribution of CVD increased from the early 1980s to the 1990s, then decreased again. The extent of socio-economic mediation of the ethnic mortality inequality peaked in 1991-94, again more notably among males. CONCLUSION Our results are consistent with a causal association between changing economic inequalities and changing health inequalities between ethnic groups. However, causality cannot be established from a historical analysis alone. Three lessons nevertheless emerge from the New Zealand experience: the lag between changes in ethnic social inequality and ethnic health inequality may be short (<5 years); both changes in the distribution of the social determinants of health and an appropriate health system response may be required to address ethnic health inequalities; and timely monitoring of ethnic health inequalities, based on high-quality ethnicity data, may help to sustain political commitment to pro-equity health and social policies.

Patterns of Disparity: Ethnic and Socio-economic Trends in Breast Cancer Mortality in New Zealand

ObjectiveTo test whether trends in breast cancer mortality varied by ethnicity and socio-economic position during the 1980s and ‘90s in New Zealand.MethodsFour cohorts of the entire New Zealand population for 1981–84; 86–89; 91–94 and 96–99 allowed direct determination of socio-economic trends in breast cancer mortality. For ethnicity, unlinked routine census and mortality data were used with adjustment factors for undercounting of Māori and Pacific deaths.ResultsMāori and non-Māori non-Pacific mortality rates changed little until mid-1990s with Māori experiencing 25% higher mortality. In 1996–99, Māori rates increased notably to become 68% higher than non-Māori non-Pacific (SRR 1.68; 95% CI: 1.49–1.90). Pacific women experienced an approximate three-fold increase in breast cancer mortality over time.There appeared to be reducing mortality among higher income and education groups but trends within socio-economic groups were not statistically significant. Nevertheless, by 1996–99, there was a significant 22% excess mortality (SRR 1.22; 95% CI: 1.01–1.49) for low compared with high-income groups.ConclusionsWidening ethnic, and probably, socio-economic disparities in breast cancer mortality are likely due to both underlying incidence and differential survival trends. Disparities are likely to increase once the full differential mortality benefits of screening impact on the population.

Do social and economic reforms change socioeconomic inequalities in child mortality? A case study: New Zealand 1981–1999

Background: Socioeconomic inequalities in child mortality are known to exist; however the trends in these inequalities have not been well examined. This study examines the trends in child mortality inequality between 1981 and 1999 against the background of the rapid and dramatic social and economic restructuring in New Zealand during this time period. Methods: Record linkage studies of census and mortality records of all New Zealand children aged 0–14 years on census night 1981, 1986, 1991, 1996, each followed up for three years for mortality between ages 1–14 years. Socioeconomic position was measured using maternal education, household income, and highest occupational class in the household. Standardised mortality rates, rate ratios, and rates differences as well as regression based measures of inequality were calculated. Results: Mortality in all socioeconomic groups fell between 1981 and 1999. Socioeconomic inequality in child mortality existed by all measures of socioeconomic position, however only trends by income suggested a change over time: the relative index of inequality increased from 1.5 in 1981–84 to 1.8 in 1996–99 (p trend 0.06), but absolute inequality remained stable (slope index of inequality 15/100 000 in 1981–84 and 14/100 000 in 1996–99. Conclusions: Dramatic changes in income in New Zealand possibly translated into increasing relative inequality in child mortality by income, but not by education or occupational class. The a priori hypothesis that socioeconomic inequalities in child mortality would have increased in New Zealand during a period of rapid structural reform and widening income inequalities was only partly supported.

Social inequalities or inequities in cancer incidence? Repeated census-cancer cohort studies, New Zealand 1981–1986 to 2001–2004

