This study used audit data to compare breast cancer survival in New Zealand at five years post-treatment across various factors to assess their effect on survival. Survival results were also compared with Australian results. The audit gained survival information from a linkage with New Zealand death records.

Comparison with Australian results

Five-year survival for Australian cases was 93%. In comparison, New Zealand survival was 90%. While the difference here is small, five-year survival is an early measure of ultimate survival. Only one third of breast cancer deaths occur within that first five years post-treatment. The 3% absolute increase in deaths seen in this study, therefore, represents an approximate 40% higher mortality for New Zealand cases overall.

The difference between countries may reflect a real difference in patient risk profiles such as ethnicity, socio-economic status, obesity and smoking rates, as well as differences in access to, and timeliness and quality of breast cancer care.

It is also possible that this difference was influenced by the method of linkage to survival data for each country. The New Zealand linkage used National Health Index (NHI) number which would have been more exact than the matching on limited patient identifiers used with the Australian linkage (The audit collects first three letters of patient surname, date of birth and patient postcode. New Zealand surgeons also have the option of recording NHI number as their Clinic Reference for the patient).


Five-year survival was 87% for Maori patients, 84% for Pacific and 91% for 'other' ethnicities. When survival calculations were adjusted to take into account other cancer and patient risk factors, Maori women still showed lower survival. For Pacific women, the survival rate improved, although it was still slightly lower than 'other' ethnicities.

Differences across ethnicity were smaller in BQA data than seen in other studies; however, this may be due to the audit's focus on early breast cancer.

Cancer and patient characteristics

Survival was lower for patients with any of the following:

  • age over 80 years or under 30 years
  • tumours located in the milk-ducts rather than other locations, such as the lobules
  • larger tumour size
  • higher tumour grade
  • lymph nodes positive for cancer cells
  • triple negative tumours (i.e. negative for all three hormonal receptor tests, therefore, will not respond to hormonal therapy)
  • vascular invasion (cancer spread to blood)
  • three or more tumours in the breast

Patient place of residence

Data indicated that survival tended to be higher in the least disadvantaged areas. Differences were small at five years; however, five-year survival is a short-term follow-up and greater differences would be expected over the long term.

Factors not affecting survival

No differences in survival according to remoteness of patient residence were noted.

Surgeon case load (i.e. number of early breast cancer patients treated in a year) and patient's method of referral (i.e. from breast screening or having presented to GP with symptoms) were not shown to affect survival after adjusting for other known risk factors.