Background
Tinnitus is an auditory perception of sound in the absence of any external or internal acoustic stimulus. Chronic subjective tinnitus is a frequent condition with a prevalence that rates between 2.4 and 20 % [
1]. A study done by the Clinical Telemedicine Department of the University Hospital Zurich (which has provided an online consultation service on health issues in all areas of medicine since 1999) shows that requests concerning tinnitus belong to the 50 most frequently raised topics [
2]. Furthermore tinnitus can be associated with the impairment of health related quality of life (HRQoL) [
3,
4] or depressive symptoms [
5‐
7]. The extent to which tinnitus affects the quality of life or is linked with depressive symptoms is highly variable [
7,
8]. Recent studies demonstrated an association between various subjective aspects of tinnitus, HRQoL and depressive symptoms [
8,
9].
The lack of objective means of measuring tinnitus necessitates the use of self-report questionnaires for the evaluation of tinnitus severity. Several questionnaires have been developed to assess tinnitus severity or tinnitus-related impairment [
8]. The
Tinnitus Handicap Inventory (THI) is the most standardized questionnaire for assessing the tinnitus related handicap [
10]. The validated German version [
11] of the
Tinnitus Questionnaire (TQ) [
12] is a widely used measure of tinnitus-related psychological and psychosomatic distress.
As described in our previous study [
9], the assessment of tinnitus severity has a particular relevance in clinical evaluation as it can help in the identification of patients who are at risk of developing psychological distress or depressive symptoms. At present, tinnitus questionnaires are widely used as measuring instruments in tinnitus research and in specialized tinnitus clinics. However, according to our own experience, in clinical practice most general practitioners, otologists and other clinicians do not use these questionnaires in a daily routine. A reason for this might be the time consuming aspect involved in the filling out and analyzing of these questionnaires. To address this time consuming aspect shortened forms of the THI have been developed and validated in different languages [
13,
14]. Furthermore, problems in the understanding of the questions by foreign language speaking patients may complicate the assessment.
Other methods for assessing the impact of a disease are graphic tools such as pictures, charts and graphs. With a simple two-dimensional visual instrument such as the
Pictorial Representation of Illness and Self-Measure (PRISM), the burden of suffering caused by an illness can be quickly and easily measured [
15]. Burden of suffering is defined as ‘a state of severe distress associated with events that threaten the intactness of the person’ [
16]. So far, PRISM has been validated as a reliable method for assessing the burden of suffering in different psychological and physical conditions such as dizziness [
17], post-traumatic stress disorder [
18], rheumatoid arthritis [
19], dermatological diseases [
20‐
23], and orofacial pain [
24]. There is a strong correlation between the patient’s self-perceived severity of illness and their perception of the burden of suffering caused by the illness, as measured with PRISM [
21‐
23]. Overall, the burden of suffering measured with PRISM seems to reflect an individual’s well-being in the context of illness [
25]. Therefore, PRISM seems to be a reliable, feasible and useful tool in the assessment of illness [
25].
To our knowledge, no previous studies have focused on the burden of suffering as assessed by the PRISM instrument in patients with tinnitus. In a prospective observational study we aimed to investigate whether PRISM can be used for this purpose. Therefore, we compared the burden of suffering, measured with the electronic version of PRISM on an iPad, with different, conventional means of measuring tinnitus severity, HRQoL and depressive symptoms. Furthermore, we aimed to investigate which sociodemographic factors contributed to the burden of suffering in tinnitus patients.
Discussion
The current study aimed at investigating the association between the burden of suffering measured with PRISM and standardized tinnitus questionnaires. A high burden of suffering was anticipated to be significantly correlated with a high tinnitus severity. As anticipated, a high burden of suffering (low SIS) was significantly correlated with a high tinnitus severity (high THI and TQ).
In addition to the investigation of the correlation between the burden of suffering and tinnitus severity, we divided the results of the burden of suffering (SIS) into three different PRISM groups (representing the different levels of tinnitus severity) to make it easier for the investigator to interpret the results in a short time. We then compared the means of THI and TQ between these groups. We could show significantly different means of THI between PRISM groups 1, 2 and 3.
Furthermore, the mean THI of PRISM groups 1, 2 and 3 classified in the corresponding grading system of THI showed a severe (group 1), moderate (group 2) and mild (group 3) tinnitus handicap, respectively. Similar to THI, the means of TQ between the three PRISM groups were significantly different from each other and each mean TQ of PRISM groups 1, 2 and 3 was classified in the corresponding grading system of TQ as severe (group 1), moderate (group 2) and slight (group 3) tinnitus, respectively. Therefore, the burden of suffering from tinnitus measured with PRISM is not only significantly inversely correlated with the tinnitus questionnaires, it could also be divided into three significantly different PRISM groups. Furthermore, the mean scores of THI and TQ of the three different PRISM groups are classified in different grades in the corresponding grading system of THI and TQ.
