Independent Life Expectancy in New Zealand 2013

Independent Life Expectancy in New Zealand

2013

Acknowledgements

This report was written by Vladimir Stevanovic with support from Michelle Liu.

The authors acknowledge valuable input from peer reviewers of this report:

Jackie Fawcett, Denise Hutana, Martin Tobias and Li-Chia Yeh (Ministry of Health)
Phillipa O’Brien, Kim Dunstan and Shari Mason (Statistics New Zealand)
Natalie Jackson (Massey University)
Peter Davis (University of Auckland).

Citation: Ministry of Health. 2015. Independent Life Expectancy in New Zealand 2013.Wellington: Ministry of Health.

Published in July 2015
by theMinistry of Health
PO Box 5013, Wellington 6145, New Zealand

ISBN978-0-478-44843-6(online)
HP 6219

This document is available at

Contents

Executive summary

Introduction

Health expectancy as a summary measure of population health

Two approaches to estimating health expectancy

How health expectancy and ILE are being applied in New Zealand

Overview of this report

How independent life expectancy has been estimated

Sources of data for this report

Methods of estimating ILE

Analysis by gender and ethnicity

Independent life expectancy at birth in 2013

ILE by gender

ILE by ethnicity

Analysis across gender and ethnic groups

Limits to data reliability

Independent life expectancy at age 65 in 2013

ILE by gender

ILE by ethnicity

Analysis across gender and ethnic groups

Limits to data reliability

Trends in independent life expectancy 1996–2013

Trends in ILE by gender

Trends in ILE by ethnicity

Analysis across gender and ethnic groups

Limits to data reliability and comparability

How well does ILE perform as an indicator?

Strengths

Limitations

Overall assessment

Conclusions

References

Appendix: Methods used to produce the findings in this report

List of Tables

Table 1:Independent life expectancy at birth, by gender and ethnicity, 2013

Table 2:Independent life expectancy at age 65, by gender and ethnicity, 2013

Table 3:Independent life expectancy at selected ages in 2013

Table 4:Independent life expectancy at birth, by gender, 1996–2013

Table 5:Change in independent life expectancy at birth, by gender, between 1996 and 2013

Table 6:Independent life expectancy at birth, by ethnicity, 1996–2013

Table 7:Change in independent life expectancy at birth, by ethnicity, between 1996 and 2013

List of Figures

Figure 1:A model of different levels of health state expectancy measures

Figure 2:Independent life expectancy at birth, New Zealand males and females, 2013

Figure 3:Independent life expectancy at birth, Māori males and females, 2013

Figure 4:Independent life expectancy at birth, non-Māori males and females, 2013

Figure 5:Independent life expectancy at age 65, New Zealand males and females, 2013

Figure 6:Independent life expectancy at age 65, New Zealand males and females, 2013

Figure 7:Independent life expectancy at age 65, non-Māori males and females, 2013

Figure 8:Life expectancy, independent life expectancy and life expectancy with dependency, New Zealand males, 2013

Figure 9:Life expectancy, independent life expectancy and life expectancy with dependency, New Zealand females, 2013

Figure 10:Independent life expectancy at birth, New Zealand males and females,
1996–2013

Figure 11:Independent life expectancy at birth, Māori males, 1996–2013

Figure 12:Independent life expectancy at birth, Māori females, 1996–2013

Figure 13:Independent life expectancy at birth, non-Māori males, 1996–2013

Figure 14:Independent life expectancy at birth, non-Māori females, 1996–2013

Executive summary

ILE is defined as the average number of years that a person can be expected to live independently, either free of any disability (functional limitation) or with functional limitations that they can manage without assistance. It is estimated using the Sullivan prevalence methodand the following Statistics New Zealanddata:

disability data from the disability surveys carried out after eachCensus of Population and Dwellings
mortality data from the complete period life tables.

Independent life expectancy in 2013

The following findings have been concluded from the results of the 2013 Disability Survey andthe 2012–2014 period life tables.

