Differences in characteristics between people with tinnitus that seek help and that do not

Knowledge on characteristics of people that seek help for tinnitus is scarce. The primary objective of this study was to describe differences in characteristics between people with tinnitus that seek help compared to those who do not seek help. Next, we described differences in characteristics between those with and without tinnitus. In this cross-sectional study, we sent a questionnaire on characteristics in different domains; demographic, tinnitus-specific, general- and psychological health, auditory and noise- and substance behaviour. We assessed if participants had sought help or planned to seek help for tinnitus. Tinnitus distress was defined with the Tinnitus Functional Index. Differences between groups (help seeking: yes/no, tinnitus: yes/no) were described. 932 people took part in our survey. Two hundred and sixteen participants were defined as having tinnitus (23.2%). Seventy-three of those sought or planned to seek help. A constant tinnitus pattern, a varying tinnitus loudness, and hearing loss, were described more frequently in help seekers. Help seekers reported higher TFI scores. Differences between help seekers and people not seeking help were mainly identified in tinnitus- and audiological characteristics. These outcomes might function as a foundation to explore the heterogeneity in tinnitus patients.


Study design and population.
For this observational study we prospectively gathered data by means of a questionnaire send to Dutch adults. They were members of the Dutch Health Care Consumer panel 12 .
"The aim of the Nivel Dutch Health Care Consumer Panel (DHCCP) is to measure, at national level, opinions on and knowledge about health care and the expectations and experiences with health care. The Consumer Panel is a so-called 'access panel' . An access panel consists of a large number of persons who have agreed to answer questions on a regular basis. In addition, many background characteristics of these persons (for example age, level of education, income, self-reported general health) are known 12 ". At the time of this study (January 2020), the panel consisted of approximately 12,000 people aged 18 years and older. "From the access panel samples can be drawn for every separate survey. It is not possible for people to sign up on their own initiative. The panel is renewed on regular basis. Renewal is necessary to make sure that members do not develop specific knowledge of, and attention for, healthcare issues, and that no 'questionnaire fatigue' occurs. Moreover, renewal compensates for panel members who, for example, have died or moved without informing the panel about the new address 12 ".
This study is part of a larger study on tinnitus prevalence, characteristics and health care usage. The study sample therefore entails those DHCCP panel members (N = 2251), who agreed to combine their survey answers with health care consumption data as registered by their general practitioner 13 . A previously published study on tinnitus prevalence was based on the same data 14 .
Outcome assessment. Logistics. A questionnaire was sent to panel members of the Dutch Health Care Consumer Panel. The questionnaire was sent via postal services or online. This depended on the preference of the panel member. The postal survey was sent on 14-01-2020. One postal reminder was sent on 30-01-2020. The online survey was sent on the 16-01-2020. Two online reminders were sent on 23-01-2020 and 30-01-2020. The survey was closed on 14-02-2020. Please find the questionnaire in Supplementary Methods S2.
Questionnaire outline. The survey was created by MR (medical doctor), AS (ENT surgeon), IS (epidemiologist) and AB (senior researcher Nivel). Characteristics among different domains were collected: demographic, tinnitus-specific, general health, psychological health, audiological, and noise-and substance behaviour. The full questionnaire can be found in Supplementary Methods S2. The overall survey structure was based on the European School for Interdisciplinary Tinnitus Research Screening Questionnaire (ESIT-SQ), which consists of two parts 7 . Part one consist of 17 questions regarding individual characteristics in people with or without tinnitus. The second part is only meant for people with tinnitus. It consists of 22 questions regarding tinnitus characteristics. The set-up of our survey was similar: a part to be answered by all participants, and a part that was specifically for those that had tinnitus. For the part to be answered by all participants, we directly used or used a variation on 13 of 17 questions of the ESIT-SQ part A. Survey items about the characteristics of tinnitus were based on the Tinnitus Sample Case History Questionnaire (TSCHQ) and the ESIT-SQ part B 7,15,16 . The TSCHQ consists of 35 questions concerning tinnitus history and tinnitus characteristics. We did not use the full versions of one of both questionnaires due to space limitations. Questions were either an exact copy of one of two questionnaires or questions/answer options were combined. Twelve questions were based or an exact copy of the 22 questions of ESIT-SQ part B. Twenty-three questions were based on or an exact copy of the 35 questions of the TSCHQ.
