Abstract
Background
Globally, societies are ageing as life expectancy increases and fertility rates decline. Many studies exploring associations between socioeconomic position (SEP) and health outcomes among the older population use conventional working-age measures of SEP (education, occupation, and income) or an area-based deprivation index.
Methods
Microdata from the 2013 New Zealand Census was used to create a new measure of socioeconomic position specifically for the population aged ≥65 years (SEP65), using information about housing tenure, wealth, and receiving government benefits. We dichotomised variables representing income, assets, and housing tenure to apportion 489,264 older people in our analytic sample into low, medium and high levels of SEP. A multivariable logistic regression was used to investigate the relationship between smoking and SEP65, controlling for age, sex, ethnicity and area-level deprivation.
Results
We report inequities between European, Māori and Pacific populations, which increase with age. We found an inverse relationship between the odds of smoking and levels of SEP65. Moreover, we further validated the SEP65 measure against the New Zealand Index of Deprivation (NZDep2013).
At the third SPINE20 summit 2022 which took place in Bali, Indonesia, in August 2022, 17 associations endorsed its recommendations.
Conclusion
Using an age-specific measure of SEP for the older population will aid policymakers in the appropriate allocation of resources according to need more effectively than working-age indicators.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
An ageing population is a feature of most developed countries; life expectancy is increasing over the long term while fertility rates continue to decrease. Older age groups now account for most health service utilisation in Aotearoa New Zealand (ANZ) (Ministry of Health, 2016). However, while life expectancy is increasing in all age groups, there are inequalities in life expectancy by gender, ethnicity and socioeconomic circumstance (Stats NZ Tatauranga Aotearoa, 2015). For example, those living in areas of high disadvantage are expected to live shorter lives than those living in areas with lower deprivation (7.5 years and 6.1 years shorter for males and females, respectively (Ministry of Social Development, 2016)). As Aotearoa New Zealand’s population continues to age, the Government will require a greater understanding of the older population’s socioeconomic position and deprivation status to make effective policy decisions and best utilise available resources.
1.1 Theoretical Background on Socioeconomic Position
Despite the evidence on inequalities in life expectancy (Dibben & Popham, 2012), research on measuring socioeconomic position in the older population is lacking (Grundy & Holt, 2001). Socioeconomic position (SEP), also known as socioeconomic status (SES), is broadly defined as the “social and economic factors that influence what position(s) individuals and groups hold within the structure of society” (Lynch & Kaplan, 2000, p. 14). Area deprivation indices are often used as a proxy measure of socio-economic levels, especially when information about individual income, education, or occupation information is not routinely available. In the UK, for example, deprivation indices such as Townsend (Townsend, 1987) and Carstairs and Morris (Carstairs & Morris, 1989) are used ubiquitously as SES measures, due to the absence of income-based information in the Census. SEP has been found to have strong associations with health outcomes (Bosworth, 2018; Chandola & Jenkinson, 2000; Darin-Mattsson et al., 2017; Exeter et al., 2005), and is a good predictor of overall health and wellbeing (House et al., 1990). For example, both in ANZ and internationally, there is a known relationship between smoking status and SEP (Blakely et al., 2006; Hill et al., 2005; Hiscock et al., 2012; Wilson et al., 2006). Smoking has been found to be four times higher for those who are most socioeconomically disadvantaged (Hiscock et al., 2012), and in the ANZ context smoking was found to vary drastically between ethnicity and socioeconomic circumstance (Hill et al., 2005). However, the majority of these studies are for the working age population and there is limited research on the relationship between smoking status and SEP for older people.
Commonly, socioeconomic position (SEP) is measured using variables such as income, occupation and education (Lynch & Kaplan, 2000). However, most older populations are not engaged in paid work (Acciai, 2018), and conventional working-age measures of SEP are less relevant for those in older age or retired. For example, income typically decreases in retirement, many older people being asset rich but income poor (Davey, 1996), and a person’s current income may not be an accurate measure of their SEP (Acciai, 2018; Kaplan et al., 1987). An older person may be retired, or if working, in a different work area or job description that for the majority of their working life. Therefore, occupation is not a suitable measure of SEP in older age. While educational attainment measures are the most likely to remain constant over time, older populations are less likely than younger populations to have received secondary and tertiary level education (Teh et al., 2014). Similar to income and occupation, education is not the most suitable measure of SEP in older age either.
A wide range of alternative indicators for measuring SEP in older population have been proposed. These include financial indicators such as liquid assets, homeownership, and lifestyle related indicators such as car access and heating. Robert and House (1996) investigated the suitability of the alternative indicators of socioeconomic status for predicting health in older age, finding that liquid assets and homeownership were suitable for predicting functional health in older age. This suitability of these alternative indicators is supported by other studies such as Connolly et al. (2010) who found accumulated wealth (based on homeownership and the value of the house) was a good measure of health status in older people. For older people, homeownership and the equity from a home is a key asset, providing both financial and social security and well-being (Ong, 2008). Indicators related to lifestyle or material items are not suitable for measuring SEP in older populations as these may be influenced by other factors, such as lifestyle choice, and may not be of a financial nature (Grundy & Holt, 2001).
