| [Download PDF]
|Year : 2007 | Volume
| Issue : 36 | Page : 55--63
Risk behaviour and noise exposure among adolescents
Margareta C Bohlin1, Soly I Erlandsson2,
1 Department of Social and Behavioral Sciences, University West; Department of Psychology, Göteborg University, Sweden
2 Department of Social and Behavioral Sciences, University West, Sweden
Margareta C Bohlin
Department of Social and Behavioural Sciences, University West, 461 86 Trollhättan
Adolescents in Western society often expose themselves to high levels of sound in gyms, rock concerts, discotheques etc. As these behaviours are as threatening to young people's health as more traditional risk behaviours are, our aim in the present study was to analyze the relationship between self-exposure to noise, risk behaviours and risk judgements among 310 Swedish adolescents aged 15-20 (167 men; 143 women). Adolescents' behaviour in different traditional risk situations correlated with behaviour in noisy environments, while judgements about traditional risks correlated with judgements regarding noise exposure. It is an interesting finding that although young women judge risk situations as generally more dangerous than young men do, they nevertheless behave in the same way. We suggest that this difference is a social and cultural phenomenon which underscores the importance of adopting a gender perspective in the analysis of risk factors. Adolescents reporting permanent tinnitus judged loud music as more risky than adolescents with no symptoms and they did not listen to loud music as often as those with occasional tinnitus. Research on hearing prevention for young people needs to acknowledge and make use of theories on risk behaviour, especially due to the existence of a relationship between adolescents' risk-taking in noisy environments and other types of risk-taking. Similarly, theories on risk behaviour should acknowledge noise as a risk factor.
|How to cite this article:|
Bohlin MC, Erlandsson SI. Risk behaviour and noise exposure among adolescents.Noise Health 2007;9:55-63
|How to cite this URL:|
Bohlin MC, Erlandsson SI. Risk behaviour and noise exposure among adolescents. Noise Health [serial online] 2007 [cited 2021 Nov 30 ];9:55-63
Available from: https://www.noiseandhealth.org/text.asp?2007/9/36/55/36981
Adolescents and young adults frequently expose themselves to loud music, often for hours at a time. Jazz and rock musicians appear, according to Kδhδri,  to have many hearing-related problems such as sensitivity to sound, tinnitus and hearing impairments. The prevalence of tinnitus and hearing impairments among young people seems to increase as a consequence of exposure to loud noise or music played at loud volumes. Young people who play musical instruments, both as amateurs and professionals, experience hearing problems as a side effect of their interest in music. , Seventeen per cent of adolescents in Sweden who play musical instruments were found to suffer from tinnitus and reported that the problem was related to their own playing as well as to visits to discotheques and concerts. 
A number of studies have recently been published on the auditory effects of music exposure at discotheques and rock concerts. , The results from previous studies indicate that concert- and discotheque-goers are routinely exposed to sound levels above 100 dBA.  In a Swedish study of 1285 subjects, 7.8% of young women and 9.6% of young men reported permanent tinnitus.  Significant age-related differences were found with respect to the prevalence rates of experienced tinnitus and noise sensitivity. Older adolescents reported hearing-related symptoms to a greater extent than younger ones, which may be due to more frequent visits to concerts and discotheques by upper secondary school students. In an Argentinean study, the effects of recreational noise exposure on hearing among adolescents over a four-year period showed that there was a tendency for an increase in the mean hearing threshold levels (HTL) for boys at 4000 Hz and 6000 Hz in both ears and while there was no difference for girls at 4000 Hz, there was a tendency towards an increase at 6000 Hz in both ears. , Participation in musical activities increased every year among the subjects with discos being the most popular activity, a finding similar to that of Olsen Widιn and Erlandsson.  However, in general, young men were exposed to higher levels of sound than young women, partly as a result of their involvement in other noisy activities such as noisy sports, the use of noisy tools and carrying out noisy tasks. The equivalent sound levels in discos ranged between 104.3 and 112.4 dBA. ,
According to Arnett,  adolescents who are involved in risky activities and who report high levels of sensation seeking, often listen to hard rock and heavy metal music in order to experience intense feelings. Holmberg and Berg-Kelly  found that unhealthy risk behaviours ( i.e. , the use of alcohol, smoking and taking anabolic steroids) are more common in clusters of young men who have been involved in pregnancies. In other words, different forms of risky behaviours seem to correlate or to express it differently, adolescents often tend to expose themselves to more than one risk. ,
The influence of power that certain genres of music can wield represents strength, hope and meaning in life.  For some adolescents, music has an existential meaning and seems to be related to life and death in a deeper sense. There is a positive relationship between adolescents' musical preferences, self-concept and self-esteem, which according to North and Hargreaves  supports the notion that musical preference is a marker of identity and lifestyle for young men and women. Studies such as Zuckerman's  indicate that individuals who expose themselves to risky situations also engage in less riskier activities such as watching scary movies, travelling, partying as well as listening to loud rock music. Further, listening to rock music is related to smoking, drinking and drug use. Hence, theories of risk-taking behaviours suggest that risk-taking adolescents are a group in society who are frequent listeners to all kinds of music whether on stereos, at concerts or at discos. ,,, Several theories developed in the field of risk-taking behaviour contribute significantly to this area of research (for example, sensation seeking, problem behaviour and decision-making theories).
