Article Text
Abstract
Background Although breast cancer is a disease associated with high socioeconomic status (SES), previous studies report higher mortality amongst women with lower SES. Breast cancer prognosis is highly related to tumour characteristics, such as oestrogen receptor (ER) status. ER+ tumours have better prognosis than ER- tumours as they are responsive to hormone treatments. This study aims to examine if socioeconomic inequalities in breast cancer mortality in Scotland differ by ER status.
Methods All women diagnosed with breast cancer in Scotland (recorded in the cancer registry) from 1997 to 2016 were followed up to the end of 2016. Median follow up time was 5.5 years, and 5, 10 and 15 years Kaplan Meier estimates for all-cause mortality were calculated by Scottish Index of Multiple Deprivation (SIMD) –an area-based measure of deprivation, stratified by ER status (82% ER+). Cox models were used to estimate hazard ratios (HR) between women in the most deprived quintile (Q1) of areas and women in the least deprived quintile (Q5) of areas stratified by ER status. Models were adjusted for age, Scottish region, tumour characteristics (grade, tumour size and whether screen-detected or not), treatment regimens and Charlson index of comorbidity. The potential for SIMD and ER status interaction was tested using likelihood ratio test.
Results Among the total of 72,217 women with breast cancer 12,923 (18%) were in Q1 and 14,980 (21%) were in Q5. There were 5,688 (44%) deaths in Q1 and 4,526 (30%) deaths in Q5. Women in Q1 had more advanced tumours than women in Q5: 43% vs 37% with poorly differentiated tumours, 45% vs 40% with tumours bigger than 2cm24% vs 30% screen detected tumours. Difference in proportions who were dead at 5, 10 and 15 years between Q1 and Q 5 were 11%, 14% and 17% respectively. Five and 10 years mortality rates were higher for ER- tumours than for ER+ but there were no differences in mortality at 15 years. Fully adjusted Cox regression models for mortality for Q1 compared to Q5 gave HR of 1.40 [95% CI: 1.30–1.50]) for women with ER+ tumours and 1.35 [95% CI: 1.19–1.53] for women with ER- tumours There was no evidence of interaction between SIMD and ER status (p value=0.375).
Conclusion Socio-economic status is inversely associated with breast cancer mortality in Scotland regardless of ER status. Future studies are needed to determine cause of death and the role of comorbidities in this population.