Mortality and mental illness

Abstract

Background: The finding by Lawrence, Holman and Jablensky (Duty to Care) that mortality among the mentally ill in Western Australia was 2.5 times that of the general population, seemingly, has great significance for public policy concerning the mentally ill. 'Mortality' could be a useful outcome measure for mental health services.

Objectives: To replicate that study in the Australian Capital territory (ACT), comparing mortality rates in the mentally ill with those of the total population.

Method: A list of all people who contacted the ACT mental health services between 1985 and 2000 was compiled. Using the national register of deaths (Australian Institute of Health and Welfare), persons known to the mental health services who died between 1990 and 2000 were identified and sex, date of birth, date of death, cause of death and place of death were noted. Using Australian Bureau of Statistics data for all deaths in the ACT, deaths in the total population and in the mentally ill population were tabulated for the period 1996-2000. With 1996 data as the base, using total population data from the ABS and mental health population data derived by amending 'the list' to remove duplications, pre1996 deaths and post1996 additions, mortality rates for the period 1996-2000 were compared.

Results: The gender and age distribution of the mentally ill population and the total population, and of deaths in those populations, were very different. One third of all deaths in the mentally ill occurred outside of the ACT. Compared with the general population, mortality in the mentally ill (including deaths outside of the ACT) was only slightly excessive for 'all causes' and ischaemic heart disease but grossly excessive for 'suicide'.

Conclusions: This study did not confirm the excessive mortality rate in the mentally ill reported by Lawrence et al. except for suicide. Many methodological issues in using population studies to attempt to measure the size of the increase were identified. Differences in method between the ACT and WA studies probably explain the differences in results. Caution is urged in using the results of mortality studies as determinants of public policy or to evaluate services.