Males have shorter lives than females but spend a higher proportion of their lives in good health. A New Zealand male born in 2013 can expect to live 3.7 years less than a New Zealand female born in the same year(79.5 yearsand 83.2 years respectively). Hecan expect to live 65.2 years independently (82.0% of his life), while a female can expect to live 66.5 years independently (79.9% of her life). Conversely, females are expected to live longer with disability for which they need support than males (16.7 years and 14.3 years respectively).
Māori males have the lowest ILE and live longest with dependency.A Māori male born in 2013 can expect to live 54.3 years, or 74.4% of his life, independently, which is the lowest ILE estimate among all groups. He also can expect to live longest with functional limitations for which he needs support (18.7 years). AMāori female born in 2013 can expect to live60.4 years independently, which is 78.4% of her life.
At the age of 65 years, New Zealanders can expect to live half of their remaining lives independently (54.1% for males and 49.5% for females).Non-Māori females aged 65 in 2013, including Pacific and Asian peoples, can expect to live longest without disability requiring assistance (another 10.7 years). A 65-year-old Māori male has, on average, the shortest remaining time of living independently (5.5 years) and the highest proportion of remaining years lived with dependency (64.3%).

Trends in independent life expectancy 1996–2013

The following findings have been concluded from the results of disability surveys and period life tables from 1996 onwards to gauge trends in independent life expectancy over 17 years.

Overall, people are living longer, but spending more time in dependent health states (morbidity). Between 1996 and 2013, independent life expectancy at birth increased for all groups apart from Māori males.In relative terms, however, all groups experienced a decrease: that is, the proportion of years they lived independently relative to their life expectancy was lower than it was for their counterparts in 1996.
Independent life expectancy for Māori males decreased in both absolute and relative terms, which indicates an absolute expansion of morbidity.
ILE for Māori females increased in absolute terms and decreased in relative terms. Among those with functional limitations, more required a medium level of support and fewer required a high level of support over time.

Conclusions

The findings on ILE are aligned with the results from the New Zealand Burden of Diseases, Injuries and Risk Factors Study (Ministry of Health 2013) and illustrate the value of ILE for health policy.

Today people in New Zealand live longer in good health, but spend proportionally more time living with dependency than before. Part of the reason for this trend is the ageing population structure because, with a higher proportion of older people in the population, the proportion of people with functional limitations is likely to be correspondingly higher.

As ILE is not keeping pace with life expectancy, these findings indicate an increasing need in the future for the treatment of long-term conditions with relatively low fatality. They also highlight substantial inequality between Māori and non-Māori groups. These issues are particularly important at a time when the health system is facing challenges associated with population ageing and increased expenditure on long-term care.

Independent Life Expectancy in New Zealand 20131

Introduction

Independent life expectancy (ILE) is one measure of health expectancy that both the Ministry of Health and the Ministry of Social Development have used as a‘headline’ or ‘peak’ health indicator. Although it is the main focus of this report, ILE is part of a broader set of indicators of health expectancy. This section presents the general context in which ILE fits. After defining health expectancy, it describes the two main ways of estimating it and outlines how health expectancy generally and ILE specifically have been applied in New Zealand. Finally, this section gives a brief overview of the content of this report.

Health expectancy as a summary measure of population health

Health expectancy is a summary measure of population health in terms of both quality and quantity (Murray et al 2002). It is a single figure calculated from both non-fatal and fatal health outcomes. As life expectancy (LE) is increasing, health expectancy indicators offer a way of assessing how many of theyears a population has gained in life expectancy are spent in good health and free from functional limitations.

The World Health Organization (WHO) has developed reliable survey instruments to measure non-fatal health states using the International Classification of Functioning, Disability and Health (ICF) (WHO 2001). The aim of the ICF, an international standard endorsed by WHO Member States, is to describe and measure health and disability. It defines different non-fatal health states based on people’s level of functioning across a range of health and health-related domains. Health expectancy is calculated by combining survey data ona population’s level of functioning with mortality data.