There is a strong link between increased life expectancy and a person living in their own home (owner-occupation), where owner-occupiers tend to have better resilience to health concerns that may affect them in the future (Smith et al., 2003). However, homeownership is not equally distributed and those who have greater social, economic and cultural capital are significantly more likely to be homeowners. Additionally, inequalities in homeownership by ethnicity and level of socioeconomic deprivation exist. In the ANZ context, while two-thirds of the total population lived in their own home in 2013, this was true for less than half of the Māori and Pacific populations (Goodyear, 2017). For older populations evidence suggests that people not living in their own homes are “multiply disadvantaged” (Yates & Bradbury, 2010) and that health and wellbeing are better among owner-occupiers than renters (Windle et al., 2006). In cases where homeowners are mortgage-free at retirement, they are more likely to live adequately on government retirement payments (Castles, 1998). For retirees categorised as either non-mortgage-free homeowners or renters, a proportion of their income will be spent on accommodation costs, leaving less income available for weekly expenses (Castles, 1998).
Further, while homeownership is a strong indicator of SEP in older age, it does not guarantee that a person will have good health in old age. Homeownership can assist in decreasing the likelihood of falling into poverty in older age (Delfani et al., 2015; Dewilde & Raeymaeckers, 2008), but the income of an individual still plays an important role as those on low incomes are less likely to own a home mortgage free and those on higher incomes are more likely to own a home mortgage free and have increased equity associated with their homes (Hancock, 1998). Therefore, owning a home in old age is unlikely to completely remove problems associated with lower SEP for those in old age, and other variables such as whether the home is owned mortgage free, and the value of the home, and important indicators of SEP (Perry, 2010). Ideally, a measure of SEP for older populations would include indicators of house value, housing income (after costs), housing tenure, and assets and savings. This gives a good indication of the overall wealth of a person in older age, and provides a measure of a person’s access to resources in older age and better reflects the outcome of a lifetime financial journey of savings and borrowing (McMunn et al., 2006) than just using indicators of homeownership or income alone. We will use these ideal variables to create an ANZ measure of SEP for those aged ≥ 65.
There is consensus that wealth and accumulated asset indicators are the most suitable for measuring SEP in those aged ≥ 65 (Acciai, 2018; Darin-Mattsson et al., 2017; Steptoe & Zaninotto, 2020). In addition, wealth-based measures of SEP reveal graded associations with health outcomes among the population aged ≥ 65, which are largely masked when traditional indicators of SEP are used (Acciai, 2018).
1.2 SEP and Social Outcomes
The majority of research investigating associations between socioeconomic position and health or social outcomes among the population aged ≥ 65 years rely on measures of SEP that are based on working-age populations (Artazcoz & Rueda, 2007; Grundy & Holt, 2001; Jones & Higgs, 2013; Krieger et al., 1997; Lopes, 2011; Robert & House, 1996).
The lack of any SEP measures specific to the older population requires urgent attention, particularly as high-income countries where their populations are experiencing population ageing. There are a small number of studies that use a post-working-age population in their SEP measure. Connolly et al. (2010) used census data from Northern Ireland and created a measure of SEP based on the indicators of housing tenure (homeownership) and house value, alongside other socioeconomic indicators related to health status, for the older population. They found that house value and tenure were strongly associated with an individual’s current health status. Also, those not living in their own home had higher mortality rates than those living in their own homes. Grundy and Holt (2001) assessed the validity of traditional and alternative indicators of SEP for those aged 55–75, and similarly Dalstra et al. (2006) examined the strength of different indicators of SEP for assessing health for the older population, distinguishing between those living in their own home and those renting. Grundy and Holt (2001) found that the most suitable variables were education or social class, when coupled with an indicator of deprivation, while Dalstra et al. (2006) found that housing tenure was a good predictor of poor health. Robert and House (1996) also used less traditional indicators of SEP and assessed health for those in older age, comparing traditional indicators of income and education to the alternative indicators of liquid assets and homeownership. Results indicated that the alternative indicators better predicted functional status while the traditional indicators better predicted self-rated health. These studies generally support the use of both traditional and alternative indicators of SEP, however housing related indicators are key to include.
This paper aims to develop a measure of SEP for the population aged ≥ 65 years using 2013 Census data. We propose a conceptual framework in which house tenure is utilised as an indicator of wealth accumulated over a lifetime of SEP and incorporate assets and wealth indicators to ascertain SEP levels among adults aged ≥ 65. Consistent with many international measures of socioeconomic position or deprivation (Atkinson et al., 2014; Cai et al., 2019; Exeter et al., 2017; Fahy et al., 2017; Hiscock et al., 2012; Milne, 2012; Pampalon et al., 2008), we then carry out validity testing by exploring the associations between SEP and the likelihood of people smoking using a multivariate logistic regression, and investigating the relationship the proposed SEP65 measure has with an area-based measure of deprivation, where the indicators are predominantly focussed on the working-age population. Given the established convention in ANZ of increasing deprivation being represented by corresponding increasing deciles, we would expect an inverse association between the SEP65 and NZDep2013. To our knowledge this is the first study in ANZ to use whole population census data to create a measure of SEP reflective of the changing circumstances of older people.
2 Data and Methods
2.1 Population and Design
Ethical approval for this study was obtained from the Host Institution’s Human Participant’s Ethics Committee (Ref UAHPEC17259). We used the STROBE cross sectional reporting. This study uses de-identified individual and household-level data from the 2013 Census of Population and Dwellings in ANZ. The eligible population was adults aged ≥ 65 as of 5 March 2013 who were present in ANZ on Census night and who participated in the 2013 Census. Data used was sourced from the Integrated Data Infrastructure (IDI). The IDI comprises routinely collected administrative data from government and non-government organisations on topics such as health, housing, education and crime, and nationally representative surveys such as the New Zealand Health Survey and the 2013 Census (Stats NZ Tatauranga Aotearoa, 2021).
