Summary
This report presents the first results from an Australian-first project, combining data from BreastScreen Australia, the National Cervical Screening Program, the National Bowel Cancer Screening Program, the Australian Cancer Database, the National Death Index, and the National HPV (human papillomavirus) Vaccination Program Register.
Breast cancers detected through BreastScreen Australia were less likely to cause death
This report compared survival outcomes of breast cancers detected through BreastScreen Australia with breast cancers diagnosed in women who had never screened.
Of the breast cancers diagnosed in women aged 50–69 in 2002–2012:
- 30,510 were detected through BreastScreen Australia
- 19,526 were diagnosed in women who had never screened.
Women diagnosed through BreastScreen Australia had a 69% lower risk of dying from breast cancer before 31 December 2015 than those who had never screened.
Even after correcting for lead-time bias (where an earlier diagnosis may not affect date of death, yet give a seemingly longer survival time) and screening selection bias (where women who choose to participate in screening may be at a lower risk of death, which would result in an increase in survival that may not be real), the risk of dying from breast cancer was still 42% lower for women diagnosed through BreastScreen Australia than for women who had never screened through BreastScreen Australia.
Cervical cancers detected through cervical screening were less likely to cause death
Cervical cancer outcomes need to be considered within the context of cervical screening that aims to detect and treat precancerous disease, thereby preventing cervical cancers. This is reflected in these data, since the vast majority of cervical cancers occurred in women who had either never screened or who were lapsed screeners (had not screened for some time).
This report compared survival outcomes of cervical cancers detected through cervical screening with cervical cancers diagnosed in women who had never had a Pap test.
Of the cervical cancers diagnosed in women aged 20–69 in 2002–2012:
- 354 were detected through cervical screening
- 1,222 were diagnosed in women who had never had a Pap test.
Women diagnosed through cervical screening had an 87% lower risk of dying from cervical cancer before 31 December 2015 than women who had never had a Pap test.
Bowel cancers detected through the National Bowel Cancer Screening Program were less likely to cause death
This report compares survival outcomes of bowel cancers detected through the National Bowel Cancer Screening Program with bowel cancers diagnosed in people who had never been invited to screen.
Of the bowel cancers diagnosed in people aged 50–69 in 2006–2012:
- 3,316 were detected through the National Bowel Cancer Screening Program
- 20,217 were diagnosed in people who had never been invited to screen.
People diagnosed through the National Bowel Cancer Screening Program had a 59% lower risk of dying from bowel cancer before 31 December 2015 than people who had never been invited to screen.
Even after correcting for lead-time bias (where an earlier diagnosis may not affect date of death, yet give a seemingly longer survival time), the risk of dying from bowel cancer was still 40% lower for people diagnosed through the National Bowel Cancer Screening Program than for people who had never been invited to screen.
Screening behaviour
This project also used the combined data to understand screening behaviour of women across the 3 national cancer screening programs. Key findings are summarised below.
- A high proportion (69%) of screened women participated in all the cancer screening programs for which they were eligible, indicating overall good screening behaviour in women who were already engaged in screening. As these data excluded women who did not screen in any program, this is an overestimate of the true proportion of eligible women who screened.
This may mean that if the barriers to screening can be broken in non-screeners, thereby engaging them to participate in at least one screening program, it might lead to participation in other cancer screening programs as well, and greater overall participation, leading to more cancers detected through screening, which have a lower risk of death, as this study shows.
- A positive screening test result that resulted in diagnostic testing in one program resulted in women being more likely to screen in another program, and to do so sooner.
These results indicate that women may be more aware of the need to screen across programs following a screening event that required diagnostic follow-up.
Women vaccinated against HPV are more likely to participate in cervical screening than unvaccinated women
Comparing cervical screening participation rates:
- In women aged 20–24, participation was 45.5% in HPV-vaccinated women and 33.1% in unvaccinated women
- In women aged 25–29, participation was 56.5% in HPV-vaccinated women and 44.3% in unvaccinated women.
This indicates that women who are vaccinated against HPV are either more aware of the need to participate in cervical screening, or are more likely to take part in healthy behaviours generally.
Summary
1. Introduction
1.1 Cancer screening programs in Australia
1.2 Cancer screening programs reduce mortality
1.3 Participation in cancer screening programs
2. Objectives
2.1 Premise of this data linkage project
2.2 Objectives of this data linkage project
3. Data and methods
3.1 Data sources
3.2 Methods
4. Preliminary findings
4.1 Assessment of the success of the data linkage
4.2 Assessment of overlap between linked data sets
5. What are the benefits of screening?
5.1 Addition of screening history to cancers
5.2 Survival of women with screen-detected versus non-screen-detected breast cancers
5.3 Survival of women with screen-detected versus non-screen-detected cervical cancers
5.4 Survival of people with screen-detected versus non-screen-detected bowel cancers
5.5 Effect of screening behaviour on breast cancer and cervical cancer outcomes
6. Who screens and what affects this?
6.1 Patterns in participation across the 3 cancer screening programs
6.2 Effect of a cancer diagnosis on screening behaviour
6.3 Effect of HPV vaccination on participation in cervical screening
7. Discussion and conclusions
7.1 Objective 1: Cancer outcomes
7.2 Objective 2: Screening behaviour
7.3 Objective 3: Enhancing screening data
7.4 Project limitations and areas for improvement
7.5 Where to from here?
Appendix A: Additional data tables
Appendix B: Positive predictive values of screening tests
Appendix C: Classifications
Appendix D: Additional statistical methods
End matter: Acknowledgments; Abbreviations; Symbols; Glossary; References; List of tables; List of figures; List of boxes; Related publications