23

Canadian Studies in Population 43, no. 1–2 (2016): 23–47.

Epidemiologic Transition in Australia: The last hundred years

Heather Booth1 Leonie Tickle Jiaying Zhao

Abstract

Mortality change in Australia since 1907 is analysed in the light of Epidemiologic Transition theory. Australia began the twentieth century in the second age of the Epidemiologic Transition, the Age of Receding Pandemics. Australia probably moved to the third, the Age of Degenerative and Man-Made Diseases before 1946, which is slightly in advance of most Western countries. Transition to the fourth, the Age of Delayed Degenerative Diseases, is clearly marked by a downturn, in about 1970, in circulatory disease mortality, concurrent with other Western countries.

Keywords: mortality, trends, decomposition, life expectancy, differentials. Australia.

Résumé

La théorie de la transition épidémiologique sert de base pour une analyse des changements de mortalité en Australie depuis 1907. Au début du XXe siècle, l’Australie était dans la deuxième phase de la transition épidémiologique, celle du recul des pandémies. Néanmoins, l’Australie entrait probablement avant 1946 dans la troisième phase, celle des maladies dégénératives, ce qui est légèrement en avance sur la plupart des pays occidentaux. La transition vers la quatrième phase, celle des maladies dégénératives retardées, est clairement marqué par un ralentissement depuis environ 1970 dans la mortalité par maladies circulatoires, en même temps que chez d’autres pays occidentaux.

Mots-clés : mortalité, tendances, décomposition, espérance de vie, écarts, Australie.

Introduction

Australia enjoys a life expectancy that is among the highest in the world. In 2011–13, life expect- ancy at birth among females was 84.3 years, and 80.1 years among males (ABS 2014a). The recently- released United Nations World Population Prospects 2015 (UNPD 2015) shows that for life expectancy at birth in 2010–15, Australian males rank eighth internationally and Australian females rank tenth. The top five ranked countries for males are Hong Kong, Iceland, Switzerland, Italy, and Israel, and for females they are Hong Kong, Japan, Singapore, Italy, and Spain.

1. Corresponding author: Prof. Heather Booth, School of Demography, The Australian National University, Canberra ACT 2601, Australia, e-mail: heather.booth@anu.edu.au; Prof. Leonie Tickle, Faculty of Business and Economics, Macquarie University, Sydney; and Dr. Jiaying Zhao, School of Demography, The Australian National University.

Canadian Studies in Population 43, no. 1–2 (Spring/Summer 2016): Special issue on Canada and Australia

24

This paper examines changing mortality in Australia since 1907, six years after the Commonwealth of Australia was created. The focus is at the national level, with some discussion of differentials. The theoretical framework of the analysis is the Epidemiologic Transition (Omran 1971; Olshansky and Ault 1986). As noted by de Looper (2015), the more recent Epidemiologic Transition in Australia has not been addressed as such, though studies of twentieth-century mortality decline there do exist (e.g., Taylor and Lewis 1998; Taylor et al. 1998; Booth 2003). This paper remedies the omission.

The paper is organized as follows. After a discussion of Epidemiologic Transition theory and a description of the data and methods employed, the paper examines trends in life expectancy at birth and at selected ages by sex. Cause of death, in conjunction with age, is then explored through a series of decompositions of temporal change in life expectancy over the course of the lengthy period con- sidered. The following section addresses age patterns of change and focuses on infant mortality, the adolescent and young adult mortality hump, and old age mortality. The penultimate section presents geographic, indigeneity and socio-economic mortality differentials, and the paper concludes with a discussion of the findings in relation to Epidemiologic Transition theory.

Epidemiologic Transition theory

The theory of Epidemiologic Transition (Omran 1971, 1983) describes health changes during the process of modernisation as a series of three successive stages of transition or ‘Ages’. The first is the ‘Age of Pestilence and Famine’, characterised by low and fluctuating life expectancy in the range 20–40 years. The second is the ‘Age of Receding Pandemics’ when life expectancy increases steadily from an average of about 30 years to 50 (Omran 1971) or 55 (Omran 1983) years, largely as a result of less frequent epidemics and the decline of infectious diseases; the underlying causes were primarily socio-economic, ‘augmented by the sanitary revolution in the late nineteenth century and by medical and public health progress in the twentieth century’ (Omran 1971, reprint p.753). The third ‘Age of Degenerative and Man-Made Diseases’ is characterised by a slow increase in life expectancy due to the balancing effects of the disappearance of infectious diseases and the rise of ‘degenerative and man-made’ or non-communicable diseases such as heart disease, stroke, cancers, and external causes. At the time of publication of the theory, the general consensus was that there was a limit to life expectancy which would soon be reached (see Meslé and Vallin 2011); for example, United Na- tions (1975) took this limit to be 75 years.

