Red X iconGreen tick iconYellow tick icon

Research Team

  • Peter Davis
  • Patrick Graham
  • Andrew Sporle
  • Tony Blakely
  • Neil Pearce

Supported by the Health Research Council


The primary aim of this research is to extend previous New Zealand work on health expectancy by analysing social variations in health expectancy trends over a 15 year period for a range of disability measures and to study socio-economic variation in health expectancy using a range of indicators of socio-economic status.


Health expectancy measures combine information on mortality and morbidity to produce estimates of the expectation of life in various health states. Two examples are disabled and disability free life expectancy. This research will inform current debates about the evolution of population health in New Zealand and in developed countries generally. Three scenarios, the so-called compression of morbidity, expansion of morbidity and dynamic equilibrium hypotheses have been proposed to describe this evolution. The compression scenario predicts an increasing proportion of the lifespan spent free of functional limitation and disability while the "expansion" hypothesis posits an increase in this proportion due to case fatality rates declining faster than disease incidence and progression rates. The dynamic equilibrium hypothesis is more complex and a simple version posits a decline in the proportion of the life span with serious functional limitations or disability but an increase in the proportion of the life span spent with less severe limitations. Our research will concentrate on variations in health expectancy trends by the key social variables of gender, ethnicity and socio-economic status. Therefore this research will contribute towards the development a more socially differentiated theory of health transition than has been available hitherto.


The proposed research involves two major components;

1. an investigation of changes in health expectancy between 1981 and 1996, by gender, ethnic group (Maori and nonMaori) and occupationally defined socio-economic status and

2. an investigation of socio-economic variation in health expectancy using non-occupationally based socio-economic indicators. The latter analysis will permit the analysis of socio-economic variation to be extended to the whole population whereas our previous analyses of this issue have been restricted to males aged 15 to 64 because of the difficulty of applying occupationally based SES measures to the wider population.

The analysis of changes in health expectancy will use comparable data on functional limitations and disability from the 1980/81 Social Indicators Survey and the 1996/7 Disability survey, together with mortality data and information on the size of the institutionalised population from the relevant time periods. The mortality, morbidity and institutionalisation information can be combined using a simple modification of conventional life table methods to generate health expectancy measures.

The analysis of variation by non-occupationally defined socio-economic status will draw mortality information from the New Zealand census cohort mortality study which is a record linkage study based on the 1991 census. To provide morbidity data from a comparable time period we will use the 1992/3 household health survey. Because of differences in the questionnaires used, the health expectancy estimates derived using this dataset will not be comparable to the 1981 and 1996 estimates. However comparison with our previous analyses of socio-economic variation in health expectancy using the 1992/3 health survey will allow us to investigate the direction and magnitude of any bias introduced by using occupationally based measures of socio-economic status.

Relevance and Significance

By focussing on social variation in trends in health expectancy this study will significantly broaden debates about the evolution of population health and the impacts of population ageing. The issues of trends in the health of the population and the extent to which improvements in health will be shared equally by all social groups has implications for future funding of health services and for wider issues of social equity. Methodologically, the project addresses important analytical issues resulting from changes to the classification of ethnicity in New Zealand.


  • Davis P, Mathers C, Graham P. Health expectancy in Australia and New Zealand. In JM Robine et al. (eds.), Health Expectancies at the Dawn of the Third Millennium (INSERM, Paris). (in press)
  • Davis P, Graham P, Pearce N. Health expectancy in New Zealand, 1981-1991: social variations and trends in a period of rapid social and economic change. Journal of Epidemiology and Community Health 53: 519-27, 1999.
  • Lightfoot R, Davis P, Finn E, Lay-Yee R, Gribben B, McAvoy B. Practice nurses in the Waikato, 1991-1992, I: Occupational profile. New Zealand Medical Journal 112: 26-8, 1999.
Back to top