The census collects demographic and economic variables, including age, sex, income, household tenure, marital status and ethnicity. We used the total response ethnicity output approach, allowing individuals to identify with multiple ethnic groups. Therefore, the total sum of individuals by ethnicity is greater than the total population. We identified nine variables from the census, which were matched to four constructs for measuring SEP in older people: house value, housing income, housing tenure, and assets and savings. In creating the SEP65 measure, a mix of individual dwelling and household level variables were used (See Table 1). The 2013 Census outputs comprised a range of categorical and continuous variables, so all variables were dichotomised: a value of 1 indicates more advantageous socioeconomic circumstances and a value of 0 indicates greater disadvantage. Similar to previous work (Morris & Carstairs, 1991; Phillimore et al., 1994; Townsend, 1987), we assigned equal weights to the variables in the SEP65.
Total household income is a variable in the census, that is derived from total personal income. Thresholds for total household income were used to assign classifications, where those living alone have a different threshold than those not living alone. The thresholds were chosen based on the estimated annual income of a single or married couple, respectively, who receive New Zealand Super and Veteran Payments (NZSVP), where earning below this amount would be classified as low SEP. The Census asked respondents to state personal income, and to list all the sources of income (e.g. wages and salaries, interest and dividends, superannuation, and government benefits such as accommodation supplements or sickness and invalid benefits). While receiving a sickness or invalid benefit indicates poor health, these variables are assigned as 1 if an individual receives them as this is used as an indicator for receiving payments, which is in turn used to indicate an individual may have low SEP.
2.2 Population Characteristics
The 2013 Census details 4,353,198 usual residents in ANZ, of whom 628,638 (14%) were aged ≥ 65 (Table 2). Overall, there were marginally more females (51%) than males (49%), although the distribution diverged with increasing age among the ≥ 65 population, with nearly double the number of females (64%) than males (36%) at ≥ 85 years.
While nearly three-quarters of the total population were of European ethnicity, Table 2 highlights substantial ethnic differences among the population aged ≥ 65. The proportion of Europeans increased to 88%, while the proportion of Māori (5%), Asian (5%) or Pacific (2%) declined to approximately one third of the proportion of people aged below 65 (Table 2). In addition, as age increases, the proportion of the European population increased to 95% for the population aged ≥ 85 and a rapid decline in the percentage of Māori as the age bands increased, from 6% aged 65–74 years to 2% aged ≥ 85, highlighting the difference in life expectancy between European and Māori. There is a similar pattern for those who identify as Pacific.
2.3 Creating the SEP65 Measure
Variables used in the creation of the SEP65 are presented in Table 1. Figure 1 outlines the rules used to ascertain levels of SEP for the population aged ≥ 65. Individuals are classified as low SEP if they live in public rental accommodation. Individuals are also classified as low SEP if they live in either private rental housing or live in their own home but have a mortgage and have low household income, no investments, no NZSVP or a sickness benefit.
The final SEP65 measure score ranges from 0 to 2, where 0 is categorised as low SEP, 1 as medium SEP, and 2 as high SEP. Table 3 in the results section shows the demographic composition of the analytical sample by SEP65 category, including the number of people in our sample that are classified by each SEP65 category.
2.4 Statistical Analysis
Statistical analysis was carried out within the IDI’s secure data laboratory environment. Counts were calculated for those aged ≥ 65, stratified by gender, age, and ethnicity. We tested the validity of the SEP65 by exploring the associations between SEP65 and an area-level measure of deprivation, the NZDep2013 New Zealand Index of Deprivation 2013 (NZDep2013), Atkinson et al., 2014). We used deciles of deprivation, with Decile 1 representing the least deprived 10% of small areas across ANZ and Decile 10 the most deprived 10%. Given the established convention of increasing deprivation with corresponding deciles, we would expect an inverse association between the SEP65 and NZDep2013.
Binary logistic regression modelling was also used to test the validity of the SEP65 measure against the smoking status of the study population. We used individuals who stated they were regular smokers in the 2013 Census as the outcome variable, and separate analyses were completed controlling for (i) SEP65; (ii) SEP65, age, sex, ethnicity; and (iii) SEP65, age, sex, ethnicity and area deprivation. We chose the regular smoker category as our outcome of interest to align with previous work validating measures of SEP/deprivation (Atkinson et al., 2014; Exeter et al., 2017; Fahy et al., 2017; Milne, 2012). Area deprivation is used in the fully adjusted model in order to determine whether SEP65 had an effect on the likelihood of an individual smoking beyond area deprivation.
3 Results
3.1 SEP65 Measure
Figure 1 demonstrates the algorithm used to apportion individuals into the SEP65 categories shown in Table 3. Overall, there were 96,816 (20%) people aged ≥ 65 categorised as low SEP, while 286,440 (59%) were apportioned medium SEP, and 106,008 (22%) of the analytic population were high SEP. There was little to no difference in SEP by gender or age, although there were proportionally more people aged ≥ 85 in the high SEP group than for other ages. The greatest differences in SEP were by ethnic group. Almost a quarter of Europeans aged ≥ 65 were classified as high SEP (24%) and over half (57%) of Pacific people in low SEP and only 2% in high SEP. This is similar for Māori who have 42% in low SEP and only 7% in high SEP.
