Article Text
PDF
Abstract
Background Hearing problems are among the top 10 most common burdens of disease and are projected to be become even more common by the year 2030. The aim of the present study was to give a current assessment of the prevalence of communication difficulties because of hearing loss and tinnitus, in the general Swedish working and non-working populations in relation to sex, age, socioeconomic status (SES) and noise exposure. How prevalence is affected by SES has not been previously established.
Methods A total of 18 734 individuals were invited to participate in the study, of which 11 441 (61%) enrolled. Of the participants, 9756 answered the questionnaire for those who work and 1685 answered the version for non-workers.
Findings The most important findings are that 31% in the working population and 36% in the non-working population report either hearing loss or tinnitus or both. The prevalence of hearing problems increases with age, is higher among men and persons with low self-rated SES, and covaries with exposure to noise at work. Severe hearing problems are already present in men and women under 40 years of age who are exposed to work-related noise.
Interpretation Prevalence of hearing problems is far more common than previously estimated and is associated with SES and noise exposure history. Hearing problems have a gradual onset that can take years to become recognised. In order to proactively intervene and prevent this deleterious, yet avoidable handicap, statistics need to be regularly updated.
- Hearing loss
- prevalence
- tinnitus
- socioeconomic status
- public health
- noise
- sex
Statistics from Altmetric.com
According to the World Health Organization (WHO), hearing problems are among the top 10 most common burdens of disease in medium and high income countries.1 2 Prevalence estimates differ between studies, and previous findings show that approximately 10–15% of the population is affected. The prevalence of hearing loss and tinnitus is age-related but also influenced by cumulate exposure to noise. Men are more often affected than women.3–5 Noise levels are predicted to increase in the future2 4 6–8 because of increased sound sources primarily from the environment and leisure time activities.3 6 8 Since hearing loss usually has a gradual onset that may take years to become recognised, statistics need to be regularly updated and compared in order to find strategies to prevent or delay this deleterious, yet avoidable handicap.
Some examples of hearing problems can include difficulties in understanding speech in noise, decreased hearing acuity, hyperacusis, mild to severe hearing loss and tinnitus. These problems are associated with communication difficulties and affect health-related quality of life, may pose a risk to workplace safety and health and negatively affect cognitive and emotional status.4 9–12 Hearing problems are often associated with various forms of comorbidity, such as tooth loss, diabetes, psychiatric conditions and cardiovascular disease.6 13–15 Apart from the human suffering, the economic toll (in terms of sick leave, rehabilitation, exclusion from work life, sickness, etc) of hearing problems is very high.4
Most prevalence studies assess only one or two dimensions of hearing problems, that is hearing loss or tinnitus and possibly severity, but not the different forms of hearing problems. It is common that individuals who suffer from hearing problems exhibit more than one symptom—for example, both hearing loss and tinnitus. Therefore, more knowledge about the prevalence of different combinations of hearing problems is needed. Generally, the prevalence of hearing problems is described in relation to sex and age. To our knowledge, there is no study that has examined the prevalence of hearing problems in relation to socioeconomic status (SES).
Thus, the aim of the present study was to assess the prevalence of both hearing loss and tinnitus in the general Swedish working and non-working population between the ages of 19 and 70. Furthermore, covariation of these hearing problems with age, sex, SES and amount of noise exposure in the work place was analysed.
Methods
Population
The Swedish Work Environment Survey is conducted biennially by Statistics Sweden and consists of subsamples of gainfully employed people, aged 16–64 years, from the Labor Force Survey. These individuals were first sampled into the LFS through stratification by country, sex, citizenship and occupation. The respondents to the Swedish Work Environment Survey 2003 and 2005 were invited to enrol in the Swedish Longitudinal Occupational Survey of Health (SLOSH).16 The study was conducted in April 2008 by Statistics Sweden, on behalf of the Stress Research Institute at Stockholm University. SLOSH consists of two questionnaires, one addressed to those who work and one to those who for any reason do not work. The respondents selected the questionnaire that corresponded to their work status at the time. A total of 18 734 individuals were mailed self-completion questionnaires in 2008, of which 11 441 (61%) responded. Of the participants, 9756 answered the questionnaire for workers and 1685 answered the version for non-workers. The non-workers in this study are representative only of those respondents who, in 2008, had temporarily or permanently stopped working since 2003 or 2005, respectively. Many were on parental leave or recently retired, and some were unemployed or on long-term sickness absence. Table 1 depicts the response rate and characteristics of responders versus non-responders. There was no difference between responders and non-responders with regard to country of birth and citizenship. For the present study, only data from the respondents were analysed.
Response rate and demographic characteristics of responders versus non-responders
Questionnaire
A comprehensive questionnaire was distributed by mail to all potential study participants. Apart from SES and demographic factors, it also included approximately 120 questions about psychosocial and physical work environment, lifestyle as well as physical and mental health.
Hearing problems were assessed with three questions: Tinnitus—“Have you during the most recent time experienced sound in any of the ears, without there being an external source (so-called tinnitus) lasting more than 5 min?” (No, Yes sometimes, Yes often, Yes all the time). Tinnitus severity—“How much do you feel that the tinnitus sounds worry, bother or upset you?” (Not at all, A little, Moderately, Severely). Hearing loss—“How difficult is it for you to (without hearing aid) hear what is said in a conversation between several persons?” (Not difficult at all, Not very difficult, Quite difficult, Very difficult). In this study, hearing loss reflects difficulties in communicating. The questions about tinnitus were slightly modified and adapted from Palmer et al.17 The questions about hearing loss were derived from Statistics Sweden and have been used in population studies for several years.
A new variable, “hearing problems”, was computed based on the existence or non-existence of either tinnitus or hearing loss or both. This consequently yielded three groups: those without hearing problems, those with either tinnitus or hearing loss or those suffering from both. The cut-off for tinnitus was “yes, sometimes” or more often, and for hearing loss, “quite difficult” or “very difficult”.
Subjective SES was assessed by means of a self-rating instrument that has previously been validated.18 Participants were given a drawing of a ladder of 10 steps with the following instruction: “Think of this ladder as representing where people stand in our society. At the top of the ladder are the people who are best off—those with the most money, most education and best jobs. At the bottom are the people who are worst off—who have the least money, least education, and the worst jobs or no job. The higher up you are on this ladder, the closer you are to people at the very top, and the lower you are, the closer you are to the people at the very bottom. Where would you place yourself on this ladder? Please place a large ‘X’ on the rung (not in between) where you think you stand”. The 10 SES positions were divided into three groups by quartile split. Hence, self-rated SES 1–5 constituted the lowest SES group, 6–7 the middle group and 8–10 represented those with highest SES.
Noise exposure was assessed with the question: “Is the noise level in your workplace so loud that it is difficult for you to hear what your colleagues say?” (Never or almost never, About 10% of the time, About 25% of the time, About 50% of the time, About 75% of the time, 90% of the time, Almost all the time or all the time). Noise disturbance was assessed with the question: “How disturbing is the noise in your workplace?” (Not disturbing at all, Barely disturbing, Somewhat disturbing, Quite disturbing, Very disturbing, Very very disturbing, Almost unbearable). These questions were derived from Tesarz and Kjellberg19 and have been validated before. Additionally, we asked if the respondents were exposed to “disturbing or tiring sounds” at work (Almost all the time, About three fourths of the time, Half the time, About one fourth of the time, Little, perhaps one tenth of the time, No, not at all).
Statistical analyses
The program SPSS V.17.0 was used for statistical analyses. Prevalence was calculated via frequency plots and crosstabs were used for calculation of χ2, Kendall's τ-b and specific prevalence within different groups. A multiple logistic regression was used to estimate the possible effect of SES on hearing problems after adjustment for age. Statistical significance was set at p<0.05 level.
Ethical approval
The Karolinska Institute ethics committee approved the research project (#2008/240-32 as well as #2006/158-31, #92-198 and #03-125), and all participants gave their informed consent to participate.
