Convergent validity of the tinnitus handicap inventory and the tinnitus questionnaire

For research into tinnitus to be robust and credible, the use of well-valid ated instruments of self-perceived tinnitus handicap as outcome measures is essential. The tinnitus handicap inventory (THI) and the tinnitus question naire (TQ) are two such instruments which are in widespread use. Both quest ionnaires were administered by mail to 100 consecutive new patients of the Cambridge Tinnitus Clinic, and completed in randomized order. These patient s had been referred by the otolaryngology team and had not undergone any ti nnitus therapy. The response rate was 78 per cent, neither questionnaire be ing more acceptable to patients than the other. The convergent validity of the instruments was high, with total and subscale scores all being signific antly correlated at the five per cent level (Spearman correlation coefficie nts). A number of subscale scores were not significantly correlated at the one per cent level however. In particular, the sleep disturbance element of the TQ was demonstrated to have some discriminant validity from the THI an d from other elements of the TQ at the one per cent significance level. The THI and TQ have been demonstrated to have high convergent validity and are both suitable for tinnitus outcome studies involving the quantification of self-perceived tinnitus handicap. For research that aims to determine the specific effect of an intervention on tinnitus-related sleep disturbance, t he TQ sleep subscale has potential utility. The hypothetical constructs of tinnitus handicap underlying the psychologist-developed TQ and the audiolog ist-developed THI have been shown to be convergent.