Shahrekord University of Medical Sciences Epidemiology and Health System Journal 2980-7891 13 1 2026 06 29 Prognostic Factors in Breast Cancer Survival: A Log-Normal Accelerated Failure Time Model Analysis From Western Iran 60 64 10.34172/ehsj.26677 EN Yousef Veisani https://orcid.org/0000-0002-9131-3744 Hassan Nourmohammadi https://orcid.org/0000-0002-5967-3395 Hojjat Sayyadi https://orcid.org/0000-0002-8348-3604 Fereshteh Monfaredi Tabar https://orcid.org/0009-0005-3731-8359 Maryam Kheiry https://orcid.org/0000-0003-2238-3295 Journal Article 10.34172/ehsj.26677 2025 12 28 2026 06 14 Introduction: Breast cancer (BC) is the most common malignancy among women worldwide. Thus, this retrospective cohort study aimed to identify prognostic factors for BC-specific survival using parametric accelerated failure time models. Methods: Data were collected from 820 female patients diagnosed between 2014 and 2020 at Shahid Mostafa Khomeini Hospital in Ilam, Iran, with follow-ups through 2023. Survival time was defined from diagnosis to BC death or censoring. The Kaplan-Meier method, along with log-rank tests, was applied in the model. Results: Among 798 patients, 72 BC deaths occurred (9.0%). The goodness of fit in the log-normal model was evaluated using the Akaike Information Criterion (AIC=574.2) and the Bayesian Information Criterion (BIC=607.0). Moreover, histological type was the only significant independent predictor of survival. Compared with infiltrating ductal carcinoma, lobular carcinoma was associated with shorter survival (time ratio=0.613; 95% confidence interval: 0.401–0.939; P=0.024). Nonetheless, age, tumor topography, diagnostic method, and tumor grade were not statistically significant predictors. Conclusion: Histological type is a key prognostic factor for breast cancer survival, with non-ductal tumors indicating poorer outcomes. The log-normal model provides the best statistical fit for this analysis.