Demographics. Demographic data were gathered when people became a member of the panel and were provided by Nivel for this study. These include data about educational level, marital status, social position (e.g. employed/unemployed/student), ethnicity, age (calculated at date of sending of the questionnaire), gender, and www.nature.com/scientificreports/ net income of the households of the participating panel members, self-reported general health and self-reported mental health.
Tinnitus classification and definitions. We assessed the presence of tinnitus with three questions. We described tinnitus as Tinnitus is the hearing of e.g. a beep, whistle, hissing, zoom or another sound without the actual presence of the sound in your surroundings. This can last for a very short amount of time or a whole day. First, the participants were asked whether they experienced tinnitus over the last year. Next, a question about duration was asked (tinnitus lasting < 5 min, 5-60 min, ≥ 60 min or continuously). The third question was about the frequency of the experienced sound (daily or almost daily, weekly, monthly, less than once a year). We subsequently defined people as having tinnitus when they experienced the sound for 5-60 min (daily or almost daily or weekly), or tinnitus for ≥ 60 min or continuously (daily or almost daily or weekly or monthly). This was based on literature and expert opinion 5 .
Tinnitus characteristics. The following items were assessed: whether the participant had sought help for tinnitus or planned to seek help and the source of the help, tinnitus pattern, subjective problem of tinnitus, acute or chronic tinnitus (< 3 months, 3-6 months, ≥ 6 months), manner of the tinnitus start, number of different sounds, pulsatile nature, whether the tinnitus varied in loudness, the pitch and location of the tinnitus, the intrusiveness of the tinnitus, influencing factors, potential causes.
Definition of help seeking tinnitus participants. We defined participants as help seeking tinnitus participants, if they had sought help in the past or planned to seek help for their tinnitus.  18 .
General health characteristics. The following items were asked in the questionnaire to assess general health: the presence of chronic pain, family history of certain diseases, and presence of certain diseases as diagnosed by a doctor.  19 . The total scores range from 0 to 21. A score of 8 or higher indicates a potential anxiety or depression 19,20 .
Audiological characteristics. The following items were assessed in the questionnaire to assess audiological characteristics: hyperacusis, presence of hearing problems, use of hearing aids/cochlear implants/sound generator or tinnitus maskers, and auditory hallucinations.
Characteristics on noise-and substance behavior. The following items were assessed in the questionnaire to assess noise and substance behaviour: the use of head-or earphones, exposure to potential damaging sound levels (subjectively judged), the use of hearing protection, smoking habits, drug use and alcohol consumption.
Data handling and ethics. Data are analyzed anonymously and the privacy of the panel members is guaranteed, as is described in the privacy policy of the Dutch Health Care Consumer Panel. This complies with the General Data Protection Regulation (GDPR). According to Dutch legislation, neither obtaining informed consent nor approval by a medical ethics committee is obligatory for conducting research through the panel (CCMO, 2020) 12  www.nature.com/scientificreports/ the different questions in the survey: tinnitus pattern, subjective problem of tinnitus, duration of tinnitus, varying loudness, tinnitus intrusiveness, TFI score and TFI grades, chronic pain, HADS-A, HADS-D, hyperacusis, hearing problems, the use of different hearing aids, auditory hallucinations, use of head/ear phones, potential damaging sound levels, use of hearing protection, gender, age and educational level.

