Measurements of plasma viscosity (PV) and erythrocyte sedimentation rate (ESR) are supposed to reflect the complex of acute phase reactants in inflammations. Both tests were prospectively studied at the rheumatology out-patient department with regard to their ability to discriminate between inflammatory and non-inflammatory rheumatic diseases in new patients (n = 235). PV and ESR were measured using the Coulter Viscometer II and the Westergren method, respectively. The Receiver Operating Characteristic curve for PV was slightly better than for ESR. However, at the higher cut-off points (PV greater than 1.81 mPa.s, ESR greater than 25 mm/l hr), sensitivities, specificities, predictive values and odds ratios were more favourable for ESR. Furthermore, since the costs of PV measurement are considerably higher, there is no reason for an implementation of PV in the daily routine of the rheumatologist at the outpatient department.