Stats NZ has a new website.

For new releases go to

www.stats.govt.nz

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Chapter 2: National trends in longevity and mortality

This chapter presents a summary of mortality and longevity trends for the total population, Māori population, and non-Māori population of New Zealand. Official complete period life tables for New Zealand were first produced for the period 1880–92, although these were for the non-Māori population only. For periods from 1950–52, complete period life tables have been prepared for the total New Zealand and Māori populations, in addition to life tables for the non-Māori population. Non-Māori life tables continue to be produced because they provide an important comparison with Māori life tables. However, it is important to note that changes to ethnic concepts, deaths measures, and population measures affect the historical comparability of these life tables, particularly those relating to the Māori population (see chapter 1, including table 1.01).

Age distribution of deaths

There were 83,801 deaths of New Zealand residents registered during the 2005–07 period, comprising 41,630 male deaths and 42,171 female deaths. Historically, more males are born than females, and females live longer, so male deaths have tended to outnumber female deaths. However, longer female life expectancy has resulted in more females surviving to older ages and closing of the gap between male and female life expectancies resulted in more female than male deaths in the 2005–07 period.

The median age of the New Zealand population is rising. As a result, the number of deaths at older ages is increasing. More than three-quarters of these deaths in 2005–07 were of people aged 65 years and over (figure 2.01). The proportion of male deaths occurring at 65 years and over increased from 57 percent in 1950–52, to 60 percent in 1975–77, and to 73 percent in 2005–07, while the percentage of female deaths occurring at 65 years and over increased from 62 to 72 percent, and then to 82 percent over the same period.

Children (those under 15 years) accounted for less than 2 percent of deaths in 2005–07, whereas they accounted for 11 percent of deaths in 1950–52 and 5 percent of deaths in 1975–77. Infants (under one year old) account for about two-thirds of the deaths of children. Those aged 15–29 years accounted for 2 percent of deaths in 2005–07, with 72 percent of these being male.

Figure 2.01

Figure, Cumulative deaths by age and sex.

New Zealand life expectancy at birth

Based on period life tables, life expectancy at birth was 78.0 years for males and 82.2 years for females in 2005–07 (table 2.01 and figure 2.02). Male life expectancy increased by 1.7 years, from 76.3 years in 2000–02, while female life expectancy increased by 1.0 year, from 81.1 years. Life expectancies have increased by almost 11 years since 1950–52, up from 67.2 years for males and 71.3 years for females.
For most periods before 1975–77, increases in male life expectancy lagged behind increases in female life expectancy. Male life expectancy increased by 1.8 years between 1950–52 and 1975–77 compared with 4.2 years for females. However, within this 25-year period, male life expectancy at birth decreased slightly between 1960–62 and 1965–67 (as a result of an increase in mortality rates from heart disease, cancer, and motor vehicle accidents). Since 1975–77, male life expectancy has increased more than female life expectancy. Male life expectancy at birth increased by 9.0 years during the 30-year period 1975–77 to 2005–07, and by 6.7 years for females. The five-yearly increase in male life expectancy (1.9 years) between 1995–97 and 2000–02 was the highest recorded since 1950–52.

Figure 2.02

Figure, Life expectancy at birth.

Figure 2.03

Graph, Five-yearly change in life expectancy at birth.

Cohort life tables

The period life tables show the mortality and survival experience of the population, based on people dying in the specific period (for example, 2005–07). The life expectancies from period life tables assume that people experience the observed mortality rates of the given period throughout their lives. In reality, death rates do not remain constant. Hence, if death rates continue to decrease, people born during 2005–07 will experience greater longevity than implied by the 2005–07 period life tables.

Statistics NZ recently developed cohort life tables covering the New Zealand population born in each year from 1876. The cohort mortality series tracks the birth cohorts (people born in each year) over their entire lifetime, by following the deaths of each cohort at each age.

The cohort life tables indicate that life expectancy at birth increased between the 1876 and 1933 birth cohorts, from 50.4 years to 70.3 years for males, and from 54.0 years to 75.4 years for females. Both the level and rate of change in life expectancy at birth are higher than implied by the period life tables, because of the progressive decline in mortality with successive birth cohorts.

Life expectancy is the average length of life of a group of people from a given age. The death of the last survivor of a birth cohort is therefore needed before life expectancy (at any age) can be calculated. Some remaining survival and mortality experience has been projected at ages above 74 years to complete the life tables for birth cohorts up to 1933. For cohorts born after 1933, other life table measures, such as death rates at different ages and proportions dying by different ages, are still available. Tables derived from cohort life tables are included in appendix 3. More information on the cohort life tables is available from the Statistics NZ website: http://www.stats.govt.nz/browse_for_stats/health/life_expectancy/cohort-life-tables.aspx.