In the field of psychology, current theories on risk behaviour are rooted in a cognitive and consequential framework and can be regarded as belonging to the broader body of theory on decision-making. Emotions are most frequently seen as components that occur after a decision has been made ( i.e. , following risky activity) and which have no direct effect on social judgements. The causes of risk behaviours and the role of emotions are frequently discussed among researchers. Loewenstein et al. ,  have chosen to study the effect of emotions on risk behaviours as a starting point for their "Risk as Feelings" theory. They focus on the role of affect when the decision is made. People may also experience fear in advance of an unknown situation and also experience a discrepancy between the fear of a risky situation and the cognitive appraisal of the same risk. People's reactions and emotions vary depending on several different factors. The behaviour of people is often driven by emotional reactions. For example, highly anxious individuals tend to judge risky situations as more threatening than those without such problems, something that according to Loewenstein et al. ,  can support the theory of "Risk as Feelings". Furthermore, a recent study about adolescent exposure to noise reveals that high anxiety and sensitivity to noise can have an impact on the use of ear protection.  Consequently, anxiety seems to have an effect on participation in risky activities. We could say that anxiety seems to work as a means of protection against risk behaviour. However, at the same time, people with high anxiety may be "overly" careful and miss out on important learning experiences.
Adolescents engage in activities leading to increased risks as these activities give them certain privileges, e.g., new experiences, higher status among peers, pleasure and satisfaction.  Activities that from an adult perspective, present a risk to adolescent health are not always the same as adolescents' own perceptions of risky activities. ,, There are correlations between what adolescents perceive to be risky activities and their degree of involvement in such activities. If a situation does not appear to be risky, the likelihood of involvement is greater. ,
Whilst many studies that have investigated risk behaviours reveal gender differences, there is currently, according to Ekervald,  a change that has been noticed among young women who now challenge young men's values and behaviours to a much greater degree than before. Byrnes, Miller and Schafer  conducted a meta-analysis of 150 studies with the aim of comparing male and female risk-taking behaviours. Their results indicate that men become involved in risky activities to a higher degree than women. However, it seems likely that gender differences vary according to context and age. Several gender differences increased with age ( i.e. , speeding) while in certain other behaviours ( i.e , smoking), differences were reduced. Gullone et al. , , developed the Adolescent Risk Taking Questionnaire (ARQ) and used it to measure adolescents' own judgements of and engagement in risky activities in an Australian sample. The results obtained from the ARQ reveal that adolescents' own perceptions of risk can be used to predict their risk behaviour. If a situation is perceived as dangerous it is unlikely that the individual will participate in that activity. Similarly, gender too has a predictive value; girls report lower levels of engagement in risk behaviours as compared to boys. ,
The purpose of the present study was to analyze the relationship between Swedish adolescents' self-exposure to high levels of noise and their involvement in risk-situations in a more traditional sense ( i.e., smoking and alcohol use). We hypothesized that a relationship would be found between self-exposure to leisure time noise and traditional risk-taking behaviours and judgements. As research indicates that in general, men are more involved in risky activities than women, we also assumed that gender plays a role in the way that young people judge and become involved in risk situations.
