BACKGROUND: Identifying characteristics of primary care practices that perform well on cardiovascular clinical quality measures (CQMs) may point to important practice improvement strategies. OBJECTIVE: To identify practice characteristics associated with high performance on four cardiovascular disease CQMs. DESIGN: Longitudinal cohort study among 211 primary care practices in Colorado and New Mexico. Quarterly CQM reports were obtained from 178 (84.4%) practices. There was 100% response rate for baseline practice characteristics and implementation tracking surveys. Followup implementation tracking surveys were completed for 80.6% of practices. PARTICIPANTS: Adult patients, staff, and clinicians in family medicine, general internal medicine, and mixedspecialty practices. INTERVENTION: Practices received 9 months of practice facilitation and health information technology support, plus biannual collaborative learning sessions. MAIN MEASURES: This study identified practice characteristics associated with overall highest performance using area under the curve (AUC) analysis on aspirin therapy, blood pressure management, and smoking cessation CQMs. RESULTS: Among 178 practices, 39 were exemplars. Exemplars were more likely to be a Federally Qualified Health Center (69.2% vs 35.3%, p = 0.0006), have an underserved designation (69.2% vs 45.3%, p = 0.0083), and have higher percentage of patients with Medicaid (p < 0.0001). Exemplars reported greater use of cardiovascular disease registries (61.5% vs 29.5%,), standing orders (38.5 vs 22.3%) or electronic health record prompts (84.6% vs 49.6%) (all p < 0.05), were more likely to have medical home recognition (74.4% vs 43.2%, p = 0.0006), and reported greater implementation of building blocks of high-performing primary care: regular quality improvement team meetings (3.0 vs 2.2), patient experience survey (3.1 vs 2.2), and resources for patients to manage their health (3.0 vs 2.3). High improvers (n = 45) showed greater improvement implementing team-based care (32.8 vs 11.7, p = 0.0004) and population management (37.4 vs 20.5, p = 0.0057). CONCLUSIONS: Multiple strategies-registries, prompts and protocols, patient self-management support, and patient-team partnership activities-were associated with delivering high-quality cardiovascular care over time, measured by CQMs.