PREDICTORS OF AN INCREASED RISK OF FUTURE HYPERTENSION IN UTAH - A SCREENING ANALYSIS

A prospective study on 1,482 adult members of 98 Utah pedigrees was carried out to determine which variables may be associated with an increased risk of hypertension incidence. After an average of 7 years of follow-up, 40 individuals had been placed on antihypertensive medications to lower blood pressure. Baseline study variables included anthropometrics, clinical chemistry measurements of blood and urine, socioeconomic and lifestyle variables, and detailed erythrocyte ion transport and concentration measurements. Age (relative risk of 4.28 for a 2 SD difference, p < 0.0001) and baseline systolic and diastolic blood pressures (relative risks of 3.55 and 3.52, respectively, both p < 0.0001) had the strongest associations with hypertension incidence. Controlling for age and baseline blood pressure, the following age- and sex-adjusted variables were associated with an increased risk of future hypertension (relative risks for a 2 SD difference, all p < 0.10): family history of hypertension (2.35); height (1.97); body mass index (2.31); abdominal girth (2.66); subscapular, suprailiac, and triceps skinfold thicknesses (2.79, 2.52, and 2.28, respectively); percent ideal body weight (2.63); log triglyceride concentration (2.02); plasma uric acid (2.16); inorganic phosphate (0.50); and passive erythrocyte sodium permeability (1.59). The final model, which included all of the age- and sex-adjusted variables (p < 0.10) in a backward elimination logistic regression analysis, consisted of age (4.78), systolic blood pressure (2.91), subscapular skinfold thickness (2.21), height (1.92), uric acid (2.06), inorganic phosphate (0.50), and family history of hypertension (1.82). None of the ion transport or concentration measurements was associated with an increased risk of hypertension. We conclude that the risk factors of age, body fat and size, plasma uric acid, low plasma inorganic phosphate, and family history of hypertension, previously shown to be significant in cross-sectional studies, prospectively increase the risk of hypertension. If other variables have important predictive roles, then they may affect only a subset of this or other large study populations.