Breast screening in NSW, Australia: predictors of non-attendance and irregular attendance

Presenter: Minh Vu, Centre for Health Economics Research & Evaluation (CHERE), University of Technology Sydney

Abstract

Rationale: Australia’s national breast screening program aims to provide equitable access to mammography services and encourages women aged between 50 and 69 to screen at two-yearly intervals. Despite considerable efforts and investment in the program there is evidence of inequitable uptake of mammography screening amongst socio-economic groups. However, there is little understanding of the determinants of screening participation. This paper explores the role of women’s time preferences in predicting screening uptake. Using Fuch’s finding that health habits are related to discount rates, we examine the impact of smoking status on screening behaviour as well as its interaction with other key determinants.

Objectives: This study compares the characteristics of women who (1) screen regularly, (2) screen irregularly and (3) have never screened. The aim of the study is to identify similarities and differences between these three groups to help policy makers gain a better understanding of the determinants of screening and develop targeted strategies to improve overall participation rates and/or reduce inequalities.

Methods: A multinomial logit (MNL) model was used to examine the role of socioeconomic status, cultural background, preferences, education and region of residence on breast screening behaviour amongst women aged 50 to 69. Data from a state-based (NSW) population health survey was used in this analysis. This cross sectional survey is conducted on an annual basis to study health risk factors and behaviors of a random sample of NSW residents, aged16 years and over.

Results: The results show that younger and older women within the target age had a higher chance of never having screened. No significant differences were found for women residing in rural or remote regions. Women who worked longer hours or have a higher education were observed to have a higher chance of screening irregularly. Compared to the lowest income group, middle income earners were significantly less likely to never screen. Further, there is evidence of a positive relationship between income and a woman’s likelihood of screening regularly. When smoking status was added to the model as a proxy for a woman’s time preference, it was found to be highly significant in increasing both the chance of never and irregularly screening. Furthermore, adding smoking status to the model reduced the significance of income in predicting women’s participation.

Conclusions: These results provide evidence of persistent and systematic variation in screening uptake and regular participation. The results also point towards the need for more targeted recruitment and retainment strategies. Providing greater access for working women may be important (e.g. longer opening hours and/or location of centres near major employment centres) as well as in some outer metropolitan areas. The results may also indicate that socio-economic differences in screening uptake may be explained by women’s time-preferences. This finding warrants further research because future gains in participation rates may be achieved more effectively through demand side strategies.