3.2 Validity Testing
We validated the SEP65 measure against two recognised indicators of deprivation: smoking status and the New Zealand Index of Deprivation (NZDep2013). While smoking status among the older population may be lower than for the general population, there is strong evidence of a graded association in New Zealand according to deprivation and by ethnic group (Boven et al., 2019; Hill et al., 2005; Hiscock et al., 2012) and it has been widely used in New Zealand (i.e. Atkinson et al., 2014; Exeter et al., 2017; Salmond et al., 2006) and abroad (i.e. Morris & Carstairs, 1991; Pampalon et al., 2012; Shohaimi et al., 2003).
Table 4 describes the distribution of area deprivation and variations in smoking patterns from the 2013 Census, by level of SEP65. The descriptive (univariate) analyses in Table 4 indicate that individuals aged ≥ 65 classified as low SEP are nearly four times more likely (11%) to be regular smokers than people in high SEP (3%). Those classified as low SEP have a similar likelihood of being an ex-smoker (30%) to those in medium (35%) or high SEP (35%). The likelihood of having never smoked regularly is higher for individuals aged ≥ 65 classified as high SEP than low SEP, but the difference is less than 10%.
3.3 Association with the New Zealand Index of Deprivation (NZDep) (Atkinson et al., 2014)
Table 4 also shows the proportion of the analytic sample in each SEP category by NZDep2013 deciles. By convention, deprivation circumstances increase with ascending NZDep2013 deciles, and an inverse association with SEP65 categories is to be expected. Of those aged ≥ 65, 45% of people in the low SEP group live in areas with an NZDep2013 index score of 8 or above, compared to only 15% of people in the high SEP category. Among individuals classified into high SEP, 42% were allocated an NZDep2013 index score of 3 or lower, while only 17% of those classified in low SEP were given the same score.
3.4 Regression Analysis
We used logistic regression to predict the likelihood of a person aged ≥ 65 being a regular smoker. It is important to note that because we used a whole-of-population cohort approach to this research, our analytic sample comprised 489,264 people aged ≥ 65 years, which means that all observations were estimated to have p < 0.001 in our logistic regression analyses and we report the Standard Errors alongside the Odds Ratios and 95% confidence intervals.
Table 5 shows the results from three separate models. The results show that the odds of being a regular smoker for the low SEP group is 3.48 times (95% CI 3.34–3.61) that of the high SEP group, which reduced marginally (but not significantly so) to 3.41 times (95% CI 3.28–3.55) when we controlled for gender, age and ethnicity. In the fully adjusted model, the more deprivation the area a person lives in, the more likely the person is to be a smoker compared to those in a low deprivation area. Those living in areas of the highest deprivation, are 2.65 times (95% CI 2.5–2.81) more likely to be smokers than those living in areas of the lowest deprivation.
4 Discussion
We present the first study to use whole population data to construct a census-based measure of socioeconomic position for the ≥ 65 population in ANZ. Theoretically driven indicators appropriate for a post-working population were used, thus addressing a significant gap in the literature and revealing the plausibility of using house tenure as a measure of SEP in older people. We highlight the significance of housing tenure as a valid indicator of accumulated wealth and the importance of owner-occupation, which may protect against poverty in later years (Dewilde & Raeymaeckers, 2008). The development of the SEP65 relies on information from the Census and uses a simple but effective method of apportioning the analytic population into three distinct levels of socioeconomic position.
Directly comparable studies to our research are limited as current measures developed for assessing SEP and SES in the ≥ 65 population are based on working-age populations, which are not reflective of circumstances experienced by people aged ≥ 65 years (Artazcoz & Rueda, 2007; Grundy & Holt, 2001; Krieger et al., 1997; Lopes, 2011; Robert & House, 1996). In studies that have focussed on older populations, key similarities to the findings of this study can be seen.
This study used a strong theoretical grounding in the selection of SEP indicators for the older population. The importance of a ‘theoretically grounded’ measure in Grundy and Holt (2001), who examined the applicability of different indicators of SEP and deprivation for a post-working population was extended by the SEP65. A key indicator used in the SEP65 is housing tenure, similar to a study by Connolly et al. (2010), where house value and tenure were used as indicators of accumulated wealth. Both the SEP65 and the work by Connolly et al. found owner-occupiers possess a significant advantage compared to those living in rental accommodation, in terms of better health or higher socioeconomic positioning. These findings highlight the importance of using a measure of SEP incorporating housing tenure and wealth when analysing the social circumstances of older people.
We reported an ethnic disparity in the distribution of the SEP65, with lower rates of wealth in Māori and Pacific populations compared to Europeans. This finding is supported by a study by Lotoala et al. (2014) who found similar inequities between European and Pacific people where Europeans scored the highest on income, assets and wealth levels and Pacific had the lowest levels of assets and wealth. The consistency of Lotoala et al.’s (2014) results and this study stress the need to address ethnic inequities among our older population using measures specific to this age group, including the SEP65.
We also reported a social gradient in the association between smoking and SEP65. Those older people in low SEP were three times more likely to be regular smokers than those in high SEP. Our observation broadly aligns with previous research using the Irish Longitudinal Study of Ageing, which found older adults in the lowest wealth quartile are nearly three times more likely to be current smokers than those in the most advantaged group (Cronin et al., 2011). Similar results were found by Lugo et al. (2013), where smoking prevalence was highest in men with limited education. While older people may be less likely than young adults to be regular smokers, we highlight the persistence of smoking inequities in old age.