The cohort life tables indicate that males and females born in the early 1930s have lived for about 70 and 75 years on average, respectively. By comparison, it was not until the mid-1970s that period life tables indicated similar life expectancies at birth. The period life tables measure the life expectancy of the population at a moment in time, based on the most recently available mortality trends, while cohort life tables provide an insight into generational changes in mortality. 

Figure 2.04

Graph, Life expectancy at birth from cohort and period life tables.

Māori and non-Māori life expectancy at birth

The 2005–07 life tables indicate that a newborn Māori boy can expect to live 70.4 years and a newborn Māori girl 75.1 years (table 2.01). This is an increase of 1.4 years for males and 1.9 years for females over the 2000–02 figures of 69.0 years for males and 73.2 years for females.

Over the decade to 2007, Māori life expectancy has increased by 3.8 years for both males and females, up from 66.6 and 71.3 years, respectively, in 1995–1997.

For non-Māori in 2005–07, a newborn boy can expect to live 79.0 years and a newborn girl 83.0 years. This is an increase of 1.8 years for males and 1.0 year for females over the 2000–02 figures of 77.2 years for males and 81.9 years for females.

Māori and non-Māori life expectancy figures for 1995–97 onwards are not directly comparable with figures for 1990–92 and earlier. This is mainly because of the impact of the new ethnicity question on the birth and death registration forms, which was implemented in September 1995. See chapter 1 for more details on the new forms.

Table 2.01

Table, Life expectancy at selected ages.

Māori and non-Māori differentials in life expectancy

As a result of differences in death rates, life expectancy at birth for non-Māori exceeded that of Māori by 8.6 years for males and by 7.9 years for females in 2005–07. For males, three-quarters of this difference is due to higher Māori death rates at ages 40–79 years. For females, three-quarters of this difference is due to higher Māori death rates at ages 50–84 years.

The Māori/non-Māori differential partly reflects different rates of diabetes and smoking, as well as socio-economic differences. Cause-of-death statistics for 2005 show age-standardised death rates from diabetes were four times higher for male Māori than male non-Māori and five times higher for female Māori compared with female non-Māori (personal communication, New Zealand Health Information Service, 2009). The 2006 Census reported that 42 percent of Māori aged 15 years and over were regular smokers, compared with 18 percent of non-Māori.

The gap between Māori and non-Māori life expectancy has narrowed. In 1995–97, it stood at 9.1 years (average of male and female). By 2000–02 it had dropped to about 8.5 years. In 2005–07, the gap was 8.2 years.

However, the observed gap between Māori males and non-Māori males life expectancy has fluctuated, dropping from 8.8 years in 1995–97 to 8.2 years in 2000–02, but increasing to 8.6 years in 2005–07. In contrast, the gap between Māori females and non-Māori females life expectancy has dropped from 9.3 years in 1995–97, to 8.8 years in 2000–02, to 7.9 years in 2005–07.

Table 2.02

Table, Difference between Maori and non-Maori life expectancy.

Official life tables are unavailable for other ethnic populations, such as the broad Asian and Pacific populations, partly because of the relatively small size of these ethnic populations, and relatively few deaths. Nevertheless, other mortality measures, such as age-standardised death rates, indicate that Māori and Pacific death rates (9.9 deaths per 1,000 population and 8.8 per 1,000, respectively) are significantly higher than for the total population (5.8 per 1,000) for the 2005–07 period. The European or Other (5.4 per 1,000) and Asian (3.7 per 1,000) ethnic groups had much lower death rates than the Māori and Pacific ethnic groups.

It is important to note that standardised death rates can only be used to compare mortality trends for populations that have been standardised against the same population. Hence standardised death rates in this report (using the estimated resident population at 30 June 1996) are not comparable with rates published elsewhere that use a different standard population.

Table 2.03

Table, Standardised death rates.

Male-female difference in life expectancy

Before 1975–77, female life expectancy generally increased faster than male life expectancy. As a result, the difference between female and male life expectancy at birth increased from 4.1 years in 1950–52 to a peak of 6.4 years in 1975–77 (table 2.04). Since then, male life expectancy has been increasing faster and in 2005–07 the difference had narrowed to 4.1 years.

Table 2.04

Table, Difference between male and female life expectancy.