Role of the funding source
The sponsor had no role in the study design, data collection, data analysis, data interpretation or writing of the report. All the authors had full access to all data in the study and had final responsibility for the decision to submit for publication.
Results
Figures 1 and 2 illustrate the specific prevalence numbers for each question about hearing problems. The prevalence of hearing problems (tinnitus or hearing loss or both) in the general Swedish working population is 31%. Of those who suffer from tinnitus sometimes, 87% are not bothered by it at all or only a little. On the other hand, of those who often experience tinnitus, 47% are quite or severely disturbed by it. Of those who suffer from tinnitus all the time, 60% are quite or severely distressed by it. The correlation between tinnitus and hearing loss was r=0.31, p<0.0001.
Prevalence of hearing problems in the general Swedish working population.
Prevalence of hearing problems among non-working respondents.
Sex differences
There was a statistically significant difference in prevalence of hearing problems between women and men in the working population (χ2=99.996df=2, p<0.0001, table 2). In general, 25% of the working population reported either tinnitus or hearing loss. The corresponding figure among men was 29% and 23% among women. For those exhibiting both tinnitus and hearing loss, the general prevalence was 6%, while it was 8% for men and 5% for women. Thus, men are clearly overrepresented among those with hearing problems.
Prevalence of hearing problems in the working and non-working populations divided by age and sex
With regard to tinnitus, there was a significant sex-related difference (χ2=148.110df=3, p<0.0001) in the working population. Overall, 4% of the working respondents suffered from tinnitus, and the prevalence was higher for men than for women (5% vs 3%, p<0.0001). Overall, 7% suffered from tinnitus all the time; the prevalence for men was 10% compared with 5% for women. With regard to hearing loss, the sex difference was somewhat smaller but still statistically significant (χ2=44.632df=3, p<0.0001). Among the 11% with quite severe to severe hearing loss, the prevalence for men was 13% compared with 10% in women.
In the non-working population, there was also a significant difference between women and men (χ2=75.925df=2, p<0.0001) and the prevalence was slightly higher than in the working population. Thus, 27% suffered from tinnitus or hearing loss (34% of the men and 23% of the women) and 9% from both (14% of the men and 5% of the women).
The prevalence of having tinnitus frequently was higher among non-workers than in the working population, that is 6% (9% for men and 5% for women), and also for those who suffered from tinnitus all the time, that is 9% (16% for men and 5% for women). The prevalence of having quite severe to severe hearing loss was also higher than in the working population, that is 15% (22% for men and 11% for women).
Age differences
In the working population, there was a statistically significant increase with age in the prevalence of hearing problems for both women and men (Kendall's τ-b 0.09 and 0.16, respectively, p<0.0001). The age gradient was more prominent in men than in women. The younger population of both men and women (40 years or younger) shows relatively high values for the prevalence of either tinnitus or hearing loss. Table 2 illustrates the prevalence of hearing loss divided by age and sex in the working population.
In the non-working respondents, there was also a statistically significant age gradient in the prevalence of hearing problems between the different age groups for both women and men (Kendall's τ-b 0.13 and 0.16, respectively, p<0.0001). In general, 27% of the non-workers suffered from either tinnitus or hearing loss: 19% of those under 40 years, 28% of those between 41 and 50 years and 30% of those older than Q years. Nine per cent of the non-workers suffered from both tinnitus and hearing loss: 2% of those under 40 years, 5% of those between 41 and 50 years, 8% of those between 51 and 60 years and 12% of those older than 60 years.
Socioeconomic status
In the working population, there was a statistically significant difference in the prevalence of hearing problems between the different socioeconomic groups (χ2=43.513df=4, p<0.0001) (see table 3). Hearing problems are more common in lower SES. The SES-related gradient is particularly pronounced for those under 40 years of age (Kendall's τ-b −0.10, p<0.001 for women and −0.12, p<0.001 for men). In the age group between 41 and 50 years, the SES gradient is still present (Kendall's τ-b −0.05, p<0.05 for women and −0.08, p<0.01 for men), while for those older than 50 years, the gradient is found only in women (–0.06 in the ages 51–60 and −0.09 in 61+, p<0.01). Hearing problems may covary with noise exposure and individuals with lower SES are usually more exposed to different stressors. To assess this, Pearson's correlation was conducted between SES and noise exposure. There was a small correlation (r=−0.17, p<0.0001, two-tailed), indicating a higher noise exposure in lower SES groups.