4.1 Wider Implications and Future Work
The SEP65 is a simple and easy-to-use tool which could be recreated within the IDI by those wanting to measure SEP for those aged ≥ 65. SEP65 could be integrated with other data in the IDI, such as health information for example, which could assist in informing societies most ‘at-risk’ groups based on health status and SEP position. This information could in turn assist in the allocation of resources to the most vulnerable members of society. However, the key drivers of SEP in older people must be addressed before improvements in health can be made. SEP and health work hand-in-hand and ethnic disparities in low SEP must be recognised and urgently given attention by those developing policies and working in Government to ensure older people can age equally. Low SEP and financial hardship in later life influence the day-to-day living standards and have a detrimental impact on mental health, including loneliness (Stephens, 2016).
This research focuses solely on information available from the 2013 Census, which is available as microdata within Statistics New Zealand’s Integrated Data Infrastructure. At present, the IDI is a repository for nationally represented surveys, routinely collected government agency data and some non-governmental organisation data. These sources integrate population, housing, income, health education, people and communities, justice, education and benefit/social services data, enabling researchers to explore research questions using de-identified data.
The plethora of data available within the IDI provides numerous opportunities for further work. For example, creating a measure that incorporates additional data sets available in the IDI, allowing for a greater understanding surrounding the different types of assets and savings older people have and enhancing the differences between low and high SEP groups. Integrating routinely collected information on the population of service providers (e.g. all publicly funded health service users) with nationally-representative survey information introduces new methodological challenges, although the inclusion of the Household Economic Survey data may provide more detailed information from a subset of the population on income, debts, savings and the wealth of older people in ANZ, while integration with the NZ General Social Survey would provide estimates of self-reported health information thus facilitating research similar to Connolly et al. (2010). The IDI also permits some linkage to information available outside the IDI, such as information about house values from a private/commercial data provider (Quotable Value) when measuring socioeconomic position in the ≥ 65 population. Combining this information with the SEP65 in future work would allow an examination of the true extent to which mortality and health related outcomes are influenced by low SEP for those in older age. We acknowledge that the cost of living and ones ability to purchase goods and services are key aspects that influence socioeconomic position (Galobardes et al., 2006). Unfortunately it was not possible to discern specific information regarding the number or type of assets or an individual’s spending behaviour from the data available from the 2013 Census data. Although linkage to Inland Revenue (the NZ Government’s tax department) was beyond the scope of this study, future research could explore data from Inland Revenue (the NZ Government’s tax agency), or other data from private data providers to better measure assets and wealth among our analytic sample.
The SEP65 excluded non-private dwellings from the analytic population as no household-level measures are relevant to older people living in institutions is available from the NZ Census, despite being an important population group to be considered in the SEP context (Grundy & Holt, 2001). However, including institutional care facilities in a future measure of SEP would change the focus of the measure from accumulated wealth and SEP to functional status and health. This could be work for an alternative measure of SEP.
We opted to assign equal weights to each indicator included in SEP65, an approach that is consistent with methods used to weight indicators for census-derived deprivation indices (Morris & Carstairs, 1991; Norman et al., 2019; Townsend, 1987). Indeed, selecting the method of weighting indicators (in the context of deprivation indices specifically) has drawn much debate (Coombes et al., 1995; Noble et al., 2004, 2006), however more recent research (Schederecker et al., 2019; Watson et al., 2019), has demonstrated that overall, the final weights applied makes little impact to the associations or rankings in the indices. Further research is required to explore the impact of different weights on the indicators used in the SEP65.
The development of the SEP65 could be further enhanced be using more advanced statistical methods, such as factor analyses to apportion the analytic sample into SEP categories. This research was part of a wider project exploring the measures of multiple deprivation among older people, and readers may be interested in the complementary work using the IDI, developing an older person’s index of multiple deprivation (Exeter et al., 2022).
5 Conclusion
The construction of a measure of SEP that is representative for the ≥ 65 population is an area of research that has been largely ignored to date. This is due to the complexities of measuring SEP in a post-working population. This study has explored these complexities and shown the relevance of using non-traditional indicators for measuring SEP in the older population. We created a measure of SEP using theoretically based indicators that splits the ≥ 65 population into three separate groups (low, medium and high SEP). An advantage of the SEP65 is the stratification by ethnicity, given the pervasive social gradient between marginalised ethnic groups (Jatrana & Blakely, 2008). A strong ethnic disparity between Europeans, Māori and Pacific people was revealed using the SEP65. The SEP65 used housing tenure and other relevant indicators to create a measure of accumulated wealth, capturing the older population’s circumstances while highlighting whether people have been ‘economically successful’ or not. The SEP65 may be replicated using census data from other countries for which microdata are available, potentially enabling cross-country comparisons. The simplicity of the SEP65 measure will encourage its use by researchers and policymakers to inform the reduction of health and wellbeing inequities experienced among New Zealand's older population.
