Tinnitus questionnaires score the burden tinnitus places on daily life, mostly distress and disruption rather than the sound, and they are not interchangeable: the THI and TFI are the trial standards, and the TQ leads in German-speaking countries.
With no scan or blood test for tinnitus, clinics and trials rely on what people report. How tinnitus is measured covers the THI and TFI in depth; this page is the wider map. Entries with no threshold for meaningful change have none in the published sources behind them.
Why there are so many questionnaires
A 2016 systematic review of 228 tinnitus treatment studies published from 2006 to 2015 found 78 different primary outcome instruments, 24 of them patient-reported tools. The Tinnitus Handicap Inventory was the most used, chosen as a primary outcome 77 times, and one study in ten did not clearly say which instrument it had used.
A 2026 analysis broke the 199 items of eight tinnitus-burden questionnaires into 83 distinct symptoms. Of those, 41 appeared in only one questionnaire, and none appeared in all eight. The closest pair, the TFI and THI, still overlapped only weakly, with a Jaccard index of 0.35. The authors concluded that the questionnaires should not be assumed to be interchangeable, and noted that none of the eight asks about dizziness or hyperacusis. They added that studies seldom justify their choice of questionnaire; legacy data, region, language and copyright access may play a part.
Their totals do move together: in a 2012 international analysis of 1,318 patients, the questionnaires studied correlated at around 0.8. They can still disagree about who improved.
Tinnitus Handicap Inventory (THI)
- Measures: the handicap tinnitus causes in daily life, in functional, emotional and catastrophic subscales. In its development study it correlated only weakly with rated pitch and loudness.
- Items and scoring: 25 questions answered yes, sometimes or no, each worth 0, 2 or 4 points, for a total of 0 to 100 (the THI entry has the development and grading studies).
- Bands: 0–16 very mild, 18–36 mild, 38–56 moderate, 58–76 severe, 78–100 catastrophic (totals are always even). They follow five grades proposed by a British working group in 2001, which called the lowest slight and said that across the general population the top two should be very rare. Among people seeking treatment they need not be: 31% of participants in the 2025 study's trial scored in them.
- Reliability: internal consistency of 0.93 in the 1996 development study. A 1998 retest study of 29 adults found that one person's score had to move 20 points or more to be a statistically significant change.
- Meaningful change: a 2011 anchor-based study of 210 patients put it at 7 points. A 2025 multi-centre study estimated 11 points after 12 weeks (estimates ranged from 7.8 to 12), and found that people who start out more severe need a larger drop.
- Languages: validated in Chinese, Danish, French, German, Italian and Portuguese, among others, and it performed comparably across four languages in the 2012 analysis.
- Used for: grading impact in clinic and as a trial outcome, although it was not originally built to measure change. A shorter screening version, the THI-S, correlates with the full inventory at 0.90.
Tinnitus Functional Index (TFI)
- Measures: severity and negative impact, and it was built specifically to detect change with treatment. Its eight subscales are intrusiveness, sense of control, cognitive interference, sleep, auditory difficulties, relaxation, quality of life and emotional distress.
- Items and scoring: 25 items, each rated 0 to 10. The average is multiplied by 10 for a total of 0 to 100.
- Bands: 0–17 not a problem, 18–31 small problem, 32–53 moderate problem, 54–72 big problem, 73–100 very big problem.
- Reliability: internal consistency 0.97, test–retest 0.78, and a correlation with the THI of 0.86.
- Meaningful change: the developers set 13 points as a preliminary criterion in 2012. The 2025 study estimated 9 points (7.3 to 9.4) and argued that 13 reflects moderate rather than minimal improvement. Earlier estimates ranged from 4.8 to 22.4 points, depending on method. In 2024, members of the development team still recommended 13 points for an individual, alongside the person's own sense of improvement.
- Languages: by 2016 it was being translated into at least 14 languages.
- Used for: trial outcomes and tracking in clinic; the 2024 VA/DoD guideline names it and the THI as validated measures for monitoring care, as the VA/DoD guideline in plain English explains.
Tinnitus Questionnaire (TQ)
- Measures: tinnitus-related distress, across emotional and cognitive distress, intrusiveness, auditory perceptual difficulties, sleep disturbance and somatic complaints. It traces back to questionnaires developed in London in 1988.
- Items and scoring: 52 statements answered true, partly true or not true, each worth 0, 1 or 2 points. The German scoring uses 40 of them, counting two twice, for a total of 0 to 84.
- Bands: in German practice, 46 or below counts as compensated tinnitus, and 47 or above as decompensated tinnitus with high burden and relevant impairment.
- Reliability: test–retest 0.94 for the total score in German inpatients.
- Meaningful change: an anchor-based study of 757 patients put it at 5 points for improvement and 1 point for worsening. Some studies use a stricter 12-point threshold.
- Languages: English original, validated in German, Dutch, French and Chinese.
- Used for: it is the most widely used questionnaire in German-speaking areas. It is also the longest of the burden questionnaires and the most distinctive: in the 2026 analysis, 52.5% of its symptoms appeared in no other scale, including the only somatic items, such as headache.
Mini-TQ
- Measures: the same distress as the TQ, as one total with no subscales.
- Items and scoring: 12 TQ items chosen in 2004 for reliability and sensitivity to change, scored 0 to 24.
- Bands: 8–12 moderate, 13–18 severe, 19–24 very severe. Lower scores count as compensated.
- Reliability: correlation with the full TQ above 0.90, and test–retest 0.89. Treatment effects came out slightly larger on the Mini-TQ than on the full TQ.
