Occupational noise exposure and hearing: a systematic review

In a Swedish cross-sectional study, Ivarsson et al. compared hearing in employees in the automotive industry, shipyards and quarries (N = 1796) (L _{ex, 8h} > 95 dB) for the period 1983–1990 with hearing in office workers (L _{ex, 8h} < 80 dB) and found that among blue collar workers over the age of 50, only 8–28 % had a normal hearing, compared to 70 % among office workers (Ivarsson et al. 1992). The results were compared with another Swedish study from 1970 to 1971. Exposed workers more than 50 years had about the same level of hearing acuity in the 1980s as in the 1970s, while exposed workers aged 20–30 had a significantly better hearing in the 1980s compared to the 1970s and about as well as nonnoise-exposed workers.

In a longitudinal study for the period 1983–1989, Lee-Feldstein compared the hearing loss in the range 2–4 kHz of 11,435 US automobile workers in five plants over a 5-year period with 331 nonexposed controls (Lee-Feldstein 1993). The aim of this study was to evaluate the efficacy of preventive measures. The noise exposure was from L _{ex, 8h} = 85 to 114 dB. In four of the five factories, the difference in hearing between exposed and controls was not significant.

In a retrospective study of male metal workers at a Swedish car factory, the hearing loss for 1964, 1972, 1980, 1987 and 1989 was compared to ISO 1999, Annex A (Bruehl et al. 1994). The study showed that during the period 1964–1989, the hearing loss decreased from about 20 to 5 dB in the age group of 20–29 years and from 30 to 10 dB in the age group 50–59 years, all values compared to ISO 1999 A. The authors attributed this improvement to better access to and use of hearing protection and lower noise exposure during the period. If the results had been compared to recent Swedish reference values (Johansson 2002) instead of the ISO 1999 Annex A, hearing among the youngest in 1989 would have been found close to normal.

Hearing loss was also found in a Russian study of the personnel in two workshops, revealing that impulse noise caused a greater hearing loss than one would expect from the estimated A-weighted noise dose (Suvorov et al. 2001). The authors suggested that an extra safety margin of 5 dB should be subtracted from the threshold value for what is considered harmful impulse noise.

Martin et al. (1975) examined hearing in 228 Canadian smelter workers in three different departments and compared it to nonexposed employees from the same factory. The daily noise exposure level was mostly below 90 dB, which was then the Canadian threshold limit value (TLV), and therefore hearing protection was not found to be necessary. The prevalence of a hearing loss, defined as the average >25 dB for the 0.5–2-kHz area, ranged from 14 to 32 % in exposed >50 years, compared to 4 % of the controls. The hearing loss was greatest in areas with impulse noise and lowest in areas with a continuous noise level, L _Aeq, of 85–90 dB.

Keatinge and Laner examined 63 factory workers younger than 40 years with a noise exposure, from 115 to 128 dB (sound pressure level, SPL), without the use of hearing protection, and with short service time, and compared their hearing with hearing in a control group. They found that most of the hearing loss occurred during the first 3 years of employment (Keatinge and Laner 1958).

Noise-exposed industrial workers (N = 47,388), 85 % men, mostly between 20 and 55 years, and with a daily noise exposure level between 86 and >105 dB, were examined in Austria. The strongest estimated effect of noise, adjusted for age, gender and other covariates, was a standardized regression coefficient of 0.08 for 4 kHz (Bauer et al. 1991). This corresponds to a 5-dB difference between nonexposed and heavily exposed subjects.

In a cross-sectional study, Somma et al. (2008) examined hearing of 184 male cement workers, with a daily noise exposure level >85 dB and compared it with 98 nonexposed controls. The study revealed a hearing loss in the 3–6-kHz range of 5 dB among the youngest, age 21–30, and 20 dB among the oldest, age 51–60, compared to the controls.

In a follow-up study of 449 male steel workers with exposures ranging from L _{Aex, 8h} < 90 to L _{Aex, 8h} > 100 dB, no changes in hearing in the 0.5–6-kHz range during a 6–8-year period could be attributed to noise exposure (Howell 1978).