Materials and Methods
The participants were 310 adolescents-143 women and 167 men (aged 15-20 years) recruited from three upper secondary schools in the west of Sweden. The schools were located in a large urban centre ( n = 83), in a medium-sized town ( n = 131) and in a small rural community ( n = 96). Participants were selected from six different theoretical and vocationally-oriented study programs (Technology, Social Sciences, Electronics and Telecommunications, Science, Catering and Child Care). The response rate was 84.2% (310 of 359 pupils), the nonrespondents being those students (15, 8%) who were absent from school on the day when the questionnaire was administered.
The "Adolescent Risk-Taking Questionnaire"  which measures both risk judgements and risk behaviours on a five-point Likert scale, is comprised of 44 items. The respondents were asked to check the box that best described how risky they believed different situations or behaviours to be. The additional formulation "to your hearing" was added to those questions regarding noise. The risk judgement scale considers the degree of risk that is perceived to attach to various situations ( e.g. , smoking, taking drugs and visiting rock concerts). Permitted response options are 'extremely risky', 'very risky', 'risky', 'not so risky' and 'not at all risky' (coded 0 for 'extremely risky' to 4 for 'not at all risky'). The risk behaviour scale measures the extent to which the subject is normally involved in these behaviours (never done, almost never done, do sometimes, do often, do very often) and coded 0 'never done' to 4 'do very often'. The respondents were instructed to check the box that best described their behaviour for each situation on the list. Two questions were added to the original scale in both judgement and behaviour; "Not eating enough" and "Injured oneself (on purpose)". Questions concerning exposure to high levels of sound designed as the original scale were added to the ARQ. Questions on hearing protection use and hearing status were selected from the scale "Hearing Symptoms description" (HSD) and "Youth Attitude to Noise Scale" (YANS) (eight items from HSD and YANS  ). The instrument contained 79 items totally. The demographic data obtained were the age, gender and school attended for each participant (six items). The ARQ was translated into Swedish by two researchers. It was also possible to contact the original creators of the instrument  in the event of any concerns related to the meaning of certain questions. In the questionnaire, nine items regarding the quality of life were also included. However, this data has not been analyzed in this article.
The principals at each school gave their permission for the study to be conducted. The teachers were informed of the purpose of the study and of the times when the questionnaires would be administered, all of which took place during lesson-time. A researcher (the first author) was present during all of the administrations in order to give instructions on how to fill in the questionnaire and to answer any questions raised by the students. In each classroom, the teacher provided assistance by referring any questions that emerged during the approximately 25 minutes it took to fill in the questionnaires to the researcher. The researcher guaranteed the participants discretion and confidentiality.
Statistics and grouping variables
Reliability for the ARQ (judgement and behaviour) was examined using Cronbach's alpha. For the risk judgement part of the questionnaire, a Cronbach's alpha level of 0.88 (24 questions) was attained while the obtained alpha level for the behavioural part (24 questions) was 0.82. When the four items on noise exposure were included in the judgement and the behaviour scale, the alpha values yielded were 0.89 and 0.83 respectively. The frequencies were unevenly distributed in several items and there was a tendency of extreme values being favoured in either direction due to engagement in certain behaviours to either a lesser or a greater degree. It is therefore not certain, given either a larger or a smaller range of options when many different risk behaviours are measured, that a more justified result would have been obtained. However, the total score for both the risk perception and the risk behaviour scales was normally distributed. Bimodal curves were not found. The total mean scores of the risk judgement and risk behaviour scales were 2.01 and 1.03 respectively. More comprehensive information on the evaluation of the ARQ can be found in Appendix 1.
Descriptive statistics were used to examine the collected data in terms of numbers, means, standard deviations and frequencies. Gender-related comparisons regarding hearing problems were examined by the use of the Chi-square test. MANOVA-tests (Multivariate analysis of variance) were conducted to analyze variation in gender for the total ARQ scales and were also calculated for each item on the risk judgement scale. Gender was used as an independent variable and the total score of the entire ARQ (noise included) as the dependent variable (MANOVA). Pearson product moment correlations were performed for the risk behaviour and risk judgement scales.
In order to study comparisons between symptom groups on risk judgment and behaviour, the response rates were compounded together. In the presentation of frequencies, the three response options in the judgement scale regarding noise: "extremely risky", "very risky" and "risky" were compounded to form the category "risky". The two response options, "not very risky" and "not risky at all" were compounded into "not risky". In the behaviour scale, the two response options, "never done" and "hardly ever done" were compounded into "low" and the three response options "done sometimes", "done often" and "done very often" were compounded to form the category "high". This procedure was undertaken in order to gain more distinct response options. The Kruskal Wallis test was then used to make comparisons between the symptom groups.
