BACKGROUND A number of alterations of ion transport and concentrations have been reported in erythrocytes from patients with hypertension and their young normotensive offspring. Of these, only sodium-lithium countertransport appears to be capable of predicting the effects of genetic variation on interindividual differences in blood pressure. OBJECTIVE As a first step toward identifying other erythrocyte traits capable of predicting the effects of genetic variation on interindividual differences in blood pressure, we evaluated whether furosemide-sensitive sodium or potassium cotransport, ouabain-sensitive sodium pump, passive permeability to sodium or potassium, or intracellular concentrations of sodium or potassium are predictors of the probability of having hypertension in a general Caucasian population; we also attempted to confirm our previous inference that sodium-lithium countertransport is a predictor of the probability of having hypertension in this population. METHODS The sample for analyses consisted of 199 unrelated men, aged 56-91 years, and 216 unrelated, nonpregnant women, aged 53-87 years, from the population of Rochester, Minnesota. Logistic regression was used to assess the relationship between predictor traits and blood pressure diagnostic status (normotension or hypertension). We assessed whether each erythrocyte trait was a predictor of the probability of having hypertension when the trait was considered by itself and when it was considered after the effects of other identified predictors of the probability of hypertension were included in the logistic regression models. RESULTS When each erythrocyte trait was considered by itself, the predictors of the probability of having hypertension were sodium-lithium countertransport in men and sodium-lithium countertransport, sodium and potassium cotransport, and intracellular sodium concentration in women. When each erythrocyte trait was considered after other predictors had been included in the logistic regression models, none of the erythrocyte traits made an additional contribution to prediction of the probability of having hypertension unless its interactions with other predictors were also included in the models. CONCLUSIONS Multiple measures of erythrocyte ion transport are associated with hypertension in the general Caucasian population; however, most of these associations are context-dependent inasmuch as they are gender-specific and dependent on levels of other predictor traits.