In a longitudinal study of 113 Egyptian male cotton workers, Moselhi et al. (1979) reported that a daily noise exposure level <85 dB gave only a slight hearing loss compared to the control group (N = 64), whereas exposure >85 dB revealed a hearing loss of >25 dB in the frequency range 0.5–2 kHz in 9.6 % of the employees.

Swedish pulp workers (N = 319), aged 26–40 years at the beginning of the study, with varying daily noise exposure levels from 80 to 100 dB, were followed with repeated audiometric examinations from 1959 and 20 years ahead. The hearing loss was most pronounced at 4 kHz, approximately 15 dB. There was only a small difference in hearing between high- and low-exposed subjects. The authors suggested that the small difference might reflect chemical exposure among the workers with low noise exposure. The importance of age as a possible cause of the hearing loss was not discussed in this study (Bergstrom and Nystrom 1986).

Egyptian metal workers (N = 88) were examined in 1980 and then again 8 years later. They were exposed to a mean background noise, L _Aeq, of 90–94 dB and impulse noise, L _{A, peak} of 112–139 dB with 20–50 “beats per minute” (bpm). They did not use any hearing protection, and the hearing loss was significant in 1980. Upon follow-up, the age-adjusted hearing loss had increased somewhat, particularly in the 0.5–2-kHz range. Most of the hearing loss in the higher frequencies (3–8 kHz) occurred in the first 10–15 years after employment (Kamal et al. 1989).

Rabinowitz et al. followed the hearing acuity in 6217 aluminum workers with daily noise exposure in four categories from <82 to >88 dB during the period from 1990 to 1996. The hearing loss in the period was lowest for the highest exposed group and normal for the lowest exposed group, compared with US reference values. The use of hearing protection among the highest exposed may partly explain the findings (Rabinowitz et al. 2007).

Hearing in 78 male aluminum workers with daily noise monitoring inside the ear protectors was followed for 4 years and compared with a control group (N = 234) matched for age, initial hearing thresholds and noise exposure. The intervention group did not develop any hearing loss, while the control group who received no intervention, had a small hearing loss. The study shows that a close follow-up with the use of hearing protection may have a beneficial effect (Rabinowitz et al. 2011).

It appears that in Western countries, the noise exposure levels in the industry have been reduced over the last few decades and this has led to a reduced hearing loss and improved hearing in noise-exposed groups in recent years (Rubak et al. 2006). The hearing loss appears to be greatest during the first years of noise exposure. There are conflicting results as regards the effect of hearing protection, but a recent US study indicates that a close follow-up of the use of hearing protection has a clear advantageous effect on hearing (Rabinowitz et al. 2011).

In an Indian study, 276 shipyard employees with daily noise exposure >90 dB were compared with 276 age- and sex-matched controls from the office staff without any occupational noise exposure. Among the exposed workers, 6 % were assessed to have a NIHL, as compared to none of the controls. The criterion for NIHL was a notched hearing loss (Bhumika et al. 2013).

Nilsson et al. examined hearing of 1492 employees at a shipyard with an exposure to noise from L _Aeq = 88–94 dB and with a lot of impulse noise, typically 2500 impulses per day with peak noise levels from 105 to 135 dB (A). Despite the fact that almost 90 % used hearing protection, about 60 % had a reduced hearing acuity. After individual age correction to the ISO 1999, there were still about 40 % with reduced hearing acuity, most of them with exposure >5 years. The authors conclude that the noise in shipyards with impulse noise is particularly damaging to hearing (Nilsson et al. 1977).

Historically, there have been noise exposure levels in the shipyards high enough to cause NIHL.

Seixas et al. studied the hearing acuity of 393 apprentices in the building trades and 62 controls (students) from 2000 to 2010. At the last follow-up, the group was reduced to 258 in the building trades and 58 controls. The daily noise exposure level was 87 dB for the noise exposed and 70 dB for the students. An exposure increase of 10 dB gave a hearing loss of 2–3 dB after 10 years at 3–6 kHz. Most of the hearing loss had already occurred before the initial examination. Self-reported data on noise outside work had no impact on hearing. The author suggests that the limit of 90 dB does not assure good enough protection to prevent NIHL (Seixas et al. 2004, 2005, 2012).