Risk judgement and risk behaviour
The analysis of the relationship between risk judgement and risk behaviour revealed a moderately high correlation between adolescents' behaviour in different risk situations and behaviour in noisy environments (r = 0.573). Weaker but nevertheless significant correlations were found between traditional risk judgement and judgement regarding noise (r = 0.399) and between risk judgement and risk behaviour (noise included, r = 0.378) [Table 1]. The total score for the judgement and behaviour scale included in the original version of the ARQ , revealed a result of r = 0.365.
For clarification, significant correlations between risk in noisy environments and risk behaviour in general found on separate items are shown in [Table 2]. Listening to high levels of music correlated with several risk behaviours such as smoking, underage drinking, staying out late, getting drunk and leaving school. Visiting discotheques correlated with smoking, underage drinking, staying out late and getting drunk. Another correlation was found between visiting a rock concert and staying out late as well as between visiting motor sport activities and speeding.
Gender: Judgement and Behaviour
MANOVA tests were conducted to investigate gender aspects in judgement and behaviour for the ARQ scale. Women judged risk situations (including risks in noisy environments) to be in general, more dangerous than was the case for men [Table 3]. The mean score for the risk judgement scale was 1.78 among women and 2.08 among men (ranging from 0, extremely risky, to 4, not risky at all) with a total score of 1.94. The difference in risk judgement between women and men was significant (13% variance explained) while there was no gender-related significance regarding risk behaviour [Table 2]. For the risk behaviour scale (ranging 0 for never done to 4 for done very often) the mean among women was 1.06 and among men 1.11 and the total score for both men and women was 1.09.
A total of eight items had been added to both the risk judgement and the risk behaviour scales (four each) and the means for these scales were calculated separately. The average value on the risk judgement scale (noise) was 1.76 for women and 1.99 for men. On the risk behaviour scale (noise), the average value was 1.61 for women and 1.63 for men (total score 1.89/1.62 respectively). The data revealed that women judge motor sport activities as more risky and engage in them to a lesser degree than men. Rock concerts are judged by women as more risky although they visited concerts just as frequently as men. Women tend to judge loud music at discotheques as more risky than men although, despite this, they visit discotheques more frequently than men [Table 3].
On the total risk judgement and behaviour scale (including noise), there was a significant difference between schools. The adolescents in the school situated in a large town judged activities to be less risky than did the students in the school in the small rural location ( F (2, 298) = 5.75, P MS = 1.08). Accordingly, adolescents from the large town school engaged in risky activities to a greater degree than those from the small rural location and the small-sized town ( F (2, 293) = 3.2, P MS = 0.43). Bonferroni posthoc testing confirmed these results.
The prevalence of hearing problems, i.e. , continuous tinnitus and perceived hearing impairment were 6.1% and 5.8% respectively. Men had continuous tinnitus to a greater degree than women and this was also true for both occasional tinnitus and perceived hearing impairment. Conversely, the women perceived that they were significantly more sensitive to noise than the men ( P P = 0.01, n = 281). The group with permanent tinnitus visited motor sport activities more frequently than the group with sensitivity to noise. Tendencies were found with respect to listening to loud music on both the noise judgement and noise behaviour scales (H = 8.72, P = 0.06, n = 280 and H = 8.94, P = 0.06, n = 281 respectively). The group with permanent tinnitus judged "listening to loud music" as more risky than those with no symptoms while the adolescents with occasional tinnitus listened to loud music more often than those with permanent tinnitus. Descriptive statistics for frequencies and percentages for each of the four noisy activities are presented in [Table 5].
Risk judgement and risk behaviour
According to previous research, it would appear that exposure to loud levels of sound including music, is not regarded among those factors usually associated with risk-taking. Zuckerman,  Arnett  and Wang  reported that individuals who expose themselves to risky situations also engage in less risky activities such as watching scary movies, travelling and partying. They also listen to rock music and heavy metal to experience intense feelings. The current results however, reveal that there is a correlation between general risk behaviour and risk behaviour regarding loud music , i.e. , high volumes experienced at concerts and discotheques. The correlation found in this study between traditional risk behaviours and risk behaviours in noisy environments highlights the need to regard noise as a risk factor. This need has been underscored further by Olsen Widιn and Erlandsson, , who reported results on the prevalence of tinnitus and both permanent and temporarily experienced noise sensitivity among adolescents in Sweden. In the aforementioned study, the use of ear protection was not sufficiently frequent to protect young people from unnecessary hearing impairments.