References
Acciai, F. (2018). The age pattern of social inequalities in health at older ages: Are common measures of socio-economic status interchangeable? Public Health, 157, 135–141. https://doi.org/10.1016/j.puhe.2018.01.002
Artazcoz, L., & Rueda, S. (2007). Social inequalities in health among the elderly: A challenge for public health research. Journal of Epidemiology and Community Health, 61(6), 466–467. https://doi.org/10.1136/jech.2006.058081
Atkinson, J., Salmond, C., & Crampton, P. (2014). NZDep2013 Index of Deprivation. https://www.otago.ac.nz/wellington/otago069936.pdf
Blakely, T., Fawcett, J., Hunt, D., & Wilson, N. (2006). What is the contribution of smoking and socioeconomic position to ethnic inequalities in mortality in New Zealand? The Lancet, 368(9529), 44–52. https://doi.org/10.1016/S0140-6736(06)68813-2
Bosworth, B. (2018). Increasing disparities in mortality by socioeconomic status. Annual Review of Public Health, 39(1), 237–251. https://doi.org/10.1146/annurev-publhealth-040617-014615
Boven, N., Exeter, D., Sporle, A., & Shackleton, N. (2019). The implications of different ethnicity categorisation methods for understanding outcomes and developing policy in New Zealand. Kōtuitui New Zealand Journal of Social Sciences Online, 15(1), 123–139. https://doi.org/10.1080/1177083x.2019.1657912
Cai, L., Wang, X.-M., Fan, L.-M., Cui, W.-L., & Golden, A. R. (2019). Socioeconomic disparities in prevalence and behaviors of smoking in rural Southwest China. BMC Public Health, 19(1), 1117. https://doi.org/10.1186/s12889-019-7455-0
Carstairs, V., & Morris, R. (1989). Deprivation: Explaining differences in mortality between Scotland and England and Wales. British Medical Journal, 299, 886–889.
Castles, F. G. (1998). The really big trade-off: Home ownership and the welfare state in the new world and the old. Acta Politica, 33, 5–19.
Chandola, T., & Jenkinson, C. (2000). The new UK national statistics socio-economic classification (NS-SEC); Investigating social class differences in self-reported health status. Journal of Public Health, 22(2), 182–190. https://doi.org/10.1093/pubmed/22.2.182
Connolly, S., O’Reilly, D., & Rosato, M. (2010). House value as an indicator of cumulative wealth is strongly related to morbidity and mortality risk in older people: A census-based cross-sectional and longitudinal study. International Journal of Epidemiology, 39(2), 383–391. https://doi.org/10.1093/ije/dyp356
Coombes, M., Raybould, S., Wong, C., & Openshaw, S. (1995). Towards an index of deprivation: A review of approaches.
Cronin, H., Oegan, C., & Kenny, R. A. (2011). Physical and behavioural health of older Irish adults. In A. Barrett, G. Savva, V. Timonen, & R. A. Kenny (Eds.), Fifty Plus in Ireland 2011 (pp. 73–154). The Irish Logitudinal Study on Ageing 2011.
Dalstra, J. A. A., Kunst, A. E., & Mackenbach, J. P. (2006). A comparative appraisal of the relationship of education, income and housing tenure with less than good health among the elderly in Europe. Social Science & Medicine, 62(8), 2046–2060. https://doi.org/10.1016/j.socscimed.2005.09.001
Darin-Mattsson, A., Fors, S., & Kareholt, I. (2017). Different indicators of socioeconomic status and their relative importance as determinants of health in old age. International Journal for Equity in Health, 16(1), 173. https://doi.org/10.1186/s12939-017-0670-3
Davey, J. A. (1996). Making use of home equity: Comparisons between Britain and New Zealand. Social Policy Journal of New Zealand, 7, 128–142.
Delfani, N., De Deken, J., & Dewilde, C. (2015). Poor because of low pensions or expensive housing? The combined impact of pension and housing systems on poverty among the elderly. International Journal of Housing Policy, 15(3), 260–284. https://doi.org/10.1080/14616718.2015.1004880
Dewilde, C., & Raeymaeckers, P. (2008). The trade-off between home-ownership and pensions: Individual and institutional determinants of old-age poverty. Ageing and Society, 28(6), 805–830. https://doi.org/10.1017/s0144686x08007277
Dibben, C., & Popham, F. (2012). Are health inequalities evident at all ages? An ecological study of English mortality records. European Journal of Public Health, 23(1), 39–45. https://doi.org/10.1093/eurpub/cks019
Exeter, D. J., Browne, M., Robinson-Chen, T., Colbert, J., Kerse, N., & Lee, A. (2022). The Older Persons’ Index of Multiple Deprivation: Measuring the deprivation circumstances of older populations in Aotearoa New Zealand. Health & Place, 76, 102850. https://doi.org/10.1016/j.healthplace.2022.102850
Exeter, D. J., Feng, Z., Flowerdew, R., & Boyle, P. J. (2005). Shrinking areas and mortality: An artefact of deprivation effects? Journal of Epidemiology and Community Health, 59, 924–926.
Exeter, D. J., Zhao, J., Crengle, S., Lee, A., & Browne, M. (2017). The New Zealand Indices of Multiple Deprivation (IMD): A new suite of indicators for social and health research in Aotearoa. New Zealand. PLoS One, 12(8), e0181260. https://doi.org/10.1371/journal.pone.0181260
Fahy, K. M., Lee, A., & Milne, B. J. (2017). New Zealand socio-econoic index 2013. https://www.stats.govt.nz/assets/Uploads/Retirement-of-archive-website-project-files/Research/New-Zealand-socio-economic-index-2013/New-Zealand-socio-economic-index-2013.pdf
Galobardes, B., Shaw, M., Lawlor, D. A., Lynch, J. W., & Davey Smith, G. (2006). Indicators of socioeconomic position (part 1). Journal of Epidemiology and Community Health, 60(1), 7–12. https://doi.org/10.1136/jech.2004.023531
Goodyear, R. (2017). A place to call home? Declining homeownership rates for Māori and Pacific peoples in New Zealand. New Zealand Population Review, 43, 3–34.
