OBJECTIVES:: Poor blood pressure control contributes to increased cardiovascular morbidity and mortality in dialysis patients. Few studies have examined sub-optimal medication use patterns, such as antihypertensive medication discontinuation and patient-related factors that could influence such behaviors, particularly in patients on dialysis. METHODS:: This analysis is based on the BRAZPD study, a national representative peritoneal dialysis (PD) cohort, collecting prospective data from 2004-2011. The cohort included 9,905 patients; of which 7,007 were PD incident. Medication non-persistence was defined as patients who had medication prescription gaps for 90 days or higher. Cox proportional hazards regression models were used to estimate associations between medication non-persistence and sociodemographic factors, adjusting for potential confounders, allowing medication status to change each month. Patients were censored if they left the study for reasons other than death (recovery of renal function, transplant, hemodialysis transfer and loss to follow-up). RESULTS:: Nearly 86% of the population was on some type of antihypertensive (ACEI=59%, ARB =35%, BB=51%, CCB=48%, and Diuretics=58%) during a mean of 17.2 months (maximum follow up of 6 years). The mean persistence (time without a 90 day or more medication gap) was 13.3 months (SD: 15.7) with an overall non-persistence rate of 44%. There were significant differences in non-persistence across medication classes with angiotensin blockers having the greatest rates of non-persistence. In multivariable models, the primary predictors of medication non-persistence were older age, having significant comorbidity and patients starting on peritoneal dialysis (all p<0.05). These predictors remained similar across different antihypertensive medication classes. CONCLUSIONS:: Antihypertensive medication non-persistence remains a significant issue in patients on PD, and according to the identified risk factors, may reflect clinical instability, especially in patients with comorbidity and among those transitioning to dialysis. Interventions are needed especially among high risk populations in order to improve medication use behaviors and in turn, reducing morbidity and mortality in dialysis patients.