Results
Study sample. Of the 2251 panel members who were invited to participate in the survey 932 (41.4%) filled out the questionnaire. The median age of the participants was 67.0 (IQR 17) years and 52.4% was female (Table 1).
Tinnitus and its characteristics. Out of the 932 participants, 216 (23.2%, 26 missing) were classified as having tinnitus based on the set criteria of duration and frequency of the experienced sound ( Table 2). Out of these 216 tinnitus participants (91.7%, 1 missing,) 198 experienced their tinnitus for 6 months or more. The total TFI-score could be calculated for 212 tinnitus participants (4 missing) and the median total score was 16.6 (IQR 21.8) ( Table 3). Audiological characteristics and characteristics on noise exposure. The presence of any hearing problem was more frequent in tinnitus participants (135 of 216 (62.5%, 2 missing)) compared to non-tinnitus participants (248 of 690 (36%, 7 missing) (combination of answer options: small-, mediocre-, severe problems and I hear nothing). Compared to participants that did not report any exposure to potentially damaging sound levels, participants with more exposure to potential damaging sound levels had higher odds to have tinnitus multiple times a week but not daily (OR 2.97 (95% CI 1.27-6.92) p = 0.012), once a week (OR 2.23 (95% CI 1.04-4.81) p = 0.041), less than once a week (OR 1.49 (95% CI 1.05-2.12) p = 0.026)) ( Table 4).

Comparison of help seekers versus non-help seekers.
Demographics. Twenty-six of 73 help seekers (HS) were female (35.6%, 0 missing), compared to 66 of 142 (46.5%, 0 missing) of non-help seekers (NHS). Compared to males, females were not more likely to seek help for tinnitus (OR 0.64 (95% CI 0.36-1.14) p = 0.129). The help-seekers had a median age of 69 (IQR 13) years, compared to 66.0 (IQR 16) years of age in the in the non-help seekers (Table 1).
Audiological characteristics and characteristics on noise exposure. Participants who judged sounds as a mediocre problem (hyperacusis) were more likely to seek help for tinnitus than not to seek help for their tinnitus Participants that had exposed themselves to potential damaging sound levels were not more likely to seek help compared to not seek help (reference: no exposure to damaging sound levels, daily: (OR 1.16 (95% CI Missing

Discussion
In this study our primary objective was to describe differences in characteristics of help seeking versus non-help seeking tinnitus participants by means of a questionnaire. It was sent to an adult sample of inhabitants of the Netherlands. Help seeking tinnitus participants had a higher median score on the TFI compared to non-help seeking tinnitus participants. These numbers illustrate that a higher distress score is more frequent in individuals who seek help. We defined participants as a help seeking participant when they planned to seek help for their tinnitus within the next month or had already sought help. We added no time limitations on how long ago in the past they sought help to this definition. Consequently, people could had already sought help years ago, and did not have an active wish for help at the moment of the questionnaire. Interestingly, the help seeking group consisted for 99% (72 of 73) out of participants that had already sought help for their tinnitus. Even though their initial tinnitus distress levels might have been higher, people were still experiencing a median score of 22.8 on the TFI, Table 2. Tinnitus participants and help seeking participants with tinnitus. Tinnitus participant were defined as experiencing tinnitus for 5-60 min daily or almost daily, or weekly or 60 min or more or continuously daily or almost daily, weekly or monthly. (These are written cursive) Please see Supplementary Table S2 for answer to type of tinnitus help, other.      www.nature.com/scientificreports/ which indicates they consider their tinnitus to be a ''small problem'' even after seeking help in the past 22 . Besides this, several questions regarding tinnitus remain; what makes people transit from 'having' tinnitus towards becoming a tinnitus patient or having tinnitus disorder 4 ? 35.6% of the help seekers, as identified in our study, had a TFI score ranging between 0 and 17, which can be interpreted as ''not a problem'' 17 . This might illustrate the controversies between experienced distress scores by these validated instruments and the willingness/need of people to seek help. We found an overlap in known risk factors for tinnitus in literature, with higher frequencies in help seekers 23 . This is to be expected since many studies that assessed tinnitus risk factors were performed in a hospital population of people with tinnitus. These samples include help seekers by definition. For example, hyperacusis and hearing loss were more common in those with tinnitus that sought help compared to those with tinnitus that did not sought help. These two are also two known risk factors for tinnitus and tinnitus distress in literature 7,23,24 .
Surprisingly, we did not find a statistically significant difference for age in help seekers and non-help seekers. Especially since advanced age is a risk factors of tinnitus 23 . We believe this might be caused by the advanced, and reasonably low variance in age of the complete sample.
In our study, we found no clinically relevant differences in anxiety or depression scores measured by the HADS between help seekers and non-help seekers. Even though the odds of having a higher score on both the anxiety and depression scale were significantly higher in the help seeking tinnitus group in our study, the median scores on both scales were all below eight. A score below eight score does not indicate a possible depression or anxiety 20 . We therefore believe that these statistically significant results are not clinically relevant. However, we did find higher frequencies of a self-reported clinical diagnosis of depression in those that had sought help. This discrepancy might be caused by the difference in timing of both questions. The HADS assesses depression or anxiety at the moment of filling out the survey. A clinical diagnosis of depression might have been made years ago. We know from literature that depression is a common risk factor for tinnitus, and was also one of the four most important variables for tinnitus subtyping 6 . The low scores on the HADS might be caused by the fact that our survey was distributed among a general population sample, rather than a hospital sample. The low scores are comparable to a population study from Norway describing similar outcomes in people with and without tinnitus 25 .
With respect to tinnitus specific characteristics, we found that 31.5% of the help seekers experience more than one sound, compared to 17.6% in the non-help seeking group. We also found help seekers to experience a varying loudness more often (65.8%) compared to those that do not seek help (40.8%). This is comparable to a study by Lilllemor et al. from 1993 in a hospital setting. They reported '' complainers'' to hear more than one sound. However, contrasting to our study they report a non-fluctuating sound to be heard by complainers more often than ''non-complainers'' 10 . These differences in characteristics could point out the way people cope with their tinnitus. One could hypothesize that varying loudness or several sounds make tinnitus more difficult to cope with.