Grundy, E., & Holt, G. (2001). The socioeconomic status of older adults: How should we measure it in studies of health inequalities? Journal of Epidemiology and Community Health, 55(12), 895–904. https://doi.org/10.1136/jech.55.12.895
Hancock, R. (1998). Housing wealth, income and financial wealth of older people in Britain. Ageing & Society, 18(1), 5–33. https://doi.org/10.1017/S0144686X97006685
Hill, S. E., Blakely, T. A., Fawcett, J. M., & Howden-Chapman, P. (2005). Could mainstream anti-smoking programs increase inequalities in tobacco use? New Zealand data from 1981–96. Australian and New Zealand Journal of Public Health, 29(3), 279–284. https://doi.org/10.1111/j.1467-842x.2005.tb00769.x
Hiscock, R., Bauld, L., Amos, A., & Platt, S. (2012). Smoking and socioeconomic status in England: The rise of the never smoker and the disadvantaged smoker. Journal of Public Health (oxford, England), 34(3), 390–396. https://doi.org/10.1093/pubmed/fds012
House, J. S., Kessler, R. C., & Herzog, A. R. (1990). Age, socioeconomic status, and health. The Milbank Quarterly, 68(3), 383–411. https://doi.org/10.2307/3350111
Jones, I. R., & Higgs, P. (2013). Class and health inequalities in later life. In Social Class in Later Life (pp. 113–132). Policy Press
Kaplan, G. A., Seeman, T. E., Cohen, R. D., Knudsen, L. P., & Guralnik, J. (1987). Mortality among the elderly in the Alameda County Study: Behavioral and demographic risk factors. American Journal of Public Health, 77(3), 307–312. https://doi.org/10.2105/ajph.77.3.307
Krieger, N., Williams, D. R., & Moss, N. E. (1997). Measuring social class in US public health research: Concepts, methodologies, and guidelines. Annual Review of Public Health, 18, 341–378. https://doi.org/10.1146/annurev.publhealth.18.1.341
Lopes, A. (2011). Ageing and social class: Towards a dynamic approach to class inequalities in old age. In M. Sargeant (Ed.), Age discrimination and diversity: Multiple discrimination from an age perspective (pp. 89–110). Cambridge University Press.
Lotoala, F., Breheny, M., Alpass, F., & Henricksen, A. (2014). Health and wellbeing of older Pacific Peoples in New Zealand. N Z Med J, 127(1407), 27–39.
Lugo, A., La Vecchia, C., Boccia, S., Murisic, B., & Gallus, S. (2013). Patterns of smoking prevalence among the elderly in Europe. International Journal of Environmental Research and Public Health, 10(9), 4418–4431. https://doi.org/10.3390/ijerph10094418
Lynch, J., & Kaplan, G. (2000). Socioeconomic position. In L. F. Berkman & I. Kawachi (Eds.), Social epidemiology (pp. 13–35). Oxford University Press.
McMunn, A., Breeze, E., Goodman, A., Nazroo, J., & Oldfield, Z. (2006). Social determinants of health in older age. Social Determinants of Health, 2, 267–296.
Milne, B. J. (2012). New Zealand Socio-economic Index 2006 (NZSEI-06): An introduction for social science researchers. New Zealand Sociology, 27(2), 117–127.
Ministry of Social Development. (2016). Life expectancy at birth – the social report. http://socialreport.msd.govt.nz/health/life-expectancy-at-birth.html
Ministry of Health. (2016). DHB spending on services for older people. http://www.health.govt.nz/nz-health-statistics/health-statistics-and-data-sets/older-peoples-health-data-and-stats/dhb-spending-services-older-people
Morris, R., & Carstairs, V. (1991). Which deprivation? A comparison of selected deprivation indexes. Journal of Public Health, 13(4), 318–326. https://doi.org/10.1093/oxfordjournals.pubmed.a042650
Noble, M., Wright, G., Dibben, C., Smith, G., McLennan, D., Anttila, C., Barnes, H., Mokhtar, C., Noble, S., & Gardner, J. (2004). The English indices of deprivation 2004 (p. 180). Office of the Deputy Prime Minister.
Noble, M., Wright, G., Smith, G., & Dibben, C. (2006). Measuring multiple deprivation at the small-area level. Environment and Planning A: Economy and Space, 38(1), 169–185. https://doi.org/10.1068/a37168
Norman, P., Berrie, L., & Exeter, D. (2019). Calculating a deprivation index using census data. Australian Population Studies, 3(1), 30–39.
Ong, R. (2008). Unlocking housing equity through reverse mortgages: The case of elderly homeowners in Australia. European Journal of Housing Policy, 8(1), 61–79. https://doi.org/10.1080/14616710701817166
Pampalon, R., Hamel, D., & Gamache, P. (2008). Recent changes in the geography of social disparities in premature mortality in Québec. Social Science & Medicine, 67(8), 1269–1281. https://doi.org/10.1016/j.socscimed.2008.06.010
Pampalon, R., Hamel, D., Gamache, P., Philibert, M. D., Raymond, G., & Simpson, A. (2012). An area-based material and social deprivation index for public health in Quebec and Canada. Canadian Journal of Public Health, 103(8 Suppl 2), S17–S22.
Perry, B. (2010). The material wellbeing of older New Zeaalnders: Background paper for the Retirement Commissioner's 2010 review. https://assets.retirement.govt.nz/public/Uploads/2016-Review-Of-Retirement-Income-Policies/Vulnerable-Groups/Heavy-Stuff/cf8b60512d/417-Vulnerable-Groups-material-wellbeing-2010-review.pdf
Phillimore, P., Beattie, A., & Townsend, P. (1994). Widening inequality of health in northern England, 1981–91. BMJ, 308, 1125–1128.