As an overall outcome measure, health expectancy is essential for assessing the performance of the health system. It also reflects the level of performance of other sectors whose actions contribute to population health outcomes. Therefore, policy makers and researchers across the social policy spectrum may find this indicator useful when reflecting on their own contribution to population health gain and the reduction of health inequalities (Ministry of Health and Statistics New Zealand 2009).

Two approaches to estimating health expectancy

The two main ways of estimating health expectancy involve:

health state expectancy measures – which are based on functional limitation thresholds
health-adjusted life expectancy – which is based on continuous weighting of non-fatal health states relative to full health.

Health state expectancy measures

Health state expectancy measurescategorise individuals (in terms of person years) into different health states based ontheir level of dependency, or functional limitation thresholds. Dependency in this context means the need for assistancewith everyday routines, which may come from another person or from a complex assistive device, and may be either intermittent or continuous. Statistics New Zealand (Statistics NZ) post-censal disability surveys classify respondents into three support need levels:

level 1 (low support need) –a person has a functional limitation but does not need assistance with any every day routines
level 2 (medium support need) –a person has a functional limitationand needs assistance but not on a daily basis
level 3 (high support need) – a person has a functional limitationand needs some types of assistance daily.

Building on this framework, a model of health state expectancy measures can be constructed (Figure1):

life expectancy (LE) – the average full life span
active life expectancy (ALE) – the average number of years it is expected people will live free of functional limitation needing daily assistance (ie, before reaching level 3)
independent life expectancy (ILE) – the average number of years it is expected people will live independently, with no functional limitation that needs assistance (ie, before reaching level 2)
life expectancy with dependency (LED) – the average number of years it is expected people will live with functional limitation that needs either daily or non-daily assistance (ie, at support levels 2 and 3)
disability-free life expectancy (DFLE) – the average number of years it is expected people will live free of any functional limitation (ie, before reaching level 1).

Figure 1: A model of different levels of health state expectancy measures

Health-adjusted life expectancy

Wherehealth state expectancies take a categorical approach with reference tofunctional limitation thresholds, health-adjusted life expectancy or healthy life expectancy (HLE) is based on continuous weighting of non-fatal health states relative to full health. Therefore, HLE is the number of years of full health that an average person can expect to live.

This method was used in the New Zealand Burden of Diseases Study (NZBD). To estimate average levels of health among survivors of disease and injury, researchers summed the NZBD’s estimates of years lived with disability (YLD) over all causes. YLD is estimated by multiplying the prevalence of the different health states by their disability weight, whichranges from 0 (full health) to 1 (equivalent to being dead). These estimates were then incorporated into the life tables to give an estimate of health-adjusted life expectancy (Ministry of Health 2013).

How health expectancy and ILE are being applied in New Zealand

The Ministry of Health recognises healthexpectancyas a key outcome indicator.An increasing number of other countries are also taking this approach, such as the United States of America (Molla et al 2001), the United Kingdom (Office for National Statistics 2004), Canada (Public Health Agency of Canada 2012) and Australia (Australian Institute of Health and Welfare 2014).

The Ministry of Health has been regularly reporting on health expectancy since 2003. It reports independent life expectancy as a ‘headline’ health indicator in both its Statement of Intent (Ministry of Health 2003a and later editions) and itsHealth and Independence Report (Ministry of Health 2003b and later editions). Independent life expectancy also features as the ‘peak’ health indicator in the Health chapter of The Social Report (Ministry of Social Development 2003 and later editions). In addition:

several research studies have been conducted on social variations and evolution of health expectancy in New Zealand (Davis et al 1999; Graham et al 2004)
the Ministry of Health used the health expectancy construct in its long-term forecasting of public health expenditure (Tobias et al 2004)
theMinistry of Health (2013) estimated health expectancy in the NZBDbased on the health-adjusted life expectancy measuresto estimate health loss.

