Article Text

other Versions

Tinnitus and hearing in 7-year-old children
  1. Jolanta Juul1,
  2. Marie-Louise Barrenäs2,
  3. Kajsa-Mia Holgers3
  1. 1Department of Otorhinolaryngology, Section for Audiology, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at University of Gothenburg, Sahlgrenska University Hospital, Göteborg, Sweden
  2. 2Department of Paediatrics, Gothenburg Paediatric Growth Research Centre (GP-GRC), Institute of Clinical Sciences, The Sahlgrenska Academy at University of Gothenburg, Queen Silvia Children's Hospital, Göteborg, Sweden
  3. 3Department of Health Sciences at Jönköping Academy, University of Jönköping, Jönköping, Sweden
  1. Correspondence to Jolanta Juul, Department of Otorhinolaryngology, Section for Audiology, Sahlgrenska University Hospital, SE-413 45 Göteborg, Sweden; jolanta.juul{at}


Background Tinnitus occurs with or without prior noise exposure (noise-induced tinnitus (NIT) and spontaneous tinnitus (ST)), and is considered a symptom related to permanent hearing impairment (HI) or temporary hearing threshold shift (TTS).

Objective To carry out a cross-sectional interview study on TTS, ST and NIT during a standard audiometric screening of 756 7-year-old children in Gothenburg.

Results 41% out of 756 children reported either NIT or ST on several occasions, 17% reported recurrent TTS and 7% failed the audiometry screening. The probability of ST was 27% for children with no HI or TTS (OR=1.23 (95% CI 1.12 to 1.34)) but 63% (OR=1.16 (95% CI 1.02 to 1.33)) if exhibiting both HI and TTS.

Conclusion This study confirms an increased occurrence of spontaneous tinnitus in children with TTS or HI and in children with both TTS and HI, in particular, but also in children with normal hearing. Possibly, tinnitus in young children correlates with stress as in adolescents and adults.

Statistics from


Studies suggest that there is an increasing prevalence of tinnitus among children. In Sweden, Holgers1 described a study of 964 7-year-olds tested in 1997, 12% of whom had experienced tinnitus, and for 2004, an occurrence of noise-induced tinnitus (NIT) was found among 53% and spontaneous tinnitus (ST) among 46% of 274 unselected Swedish school children aged 9–16 years.2 Among these pupils, 2% had tinnitus always, 14% every day and 23% were annoyed by tinnitus sometimes or often. These results were higher than the study performed in 2003–4 on 671 teenagers aged 13–16 years, 28% of whom had experienced NIT, 7% ST and 28% both.3 In that study, 18% reported that tinnitus occurred often or always and in 13% tinnitus caused annoyance often or always. The aim of this investigation was to replicate our previous investigation on the youngest school children and thus to study experience of NIT and ST in a large unselected sample of healthy 7-year-old school children from Gothenburg, with a particular focus on children without hearing loss, be it temporary hearing threshold shift (TTS) or permanent hearing impairment (HI).

Subjects and methods

In Gothenburg, all 7-year-old children (∼5000 each year) undergo an audiometric 20 dB HL screening at the 0.5, 1, 2, 3, 4, 6 and 8 kHz frequencies as part of the regular school health services. This study included the first consecutive 756 children (366 girls and 390 boys, born in 1999) tested during the year 2006 and used the same design and tinnitus questions as in 1997. HI was defined as a threshold worse than 20 dB. Structured interviews were held individually by the audiologist performing the audiometry, who also made sure that the child had understood the questions. Three questions Q1–Q3 with a three alternative response format (no (1) never (2) yes, once (3) yes, several occasions) were presented. “After listening to loud music or other loud sounds or noise, have you noticed that your hearing is worse?” (Q1) was regarded as indicating an experience of TTS. “After listening to loud music or other loud sounds or noise, have you heard any ringing, buzzing or other sort of sounds in your ear even after that the loud music or noise has been turned off?” (Q2) and “Have you heard a ringing, buzzing or other sort of sound in your ears without first having listened to loud music or other loud sounds?” (Q3) were interpreted as indicative of NIT and ST, respectively. For all these three variables, in order to be regarded as a case, the symptom had to have occurred on several occasions.

What is already known on this topic

  • ST in adults is an area generating many studies with highly different entry points, from neurophysiology to psychology.

  • In children, the symptom often goes unnoticed and when signalled, already reflects a degree of annoyance.

  • It is still unclear how to address this symptom, gathering of data on subjective experiences being difficult in this young group.

What this study adds

  • We present prevalence numbers in a sample representative of a general paediatric population and results point to a steady increase of ST.

  • In normal hearing children without signs of ear pathology, noise exposure seems insufficient as sole explanatory model for the presence of ST.

  • Stress and/or incipient mood disorders could represent complementary factors influencing this potentially disturbing symptom.

Data were analysed using SPSS 19.0 for Windows. Logistic regression was used to determine the most influential parameters for the presence of tinnitus. Where applicable, t tests were performed. A two-tailed p value <0.05 was considered as significant. This study was approved by the ethical committee in Gothenburg and performed according to the Helsinki declaration.