Gender: Judgement and Behaviour
According to previous studies , women judge certain types of behaviour as more risky than men and do not take part in risky activities to the same extent. In the Australian study conducted by Gullone et al.,  while judgements about risky behaviour were more consistent between men and women (with the exception of reckless behaviour, which men judged as less risky), men engaged in various risk behaviours more frequently than women. Our results reveal that young women judge certain behaviours to be more risky than do young men although surprisingly, they tend to involve themselves in risky activities just as much as their male counterparts. There are fewer significant gender differences in the behaviour scale than in the judgement scale. With reference to our results on risk behaviour concerning noise, the results reveal that young women judge loud music at discotheques as more risky than men although they visit discotheques more frequently than their male counterparts. Rock concerts are judged by women to be more risky although they visit concerts to the same degree as men.
One way of explaining the discrepancy between men's and women's evaluations of risk and their actual risk behaviours is by means of a social and cultural framework. The path towards gender equality in Sweden has been uneven. At certain times, great progress has been made whereas at others, gains that have been made have been subsequently eroded, a process that feminists have termed 'backlash'. At the beginning of the new century, we have witnessed a period of extensive activity in the quest to achieve equality between the sexes and this may well have had an impact on risk behaviours among young men and women. According to Ekervald,  young women in Sweden now tend to involve themselves in the same risky activities as men, a contention supported by the present study. It is possible that equality in behaviour still hasn't influenced judgements on and attitudes towards risk behaviours as young people are still strongly influenced by social norms and patterns of socially accepted behaviour. Due to traditional values, the norms for women still seem to be stricter than the corresponding norms and values for men. This creates inconsistency and contradictions in the lives of young women. Actually, if we assess judgement in relation to behaviour, women in fact seem to expose themselves to higher levels of risk than is the case for men. The fact that some women participate in a risky activity even though they judge it to be more dangerous than men might do, is a finding that is in line with the 'Risk as Feelings' theory which posits that people often experience a discrepancy between the fear for a risky situation and the cognitive appraisal of the same risk. People's reactions and emotions vary depending on several different factors such as anxiety or curiosity. 
The adolescents in the school situated in a large town judged activities to be less risky and engaged in risky activities to a greater degree than students in the school in the small rural location. This result may be derived from the fact that a large town offers a greater scope and variety for musical and noise-related activities than do small and rural communities.
The effects of noise
Leisure time noise such as high levels of music at concerts and discos, is a common occurrence in young people's everyday lives and there is no doubt that noise exposure is one of the causes of hearing impairment, tinnitus and hypersensitivity to noise. Impaired hearing has a negative effect on the quality of life and may lead to psychological consequences such as depression, difficulties in concentrating and emotional problems in relationships with other family members, friends and colleagues.  Our results revealed that young people suffering from permanent tinnitus judged "listening to loud music" as more risky than those with no symptoms while young people with occasional tinnitus judged "listening to loud music" as less risky than those reporting permanent tinnitus. These results support the Health Belief Model by Rosenstock,  which suggests that a trigger, in this case, permanent tinnitus, could prompt a person to adopt more health-oriented behaviours. The group with permanent tinnitus indulged in motor sport activities significantly more frequently than the group with sensitivity to noise. This result may be gender-related; men visit motor sport activities more frequently than women and it is also more common that men suffer from permanent tinnitus.
Due to the existence of a relationship between adolescents' risk-taking in noisy environments and their voluntary exposure to other types of risks, research dealing with hearing prevention among young people needs to acknowledge and make use of theories on risk-taking. Likewise, theories on risk-taking should acknowledge exposure to noise as a particular form of risk factor. The obvious health risks related to young peoples' lifestyles and musical preferences can also be connected to other recreational activities. According to Serra et al.  and Biassoni,  who measured the effects of recreational noise exposure on the hearing of adolescents over a four-year period in their Argentinean study, a tendency of increased hearing threshold levels was observed in both genders. However, young men were in general, exposed to higher levels of sound than young women, partly because of their involvement in other noisy activities such as noisy sports, the use of noisy tools and carrying out noisy tasks. In view of the fact that the selection of participants in this study has been limited to 310 individuals in the Southwest of Sweden, it would be worthwhile to conduct additional studies. In particular, our results underscore the importance of analyzing gender in studies with a focus on risk behaviour as a means of achieving an awareness of social changes that take place and have an influence on adolescents' lives. Another limitation of this study is the use of an instrument developed in a different cultural context. Even though the researchers carried out a very careful translation of the instrument from its original language, indeed with the assistance of its original creator, it is inevitable that some biases will occur. One example of such a bias might be that we cannot be sure that all risk situations have the same meaning in both Sweden and in Australia and therefore it is possible that some results depend more on the culture and the language than on the actual judgement or behaviour.