Robert, S., & House, J. S. (1996). SES differentials in health by age and alternative indicators of SES. Journal of Aging and Health, 8(3), 359–388. https://doi.org/10.1177/089826439600800304
Salmond, C., Crampton, P., King, P., & Waldegrave, C. (2006). NZiDep: A New Zealand index of socioeconomic deprivation for individuals. Social Science and Medicine, 62(6), 1474–1485. https://doi.org/10.1016/j.socscimed.2005.08.008
Schederecker, F., Kurz, C., Fairburn, J., & Maier, W. (2019). Do alternative weighting approaches for an Index of Multiple Deprivation change the association with mortality? A sensitivity analysis from Germany. BMJ Open, 9(8), e028553. https://doi.org/10.1136/bmjopen-2018-028553
Shohaimi, S., Luben, R., Wareham, N., Day, N., Bingham, S., Welch, A., Oakes, S., & Khaw, K. T. (2003). Residential area deprivation predicts smoking habit independently of individual educational level and occupational social class. A cross sectional study in the Norfolk cohort of the European Investigation into Cancer (EPIC-Norfolk). Jornal of Epidemiology & Community Health, 57(4), 270–276. https://doi.org/10.1136/jech.57.4.270
Smith, S. J., Easterlow, D., Munro, M., & Turner, K. M. (2003). Housing as health capital: How health trajectories and housing paths are linked. Journal of Social Issues, 59(3), 501–525. https://doi.org/10.1111/1540-4560.00075
Stats NZ Tatauranga Aotearoa. (2015). New Zealand period life tables: 2012–14. https://www.stats.govt.nz/information-releases/new-zealand-period-life-tables-201214
Stats NZ Tatauranga Aotearoa. (2021). Data in the IDI: June 2021. https://www.stats.govt.nz/assets/Uploads/Retirement-of-archive-website-project-files/Sitetree-pages/Data-in-the-IDI/Data-in-the-IDI-June-2021.pdf
Stephens, C. (2016). From success to capability for healthy ageing: Shifting the lens to include all older people. Critical Public Health, 27(4), 490–498. https://doi.org/10.1080/09581596.2016.1192583
Steptoe, A., & Zaninotto, P. (2020). Lower socioeconomic status and the acceleration of aging: An outcome-wide analysis. Proceedings of the National Academy of Sciences U S A, 117(26), 14911–14917. https://doi.org/10.1073/pnas.1915741117
Teh, R., Kerse, N., Kepa, M., Doughty, R. N., Moyes, S., Wiles, J., Wham, C., Hayman, K., Wilkinson, T., Connolly, M., Mace, C., & Dyall, L. (2014). Self-rated health, health-related behaviours and medical conditions of Maori and non-Maori in advanced age: LiLACS NZ. The New Zealand Medical Journal, 127(1397), 13–29.
Townsend, P. (1987). Deprivation. Journal of Social Policy, 16(2), 125–146. https://doi.org/10.1017/S0047279400020341
Watson, V., Dibben, C., Cox, M., Atherton, I., Sutton, M., & Ryan, M. (2019). Testing the expert based weights used in the UK’s index of multiple deprivation (IMD) against three preference-based methods. Social Indicators Research, 144(3), 1055–1074. https://doi.org/10.1007/s11205-018-02054-z
Wilson, N., Blakely, T., & Tobias, M. (2006). What potential has tobacco control for reducing health inequalities? The New Zealand situation. International Journal for Equity in Health, 5(1), 14. https://doi.org/10.1186/1475-9276-5-14
Windle, G. S., Burholt, V., & Edwards, R. T. (2006). Housing related difficulties, housing tenure and variations in health status: Evidence from older people in Wales. Health & Place, 12(3), 267–278. https://doi.org/10.1016/j.healthplace.2004.08.010
Yates, J., & Bradbury, B. (2010). Home ownership as a (crumbling) fourth pillar of social insurance in Australia. Journal of Housing and the Built Environment, 25(2), 193–211. https://doi.org/10.1007/s10901-010-9187-4
Acknowledgements
OH received a Masters Thesis scholarship supported by the Marsden Fund Council from Government funding, managed by Royal Society Te Apārangi to conduct this research.
Statistics NZ disclaimer
Access to the data used in this study was provided by Stats NZ under conditions designed to give effect to the security and confidentiality provisions of the Statistics Act 1975. The results presented in this study are the work of the author, not Stats NZ or individual data suppliers. These results are not official statistics. They have been created for research purposes from the [Integrated Data Infrastructure (IDI) which is carefully managed by Stats NZ. For more information about the IDI please visit https://www.stats.govt.nz/integrated-data/.
Funding
Open Access funding enabled and organized by CAUL and its Member Institutions. This research was part of a wider study supported by the Marsden Fund Council from Government funding, managed by Royal Society Te Apārangi (ref 15-UOA-222).
Author information
Authors and Affiliations
Contributions
DE, OH and NS designed the work, acquired and analysed the data for the work, and all authors contributed to the interpretation of the data for the work. DE and JC drafted the work, and all authors revised it critically.
Corresponding author
Ethics declarations
Conflict of interest
None.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Exeter, D.J., Healey, O., Colbert, J. et al. Developing SEP65: A Census-Derived Index of Socio-Economic Position Specifically for the Older Population in Aotearoa New Zealand. Soc Indic Res 169, 973–991 (2023). https://doi.org/10.1007/s11205-023-03190-x
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11205-023-03190-x