Overview of this report

The next section summarises the sources and methods for the data presented in this report (with more detail available in the Appendix). The following sections then set out recent data on independent life expectancy in 2013 and longer-term trends in this indicator from 1996 to 2013. The final sections analyse the performance of ILE as an indicator and draw overall conclusions from the findings in this report.

How independent life expectancy has been estimated

The first health expectancy tables in New Zealand were constructed using disability prevalence rates from the 1996 Household Disability Survey and its companion Disability Survey of Residential Facilities (Tobias and Cheung 1999). These prevalence rates were based on a functional limitation concept of disability and corresponded closely with the ICF’s core health and health-related domains. The survey was repeated after the 2001, 2006 and 2013 Censuses of Population and Dwellings. In each case, the survey was designed (with specific attention to drift and fielding of household and institutional components) so that trends in health expectancy and differences between population subgroups and regions could be estimated (Ministry of Health and Statistics NZ 2008).

Sources of data for this report

All data for this report were supplied by Statistics NZ.

Post-censal disability surveys (1996, 2001, 2006 and 2013, as noted above) provided data on:

disability rates –more precisely described in the context of this report as functional limitation rates by level of support needed
the number of respondents, which were needed to calculate the standard error of ILE.

Abridged life tables for the total, Māori and non-Māori populations by sex came from the complete period life tables 1995–1997, 2000–2002, 2005–2007 and 2012–2014 (Statistics NZ 2015b).

Methods of estimating ILE

The ILE prevalence estimates by 10-year age group were smoothed using kernel smoothing (Wand and Jones 1994) to produce estimates by five-year age group. The smoothing was conducted using statistical software Stata version 12.1. ILE and the other health state expectancy measures were calculated by the Sullivan observed prevalence method (Sullivan 1971). The two main inputs were age-specific mortality obtained from Statistics NZ’s abridged life tables and smoothed disability prevalence rates by support need level and five-year age group. ILE and its standard error were calculated using the formulae embedded in the spreadsheet. (For a full explanation of these formulae, see Jagger et al 2006.)

See the Appendix for a more detailed description of the methods used to estimate ILE.

Analysis by gender and ethnicity

Separate estimates have been calculated for New Zealand males and females overall, Māori males and females and non-Māori males and females. In this report, ‘non-Māori’ refers to those respondents who identified with an ethnic group other than Māori. European/Pākehāare included in the non-Māori category. This category also includes Asian and Pacific peoples, for whom separate estimates could not be calculated accurately because the number of respondents was too small.

Independent life expectancy at birth in 2013

This section presents estimates of independent life expectancy for New Zealanders born in 2013. Estimates are presented separately by gender and ethnicity (Māori and non-Māori)before an overall analysis of the findings and limits to data reliability are discussed.

ILE by gender

The New Zealand life expectancy at birth for babies born between 2012 and 2014 was 79.5 years for males and 83.2 years for females, which is a difference of 3.7 years.

Applying the2013 Disability Survey results to 2012–2014 period life tables, a male New Zealander born in 2013 can expect to live for:

an average of 65.2 years independently, which is 82.0% of his life
another 14.3 years with some level of disability requiring support (life expectancy with dependency or LED). For 10.2 years of this time, he will needsome assistance but not every day and for the other 4.1 years he will need daily assistance due to his functional limitations.

Also based on the2013 Disability Survey results and 2012–2014 period life tables, a female New Zealander born in 2013 can expect to live for:

an average of 66.5 years independently, which is 79.9% of her life
another 16.7 years with functional limitations that requiresupport. During this time, she will need non-daily assistance for 10.7 years and daily assistance for 6.0 years.

At birth, therefore, females can expect to live independently for 1.3 years longer than males. They can also expect to live 2.4 years longer with disability requiring assistance than males (Figure2).