Retrospective study of reasons for improved survival in patients with breast cancer in east Anglia: earlier diagnosis or better treatment.

OBJECTIVES: To investigate the recent fall in mortality from breast cancer in England and Wales, and to determine the relative contributions of improvements in treatment and earlier detection of tumours.
DESIGN: Retrospective study of all women with breast cancer registered by the East Anglian cancer registry and diagnosed between 1982 and 1989.
SUBJECTS: 3965 patients diagnosed 1982-5 compared with 4665 patients diagnosed 1986-9, in three age groups 0-49, 50-64, > or = 65 years, with information on stage at diagnosis and survival. MAIN OUTCOME MEASURES: Three year relative survival rates by time period, age group, and stage; relative hazard ratios for each time period and age group derived from Cox's proportional hazards model, adjusted for single year of age and stage.
RESULTS: Survival improved in the later time period, although there was little stage specific improvement. The proportion of early stage tumours increased especially in the 50-64 year age group, and adjustment for stage accounted for over half of the improvement in survival in women aged under 65 years.
CONCLUSION: Over half of the drop in mortality in women aged under 65 years seems to be attributable to earlier detection of tumours, which has been observed since the mid-1980s. This could have resulted from an increase in breast awareness predating the start of the breast screening programme.