Strengths and weaknesses.
A strength of the presented study is the large quantity of data regarding tinnitus and individual characteristics, collected from a sample from the general Dutch population. We created unique data about people with tinnitus that seek help versus those that do not. There are several limitations applicable to this study. The first is that, while the study was set out in a sample of the Dutch population, in terms of age the individuals that responded were not representative of the Dutch population 26 . This might be due to the fact only panel members who gave permission to combine their answers of the survey with health care consumption data as registered by their general practitioner were invited for the survey 13 . The lack of representability may also partly due to the response rate of 41.4%. The response rate might have been influenced by the lengthiness of the questionnaire (with a maximum of 8 pages) or the topic of the questionnaire. This could have made people with tinnitus more inclined to fill out the questionnaire. Due to space limitations we had to take decisions on which questions to include. Still, we did include a validated tinnitus distress measures (the TFI) and a validated anxiety and depression measure to assess these variables of importance for subtyping 6,18,19,22 . Another limitation is our definition of tinnitus. We based it on frequency and duration, but tinnitus distress was not included in our definition.

Future perspectives.
Tinnitus heterogeneity is one of the main impediments that hinder the search for a curative tinnitus treatment 27 . The presented outcomes might help to gain insight in the issue of heterogeneity. However, we believe that the only way to succeed in disentangling this heterogeneity, possibly with subtypes or prediction models, is with interdisciplinary and collaborative research with sound methodology and large datasets 3 . The first steps in multidisciplinary cooperation in research as well as training have been taken, such as programs like ESIT, Tinnitus Assessment Causes Treatment (TINACT) and Unification of Treatments and Interventions for Tinnitus Patients (UNITI) 26-30 .

Conclusions
This study pioneered in describing individual characteristics in the general population between people with tinnitus that sought help versus those who did not. Differences between groups were mainly identified in tinnitus characteristics and audiological characteristics. The outcomes of this study could serve as an initial step to detangle the heterogeneity in tinnitus patients.

Data availability
The datasets presented in this article are not readily available because the Dutch Health Care Consumer Panel has a program committee, which supervises processing the data of the Dutch Health Care Consumer Panel and decides about the use of the data. This program committee consists of representatives of the Dutch Ministry of www.nature.com/scientificreports/