Dutch construction workers (N = 29,644) with a daily noise exposure of 87–96 dB were examined longitudinally with respect to hearing and compared with an internal unexposed group. The hearing loss at 4 kHz ranged from 0 dB for the age group <25 years to about 7 dB for the 55–64 year group, compared with the internal control group, which corresponds very well to ISO 1999, Annex B (unscreened) (Leensen et al. 2011).

Engdahl and Tambs found that male construction workers may be one of the groups with the most pronounced NIHL with an average hearing loss of 9 dB in the 3–6-kHz frequency range and 6 dB for the average of 0.5, 1, 2 and 4 Hz, compared to nonnoise-exposed male teachers (Engdahl and Tambs 2010).

Overall, the literature indicates that NIHL is a frequent diagnosis in the construction industries, but we have found surprisingly few high-quality studies published from the industry. This is unfortunate, considering the potentially high noise exposure levels.

Morken et al. examined the incidence of NIHL among offshore workers on the Norwegian continental shelf, reported to the Petroleum Safety Authority (PSA) from 1992 to 2003. The study revealed a significant increase, from 1/1000 employed in 1992 to 9/1000 in 2003 (Morken et al. 2005). The majority of the cases were reported among mechanics, surface treatment workers, electricians, process technicians and roughnecks, most of them aged 50–59 years.

In 2002, Zacchariasen and Knusden stated that there is a problem with high noise exposure in the Norwegian offshore industry (Zachariassen and Knudsen 2002). Nistov et al. (2012) in a later study reported that there is a high noise exposure level, a risk of NIHL and a need for preventive measures in this industry.

Ross et al. (2010) however, found in a recent study that offshore workers except for divers had a normal hearing compared with a nonexposed population and so did Johnson and Gann (1991) in a former longitudinal study.

There is a great deal of concern over the noise exposure level and the perceived risk of NIHL in the offshore sector. The number of studies is limited, but the evidence suggests that offshore workers as a group have a relatively normal hearing. More and larger longitudinal studies are needed.

Previous cross-sectional studies of divers have not found any extra hearing loss compared to the general population, despite noise exposure, long diving experience and experienced barotrauma (Brady et al. 1976). More recent prospective studies of professional divers have shown that diving can cause impaired hearing, however (Molvaer and Lehmann 1985; Molvaer and Albrektsen 1990; Skogstad et al. 2000). Even among divers who do not report that they have been exposed to noise, excessive hearing loss has been found (Edmonds 1985). In most of these studies, exposure to noise has not been measured, but self-report and the number of dives are used as a proxy measure of exposure.

Divers can be exposed to both airborne and waterborne noise during their work, and high noise levels in the working environment of the diver can affect hearing (Summitt and Reimers 1971). It is very difficult to measure the noise submerged (Nedwell and Parvin 1994), and hearing under water is more characterized by bone conduction than air conduction (Hollien 1993). Since the impedance between water and air is different, the sound intensity will be lower in water at a given sound level. Noise from the air supply to the helmet, in addition to the noise from the pressure chamber and hydraulic tools, will add to the total noise exposure (Molvaer and Gjestland 1981; Curley and Knafelc 1987). Use of tools and explosions can generate impulse noise, even in water.

Solvents and gases, such as carbon monoxide, and heavy metals, such as lead, arsenic and mercury, might also affect hearing in divers (Phaneuf and Hetu 1990). Moreover, decompression sickness and barotrauma may be detrimental to hearing (Edmonds et al. 1992). Decompression sickness may occur by the formation of gas bubbles in the small blood vessels and the fluid in the inner ear (Shupak et al. 1991). Barotrauma of the inner ear may occur due to problems equalizing the pressure in the middle ear. This can happen during diving, when increased ambient pressure leads to a relatively low pressure in the middle ear (Edmonds et al. 1992). In addition, head injury and infections of the ear by Pseudomonas aeruginosa may affect hearing in divers (Edmonds et al. 1992; Ahlen et al. 1998). Hearing loss in the low-frequency ranges, 0.5–2 kHz, is described in two longitudinal studies among divers with and without noise exposure (Harashima and Iwasaki 1965; Haraguchi et al. 1999). In both cross-sectional and prospective studies, hearing loss in the high-frequency area has been associated with exposure in terms of number of dives (Coles 1976; Molvaer and Lehmann 1985; Zulkaflay et al. 1996; Skogstad et al. 2009; Ross et al. 2010). Some studies of divers’ hearing have revealed better hearing than the reference group even after 6 years of follow-up (Molvaer et al. 1983; Molvaer and Lehmann 1985; Skogstad et al. 2005). This may be due to selection mechanisms, as divers generally have better health and hearing than others due to screening procedures before hire.

