Tinnitus Handicap Inventory

Timothy C. Hain, MD • Page last modified: July 30, 2022Questionnaire index

There are innumerable questionnaires for tinnitus, and the THI is a popular one. The THI is also called the "THQ" for tinnitus handicap questionnaire.

A PDF version of the THI is here.Online version

Overview of the THI test.

The tinnitus handicap inventory was developed by the same authors who developed the dizziness handicap inventory, "To develop a self-report tinnitus handicap measure that is brief, easy to administer and interpret, broad in scope, and psychometrically robust." (Newman et al, 1996)

The THI is not a diagnostic test -- rather it produces scores related to the perceived tinnitus handicap -- a subjective quantity.

The THI is a 25 item test, and the maximum score is 100.

Salviati et al (2013) reported that scores above 36 suggest psychiatric distress.

Wakabayashi et al (2020) suggested that while there is a total score, followed by "emotional, functional and catastrophic subscales", the data were "extremely biased to one factor". Baguley and Andersson (2003) came to a similar conclusion, and wrote "The factor analysis yielded strong support for a unifactorial structure of the scale, with a majority of items loading on the first factor, and high internal consistency of the total score. The authors recommend use of the total score in research and in clinical practice." So it would seem that there is no need to dwell much on the subscales, but the total scale is sufficent.

Some aggregate data:

We also have a large repertoire of dizzy patients in our clinic practice with tinnitus, where we have obtained the THI. For these, the distribution is much different than for the DHI (dizziness handicap inventory) . This data is from the beginning of 2018(n=854). These are mainly people who are dizzy, and whose chief complaint is not tinnitus at all. There may be a few who are mainly being seen for tinnitus however. Note that this distribution is not a "normal" distribution. Rather it is skewed towards high numbers. This means it does not have very good statistical properties. This highly skewed distribution is similar to that of the "MIDAS" questionnaire for migraine.


General References: