OBJECTIVE: Recent data suggest that as many as 50% of older adults may have hypertonic plasma, an indicator of cell dehydration that predicts a range of adverse outcomes. To determine if a prevalence of this magnitude could be real, this study used nationally representative data to estimate the prevalence of hypertonicity, and to test for biologically plausible associations between hypertonicity, older age, glucose dysregulation, hemoconcentration, reduced bioelectrical impedance, and water intake. DESIGN: Cross-sectional. SUBJECTS: Community-dwelling adults (aged 20 to 90 years) who gave blood as part of the Third National Health and Nutrition Examination Survey (N = 14,855). STATISTICAL ANALYSES: Plasma tonicity was estimated from glucose, sodium, and potassium values. The weighted prevalences of mild (295 to 300 mmol/L) and overt hypertonicity (> or = 300 mmol/L) were estimated by age, sex, race/ethnicity, fasting, and glycemic status. Hyper- and normotonic (285 to 295 mmol/L) groups were compared with respect to elevated blood chemistry values, bioelectrical impedance analysis parameters, as well as total water intake (grams, % Adequate Intake, and grams per kilogram body weight), using multivariable models that adjusted for age, sex, race/ethnicity, and survey design. RESULTS: Mild and overt hypertonicity were observed in 40% and 20% of the sample, respectively. Hypertonicity was positively associated with older age, Hispanic and African-American race, impaired glucose tolerance, diabetes, and hemoconcentration, and inversely associated with bioelectrical impedance analysis parameters. Hypertonicity was associated with greater total water intake in younger adults, but decreased intake in older adults. CONCLUSIONS: Clinicians and researchers should be alert to hypertonicity and its causes in older adults.