Case-mix analysis and variation in rates of non-surgical treatment of older women with operable breast cancer

Background Non-surgical management of older women with oestrogen receptor (ER)-positive operable breast cancer is common in the UK, with up to 40 per cent of women aged over 70 years receiving primary endocrine therapy. Although this may be appropriate for frailer patients, for some it may result in treatment failure, contributing to the poor outcomes seen in this age group. Wide variation in the rates of non-operative management of breast cancer in older women exists across the UK. Case mix may explain some of this variation in practice. Methods Data from two UK regional cancer registries were analysed to determine whether variation in treatment observed between 2002 and 2010 at hospital and clinician level persisted after adjustment for case mix. Expected case mix-adjusted surgery rates were derived by logistic regression using the variables age, proxy Charlson co-morbidity score, deprivation quintile, method of cancer detection, tumour size, stage, grade and node status. Results Data on 17 129 women aged 70 years or more with ER-positive operable breast cancer were analysed. There was considerable variation in rates of surgery at both hospital and clinician level. Despite adjusting for case mix, this variation persisted at hospital level, although not at clinician level. Conclusion This study demonstrates variation in selection criteria for older women for operative treatment of early breast cancer, indicating that some older women may be undertreated or overtreated, and may partly explain the inferior disease outcomes in this age group. It emphasizes the urgent need for evidence-based guidelines for treatment selection criteria in older women with breast cancer.