Abstract only Background: Responses to new anti-hypertensive therapy (AHT) can differ. Cluster analysis of longitudinal systolic blood pressure (SBP) data allows identification of individuals with similar trajectories, which enables hypothesis generation to potentially explain the observed patterns. We examined electronic health data from U.S. Veterans initiating AHT from 2002-2009. Methods: SBP was tracked for 1 year before and up to 2 years after AHT initiation. Subjects were clustered using a K-means approach with the mean of the squared Euclidean distances as a distance metric and evaluated for the optimal number of clusters. Thin-plate splines were used to display the smoothed trajectories of each group with predicted SBP measures over time. Results: A total of 45,598 subjects contributed 783,852 SBP measurements. Five clusters of subjects with similar trajectories were produced, providing visualization of SBP Figure 1 prior to initiation, immediately after starting therapy, and after longer treatment duration. For example, Group 1 had the lowest mean age, lowest HDL-C, lowest LDL-C, highest triglycerides, lowest baseline SBP, lowest insulin use, and highest statin use. Conclusions: Trajectory clustering for SBP identifies distinct response groups that differ in response to therapy, laboratory measures, and medication use. Future analyses can examine anti-hypertensive medication use, compliance and genetic factors to identify potential causes for these trajectories. These cluster analyses can provide new analytical approaches related to risk factor diagnosis and treatment.