Background: Hypertension (HT) occurs in 1-3% of adolescents and is associated with long term cardiovascular morbidity. Although blood pressure (BP) is routinely measured, HT in adolescents is often missed during outpatient visits. This study sought to evaluate whether “TeenBP”, an electronic health record (EHR) linked, webbased clinical decision support tool, could improve recognition and early management of HT in adolescents. Methods: We randomized 20 primary care clinics to receive the CDS or to continue usual care (UC). At intervention sites, TeenBP was activated at the point of care, when a BP was elevated. TeenBP graphically displayed systolic and diastolic BP values and percentiles over the prior 2 years, identified patients meeting criteria for HT (3 or more BPs ≥95th percentile), provided a summary of comorbidities and medications that may affect BP, and offered patient-specific order sets. HT recognition, within 6 months of meeting criteria, was defined as adding HT or elevated BP (EBP) to the problem list or clinical notes, indicating HT or EBP as an outpatient visit diagnosis, prescribing a medication to lower BP, or diagnostic testing related to HT, identified through chart review. Generalized linear mixed models were used to test the effect of the intervention. Results: Among 21,618 patients age 10-17 years with a primary care visit at the 20 study clinics over one year (May 2014 – April 2015), 315 (1.5%) met criteria for new onset HT. Most BPs were modestly elevated, < 99th percentile. Clinical recognition of HT within 6 months occurred for 36.2% (27.7 – 45.8) of patients in UC clinics and 68.0% (56.6 – 77.6) in the CDS clinics (p=0.0003). Clinical recognition of HT in the TeenBP clinics was most often met by adding HT or EBP as an outpatient visit diagnosis (52.2%) or to the text of the clinical note (52.4%). Within 6 months of meeting criteria for HT, <10% of TeenBP or UC subjects had an echocardiogram or renal ultrasound and only one patient initiated an antihypertensive medication. Conclusion: We observed a large and statistically significant beneficial effect of this clinical decision support system on recognition of new onset hypertension, without increasing diagnostic workups or early initiation of antihypertensive medication.