Overall, there is evidence that diving can cause a moderate hearing loss in both the low- and the high-frequency ranges. This applies to both divers using air as the breathing gas and saturation divers. In the early years of the career, it appears that divers hear better than average. This may be due to health selection.

In a cross-sectional study of firefighters in Phoenix and Fort Worth, USA, Clark and Bohl found that firefighters with an estimated equivalent noise exposure level of 80–90 dB (A) had a normal hearing, compared to ISO 1999. A follow-up study showed that the hearing loss in this group was somewhat less than expected from the ISO 1999 (Clark and Bohl 2005).

Similar findings were reported in a small study of 171 Korean firefighters with exposure to noise from emergency vehicles at L _{ex, 8h} = 99–108 dB (Kim et al. 2011) and in a study of 100 Iranian firemen who were compared with 100 nonexposed controls (Assadi et al. 2013).

Kales et al. (2001) compared hearing in 319 firefighters with ISO 1999, Annex A. Firefighters <40 years old had a nearly normal hearing, while the oldest (>50 years) had a hearing loss of 20–30 dB more than expected in the 3–6-kHz range.

Since younger firefighters seem to have a normal hearing, there is no basis for claiming that this group nowadays has a great risk of hearing loss caused by noise exposure in the workplace, but high-quality longitudinal studies are missing.

A cross-sectional study of the incidence of hearing loss and tinnitus in 204 infantry officers exposed to impulse noise from various weapons with peak levels up to 185 dB (SPL), revealed a significant hearing loss compared with ISO 1999 (Christiansson and Wintzell 1993).

Segal et al. studied the development of hearing in 841 men aged 20–40 years with terminated exposure, but who had been exposed to noise trauma (shooting/explosions) and who had applied for financial compensation. Hearing in this group was compared with hearing in 150 men with continued exposure. Hearing in the noise trauma group stabilized after approximately 1 year, whereas the group with continued exposure had a continued hearing loss (Segal et al. 1988).

In three large studies of 87,000 to over 140,000 US military personnel, Helfer et al. (2005, 2010) and Helfer (2011) reported a higher than expected incidence of hearing loss among infantrymen, in those with active war experience, in men compared to women and in those over 40 years. Unfortunately, high-quality exposure data are missing.

In a prospective study of 804,535 soldiers returning from Iraq and Afghanistan in 2003–2009, the incidence of reported cases of hearing loss increased significantly during the period, from no cases in 2006 and earlier to about 5 cases per 100 in the first quarter of 2009 (Helfer et al. 2011). This was attributed to increased focus on hearing loss since 2006, showing that the extent of hearing loss may be significant for military missions.

Similar findings have been reported from a follow-up hearing study of 747 recruits with a service time of 11.7 months, who were compared with nonexposed controls (Muhr et al. 2006). Of the artillery recruits, 17 % had a “significant hearing loss” of >15 dB in at least one frequency, compared to 2.9 % of the controls. The authors conclude that recruits still lose hearing during service despite comprehensive preventive measures.

Hearing in pilots in the Finnish Air Force was examined by Kuronen et al. (2004) in a cross-sectional study. The equivalent noise exposure level in the cockpit was 90–100 dB (A), but the exposure periods were relatively short. Comparing with ISO 1999, the pilots heard clearly better than normal. Health selection criteria for recruitment to the pilot profession were the most probable cause.

Ribak et al. (1985) reported that hearing in 777 pilots/navigators in the Israeli Air Force was reduced with increasing age, but no significant impact of the type of aircraft or the number of hours flown could be demonstrated.