Closing of the gap between male and female life expectancy between 1995–97 and 2005–07 (from 5.3 to 4.1 years) is due to greater increases in non-Māori male life expectancy. Female life expectancy is higher than male life expectancy among both Māori and non-Māori. For the Māori population, male life expectancy increased more than female life expectancy between the periods 1995–97 and 2000–02. However, Māori male life expectancy increased by a smaller margin between 2000–02 and 2005–07. As a consequence, the Māori female longevity advantage was 4.7 years in both 1995–97 and 2005–07. Among non-Māori, females born in 2005–07 can expect to outlive males by 4.0 years, down from 4.8 years in 2000–02 and 5.2 years in 1995–97.

The closing gap between male and female life expectancy has also been observed overseas. In Australia, the gap has lessened from about seven years in the 1970s and early 1980s to just under five years in the mid-2000s (Australian Bureau of Statistics, 2008). Similarly, in the United Kingdom the difference between male and female life expectancy at birth has decreased from about six years to about four years over the last 25 years.

International comparison of life expectancy

In 2005–07, New Zealanders’ life expectancy at birth was 82.2 years for females and 78.0 years for males. This was slightly below the OECD median of 82.3 years for females, and slightly above the OECD median of 77.1 years for males (OECD, 2008). Of 30 OECD countries, New Zealand was ranked 17th for females and 9th for males (based on estimated 2006 figures for countries without 2006 figures). In 1960–61, New Zealand’s ranking was 9th for females and 7th= for males. Through the 1960s, 1970s, and 1980s, longevity improved faster in many other OECD countries than in New Zealand. Since the 1980s, faster-than-average gains in life expectancy in New Zealand, particularly for males, have improved New Zealand's relative position.

Amongst the OECD countries, Japanese women had the highest life expectancy (85.8 years in 2006). For men, Iceland had the highest life expectancy (79.4 years in 2006). Differences in life expectancies between OECD countries vary by up to 12 years for women and about 10 years for men.

Table 2.05

Table, Life expectancy at birth for OECD countries.

Age contribution to longevity differences

It is possible to determine the contribution that each age group has made to longevity differences, between periods or between populations (for example, Māori and non-Māori). This is because ages do not contribute equally to life expectancy at birth, with the youngest ages contributing relatively more. This comparison involves calculating and comparing ‘hypothetical or temporary’ life expectancies at each age. Hypothetical life expectancy is the average number of years that a group of people will live from age x to x + i years (where i is the age interval). For further methodological details on this process see Arriaga (1984).

Two-thirds of the gains in life expectancy between 2000–02 and 2005–07 were due to the reduction in death rates among older ages (60–84 years). Reduced death rates among males and females aged in their 50s also made a significant contribution. Among females, those aged 15–19 years experienced a small increase in death rates, meaning these age groups made a small negative contribution to the longevity gain. For males, all age groups made a positive contribution to the longevity gain.

Figure 2.05

Graph, Age contribution to increase in life expectancy at birth.

Over a longer 30-year period, 1975–77 to 2005–07, New Zealand life expectancy at birth increased by 9.0 years for males and by 6.7 years for females. There were reductions in death rates in all age groups. The main contribution was from the reduction in death rates among late working and retirement ages (50–79 years). Reduced death rates among infants, men and women in their 40s, and women aged 80 years and over were also significant.

Figure 2.06

Graph, Age contribution to increase in life expectancy at birth.

Among Māori, most of the gains in life expectancy between 2000–02 and 2005–07 were from the reduction in death rates among late working and retirement ages (50–79 years). There were also reductions in death rates for infants, men and women aged 80 years and over, and women in their 40s. However, small increases in death rates were observed at some other ages.

Figure 2.07

Graph, Age contribution to increase in life expectancy at birth.

Ethnic differentials

Māori experience higher death rates than non-Māori at all ages. Māori die at more than double the rate of non-Māori among males aged 2–6 and 28–72 years, and females aged 2–4, 12–17, and 25–75 years.

Figure 2.08

Graph, Ratio of Maori to non-Maori proporation dying within a year.

About one-half of the difference between Māori and non-Māori life expectancy in 2005–07 was due to lower non-Māori death rates at ages 55–74 years. For males, three-quarters of the difference in longevity is due to higher Māori death rates at ages 40–79 years. For females, three-quarters of the difference in longevity is due to higher Māori death rates at ages 50–84 years.

Figure 2.09

Graph, Age contribution to difference between Maori and non-Maori life expectancy at birth.

