What potential has tobacco control for reducing health inequalities? The New Zealand situation

Abstract

In this Commentary, we aim to synthesize recent epidemiological data on tobacco and health inequalities for New Zealand and present it in new ways. We also aim to describe both existing and potential tobacco control responses for addressing these inequalities. In New Zealand smoking prevalence is higher amongst Māori and Pacific peoples (compared to those of "New Zealand European" ethnicity) and amongst those with low socioeconomic position (SEP). Consequently the smoking-related mortality burden is higher among these populations. Regarding the gap in mortality between low and high socioeconomic groups, 21% and 11% of this gap for men and women was estimated to be due to smoking in 1996-99. Regarding the gap in mortality between Māori and non-Māori/non-Pacific, 5% and 8% of this gap for men and women was estimated to be due to smoking. The estimates from both these studies are probably moderate underestimates due to misclassification bias of smoking status. Despite the modest relative contribution of smoking to these gaps, the absolute number of smoking-attributable deaths is sizable and amenable to policy and health sector responses. There is some evidence, from New Zealand and elsewhere, for interventions that reduce smoking by low-income populations and indigenous peoples. These include tobacco taxation, thematically appropriate mass media campaigns, and appropriate smoking cessation support services. But there are as yet untried interventions with major potential. A key one is for a tighter regulatory framework that could rapidly shift the nicotine market towards pharmaceutical-grade nicotine (or smokeless tobacco products) and away from smoked tobacco.

Figure 1

Age standardised lung cancer mortality rates in New Zealand by ethnicity and household income, males and females (per 100,000 population). Source: Data derived from: [17]. The bars indicate 95% confidence intervals. Note the different age range for ethnicity and household income. The ethnic mortality rates were calculated using adjustment factors (from the NZCMS) for historic undercounting of Māori and Pacific deaths [2, 3], and the income mortality rates were calculated directly from linked census-mortality data. Rates by household income are standardised or both age and ethnicity. Ethnicity definitions: The definition of ethnicity progressively changed from fractionated ethnic origin in the 1981 census (eg, 7/8 European, 1/8 Māori), to multiple self-identified ethnicity in 1996 elicited by the question: "Tick as many circles as you need to show which ethnic group(s) you belong to". This change in the question and secular trends in how people viewed their own ethnicity led to a disproportionate increase in the Māori population (than expected on the basis of demographic projections alone). However, trends in mortality rates shown above are largely unaffected, as the numerators have been adjusted to be consistent with the denominators.

Figure 2

The contribution of active tobacco smoking to 45–74 year old age-standardised mortality rates, and gaps in mortality rates, in 1996–99, by ethnicity and education (with the latter as a marker for SEP). Sources: Data derived from: [75] and [27]. nMnP – non-Māori non-Pacific (ie, mainly "New Zealand European" ethnicity). See the footnotes to Table 1 for ethnicity definitions.

Figure 3

Simplified causal/intervention model for pathways between ethnicity and socioeconomic position to mortality. * Direct interpersonal racism and institutional racism probably has a diffuse impact on many causal processes represented by this diagram, including the unequal distribution of socioeconomic resources, the quantity and quality of "stress" and "psychosocial resources", "access to/access through the health system", and patterns of drug use – including smoking. There are New Zealand specific data on racism and health and racism and smoking [33, 34].