Hearing in 525 pilots aged 20–40 years in the French Air Force was examined in a cross-sectional study (Raynal et al. 2006). “Abnormal hearing” was detected in 19 % of the 20–30-year-old pilots and 38 % in those aged 30–40 years. Transport pilots had a slightly better hearing than the fighter and helicopter pilots. All groups had an audiometric notch at 6 kHz. Their hearing was not compared to any reference material, and noise exposure data were not provided.

In a prospective study, 512 pilots in the French Air Force aged 20–40 years, exposed to a noise level from 90 to 140 dB (A) (SPL), were followed over a 3-year period. Hearing was measured by pure tone audiometry and otoacoustic emissions (OAE) (Job et al. 2009). The author concluded that the use of OAE to some extent could predict who was at risk of developing reduced hearing.

Trost and Shaw (2007) compared the first and last audiograms of 267,568 persons in the US Navy during the period 1982–2004 and found that the risk of developing a hearing loss of 10 dB or more in the 2–4-kHz range was increased more for every year serving on a warship (relative risk, RR 1.062) compared to service ashore (RR 1.035). The daily exposure level was >84 dB.

Overall, it appears that the military work experience is a significant risk factor for hearing impairment.

In a prospective study, Wagstaff and Årva compared hearing in Norwegian pilots and helicopter pilots over a 2–3-year period with air traffic controllers and reference material (ISO 1999). The hearing loss in the 3-, 4- and 6-kHz ranges was slightly greater than expected according to ISO, but not compared to the air traffic controllers. Helicopter pilots had a normal hearing despite a significantly higher noise exposure, L _Aeq, of 90–95 dB, compared to 80–85 dB for fixed wing pilots (Wagstaff and Årva 2009).

Qiang et al. followed 3019 pilots aged 45–54 years flying smaller aircrafts over a 10-year period. The noise exposure in smaller aircrafts was assessed to be slightly higher than in ordinary civil aviation. The flight time was used as a proxy measure of cumulative noise exposure. When adjusted for age, there was a positive trend for the effect of flying hours, but it was not statistically significant (Qiang et al. 2008).

In a cross-sectional study, Kidera and Gaskill compared hearing in 1443 US airline pilots with a reference material. The pilots had a normal hearing (Kidera and Gaskill 1974).

Lindgren reported that hearing in Swedish pilots (N = 664) and cabin crew (N = 936) with a daily noise exposure level of 75–81 dB was normal, compared to a Swedish reference population (Lindgren et al. 2008, 2009).

In a cross-sectional study, Smedje et al. compared hearing in 327 aircraft mechanics with a Swedish reference population. The daily noise exposure was from 70 to 91 dB with A-weighted maximum levels at 119 dB (Smedje et al. 2011). A median hearing loss of 2–3 dB in the group of 35–39 years was found. Other age groups had normal or slightly better than normal hearing.

Overall, civil aviation personnel with the current hearing conservation measures are generally not exposed to a noise level that is detrimental to hearing.

In a cross-sectional study of 9778 male railway employees, a hearing loss of 2–7 dB was found in the frequency range above 2 kHz, compared to ISO 1999, Annex B (Kryter 1991). The use of firearms was reported to be associated with a greater hearing loss. In another cross-sectional study, Clark and Popelka (1989) found that hearing in 9427 trainmen did not differ from that of a reference population (ISO 1999). The daily noise exposure was on average 78 dB and ranged from 61 to 89 dB.

Virokanna et al. (1994) found that Finnish railway track maintenance workers with a daily noise exposure level of 94 dB (A) had a smaller hearing loss than expected, compared to the ISO 1999. This was attributed to the use of hearing protection.

Overall, there is no basis for claiming that this group runs a high risk of hearing loss caused by noise exposure in the workplace, but longitudinal studies to strengthen this conclusion are not available.

In a cross-sectional study, hearing in 60 US farmers from Iowa was compared with hearing in 60 age- and sex-matched nonnoise-exposed controls. The farmers had significantly reduced hearing compared to the controls (Plakke and Dare 1992). Stewart et al. (2003) reported that US part-time farmers had reduced hearing compared to full-time farmers and attributed this to more noisy and low-quality farming machinery in part-time farmers. Renick et al. (2009) examined hearing in 212 American children and young people in agriculture and found that they had a higher hearing threshold and twice as many audiometric notches as a comparable reference population.