BackgroundWe examine incidence trends for 18 adult cancers, by ethnicity and socioeconomic position in New Zealand.MethodsThe 1981 to 2001 censuses were linked to subsequent cancer registrations, giving 47.5 million person-years of follow-up.ResultsEthnicity: Pooled over time, differences were marked: Pacific and Māori rates of cervical, endometrial, stomach and pancreatic cancers were 1.5–2.5 times European/Other rates; Māori, Pacific and Asian rates of liver cancer were 5 times European/Other; European/Other rates of colorectal, bladder and brain cancers were 1.5–2 times the rates of other groups and melanoma rates 5–10 times higher; Pacific and Asian kidney cancer rates were half those of Māori and European/Other.Over time, Māori and Pacific rates of cervical cancer fell faster and Māori rates of colorectal and breast cancer increased faster, than European/Other rates. Male lung cancer rates decreased for European/Other, were stable for Māori and increased for Pacific. Female lung cancer rates increased for all ethnic groups.Income: Other than lung (rate ratio 1.35 men, 1.56 women), cervical (1.35) and stomach cancer (1.23), differences in incidence by income were modest or absent.ConclusionsTobacco explains many of the social group trends and differences and constitutes an inequity. Cervical cancer trends are plausibly explained by screening and sexual practices. Faster increases of colorectal and breast cancer among Māori are presumably due to changes in dietary and reproductive behaviour, but the higher Māori breast cancer rate is unexplained.Ethnic differences in bladder, brain, endometrial and kidney cancer cannot be fully explained.

Trends in absolute socioeconomic inequalities in mortality in Sweden and New Zealand. A 20-year gender perspective

BackgroundBoth trends in socioeconomic inequalities in mortality, and cross-country comparisons, may give more information about the causes of health inequalities. We analysed trends in socioeconomic differentials by mortality from early 1980s to late 1990s, comparing Sweden with New Zealand.MethodsThe New Zealand Census Mortality Study (NZCMS) consisting of over 2 million individuals and the Swedish Survey of Living Conditions (ULF) comprising over 100, 000 individuals were used for analyses. Education and household income were used as measures of socioeconomic position (SEP). The slope index of inequality (SII) was calculated to estimate absolute inequalities in mortality. Analyses were based on 3–5 year follow-up and limited to individuals aged 25–77 years. Age standardised mortality rates were calculated using the European population standard.ResultsAbsolute inequalities in mortality on average over the 1980s and 1990s for both men and women by education were similar in Sweden and New Zealand, but by income were greater in Sweden.Comparing trends in absolute inequalities over the 1980s and 1990s, mens absolute inequalities by education decreased by 66% in Sweden and by 17% in New Zealand (p for trend <0.01 in both countries). Womens absolute inequalities by education decreased by 19% in Sweden (p = 0.03) and by 8% in New Zealand (p = 0.53). Mens absolute inequalities by income decreased by 51% in Sweden (p for trend = 0.06), but increased by 16% in New Zealand (p = 0.13). Womens absolute inequalities by income increased in both countries: 12% in Sweden (p = 0.03) and 21% in New Zealand (p = 0.04).ConclusionTrends in socioeconomic inequalities in mortality were clearly most favourable for men in Sweden. Trends also seemed to be more favourable for men than women in New Zealand. Assuming the trends in male inequalities in Sweden were not a statistical chance finding, it is not clear what the substantive reason(s) was for the pronounced decrease. Further gender comparisons are required.

Ethnic and socioeconomic trends in breast cancer incidence in New Zealand

BackgroundBreast cancer incidence varies between social groups, but differences have not been thoroughly examined in New Zealand. The objectives of this study are to determine whether trends in breast cancer incidence varied by ethnicity and socioeconomic position between 1981 and 2004 in New Zealand, and to assess possible risk factor explanations.MethodsFive cohorts of the entire New Zealand population for 1981-86, 1986-1991, 1991-1996, 1996-2001, and 2001-2004 were created, and probabilistically linked to cancer registry records, allowing direct determination of ethnic and socioeconomic trends in breast cancer incidence.ResultsBreast cancer rates increased across all ethnic and socioeconomic groups between 1981 and 2004. Māori women consistently had the highest age standardised rates, and the difference between Māori and European/Other women increased from 7% in 1981-6 to 24% in 2001-4. Pacific and Asian women had consistently lower rates of breast cancer than European/Other women over the time period studied (12% and 28% lower respectively when pooled over time), although young Pacific women had slightly higher incidence rates than young European/other women. A gradient between high and low income women was evident, with high income women having breast cancer rates approximately 10% higher and this difference did not change significantly over time.ConclusionsDifferences in breast cancer incidence between European and Pacific women and between socioeconomic groups are explicable in terms of known risk factors. However no straightforward explanation for the relatively high incidence amongst Māori is apparent. Further research to explore high Māori breast cancer rates may contribute to reducing the burden of breast cancer amongst Māori women, as well as improving our understanding of the aetiology of breast cancer.

