Objective: Although breast self-examination has been shown to be the least-expensive, less time-consuming, and non-invasive screening method, still there is a gap in practice. Furthermore, the information among more risky population which attends anti-retroviral therapy is too limited. Therefore, the aim of this study is to assess breast self-examination practices and its associated factors among women who attended the anti-retroviral therapy clinic in Bahir Dar city administration, Northwest Ethiopia. Methods: An institution-based cross-sectional study was conducted among 660 women who attended anti-retroviral therapy clinics from March 1 to March 30, 2020. A systematic random sampling technique was used to select study participants. Pre-tested interviewer-administered questionnaires were used to collect data. Data were entered in to EPI data 3.1 and exported to Statistical Package for the Social Sciences version 23.00 software for analysis. A binary logistic regression model was fitted to identify factors associated with self-breast examination. Variables with a p value less than 0.2 in bivariable regression were candidates for multivariable regression. Adjusted odds ratios with a 95% confidence intervals and p values less than 0.05 were used to determine the association between independent and dependent variables. Hosmer–Lemeshow Test was used to determine model fitness. Result: Among 641 study participants, 224 (34.9%) have ever practiced breast self-examination. Women who attended college or above (adjusted odds ratio = 4.04, 95% confidence interval (1.65,9.90)), rich (adjusted odds ratio = 6.64, 95% confidence interval (2.72,16.20)), knowledgeable about signs and symptoms of breast cancer (adjusted odds ratio = 5.13, 95% confidence interval ( 2.55,10.31)), risk factors for breast cancer (adjusted odds ratio = 3.62, 95% confidence interval (1.85,7.07)), positive attitude toward breast self-examination (adjusted odds ratio = 2.76,95% confidence interval (1.41,11.84)), family history of breast cancer (adjusted odds ratio = 3.68,95% confidence interval (1.14,11.84)), and knowledge about breast self-examination technique (adjusted odds ratio = 2.64, 95% confidence interval (1.23,5.66)) had higher odds of breast self-examination practice. Conclusion: The practice of breast self-examination was low. During their regular visits for other issues, education and information dissemination about the benefits and techniques of self-breast examination is recommended.