A prospective Norwegian study on various occupations and hearing revealed a binaural hearing loss of 5.3 dB in the 3–6-kHz area among farmers, compared with a reference group (teachers) (Engdahl and Tambs 2010). Hwang et al. (2001) also reported an increased risk of hearing loss in farmers associated with noise exposure in an interview survey of 1622 farmers. A report from New Zealand assessed farmers as a high-risk group for occupational hearing loss (Thorne 2006).

In conclusion, farmers seem to be at risk of developing NIHL.

In 1994, Palin published a review article on hearing in classical musicians (Palin 1994). The author discusses findings from a total of 11 studies published between 1960 and 1992. In several of the studies, the participation rate was low and the measured noise levels are described as “unlikely harmful to hearing”. In spite of this, Palin concludes that the overall opinion is that musicians are at risk of developing NIHL.

Since 1994, several cross-sectional studies of classical musicians have been published (Kahari et al. 2001; Emmerich et al. 2008; Hamdan et al. 2008; Jansen et al. 2009; Pawlaczyk-Luszczynska et al. 2011; Toppila et al. 2011). Few of the studies have measured the exposure levels in the orchestra of the musicians, but some studies show equivalent noise levels of 80–90 dB (A) (Emmerich et al. 2008; Jansen et al. 2009; Pawlaczyk-Luszczynska et al. 2011), and the number of hours of exposure was also carefully recorded (Hamdan et al. 2008). Most of the studies find that the musicians do not get any more hearing loss than the controls. However, there are studies indicating that the hearing loss among musicians is greater than expected for age (Emmerich et al. 2008). Brass players and percussionists may be more affected than other musicians (Pawlaczyk-Luszczynska et al. 2011), and one study suggests a hearing loss in the 6-kHz range for this group (Jansen et al. 2009). Otoacoustic emissions (TEOAE) were recorded in a study of singers. Singers with normal hearing had a lower “signal-to-noise ratio” than controls, which may indicate subclinical cochleae dysfunction (Hamdan et al. 2008). In a follow-up study over 16 years of 56 musicians, male musicians lost 0.7 dB per year in the 3–8-kHz range, while the female musicians lost 0.4 dB per year. The losses were not greater than in a control group (Kahari et al. 2001). Similar results were reported in another study from the Nordic countries, where 135 musicians were examined after 3 and 8 years. After 8 years, the hearing loss was close to normal (Ostri and Parving 1991).

In a cross-sectional study of rock musicians from 1978, the prevalence of hearing impairment was found to be remarkably low (Axelsson and Lindgren 1978). In a follow-up study of 53 Swedish rock musicians 16 years after the initial investigation, the hearing loss was slightly less than expected, despite an equivalent exposure level of 90–105 dB from 20 to 25 h/week (Axelsson et al. 1995). Drummers had a slightly greater hearing loss than other musicians. The author suggests that a positive attitude to the music may have a protective effect on hearing. In another study, a group of drummers was examined by otoacoustic emissions. The amplitudes of DPOAE (6000 Hz) were missing to a greater extent than in the controls, suggesting a hearing loss (Pride and Cunningham 2005).

Other studies of rock/jazz musicians have uncovered equivalent exposure levels from 111 to 129 dB (A) (Kähäri et al. 2003) and 83–90 dB during practice, and 90–96 dB during concerts (McIlvaine et al. 2012). Equivalent levels at 95–108 dB (A) were measured at rock concerts in Sweden (Almstedt et al. 2000). Subjective hearing impairment was reported in 75 % of the musicians (Kähäri et al. 2003). Temporary problems, such as tinnitus and hyperacusis in adolescents attending rock concerts, are also being reported, but an excess risk of permanent hearing loss could not be demonstrated (Almstedt et al. 2000).

The noise exposure level is most pronounced in rock/jazz musicians, who also have more hearing complaints and a greater degree of hearing loss than in the case for classical musicians. A possible elevated risk of hearing loss is quite modest for musicians as a group.

Gärding (1980) studied noise and hearing in 17 Swedish nurseries with 79 employees. The equivalent noise exposure was on average 83 dB (A). Employees with more than 11 years of employment heard worse than those with less seniority, but age and noise exposure levels were not adjusted for.