Bias in relative survival methods when using incorrect life‐tables: Lung and bladder cancer by smoking status and ethnicity in New Zealand

Relative survival and excess mortality approaches are commonly used to estimate and compare net survival from cancer. These approaches are based on the assumption that the underlying (non‐cancer) mortality rate of cancer patients is the same as that of the general population. This assumption is likely to be violated particularly in the context of smoking‐related cancers. The magnitude of this bias has not been estimated. The objective of this article is to estimate the bias in relative survival ratios (RSRs) and excess mortality rate ratios (EMRRs) from using total population compared to correct subpopulation specific life‐tables. Analyses were conducted on 1996–2001 linked census–cancer data (including smoking status) for people with lung and bladder cancer, using sex‐specific (standard practice), sex‐ and ethnic‐specific, sex‐ and smoking‐specific and sex‐, ethnic‐ and smoking‐specific life‐tables. Five‐year RSRs using sex‐specific life‐tables, compared to fully stratified life‐tables, were underestimated by 10–25% for current smoking and Maori populations. For example, the current smoker male bladder cancer RSR was 0.700 for sex‐specific life‐tables, compared to 0.838 for fully stratified life‐tables. Similarly, EMRRs comparing current to never smokers and Maori to non‐Maori were overestimated using sex‐specific life‐tables only: modestly only for lung cancer, but markedly for bladder cancer. For example, the EMRR comparing current to never smokers with bladder cancer in a fully adjusted regression model was 1.475 when using sex‐specific life‐tables only, but reduced to 1.098 when using fully stratified life‐tables. Substantial bias can occur when estimating relative cancer survival across subpopulations if non‐matching life‐tables are used.

Ethnic and socioeconomic trends in testicular cancer incidence in New Zealand

Ethnic differences in testicular cancer incidence within countries are often sizeable, with white populations consistently having the highest ethnic‐specific rates. Many studies have found that high socioeconomic status is a risk factor for testicular cancer. The objectives of this article are to test whether trends in testicular cancer incidence have varied by ethnicity and socioeconomic position in New Zealand between 1981 and 2004. Five cohorts of the entire New Zealand population for 1981–1986, 1986–1991, 1991–1996, 1996–2001 and 2001–2004 were created, and probabilistically linked to cancer registry records, allowing direct determination of ethnic and household income trends in testicular cancer incidence. There were more than 2,000 cases of testicular cancer over the study period. We found increasing rates of testicular cancer for all ethnic and income groups since 1990s. Maori had higher rates, and Pacific and Asian lower rates than European/other men with rate ratios pooled over time of 1.51 (95% CI 1.31–1.74), 0.40 (95% CI 0.26–0.61) and 0.54 (95% CI 0.31–0.94), respectively. Overall, men with low incomes had higher risk of testicular cancer than those with high incomes (pooled rate ratio for lowest to highest income groups = 1.23; 95% CI 1.05–1.44). There was no strong evidence that disparities in testicular cancer incidence have varied by ethnicity or household income over time. Given the lack of understanding of the etiology of testicular cancer, the unusual patterns identified in the New Zealand context may provide some etiological clues for future novel research.

Patterns of Cancer Care Costs in a Country With Detailed Individual Data

Objective:To determine health system expenditure on cancers by time since diagnosis using data for an entire country. Methods:New Zealand cancer registry data was linked to hospitalization, pharmaceutical, outpatient, general practice, laboratory, and other datasets, with costs ascribed to each event occurring in 2006–2011. “Excess” cancer costs were estimated by subtracting “expected costs” for citizens without cancer from the “total cost” for cancer patients (

Changing socioeconomic inequalities in cancer incidence and mortality: Cohort study with 54 million person‐years follow‐up 1981–2011

2011 inflation-adjusted). Gamma regressions were used to estimate costs per person-month. Results:For first adult cancer diagnosed that the excess cost per person was between US