Prevalence of hearing problems in the general Swedish working population divided by socioeconomic status
Multiple logistic regression showed that SES (low, medium, high) is associated with hearing problems (0 indicates “no” and 1 “yes”) among women and men in the working population even after adjusting for age (odds ratio (OR)=0.8, p<0.0001 for both women and men). In this case, 0.8 means that when the highest SES quartile is compared with the middle category or the middle category to the lowest category, the likelihood that the more socioeconomically favoured group will have hearing problems is 80% compared with the less favoured group after adjustment for age. Among non-working respondents, the association between SES and hearing problems is significant only in women when adjusting for age (OR=0.8, p<0.05). Overall (combining women and men), there was no statistically significant difference in the prevalence of hearing problems between the different socioeconomic groups among non-working respondents.
Noise exposure at work
Table 4 illustrates the prevalence of hearing problems in relation to noise exposure and noise disturbance at work. There was a statistically significant difference in the prevalence of hearing problems between individuals with different noise exposures at work (Kendall's τ-b 0.16, p<0.0001), how much one was disturbed by the noise (Kendall's τ-b 0.16, p<0.0001) as well as how much the individuals were exposed to disturbing and tiring sounds (Kendall's τ-b 0.12, p<0.0001). The prevalence of noise exposure was more common among men than women (Kendall's τ-b 0.16, p<0.0001) and decreased with age among men (Kendall's τ-b −0.04, p<0.01). The covariation between noise exposure and hearing problems was statistically significant for both women and men in all age groups (p<0.01). Men were also more disturbed by noise than women (Kendall's τ-b 0.07, p<0.0001, data not shown), but the disturbance decreased with age in men (Kendall's τ-b −0.03, p<0.01). The covariation between noise disturbance and hearing problems was statistically significant for both women and men in all age groups (p<0.01).
Prevalence of hearing problems in the general Swedish working population in relation to noise and sound exposure at work
Exposure to disturbing and tiring sounds was different among men and women, with more men reporting being exposed 10–50% of the time and more women reporting 75–100% or not at all (χ2=45.632df=5, p<0.0001). This exposure decreased significantly with age for both women and men (Kendall's τ-b −0.06 and −0.07, respectively, p<0.0001). The covariation between exposure to disturbing/tiring sounds and hearing problems was statistically significant for both women and men in all age groups (p<0.01) apart from among women 61 years of age or older.
Risk groups
The prevalence of hearing problems increases with age and it is higher in men as well as among those with lower self-rated SES (see tables 2 and 3). Of particular interest is the finding that approximately every fifth young adult (≤40 years of age) has either tinnitus or hearing loss. These findings are notably high and can be an early indicator for a decreased capacity of auditory and communicative abilities of the middle-aged population.
Discussion
The aim of the present study was to assess the prevalence of two common hearing problems, that is hearing loss and tinnitus, in the general Swedish working population. The most important findings are that 31% in the working population and 36% among the non-working respondents report either hearing loss or tinnitus or both. Prevalence increases with age, is higher among men and covaries with SES and exposure to noise at work. Thus, lower SES implies more hearing problems as well as more work-related noise. Severe hearing problems are already present in men and women under 40 years of age who are exposed to work-related noise. The present study is one of the largest in the Western world with regard to assessment of hearing problems, and it is unique in the sense that it covers both tinnitus and hearing loss.
The prevalence of both hearing loss and tinnitus in the present study is higher than that reported in previous studies. In the present study, we assessed both hearing loss (defined by its inherent effect to obstruct communication) and tinnitus, whereas other studies have generally focused on only one of these conditions. Furthermore, differences in assessment may also be an explanation. Regardless of reasons, the large prevalence figures of this study combined with the estimations and predictions of the WHO and other studies1 2 4 6–8 point out that the increase in prevalence of hearing problems should be taken seriously. To change this negative trend, tailored preventive interventions need to be implemented at individual, organisational as well as societal levels.