- Meaningful change: the development study's summary gives no threshold.
- Versions: developed on German patients, and several short versions of the German TQ now exist, including a newer 15-item form, so check which one a study used.
- Used for: rapid, economical assessment.
Tinnitus Reaction Questionnaire (TRQ)
- Measures: psychological distress linked to tinnitus, in four factors: general distress, interference, severity and avoidance.
- Items and scoring: 26 items, each scored from 0 (not at all) to 4 (almost all of the time), for a total of 0 to 104.
- Reliability: test–retest 0.88 and internal consistency 0.96, across three samples totalling 156 people. It showed moderate to high correlations with anxiety and depression measures, from 0.58 to 0.87.
- Bands and meaningful change: the development study's summary gives neither.
- Languages: English, developed in Sydney, and French, where the effects of language were minor.
- Used for: distress-focused research. In the 2016 review it was the preferred primary outcome questionnaire in Oceania, though few trials came from there; the THI led in every other region. It is one of three instruments pooled in the wait-list analysis in what happens if you do nothing.
Three less common instruments: THQ, TPFQ and TSI
- Tinnitus Handicap Questionnaire (THQ), 1990: 27 items in three factors: the physical, emotional and social consequences of tinnitus; hearing ability; and the person's view of tinnitus. Its developers recommended scoring only the first two, because the third had low internal consistency.
- Tinnitus Primary Function Questionnaire (TPFQ), 2014: built around the activities tinnitus most impairs: emotions, hearing, sleep and concentration. It has 20 items and a 12-item version, and it correlated with the THQ at 0.77.
- Tinnitus Severity Index (TSI): measures the negative effects of tinnitus on life and scores from 12 to 60.
Single rating scales: VAS and NRS
Single-item scales ask for one rating of loudness, annoyance or a similar quality, on a visual analogue scale (VAS) or a numeric rating scale (NRS).
- A 2012 analysis found VAS loudness and annoyance ratings reliable on retest, at 0.8 and 0.79, and they correlated with the TQ at up to 0.67. It put meaningful change at 10 to 15 points on its own scales, whose range the summary does not give, so the figure cannot be applied to other scales.
- Rated loudness is not measured loudness. In 140 patients, VAS loudness tracked loudness matched in a sound booth only in people with normal hearing.
- Loudness was the most commonly reported primary outcome domain in the 2016 review (14%), and a numeric rating scale was the most common way to measure it (8%).
- The 2024 VA/DoD guideline suggests against psychoacoustic measures, such as loudness matching and minimum masking level, for monitoring whether management works.
- A 2018 consensus exercise with 719 participants made loudness a core outcome only for drug trials; tinnitus intrusiveness was core for all three treatment types studied.
ESIT-SQ: a profile, not a score
The European School for Interdisciplinary Tinnitus Research Screening Questionnaire, published in 2019, is a case-history instrument rather than a severity scale. It records risk factors, medical and ear history and tinnitus characteristics, and people without tinnitus can answer it too, so researchers can compare the two groups.
- Structure: 56 closed questions. Part A has 17 for everyone; Part B adds 22 for people who have had tinnitus lasting more than five minutes in the past year; an optional Part O has 17 more.
- Scoring: it is not built to produce a severity total. Only some items give a measure, including one general question on the impact of tinnitus.
- Languages: written in English, then translated into Dutch, German, Italian, Polish, Spanish and Swedish, with validated French and Portuguese versions and an Albanian translation since.
- Used for: profiling and subgrouping people with tinnitus in research.
Same patients, different verdicts
In a 2021 study, 210 people with chronic tinnitus filled in the TQ, THI and TFI before and after a 7-day intensive treatment programme, though not everyone had complete scores on all three. The totals agreed well, with intraclass correlations of 0.72 to 0.83. But each questionnaire's own threshold for improvement (12 points on the TQ, 7 on the THI, 13 on the TFI) picked out different people: about 20% were responders on the TQ, 19% on the TFI and 38% on the THI. Agreement on who had responded was only fair to moderate.
The 2025 study found a related trap. Patients who said their tinnitus had not changed still improved by an average of 8.7 THI points and 5.9 TFI points. Scores can drift down without anyone feeling better, which is why the first check in reading a tinnitus study is whether a change was measured against a control group.
So when one trial reports a responder rate on the THI and another on the TFI, the difference may be the ruler rather than the treatment.
Which questionnaire you are likely to meet
- Most trials and many clinics: the THI or TFI.
- German-speaking clinics and research: the TQ or a Mini-TQ.
- Research profiling: the ESIT-SQ.
Ask for a baseline score and the same questionnaire at follow-up, because a score on one instrument cannot be read as a score on another. Judge any change against that instrument's own threshold, and against your starting severity. What a first visit involves is set out in your first audiology appointment and how tinnitus is diagnosed.
A high score on any of these is a reason to ask for help, not a diagnosis. For treatments aimed at distress rather than the sound, see CBT for tinnitus and the treatment comparison. Some symptoms need prompt medical attention whatever a questionnaire says; the when to seek care checklist lists them.
The impact self-check is not one of these
The impact self-check on Tinnitus Clarified asks 15 original questions about concentration, sleep, mood and sense of control, each answered never, sometimes or often, and gives a total out of 30 in one of four bands. Its page states that the questions have not been through the testing that validates a clinical questionnaire. Use it to find words for how tinnitus affects you before an appointment, not as a THI or TFI score.