Assessment of smell function in clinical routine is often limited due to a lack of time and/or costs of the personnel administering the test. The aim of the present study was to validate a procedure allowing for self-administered olfactory testing in a clinical setting. Seventy-four healthy subjects (13 male, 61 female) from 18 to 30 years of age (mean 20.3 years) were tested on 2 days (interval 7-21 days, mean 8.7 days) with 16 odors of the "Sniffin' Sticks" identification test kit. On one occasion, the test was administered by an examiner. On another occasion, subjects administered the test to themselves, with the odors being identified after they had been "painted" on a sheet of paper. No significant differences were obtained between the results from both test procedures. With a maximum score of 16, assisted testing yielded a mean score of 13.7 [standard deviation (SD) 1.3] while the self-administered procedure yielded an average score of 13.8 (SD = 1.5) (P = 0.72). The mean difference between the assisted and the self-administered smell test procedures was 0.05 (SD = 1.28). The 95% confidence interval of differences ranged from -2.51 to 2.61. These results suggest that odor identification with the Sniffin' Sticks can also be administered by the subjects themselves.