Background: Breast cancer is the commonly diagnosed cancer in women worldwide. Although not a HIV-associated cancer, it is the commonest cancer in women living with HIV. Sadly, these women experience higher mortality rates than their HIV-uninfected counterparts. However, the underlying mechanisms for observed breast cancer survival disparities by HIV status are poorly understood because existing research has been limited by either small sample sizes or lack of granularity needed to properly investigate potential survival determinants. Aim: This thesis investigated the impact of HIV on the whole breast cancer care continuum from symptom(s) recognition, confirmatory diagnosis, to post-diagnosis journey (survival) in HIV-infected women in sub-Saharan Africa, relative to HIV-uninfected breast cancer women. Methods and results: Because HIV status was self-reported in most ABC-DO study sites (main PhD data source), I first assessed the validity and accuracy of self-reported HIV-positive status among 17,445 adults who tested and had been informed of their HIV test results in the Karonga Health Demographic Surveillance System (chapter 4). I compared self-reported HIV status and HIV test results, and estimated sensitivity, specificity, positive and negative predictive values for self-reported HIV status. I found very high level of agreement between HIV-self-reports and HIV-test results. Among true HIV-positive individuals, those who were older and had been interviewed in clinic settings were more likely to accurately report their HIV-positive status. I then examined differences in patient’s journey to breast cancer diagnosis, and breast tumour characteristics at diagnosis between HIV-infected and HIV-uninfected women using data from ABC-DO study (chapter 5). There was no evidence for an association between HIV status and the time interval from first breast cancer symptoms to definitive diagnosis, and tumour characteristics (grade, stage, and receptor subtypes) at breast cancer diagnosis. I further investigated differences in overall survival after a breast cancer diagnosis by HIV status, and their determinants among ABC-DO women (chapter 6) and showed that HIV-infected women had increased risk of all-cause mortality compared with HIV-uninfected women (3-year overall survival = 46% vs 55%; adjusted hazard ratio (aHR) 1.42, 95% CI: 1.15 – 1.74). The HIV survival differential was stronger among women with non-metastatic breast cancer (52% vs 65%, aHR 1.65: 1.30 – 2.10). Finally, using UNAIDS mortality data to generate HIV-specific life tables (for relative/net survival estimation) (chapter 7), I showed that HIV-infected women still experienced poorer net survival than HIV-uninfected women. However, there were only a slight difference between mortality rate ratios from overall and net survival estimates (aHR 1.41: 1.13 – 1.75 vs. 1.38: 1.09 – 1.76 respectively), suggesting that the observed survival disparities were not primarily driven by higher background mortality associated with HIV/AIDS in HIV-infected patients. Conclusion: In summary, I found no association between HIV status and time to breast cancer diagnosis and tumour characteristics at cancer diagnosis. HIV-infected women were at an increased risk of all-cause mortality compared with HIV-uninfected women, even after considering differences in HIV-associated background mortality between the two groups. To prevent more deaths in this unique population, more studies with more granular data on both HIV and breast cancer are needed to evaluate the influence of treatment, treatment-related toxicities, and HIV-related factors.