In response to renewed mortality decline from the 1970s, Olshansky and Ault (1986) proposed a fourth ‘Age of Delayed Degenerative Diseases’ characterised by the decline of cardiovascular and other non-communicable diseases at increasingly older ages, due to advances in medical technology and improved health programs. Rogers and Hackenberg (1987) also proposed a fourth ‘hubristic’ (or ‘hybristic’) stage giving prominence to the decline of social pathologies arising from individual behav- iour and lifestyle, which are driven by ‘hubris’ or notions of excessive self-confidence and invincibility. These two proposed extensions of Omran’s Epidemiologic Transition theory address different aspects of the same stage.2 The Epidemiologic Transition theory has been criticised by Robine (2001) and by Meslé and Vallin (2006), particularly in regard to the distinction between the third and fourth Ages.

Omran (1971) defined three models of Epidemiologic Transition, in recognition of the differing dates of onset and speeds of transition among countries. The Classical or Western model applies to the populations of Europe and North America. Compared with this, the Accelerated model involves

2. Proposals of fifth and sixth stages exist; these are not considered.

Booth et al.: Epidemiologic Transition in Australia – The last hundred years

25

a more rapid transition such as occurred in Japan. The Contemporary or Delayed model applies to the populations of developing countries.

While Epidemiologic Transition theory was developed to explain global patterns, it can be used in the study of mortality decline in individual countries (e.g., Caselli, Meslé and Vallin 2002; Lussier, Bourbeau and Choinière 2008). Several exceptions to the overall theory have been identified (Caselli, Meslé and Vallin 2002). A limitation is that Omran did not provide clear guidelines to determine when successive Ages begin and end (Mackenbach 1994). It has been argued that the approach is overly broad, and that there is a need to take greater account of how population subgroups experi- ence epidemiologic transitions differently (Gaylin and Kates 1997).

The early years of the Epidemiologic Transition in the settler3 population of Australia have been comprehensively documented by de Looper (2015) who notes that the Age of Pestilence and Famine was absent in Australia.4 Though there was no shortage of epidemics in the second half of the nineteenth century, famine was almost entirely absent, and life expectancy was always above the defining threshold of 40 years for transition to the second stage (Omran 1971). Thus, the second Age of Receding Pandemics characterises the start of the ‘truncated’ Epidemiologic Transition in Australia, confirmed by life expectancies in the 1860s of 45 years for males and 49 years for females. Further, de Looper (2015) concluded that, although there was rapid mortality decline in the period 1885 to 1903, there was no evidence of transition to the third Age of Degenerative and Man-Made Diseases because the major causes of death (infectious diseases, non-communicable diseases, and external causes) declined proportionately. Commencing in 1907, this analysis therefore begins in the second Age of Receding Pandemics.

Data and methods

Data for international trends and comparisons of life expectancies are from the Human Mortal- ity Database (HMD 2015). The analyses use five-year averages from 1920–24 to the present. For Aus- tralia, HMD covers 1921 to 2011, so that the first period is 1921–24.The countries for comparison are Canada, England and Wales, France, Japan, and the United States, selected on the basis of high income and either historical links and cultural similarities to Australia (Canada, England and Wales, United States) or recent leading-edge mortality experience (France, Japan). HMD data are also used for the examination of trends at specific ages.

Australian cause of death data are from the Australian Institute of Health and Welfare (AIHW) General Record of Incidence of Mortality (GRIM) books (AIHW 2015a), which contain mortality rates by five-year age groups for ages 0 to 84 and for the age 85+, from 1907 to 2012. Over this per- iod, the International Classification of Diseases underwent numerous revisions (WHO 1992), leading to inconsistencies in cause of death classification and discontinuities in time series of data. These potential problems have been largely mitigated in this analysis by considering only the major cause of death categories. The six major causes of death employed are infectious diseases, neoplasms, circulatory diseases, respiratory diseases, external causes, and ‘all other’ causes. Note that in the early part of the century,

3. The first British settlers arrived in Australia in 1788. Comprehensive mortality data have been compiled from 1856, when registration began, by de Looper (2015).

