Guidelines informing chemotherapy regimen selection are based on clinical trials with participants who do not necessarily represent general populations with breast cancer. Understanding who receives non-guideline regimens is important to understanding real-world chemotherapy administration and how it relates to patient outcomes. Using data from the Optimal Breast Cancer Chemotherapy Dosing (OBCD) cohort study, based at Kaiser Permanente Northern California (2006-2019) and Kaiser Permanente Washington (2004-2015), we use logistic regression to examine the associations between patient characteristics and receipt of non-NCCN-guideline chemotherapy among 11,293 women with primary stage I-IIIA breast cancer receiving chemotherapy. Use of non-guideline regimens was strongly associated with several factors, including older age (OR≥80 vs 18-39: 5.25, 95%CI: 3.06-9.00)(p-trend=0.002) and human epidermal growth factor-2 status (ORHER2+ vs HER2-: 3.44; 95%CI: 3.06-3.87) and was less likely in women with larger tumor size (OR>5cm vs 0.1-≤0.5cm: 0.56; 95%CI: 0.36-0.87)(p-trend=0.01) and diagnosed in later years (OR2012-2019 vs 2005-2011: 0.80; 95%CI: 0.71-0.90). Factors associated varied by type of non-guideline regimen. For example, women with comorbidity and older age were more likely to receive non-guideline drug combinations in particular, while women with larger tumor size were less likely to receive non-guideline administration schedules. Non-guideline chemotherapy regimens are more likely in certain patient populations. These associations highlight that vulnerable patient populations may be less likely to receive guideline care and thus real-world studies are essential to understanding how the use of non-guideline regimens impacts patient outcomes in these groups.