OBJECTIVES: To review trials of physician-centered interventions to reducing systolic blood pressure (SBP) and diastolic blood pressure (DBP). METHODS: Systematic review and meta-analysis. We searched MEDLINE, EMBASE and Cochrane Central for all-language articles up to September 2014. We included randomized controlled trial (RCT) of physician-centered interventions for hypertension compared with usual care or minimal intervention in primary care patients. Data were pooled using a random effect meta-analysis model. The effect were expressed as the weighted mean difference (WMD).  RESULTS: Twenty-five trials of 7595 citation were included.  Seventeen studies were cluster RCT, one trial was factorial and cluster trial. The remaining seven studies were randomized at individual patient level; five of them used a two-by-two factorial design. Two studies did not report any estimates of variance. Overall, 23 trials (43.489 participants) was contribute to the meta-analysis. The physician-centered intervention were categorized as computer decision support (6 trials), stepped treatment algorithm (6), Medical Education (4), Audit and feedback (3) and Multifaceted (4). Methodological quality of included studies was rather low.  Only interventions that the main focus were stepped treatment algorithm showed significant reductions in blood pressure: weighted mean difference, systolic - 4.2 mmHg; 95% confidence interval -5.3 to -3.2; I², 80.1% and diastolic -1.6 mmHg; 95% confidence interval -2.8 to -0.49; I². 93.4%. For the remaining five categories did not show to reducing blood pressure. Subgroup analyses by study design explained considerable heterogeneity in stepped treatment algorithm effect. CONCLUSIONS: Physician-centered interventions based in stepped treatment algorithm showed significantly reductions of systolic and diastolic blood pressure. The magnitude of reduction in blood pressure is likely to prevent stroke and death in patients.