The name is absent

DRAFT

than $40k). In addition, Taylor et al (2003) use administrative data to measure
participation. As recognised by the authors, these data do not capture mammograms
funded under the Medicare program. Whilst this program is intended to fund diagnostic
mammograms only, it is likely that sometimes it used for screening purposes.

From an economics perspective, there are two potential reasons that could explain the
systematic variation amongst women in different socio-economic groups. First, the
opportunity cost of screening may differ across groups. For example, the cost of travel
and time away from everyday activities and duties may vary amongst women. However,
it is not immediately obvious which socio-economics groups face higher opportunity
costs. On one hand women in higher socio-economic groups may face a higher
opportunity cost of time away from work due to higher wage rates. On the other, women
in lower socio-economic groups may have less flexible workplace arrangements and
therefore find it more difficult to make appointments.

The second possible reason for variation is women’s perceptions of the utility (and
disutility) associated with breast screening. For example, some women may feel more
strongly than others about the short-term inconvenience and discomfort of mammograms.
Furthermore, women’s perceptions of the long term benefits of mammography may also
vary. Such factors may explain why some women screen and others do not. However, it
does not explain why we observe systematic variation amongst socio-economic groups.
For systematic variation to occur on the basis of differences in preferences a second
condition needs to be met. That is, there would need to be some homogeneity of
preferences amongst similar (socio-economic) groups and heterogeneity of preferences
amongst different groups. Such a situation would arise if, for example, women in higher
socio-economic groups are more risk averse than those in lower SES groups. There is a
considerable theoretical and empirical body of work that supports the notion that SES
affects health behaviours (Lantz, House, Lepkowski, Williams, Mero, & Chen, 1998;
Lantz, Lynch, House, Lepkowski, Mero, Musick et al., 2001; Singh, Miller, & Hankey,
2002; Wardle, McCaffery, Nadel, & Atkin, 2004). Link et al (1998) argue that SES

11

<<   12   13   14   15   16   17   18   >>

More intriguing information