We examined clinical predictors of hypertension onset among individuals with new-onset hypertension in the community and developed a risk score that included the following components, which are easily assessed in a physician's office in primary care settings: age, sex, systolic blood pressure, diastolic blood pressure, body mass index, parental hypertension, and cigarette smoking. Our risk score for hypertension had very good discrimination and calibration, and assessment of overoptimism suggests that the model will probably do well when applied to a different sample. Physicians can use the hypertension risk score to measure an individual's estimated risk for hypertension, to inform patients of their risks and help guide their choice of nonpharmacologic measures to prevent hypertension, and to aid in clinical counseling and decision making. In addition, the risk prediction score may be useful in designing interventions to prevent high blood pressure by aiding in selection of participants at the highest risk for hypertension, who are most likely to benefit from treatment. Given that hypertension was measured on only 1 occasion, however, our score may somewhat overestimate an individual's true risk for hypertension.