As previously stated in other studies [
8,
9] we could also demonstrate a significant association of tinnitus severity (THI and TQ) with depressive symptoms (BDI) and the global and psychological HRQoL (WHOQOL-BREF). Additionally, significant inverse correlations were found between the burden of suffering from tinnitus (low SIS), depressive symptoms (high BDI), and global and psychological HRQoL (WHOQOL-BREF). Moreover, we could demonstrate that our entire study population had a significantly lower mean of global and psychological HRQoL (WHOQOL-BREF) when compared with a German norm population [
30]. We can, therefore, state that suffering from tinnitus influences the HRQoL of our study population to a significant degree. However, we cannot say whether low subjective HRQoL may have fostered symptoms of tinnitus or vice-versa.
Although PRISM was performed with an electronic version on an iPad instead of the paper or conventional magnetic board version, the association between burden of suffering (SIS) with tinnitus questionnaires (THI and TQ), depressive symptoms (BDI) and HRQoL (WHOQOL-BREF) is consistent with other studies on PRISM reflecting a subjective perception of an illness [
17‐
24,
33‐
35].
The present findings also suggest that the burden of suffering is not gender-related although a higher percentage of males sought medical advice concerning tinnitus symptoms (62.6 %) and took part in our survey (64.9 %). This observation is confirmed by other studies which also reported a higher prevalence for males perceiving tinnitus than for females [
36,
37]. The explanation for this phenomenon is the higher degree of hearing loss amongst males, especially after the age of 50, which is most likely due to occupational noise exposure [
38]. A difference to previous studies on PRISM [
17,
21], is that some socioeconomic variables played a role in the burden of suffering in patients with tinnitus. The burden of suffering was significantly lower (higher SIS) for patients in a partnership compared with patients living outside a partnership. Furthermore, patients with a university degree had a significantly lower burden of suffering (higher SIS) compared to patients with no degree or basic school education. This is in line with another previous study which demonstrated a negative association of tinnitus with the degree of education and the number of household members [
37,
39]. Indirectly, this finding could be an indicator for the better coping strategies of patients who have a partner or a higher education. The burden of suffering was not correlated with age and duration of tinnitus in the current sample.
Our data illustrated that the burden of suffering measured with PRISM showed a significant correlation with the conventionally used tinnitus questionnaires. This outcome is consistent with our expectations and previously published studies on PRISM [
17,
31]. Consequently, PRISM as a multi-factorial construct is reliably assessing the burden of suffering in patients with tinnitus, as well as those illnesses already investigated [
17‐
24,
33,
34]. Similar to a recent study [
24], the PRISM results were divided into three different PRISM groups in order to give a direct impression of the burden of suffering in tinnitus patients. For this type of analysis the SIS is not important, as a simple optical discrimination of the relationship between one circle and the other is sufficient for the result (inside, overlapping and outside). These groups were significantly discriminant between the different levels of tinnitus severity. Notably, PRISM reliably distinguished between patients with mild compensated (group 3) versus severe, probably decompensated tinnitus (group 1). Taking into account that the mean duration of explanation and the execution of PRISM was less than two minutes, PRISM therefore may be a useful screening tool for use in routine medical consultations proposing a high burden of suffering, as an indicator for a decompensated chronic tinnitus situation. To qualify PRISM as screening tool further studies should address this issue.
At this point, we want to acknowledge the limitations of our study. Firstly, the patient population suffering from tinnitus was partly preselected as, for the most part, the patients were sent to our clinic after having been seen by a general practitioner. Secondly, the majority of the patients did not perform PRISM and the online survey on the same day, as we accepted a timeframe of two weeks between them for completion. Additionally, 13 % of the tinnitus patients in our initial population were non-German speakers and had to be excluded due to the study design. Furthermore, the answers on PRISM as a multi-factorial construct are generally influenced by other uncontrollable factors like coping strategies or personal resilience. Finally, all patients with tinnitus presumably having different underlying origin of tinnitus were included in our study. In most cases the underlying origin of tinnitus was unknown and it is therefore possible that different pathologies may have affected the PRISM results.
In conclusion, PRISM cannot completely replace the standardized tinnitus questionnaires because those provide more specific information related to different aspects like emotional distress, auditory perceptual difficulties and sleep disturbances caused by tinnitus [
11]. However, our study demonstrated that PRISM is a simple approach that gives a global impression of burden of suffering in patients with tinnitus and is easily applicable in tinnitus patients in clinical contexts. It should also have great value in assessing treatment associated changes. Further to this, it might also act as an instrument for evaluating the effectiveness of treatment as demonstrated in other conditions [
19,
21‐
23]. Future research studies should attempt to quantify the effects of tinnitus therapy by considering the change of SIS.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
NP designed data-collection tools, collected data and monitored data collection for the whole trial, wrote the statistical analysis plan, cleaned and analyzed the data, contributed to the interpretation of data, drafted and revised the paper. SW initiated the collaborative project, designed data-collection tools, monitored data collection for the whole trial, wrote the statistical analysis plan, cleaned and analyzed the data, contributed to the interpretation of data and revised the drafted paper. TK initiated the collaborative project, collected data, contributed to the interpretation of data and revised the drafted paper. LH collected data for the whole trial and revised the drafted paper. SSW contributed to the interpretation of data and revised the drafted paper. MM contributed to the interpretation of data and revised the drafted paper. SB designed data-collection tools, contributed to the interpretation of data and revised the drafted paper. Furthermore he has a lot of experience in statistically analyzing questions about PRISM and supported NP and SW in writing the statistical analysis plan. All authors read and approved the final manuscript.