Breast Cancer in sub-Saharan Africa: Determinants of Stage at Diagnosis and Diagnostic Delays in Women with Symptomatic Breast Cancer.

Background: Breast cancer is the most common female cancer worldwide and in sub-Saharan Africa (SSA). Breast cancer incidence in SSA is relatively low but, as survival from the disease in the region is poor, mortality rates are as high as in high income countries. Late stage at diagnosis, and delays in a woman’s journey to a cancer diagnosis, are features known to contribute to poor breast cancer survival rates. There has been little focus on the factors affecting stage at diagnosis and the determinants of diagnostic delays in SSA despite previous studies highlighting the importance of early detection and treatment in breast cancer control. Aims and Methods: The main objectives of this thesis are: (i) To conduct a systematic review and meta-analysis of stage at breast cancer diagnosis in SSA to examine trends over time and examine possible sources of variation across the region. Random-effects meta-analyses were performed to investigate between-study heterogeneity in percentage of late-stage disease (stage III/IV) breast cancer, and meta-regression analyses were carried out to identify possible sources of variation. Percentages of Black women with late-stage breast cancer in SSA were compared with equivalent estimates for US Black and White women using the Surveillance, Epidemiology and End Results Database. (ii) To design and conduct a study, the Nigerian Integrative Epidemiology of Breast Cancer (NIBBLE) study, (iii) to investigate determinants of late stage at breast cancer diagnosis and diagnostic delays at six tertiary and secondary health facilities in Nigeria. Ordinal logistic regression was used to examine associations of socio-demographic, breast cancer awareness, health care access and clinical factors with the odds of later stage disease. Linear regression analyses were performed to examine the association of these factors with time from noticing symptoms to diagnosis (total delay), and its two main 4 components: pre-contact delay (i.e. time from symptoms to first contact with any care provider including traditional healers) and post-contact delay (i.e. time from first contact to diagnosis). Results: (i) Systematic review: 83 studies were eligible representing 26,788 women from 17 SSA countries. There was wide variation in percentage of late stage (median 74.7%, range 30.3-100%, I2=93.3% p<0.0001). Late stage at diagnosis was notably higher in Black vs non-Black women in SSA and higher for populations from mixed (urban and rural) settings than from urban settings. The percentage of women with late stage breast cancer decreased over time but it was still higher than in US White and Black women 40 years previously. (ii) Findings from NIBBLE: 300 breast cancer patients were recruited, 67.7% with late stage (III/IV) at diagnosis. Multivariate analyses showed lower educational level (odds ratio (OR) 2.35; 95% confidence interval (CI) 1.04, 5.29), not believing in a cure for breast cancer (OR 1.81; 95% CI 1.09, 3.01), Muslim religion (OR 0.46; 95% CI 0.22, 0.94) and living in a rural area (OR 2.18; 95% CI 1.05, 4.51) to be significantly associated with later stage. No associations were found between later stage and age at diagnosis, tumour grade or oestrogen receptor status. Women diagnosed in stages III/IV self-reported, on average, 36% longer total delay times than those in stages I/II. Median (IQR) for pre-, post and total delays were 2.6 (0.6, 8.3), 3.1 (0.79, 8.7) and 7.8 (3.3, 18.7) months, respectively, for all women who presented with suspicious symptoms (n=430). In fully-adjusted analyses, post-contact delays in all women with symptoms were associated with lack of a personal income (OR 1.49; 1.04, 2.00), no previous history of benign breast disease (OR 0.61; 0.42, 0.89) and having 5 or more children (OR 1.88; 95% CI 0.96, 3.67) whilst total delay was inversely associated with presentation at a secondary facility (OR 0.68; 95% CI 0.51, 0.92) and no previous history of benign breast disease (OR 0.64; 0.47, 0.88). Post-contact and total delays were both positively associated with the total number of providers visited before a diagnosis (P for trend (Pt)=0.014 and Pt