The authors are grateful for the funding provided by Stinger foundation, The Swedish Association for the Hearing Impaired (HRF) and the department of Social and Behavioural Studies, University West.
|1||Kδhδri KR. The Influence of Music on Hearing. G φteborgs Universitet: G φteborg; 2002.|
|2||Hellqvist, C. High levels of noise, adolescent behavior, knowledge and attitudes, 2002.|
|3||Serra MR, Biassoni EC, Richter U, Minoldo G, Franco G, Abraham S, et al . Recreational noise exposure and its effects on the hearing of adolescents, Part I: An interdisciplinary long-term study . Int J Audiol 2005;44:65-73.|
|4||Biassoni EC, Serra MR, Richter U, Minoldo G, Franco G, Abraham S, et al . Recreational noise exposure and its effects on the hearing of adolescents. Part II: Development of hearing disorders . Int J Audiol 2005;44:74-85.|
|5||Clark WW. Hearing: The effects of noise. Otolaryngol Head Neck Surg 1992;106:669-76.|
|6||Olsen SE, Erlandsson SI. Youth Attitudes to Noise Scale (YANS). In : Psychological aspects of adolescents' perceptions and habits in noisy environments. Widιn SE, editor. Doctorial dissertation. Gothenburg University: Gothenburg; 2004a. |
|7||Arnett J. The Soundtrack of Recklessness - Musical preferences and reckless behaviour among adolescents. J Adol Res 1992;7:313-31.|
|8||Holmberg L, Berg-Kelly K. Health, health-compromising behaviour, sexuality and involvement in pregnancy among 18-year-old Swedish males: A cross-sectional survey. Acta Pediatr 2002;91:838-43.|
|9||Wang MQ. Selected lifestyle and risk behaviours associated with adolescents smoking. Psychol Rep 2001;88:77-82. |
|10||Zuckerman M. Are you a risk taker? Psychol Today 2000;33:52-6. |
|11||Bossius T. Transformation of identity. Black Metal, belly dance and homelessness. In : Sernhede, O, Johansson T, editors. Bokf φrlaget Daidalos: G φteborg; 2001. p. 183-209.|
|12||North AC, Hargreaves DJ. Music and adolescent identity. Music Educ Res 1999;1:1.|
|13||Olsen Widιn SE, Erlandsson SI. Self-Reported Tinnitus and Noise Sensitivity among Adolescents in Sweden. Noise Health 2004b;7:29-40. |
|14||Loewenstein GF, Hsee CK, Weber EU, Welch N. Risk as feelings. Psyhol Bull 2001;127:267-86.|
|15||Siegel AW, Cousins JH. Adolescents perceptions of the benefits and risks of their own risk taking. J Emotional Behavl Disord 1994;2:89-99.|
|16||Gullone E, Moore S, Moss S, Boyd C. The adolescent risk-taking questionnaire. development and psychometric evaluation. J Adol Res 2000a;15:231-50.|
|17||Gullone E, Moore S. Adolescent risk-taking and the five-factor model of personality. J Adol 2000b;23:393-407.|
|18||Ekervald H. Adolescent research - Separate worlds and values. In : Nygεrd B, editor. Ungdomsforskning Stockholm; FAS: 2002. p. 14-7.|
|19||Byrnes JP, Miller DC, Schafer WD. Gender differences in risk taking: A meta-analysis. Psychol Bull 1999;125:367-83.|
|20||Erlandsson SI, Hallberg LR. Prediction of quality of life in patients with tinnitus. Br J Audiol 2000;34:11-20.|
|21||Rosenstock IM. The Health Belief Model and preventive health behaviour. Hlth Education Monographs 1974;2:356-86.|