DRAFT
women in the target age group, participation rates should not be systematically related to
any ethnic, economic or geographic factors.
The most recent report on the program’s performance, using administrative data, reveals
that in 2004 participation amongst women in the target age range was 55.6% nationally.
Whilst international comparisons are difficult, Australia’s participation rates appear to be
low by international standards. In 2003/04, Australia’s breast screening rate was well
below the OECD average of 69.6%.
Previous reports had shown a positive trend amongst women in the target age group.
Between 1996 and 2002 participation rates increased from 50.4% to 57.1% but since
2002 there has been a downward trend (Australian Institute of Health and Welfare, 2007).
Compared to the national average, screening rates in NSW have been consistently lower
but have exhibited similar trends over time. The latest figures show a participation rate of
50.1% for 2003-04, after having reached 53% in 2001-02.
Administrative data from the Breast Screen Program also show significant regional
variation in participation rates. Women living in metropolitan regions were more likely
to screen than women living in rural regions. Despite the program offering free
mammograms at the point of service, administrative data also show that women residing
in the most disadvantaged socio-economic areas were less likely to screen than women
living in more affluent areas. There was no discernable pattern amongst any of the other
socio-economic groups (Australian Institute of Health and Welfare, 2007). However,
results using administrative data are limited by the use of ecological socio-economic
status (SES), and are therefore likely to reduce interpersonal variability by assigning
regional average SES to individual women.
Analysis using unit record data has revealed a clearer pattern of utilisation. A study using
1997 and 1998 NSW Health Survey data found that SES, measured by imputed income,
was positively and significantly related to the likelihood of screening (Birch, 2007). This
study was restricted by the lack of income data in the 1997/98 NSW Health Surveys -