According to the WHO, the sixth most common burden of disease in the high income countries is hearing loss with adult onset preceded by—for example, ischaemic heart disease, depression and Alzheimer's disease.7 In fact, hearing problems are the most frequent sensory deficit in humans, with an estimated 250 million affected individuals worldwide.1 Furthermore, hearing loss is projected to become even more common and continue to be among the top 10 causes of burden of disease in middle and high income countries in 2030.2 This predicted that increase in hearing problems is partly explained by an ageing population in the middle and high income countries and by increasing noise exposure.
As previously found and further supported by the present study, men are afflicted by hearing loss and tinnitus more than age-matched women.3 5 8 One obvious explanation for the sex-related susceptibility to hearing problems could be attributable to the protective nature of oestrogens. One major effect of oestrogens on the central nervous system is to protect against cell death either by itself or by interacting with neurotrophins or neurotransmitters.20 21 It has been clearly demonstrated that oestradiol plays a significant role in regulating hearing sensitivity and perception in women. Changes in auditory perception and auditory latencies are susceptible to the menstrual cycle,22 23 as are the symptoms of Meniere's disease, including vertigo and hearing loss.24 The findings are demonstrating that oestrogens can have prominent effects on auditory function in young women. The findings from the present study can partially be explained by a protective effect of oestrogens; however, it would be expected that the prevalence of hearing problems should increase disproportionally in post-menopausal women. Such a disproportionate increase is, however, not found. The overall protective effect that oestrogens play with increasing age will be more complex when the role of the oestrogen receptors is taken into consideration. The action of the two different oestrogen receptors, α and β, can be antagonistic, synergistic or separate actions during ageing.
We have chosen to measure SES by means of a self-rating instrument. This method has been established and validated in a previous research.18 Self-rated SES has a strong relationship with more objective, established measures of SES such as occupational position, education, income and household wealth. However, the predictive ability of subjective SES on ill-health is stronger than that of objective measures.18 We preferred to use the self-rating because it has proved to be valid and feasible. Indeed, our findings clearly demonstrate a significant association between SES and prevalence of hearing problems. Subjects in lower SES groups are more likely to have been exposed to higher noise levels. In addition, cohort studies have shown that subjects from lower SES strata have higher prevalence of disturbed hypothalamic–pituitary–adrenal axis regulation.25 Since self-ratings are more subjective in nature, there may be a risk that associations may be inflated by inverse relationships. A person with pronounced hearing loss and difficulties communicating may be less likely to achieve a high societal position and/or may consider his/her social position to be lowered by this disability. Accordingly, our findings may be both attributable to increased noise exposure and increased vulnerability in lower SES groups. We plan a prospective study based upon the same cohort in order to disentangle these possibilities. The SES results point at the significance of preventive measures directed towards these more vulnerable groups.
The fact that there was no SES-related difference in the prevalence of hearing problems in the non-working population might indicate that non-working individuals are more similar as a group with regard to SES and are equally vulnerable regardless of social status. More studies are needed to evaluate the overall stress factors related to non-working populations. However, the rather special composition of the non-working SLOSH respondents makes generalisations very difficult. Furthermore, the difference between age, SES and sex on the one hand and the likelihood of moving into the non-working group on the other creates both power and interpretation problems.
With regard to hearing loss, about 36% in the working population and 33% in the non-working population stated that it was “not very difficult” to hear what is said in a conversation between several persons. The fact that these individuals did not choose the alternative “not difficult at all” may reflect premature signs of hearing loss. If this would be the case, these individuals would constitute a risk group. Consequently, this needs to be assessed in future studies.
Strengths and limitations
Two of the major strengths of this study is the large sample size and the fact that the sample is representative of the general Swedish working population.