Death rates by age

Total population

The New Zealand life tables are comparable with those of other OECD countries. The total population life tables are characterised by relatively high death rates in the first year of life (about 5 deaths per 1,000 population). Death rates decrease to around 10 deaths per 100,000 at ages 7–11 years then increase to a hump at around 20 years, with markedly higher death rates for males than females (figure 2.10). Death rates then change little until the mid-30s for males and the late 20s for females, when they begin to increase gradually with age, reaching 1 death per 100 people for males aged in the early 60s and females aged in the late 60s. For males aged in the mid-80s and females aged in the late 80s, death rates have reached 1 death per 10 people.

Figure 2.10

Graph, Proportion dying within a year.

Females experience lower death rates than males at all ages. For the total population, males die at more than twice the rate of females at ages 18–30 years, and at more than three times that of females at ages 23–27 years. Among Māori, males die at more than twice the rate of females at ages 18–29 years.

Figure 2.11

Graph, Ratio of male to female proportion dying within a year.

In 2005–07, life expectancy at birth for females exceeded that of males by 4.1 years for the total New Zealand population, by 4.7 years for Māori, and by 4.0 years for non-Māori. For the total and non-Māori populations, two-thirds of the difference is due to higher male death rates at ages 55–89 years. For the Māori population, two-thirds of the difference is due to higher male death rates at ages 40–79 years.

Figure 2.12

Graph, Age contribution to difference between male and female life expectancy at birth.

Change between 1950–52 and 2005–07

Between 1950–52 and 2005–07, male and female death rates decreased at all ages. However, the decreases were not uniform across age or time (figures 2.13–2.15). Between 1950–52 and 1975–77, male death rates actually increased slightly at ages 17–21 years, and at some ages between 48 and 88 years. The increase in male death rates at around 20 years accentuated the 'accident hump' at age 15–29 years, so-called because accidents (especially motor vehicle accidents) were the leading cause of death at those ages. An accident hump was also evident among female death rates at around 20 years in 1975–77, although this was smaller than for males. This hump is still apparent in 2005–07 but is less pronounced for males and more pronounced for females than it was 30 years earlier.

Figure 2.13

Graph, Proportion dying within a year.

Figure 2.14

Graph, Proportion dying within a year.

Figure 2.15 highlights changes in death rates over time at selected ages. For infants (under 1 year of age), death rates have dropped steadily for both males and females. For females aged 60, 70, and 80 years, death rates have dropped steadily. In contrast, for males at ages 60, 70, and 80 years, significant decreases in death rates were not achieved until the 1980s. 

Figure 2.15

Graph, Proportion dying within a year.

Between 1950–52 and 1975–77, death rates for children (under 15 years of age) decreased, on average, by just under 2.0 percent per year (figure 2.16). Female death rates declined more than male death rates at almost all ages above 15 years. For females aged between 15–39 years, death rates decreased by an average of between 1.0 and 1.7 percent per year, and close to 1.0 percent for those aged between 40–95 years. For males, the largest percentage decreases occurred under 15 years of age and for men aged between 29–36 years, while little change was recorded between 40 and 90 years of age. Male death rates at ages 17–21 years increased.

Figure 2.16

Graph, Average decrease in death rates per year.

The pattern of change in age-specific death rates between 1975–77 and 2005–07 was broadly similar for males and females (figure 2.17). For both sexes, all ages experienced declines in death rates over this period. As for 1950–52 to 1975–77, the largest percentage decreases in death rates between 1975–77 and 2005–07 occurred in the youngest ages. Over this 30-year period, death rates declined by 1.8 to 2.4 percent per year under 13 years of age. The smallest percentage decreases for males were between 25–34 years, which conversely had high percentage decreases in death rates for the earlier period between 1950–52 and 1975–77. However, having shown little change between 1950–52 and 1975–77, male death rates for middle and older ages (40–85 years) decreased significantly over the later period. Female death rates between 1975–77 and 2005–07 were more consistent across age than for males, averaging about 1.5 percent per year.

Figure 2.17

Graph, Average decrease in death rates per year.

Māori and non-Māori

Māori experience higher death rates than non-Māori at all ages for both sexes (figure 2.18). Māori female death rates were higher than non-Māori males, except for the late teens and twenties (17–28 years). Māori death rates are up to three times higher than non-Māori, with the largest differences occurring around 40–60 years for males, and around 14, 30, and 60 years for females.

Figure 2.18

Graph, Proportion dying within a year by age and sex.

