The aim of this study was to validate the results of diagnostic pure-tone audiometry performed in a typical practice setting by comparing with test results obtained in a standardized audiological setting in accordance with the ISO standards. In a single-blinded crossover design, 119 persons were tested (0.25-8 kHz) in both settings. The mean deviations as a function of frequency were in the order of less than 2 dB (0.5-4 kHz) and otherwise up to 4 dB; the practice setting representing the poorer thresholds. The validity of the practice audiometry at three criteria of hearing impairment (0.5-4 kHz) was characterized by sensitivity (82-100%), specificity (95-99%); positive predictive values (75-90%) and negative predictive values (98-100%) focusing on the better ear. It is concluded that pure-tone audiometry of appropriate validity can be performed in general practice and that it is useful in selecting patients with no need of further audiological examination. Guidelines are needed.