The reliability and validity of the vestibular autorotation test (VAT) in a clinical sample of subjects with complaints of dizziness

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Physical Therapy

First Committee Member

Kathryn E. Roach, Committee Chair


Purpose. To establish intra-rater and inter-rater reliability and diagnostic validity of the VAT in patients complaining of dizziness. Methods. Ninety-eight (N = 98) subjects participated from a patient testing pool within a tertiary care setting that specializes in assessment and management of dizziness. Reliability. A sub-sample of subjects (n = 69) performed 6 sequential horizontal VAT trials. In 49 of these subjects, an additional rater performed a second series of 6 trials. Diagnosis. Sensitivity and specificity of the VAT were calculated against diagnosis as determined by caloric testing (n = 92) and rotary chair testing (n = 47). In patients with unilateral caloric asymmetry (>25%) (n = 37), comparison was made to asymmetry as determined by the VAT (>5%). Results. Reliability. Intra-rater reliability for gain values ranged from 0.95(95%CI 0.93--0.97) to 0.82 (95%CI 0.75--0.88) when 6 trials were utilized to calculate these values. Phase and asymmetry reliability was found to be influenced by the number of practice trials a subject performed. The last three trials for the VAT variables phase and asymmetry revealed good reliability when subjects were given three trials to practice, ICC range: Phase 0.96(95% CI 0.92--0.98) to 0.88(95%CI 0.84--0.91); Asymmetry 0.88(95%CI 0.72 --0.91) to 0.73(95%CI 0.61--0.81)). Inter-rater reliability for gain ranged from 0.95 (95%CI 0.93--0.97) to 0.82 (95% CI 0.75--0.88). Inter-rater reliability for phase ranged from to 0.91(95%CI 0.87--0.94) to 0.50 (95%CI 0.06--0.66). Inter-rater reliability for asymmetry ranged from 0.91(95%CI 0.86--0.94) to 0.72(95%CI 0.59--0.81). Diagnosis. Maximum sensitivity was achieved when the VAT diagnostic criterion of two or more mean gain values +/-1 s d or phase were used to define abnormal VAT performance, sensitivity 0.81(95%CI 0.73--0.89); specificity was 0.28(95%CI 0.18--0.37)). Specificity was maximized using 2 or more mean gain & phase values in series, specificity 0.83(95%CI 0.76--0.90); sensitivity 0.33(95%CI 0.23--0.43). Conclusions. Improved reliability of the VAT can be achieved by allowing subjects a minimum of three practice trials; second, subjects must produce head movements of ≥3.9 Hz. It appears that the VAT in isolation cannot produce a definitive diagnosis of peripheral vestibulopathy.


Health Sciences, Audiology; Health Sciences, Rehabilitation and Therapy

Link to Full Text