Infant death rates

The proportion of infants dying in the first year of life has fallen markedly over the last 50 years. For males, the infant death rates decreased from 31.1 deaths per 1,000 population in 1950–52, to 16.9 in 1975–77, and further to 5.6 in 2005–07 (table 2.06). Females experienced a similar improvement, from 25.4 deaths per 1,000 population in 1950–52, to 12.3 in 1975–77, and further to 4.5 in 2005–07. The improvement in infant mortality is due to reductions in both the endogenous (related to the birth process) and exogenous (related to external circumstances) components of mortality in the first year of life.

Table 2.06

Table, Infant death rate.

The proportion of Māori infants dying in the first year of life is 1.6 times higher than the non-Māori rate for males and 1.8 times higher for females. Male Māori infant death rates have dropped from 10.7 per 1,000 population in 1995–97 to 7.5 in 2005–07. Female Māori infant deaths rates dropped from 9.6 per 1,000 population to 6.5 over the same period.

The infant death rates are based on the probability of dying in the first year of life (symbolised by q0 in the life tables). These rates will differ slightly from infant mortality rates (infant deaths per 1,000 live births) published elsewhere. For details on the method used to calculate q0 see chapter 3.

Chance of survival

Based on the 2005–07 life tables, 95 percent of newborn baby boys and 97 percent of girls can expect to reach 50 years of age (figure 2.19). About 86 percent of newborn boys and 90 percent of newborn girls can expect to live to 65 years of age. After that, the chance of survival decreases rapidly, with a faster decline for males than for females. About half (55 percent) of newborn boys are expected to reach 80 years of age, compared with 68 percent of newborn girls. Girls (11 percent) are twice as likely as boys (6 percent) to survive to 95 years. The chance of reaching 100 years is about 1 in 110 for newborn boys and 1 in 50 for newborn girls.

Figure 2.19

Graph, Proportion surviving from birth to selected ages.

According to the 2005–07 life tables, 90 percent of newborn Māori baby boys and 94 percent of girls can expect to reach 50 years of age. As age increases, the decrease in the chance of survival is more pronounced for Māori than for non-Māori. More than half (58 percent) of newborn non-Māori boys are expected to reach age 80 years, compared with 31 percent of newborn Māori boys. For females, 70 percent of newborn non-Māori girls are expected to reach age 80 years, compared with 44 percent of newborn Māori girls.

Table 2.07

Table, Chance of survival from birth to selected ages.

Causes of death

Diseases of the circulatory system (for example, heart disease, strokes) are the major cause of death in New Zealand. In the period 2005–06 (the latest two years for which cause of death data has been processed by the New Zealand Health Information Service), 38 percent of all deaths were due to this cause (figure 2.20). Neoplasms (mainly cancer) was the next biggest cause of death, accounting for 29 percent of deaths, followed by diseases of the respiratory system (8 percent), and external causes of death such as accidents and violence (6 percent). Diabetes, diseases of the nervous and digestive systems, mental disorders, and other causes made up the remaining 18 percent.

Figure 2.20

Graph, Deaths by cause.

Eighty-eight percent of deaths from circulatory diseases in 2005–06 were of people aged 65 years and over. Over the last 25 years, mortality rates from circulatory diseases have dropped by 51 percent for males aged 65 years and over, and 35 percent for females aged 65 years and over. For those aged 45–64 years the decreases have been greater, at 72 and 74 percent for males and females, respectively (table 2.08).

Mortality rates from cancer for men aged 65 years and over increased slightly between 1980–82 and 2000–02, but dropped in 2005–06. Mortality rates from cancer for women aged 65 and over have increased over the last 25 years – up 11 percent compared with 1980–82. However, mortality rates from cancer in the 65 years and over age group are about one-third higher in men than in women. Cancer is the leading cause of death for people aged 45–64 years, accounting for 46 percent of deaths in 2005–06.

For people aged 15–24 years, external causes accounted for about 79 percent of male deaths and 60 percent of female deaths in 2005–06. Male deaths outnumbered female deaths from this cause by more than 3 to 1 among those aged 15–24 years. Mortality rates from external causes have dropped by 38 percent for males and 35 percent for females aged 15–24 years since 1980–82.

Figure 2.21

Graph, Mortality rates of 65+ age group.

Detailed statistical and analytical information on the underlying causes of deaths registered in New Zealand are available from the New Zealand Health Information Service:

www.nzhis.govt.nz/moh.nsf/pagesns/528

www.nzhis.govt.nz/moh.nsf/pagesns/530

Table 2.08

Table, Morality rates by cause of death, age and sex.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Top
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+