AIM:
To assess whether the proportion of patients aged 70 and over with ER+ operable breast cancer in England who are treated with surgery has changed since 2002, and to determine whether age and individual level factors including tumour characteristics and co-morbidity influence treatment choice.
METHODS:
A retrospective cohort analysis of routinely collected cancer registration data from two English regions (West Midlands, Northern & Yorkshire) was carried out (n = 17,129). Trends in surgical use over time for different age groups were assessed graphically and with linear regression. Uni- and multivariable logistic regressions were used to assess the effects of age, comorbidity, deprivation and disease characteristics on treatment choice. Missing data was handled using multiple imputation.