OBJECTIVE: To evaluate, among adolescents 10 to 17 years of age with an incident hypertensive blood pressure (BP; >/=95th percentile) at a primary care visit, whether TeenBP, a novel electronic health record-linked clinical decision support tool (CDS), improved recognition of elevated BP, and return for follow-up BP evaluation. METHODS: We conducted a pragmatic cluster randomized trial in 20 primary care clinics in a large Midwestern medical group. Ten clinics received the TeenBP CDS, including an alert to remeasure a hypertensive BP at that visit, an alert that a hypertensive BP should be repeated in 1 to 3 weeks, and patient-specific order sets. In the 10 usual care (UC) clinics, elevated BPs were displayed in red font in the electronic health record. For comparisons between CDS and UC we used generalized linear mixed models. RESULTS: The study population included 607 CDS patients and 607 UC patients with an incident hypertensive BP. In adjusted analyses, at the index visit, CDS patients were more likely to have their hypertensive BP on the basis of >/=2 BP measurements (47.1% vs 27.6%; P = .007) and to have elevated BP (International Classification of Diseases, Ninth Revision code 796.2) diagnosed (28.2% vs 4.2%; P < .001). In a multivariate model adjusted for age, sex, systolic BP percentile, and visit type, rates for repeat BP measurement within 30 days were 14.3% at TeenBP CDS clinics versus 10.6% at UC clinics (P = .07). CONCLUSIONS: The TeenBP CDS intervention significantly increased repeat BP measurement at the index visit and recognition of a hypertensive BP. Rates for follow-up BP measurement at 30 days were low and did not differ between TeenBP and UC subjects.