Potentially avoidable deaths were classified by ICD-9 up to 1998 and by ICD-10 from 1999 onwards. Data are reported by year of death. Rates were age-adjusted using the standard Australian population as at 30 June 2001. A Poisson regression model was fitted to assess differences in the slope of the trend lines.
Source:
ABS mortality data, ABS population estimates and ABS Socio Economic Indices for Areas (HOIST), avoidable mortality fractions from Tobias et al., 2001. Centre for Epidemiology and Research, NSW Health.
Classification of premature deaths as 'potentially avoidable' and 'unavoidable' gives an indication of the potential scope for improving health and closing health gaps. Deaths classified as 'avoidable' are those that could potentially be avoided through the activities of the health and related sectors (Tobias and Jackson, 2001). Here, we use a summary measure, developed in New Zealand, of avoidable mortality, which incorporates a range of causes of death before the age of 75 years, including selected cardiovascular diseases, cancers, communicable diseases, and injuries. This measure is similar to others used more widely to monitor health system performance in other countries (Tobias and Jackson, 2001; Holland et al., 1994; Wood et al., 1999).
In 1980 in New South Wales, deaths from potentially avoidable causes accounted for 75.0% of all premature deaths (before the age of 75 years), falling to 61.1% of all premature deaths in 2000. Over the 20 year period, rates of premature death fell by 40.7% and rates of avoidable death fell by 51.8%. The fall in avoidable death rates was higher for males (54.3%) than for females (48.4%) between 1980 and 2000. The fall was also the greatest in those from the highest SES group (62.2% in males and 54.8% in females), compared to the lowest (53.4% in males and 48.2% in females).
In absolute terms, the gap in potentially avoidable mortality between the highest and lowest SES groups narrowed between 1980 and 2000, with rate differences of 124.7 and 56.3 per 100,000 in males and females in 1980, falling to 96.1 and 43.5 per 100,000 in males and females in 2000. In the period between 1980 and 2000 the relative difference in rates between the lowest and highest SES groups for males increased from 34% higher in the lowest SES group at the beginning of the period, to 63% higher in 2000 and from 27% to 40% higher in females over the period. This relative increase in the gap between males in the highest and lowest SES groups over the twenty-year period was highly significant (p < 0.0001) and in females it was moderately significant (p < 0.02).
The individual diseases contributing most to the burden of potentially avoidable mortality were ischaemic heart disease, stroke, and lung cancer in 1980 and ischaemic heart disease, lung cancer and colorectal cancer in 2000. For ischaemic heart disease, the rates decreased relatively more quickly over the period for the highest SES group than for the lowest SES group.
The results overall show that, in New South Wales, the burden of potentially avoidable mortality has decreased overall and across all SES groups between 1980 and 2000; however the rate of reduction has been higher in the highest SES group compared to both the lowest SES group and the rest (middle 60%) of the population.
The data for Hunter New England for the period 1994 to 2003 indicate little change over time in the pattern for both males and females of people from the lowest socioeconomic status quintile consistently having worse avoidable premature death rates (among the population aged less than 75 years) compared with those from the highest socioeconomic status group.
For more information:
Tobias M, Jackson G. Avoidable mortality in New Zealand, 1981-97. Aust NZ J Public Health 2001;25:12-20.
Holland WW, Fitzgerald AP, Hildrey SJ, Phillips SJ. Heaven can wait. J Public Health Med 1994;16(3):321-30.
Wood E, Sallar AM, Schechter MT, Hogg RS. Social inequalities in male mortality amenable to medical intervention in British Columbia. Soc Sci and Med 1999;48(12):1751-8.
Although this page can be printed directly from your Web browser,
a higher quality version of this entire page (graph, table and text)
is available as an Acrobat PDF file which can be printed or viewed on screen
using free software.
The data associated with this page are available for download as a CSV file which can be
imported into many software packages. The graph is available for download as an EPS (Encapsulated PostScript) file
and as an EMF (Enhanced Metafile Format) file. Files in these formats can be imported into most word processing,
presentation and graphics software packages.
Rider:
The information presented in this report result from analyses of a variety of social and health focused datasets. These datasets originate from a variety of sources including Hunter New England Health, the NSW Department of Health, and the Australian Bureau of Statistics. The timing of the release of these data to third parties is controlled by the owner of these data. It is therefore possible for these organisations to publish data that they have not yet made available to Hunter New England Population Health for analysis and release. Users should therefore check the publications of these organisations as it is possible that they may have published even more up to date information on Hunter New England than those available in this report. As this report is an ongoing project, the indicators presented will be updated as soon as possible after the release of all datasets to Hunter New England Population Health.
Copyright notice:
This work is copyright Hunter New England Area Health Service, 2005. It may be reproduced
in whole or in part, subject to the inclusion of an acknowledgement of the source. Commercial usage or sale is prohibited.
Hunter New England Population Health. Health in Hunter New England. Hunter New England Area Health Service, 2005. Available at:
http://www.hnehealth.nsw.gov.au/HHNE/ses/ses_avmlomidhi.htm. Accessed (insert date of access).
Top of page
Table of contents