4. Smith (1980) and Gray (1985) suggest that the historic Indigenous population was stationary prior to settlement. Therefore, this population would not have been subject to the fluctuation defining the Age of Pestilence and Famine. Thus, neither the Indigenous nor settler population appears to have experienced Omran’s first Age.

Canadian Studies in Population 43, no. 1–2 (Spring/Summer 2016): Special issue on Canada and Australia

26

many ‘all other’ causes of death were indicated as ‘ill-defined’, but this classification was reduced to near zero by 1960 (Lancaster 1990).

The cause of death analysis uses standardized mortality rates and life expectancy decomposition. Standardized mortality rates are computed by sex and the six major causes of death, using the 1981 total Australian population (both sexes) by five-year age groups as the standard. Life expectancy decomposition uses the Arriaga (1984) method to attribute differences in life expectancy at birth to mortality change by age and major cause simultaneously. To facilitate discussion of the Epidemio- logic Transition, decomposition analyses were conducted for four periods:5 1922–46, 1946–70, 1970– 94, and 1994–2011. The periods were identified on the basis of internal consistency of patterns; that they are of roughly equal length assists in their comparison.

Trends in life expectancy

Australia in international context

Figure 1 compares historic male and female Australian life expectancies at birth with those of Canada, England and Wales, France, Japan, and the United States. The upward trends confirm the experience of Epidemiologic Transition. Deaths of Australian military personnel during World War II were excluded from national mortality statistics (Taylor et al. 1998), accounting for the absence of a downward spike in male life expectancy observed for some other countries.

In 1921–24, life expectancy in Australia was the highest among the six selected countries. Figure 1 shows that for both males and females, Australian life expectancy exceeded that of the second- highest of this group by as much as four years. Over the next two to three decades, this advantage diminished, and in the 1950s and 1960s Australia fell behind other countries. In the 1960s, for males, only US life expectancy was less than Australian life expectancy, while for females Australian life ex- pectancy was as low as any other at this time. Australian life expectancy has since recovered, ranking first among the selected group for males, and third for females, in the most recent period.

It may be observed in Figure 1 that the life expectancies of the selected countries largely con- verged in the 1960s. Convergence among the five Western countries persisted for about two decades, and coincided with the period when Japan overtook the Western countries. In recent decades, all six countries have tended to diverge, with differences of as much as five years occurring in the most recent decade. This pattern of convergence and divergence is consistent with wider trends (Meslé and Vallin 2011). Cardiovascular diseases, which are of key importance in the transition from the Age of Degenerative and Man-Made Diseases to the Age of Delayed Degenerative Diseases, played a dominant role in both the convergence and divergence of countries over the entire period (Meslé and Vallin 2011).

Japanese life expectancy increased rapidly in the post–World War II years, in keeping with its characterization as undergoing Accelerated Epidemiologic Transition (Omran 1971; Zhao et al. 2014). Japan has been a leader in life expectancy since the 1970s, particularly for females. In 2011, Japanese females had a 1.6 year advantage over Australian females, though Japanese males were at a slight disadvantage compared with Australian males (HMD 2015). In contrast, the United States has generally ranked last—since the mid-1960s for males, and since the early 1990s for females (Figure 1); this has been attributed to higher prevalences of smoking, obesity, and violence, as well as restricted access to health care (Caselli et al. 2014: 231).

5. 1921 was omitted owing to a large increase in that year.

Booth et al.: Epidemiologic Transition in Australia – The last hundred years

27

The sex difference in life expectancy at birth for the selected countries is shown in Figure 2. Ignoring war-related deviations, all countries experienced a general increase in the sex difference, followed by a downturn, as male improvements began to exceed female improvements. The turning point differs among countries, occurring first for England and Wales followed by the United States, Canada, and Australia, France, and finally (only around a decade ago) Japan. This same pattern has been found for high-income countries more generally, and has been attributed in most countries primarily to sex differences in the age pattern of mortality, rather than declining sex ratios in mortal- ity (Glei and Horiuchi 2007). A decomposition analysis of the G7 countries over the three decades to 2000 found that the main causes of death contributing to narrowing of the sex difference were circulatory diseases and accidents, violence, and suicide (Trovato and Heyen 2006).

45

50

55

60

65