Descriptive results are presented in table 1. Of the 706 children with normal hearing, 288 children (40.8%) reported having experienced tinnitus in some form. The corresponding occurrence among the 50 children with HI was 58.0% (29 children). Among the 50 children with HI (32 boys and 18 girls), 16 (32%) exhibited a normal tympanometry test—that is, a sensorineural hearing loss. The most common configuration was a high-frequency loss. The mean pure tone average on 0.5, 1 and 2 kHz did not differ between children with and without tinnitus—20.6 dB and 20.2 dB, respectively (t test; p=0.94) table 1.

Table 1

Distribution of interview data according to hearing (normal/impaired) among subjects who responded to all three items spontaneous tinnitus, noise-induced tinnitus and temporary hearing threshold shift (TTS) (n=706)

According to the logistic regression model estimations (figure 1), the probability of ST was 41% for children with HI, 46% for those reporting TTS, 63% for children with both TTS and HI, 63% and 27% among children with neither TTS nor HI. The probability for NIT in children who experienced TTS was estimated as 59% compared with 17% in children with no TTS. HI did not affect NIT significantly. There is, however, some uncertainty about the estimations, since probability levels varied by about ±10% as shown by the CIs (for OR, see figure 1 legend). Gender had no impact on tinnitus or TTS (χ2 test p=0.984 and p=0.229, respectively), and TTS did not correlate with HI (p=0.579).

Figure 1

Observed (open triangle) and estimated (filled circle) probability for spontaneous tinnitus and noise-induced tinnitus (NIT). Error bars represents 95% CI for estimation based on the logistic regression models for ST constant; β=−1.021 (SE=0.628); temporary hearing threshold shift (TTS) β=0.888 (SE=0.199): OR (95% CI)=2.43 (1.65 to 3.59): p<0.000; hearing impairment β=0.666 (SE=0.301): OR (95% CI)=1.95 (1.08 to 3.51): p<0.000; NIT constant; β=−1.574 (SE=0.725); TTS β=1.948 (SE=0.039): OR (95% CI)=7.01 (4.66 to 10.59): p<0.000.


Our main finding was that the high 2003 tinnitus prevalence figures in teenagers were confirmed also among 7-year-olds with normal hearing (ie, 41%). In this study, the estimated occurrence of ST among children who had both hearing loss and who reported TTS was as high as 63%, but for the general population the risk could span between 47% and 76%. Our interpretation of the relationship between the experience of TTS and ST and also NIT, is that TTS reflects inner-ear sensitivity and possibly, in accordance with data in adults, upcoming permanent hearing threshold shifts, but not necessarily noise-induced since a high noise exposure is questionable in this very young age group. This study showed that NIT was not affected by HI, and the mean pure tone average did not differ between children with and without tinnitus. Possibly, children with already confirmed HI avoid noise exposure—that is, had positive effects of hearing conservation efforts.

A quite high occurrence of ST (27%) was also present among children with normal hearing and no TTS. This suggests that factors other than noise and HI also play a part in young children—for example, stress and mood swings, a correlation found in adolescents and adults.4 We are aware of several neurophysiological explanations to tinnitus such as high serotonin vulnerability or deafferentation,5 but in our opinion, the two explanatory models are not mutually exclusive. If TTS is to be regarded as a reversible form of deafferentation, then TTS might represent one end of the scale and permanent hearing loss due to irreversible damage the other. On the other hand, the group with ST and no signs of other hearing damage might represent a symptomatology related to a dysfunctional transmitting of nerve signals owing to a faulty serotonin discharge. As argued above, we believe stress, depression and anxiety disorders are important contributing factors to tinnitus also among children. However, as all variables are greatly intertwined, the uncertainty about causality is high and conclusions should be interpreted with care.

Benefits of this study were the large sample size, that the study design was a replicate, that results were consistent with previous studies2 and that the population under investigation was representative of the general paediatric population. Also, for the first time, a clear distinction between the often-overlapping symptoms NIT and ST was made. Main shortcomings are due to ethical and methodological aspects. It was not possible to interview 7-year-old children on tinnitus degree/annoyance, or sensitive and serious psychological or family matters in the setting of a hearing screening programme. That could have given rise to questions for which the child was not prepared or sufficiently mature to handle. We are aware, that we do not have evidence to suggest that ST in 7-year-olds with normal hearing is linked to any pathology, and that our findings may just constitute an expression of increased awareness of, or worry about, different hearing sensations or phenomena. According to our clinical experience and earlier studies, annoying tinnitus is rare in this young age group.2 To answer the question whether these subjects will develop mood disorders rather than HI or both, it would be useful to continue to monitor this study sample longitudinally.


The authors are grateful to Mats Gunnarsson of the R&D department at the Varberg Hospital, County of Halland for statistical expertise. The authors will warmly remember all the assistance from the late Mrs Margareta Magnusson.


View Abstract


  • Funding This study was supported by Swedish Research Council funding for clinical research in medicine, nr ALFGBG-2766 and the funds were used for reimbursing the audiologist for the extra time spent performing the interviews, in addition to carrying out screening within a national programme.

  • Competing interests None.

  • Ethics approval This study was approved by the ethical committee in Gothenburg (reference number 125-04) and performed according to the Helsinki declaration.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.