The assessment of hearing loss and tinnitus via questionnaires has its advantages and disadvantages, and there is no consensus as to which method is most valid. For example, the subjective evaluation of hearing problems may be difficult to interpret when there is a mild hearing loss. However, situations where an individual constantly has difficulty understanding speech, speech in background noise or difficulty understanding telephone conversations, etc are some examples of individuals illustrating hearing problems. Individuals having these difficulties are usually well aware of their problem. It must be pointed out that an individual who has difficulties in understanding speech in background noise often shows a completely normal audiogram. Thus, there are also limitations to audiological testing. This is not to say that an audiological test would not complement our subjective ratings, but we argue that addressing the question if there are “hearing problems” is acceptable. In fact, rating scales commonly used to assess hearing problems4 26 showed that the single question “Do you feel you have a hearing loss?” was the most sensitive to assess hearing loss compared with pure-tone air conduction audiometry.
Predicting the future
The results of the present study indicate that hearing loss and tinnitus affect more than 30% of the Swedish population. The prevalence of hearing loss or tinnitus was more than 20% for the 40 years and younger group. An important risk factor for hearing loss or tinnitus in young adults and adolescents is exposure to noise in the work place as well as exposure to personal listening devices. Personal listening devices can easily produce sound intensities exceeding 100 dBA. Safety measures suggest 8 h of unprotected hearing for sound levels up to 90 dBA. The majority of auditory scientists are in agreement that long-term exposure to high levels of sound (music, noise, etc) can contribute to hearing problems later in life. Since the development of hearing problems is slow, incremental and irreversible, it can be predicted that the prevalence of hearing loss and tinnitus for the 40 years and younger group will be significantly greater than the present-day 60-year-old group. Moreover, it is reasonable to assume that the hearing complaints by the 40 years and younger group were incurred at least 10 years earlier, further indicating that there is a dire need for effective protection strategies and campaigns aimed at protecting hearing. We found that noise exposure is more common among younger men and that exposure to disturbing and tiring sounds is reported more in younger ages by both women and men. This supports the prediction that hearing problems could become more common in the future.
Conclusion
Hearing problems are a public health issue, and the prevalence is by far more common than previously estimated. In this study, we show that hearing problems are associated with age, sex, SES and noise exposure history. In order to proactively intervene and prevent this deleterious, yet avoidable handicap, statistics need to be regularly updated.
What is already known on this subject
Hearing problems are highly prevalent in the general population.
Hearing loss increases with noise exposure and age and affects men to a greater extent than women.
Tinnitus and hearing loss are often comorbid.
Hearing problems are predicted to become increasingly common in the general population.
What this study adds
The present investigation assessed the prevalence of both tinnitus and hearing loss simultaneously and is the largest study of its kind. It shows that hearing problems are by far more prevalent than previously estimated.
There is a clear socioeconomic gradient in the prevalence of hearing problems, where lower status is more affected.
Approximately every fifth young adult (≤40 years of age) has either tinnitus or hearing loss, indicating that future hearing problems are on the rise.
Acknowledgments
Funding for the epidemiological study was provided by the Swedish Council for Working Life and Social Research (FAS). DH is supported by a grant from FAS Centre for Research on Hearing Problems in Working Life. BC received support from the Swedish Research Council, FAS, the Karolinska Institute and Tysta Skolan. The authors wish to thank Agneta Viberg for excellent technical assistance. We also wish to thank the participants in the SLOSH study and the entire SLOSH study team: Constanze Leineweber, Linda M Hansson, Gabriel Oxenstierna, Anna Nyberg, Cecilia Stenfors and Martin Hyde.
References
Footnotes
Funding Funding for the epidemiological study was provided by the Swedish Council for Working Life and Social Research (FAS). DH is supported by a grant from FAS Centre for Research on Hearing Problems in Working life. BC is supported from the Swedish Research Council, FAS, the Karolinska Institute and Tysta Skolan.
Competing interests None.
Ethics approval This study was conducted with the approval of the Karolinska Institute ethics committee.
Provenance and peer review Not commissioned; externally peer reviewed.