Rubak et al. (2006) compared hearing in a nursery staff with a group not exposed to noise and found that the nursery staff had a normal hearing.

Results from a Norwegian survey revealed normal hearing in a group of 165 nursery workers (Engdahl and Tambs 2010).

Overall, the literature suggests that the noise exposure is too low to cause any hearing loss among nursery staff, and their hearing does not differ from nonexposed controls.

Lesage et al. conducted a cross-sectional study of hearing in 887 French policemen compared with 805 office workers, using medical records. No noise measurements were available. The risk of hearing loss exceeding 30 dB at 4 kHz was elevated, odds ratio (OR) 1.41 (1.06–1.90), especially for the use of police motorcycles (Lesage et al. 2009).

Hearing in Australian mining workers was tested in 1985–1988 in a cross-sectional study (N = 8774). The group had an average daily noise exposure of 90 dB. More than 40 % of them had a hearing loss that gave the right to workers’ compensation. This was less than in the period from 1982 to 1985, where 56 % had a right to compensation. The hearing loss was associated with age and exposure to noise. The use of hearing protection appeared to have some preventive effect (Leigh and Morgan 1990).

Rubak et al. compared hearing in Danish workers in noise hazardous work with nonnoise-exposed and found that the risk of hearing loss >20 dB at 2–4-kHz area was tripled by exposure to noise for more than 20 years. For employees <30 years of age, or who started in noise-exposed work after 1990, there was no increased risk of hearing loss (Rubak et al. 2006).

In a Norwegian prospective study, the mean binaural age-adjusted hearing loss, 3–6 kHz, in men was greatest among woodworkers (11.2 dB), miners (10.9 dB), construction workers (9.2 dB), military personnel (8.2 dB) and farmers (5.3 dB), compared to teachers (Engdahl and Tambs 2010). In younger men and women, the impact of occupational exposure was much smaller.

In three small cross-sectional studies, Fuente et al. reported that solvent exposure is associated with hearing loss (Fuente and McPherson 2007a, b; Fuente et al. 2009; Fuente and Hickson 2011). The same was reported in studies by Jacobsen et al. (1993) in a Danish cross-sectional study of 3282 men, by Botelho et al. (2009) among 155 Brazilian steel workers and by Kim et al. (2005) among 542 men from the aerospace industry. A Polish cross-sectional study of 3741 men from 24 factories also found that solvent exposure was associated with hearing loss (Dudarewicz et al. 2010). In a cross-sectional study of 393 American apprentices, a 1-year solvent exposure was estimated to cause a hearing loss of 0.6 dB (Seixas et al. 2004).

Chang et al. compared hearing in 58 employees with exposure to noise and toluene with 58 employees with noise exposure only and 58 nonexposed workers. The prevalence of hearing loss >25 dB was higher in the noise + toluene group compared to the noise group and lowest in the nonexposed group (Chang et al. 2006).

Carbon disulfide was shown to reduce hearing in the low frequencies in a study of 346 rayon wool workers, where 105 were exposed to equivalent noise levels of 80–90 dB, 132 were exposed to a combination of noise and carbon disulfide, and the rest were not exposed (Chang et al. 2003). The hearing loss was about 10 dB in both exposed groups compared to the nonexposed.

Sliwinska-Kowalska et al. (2005) reported that the combination of noise and solvents, such as styrene, xylene, n-hexane and toluene, caused a hearing loss in a study of 1117 workers from different industries. Another study of 290 solvent-exposed workers and 213 controls from a plastic boat factory revealed that various combinations of styrene, toluene and noise gave significantly increased hearing loss and that the effects were at least additive, and perhaps synergistic (Sliwinska-Kowalska et al. 2003). In a study of 701 shipyard workers exposed to noise, xylene and toluene, the combination of exposures on hearing was found to be additive (Sliwinska-Kowalska et al. 2004).

In an Australian study from a ceramic factory, solvent-exposed workers got audiometric notches earlier than those who were not solvent exposed, Safia Beshir et al. (2011).

Morata et al. (1993) reported that the combination of noise and solvents increased the hearing loss compared with exposure to noise and solvents separately, i.e., a synergistic effect.

However, no detrimental effect on hearing was found from exposure to styrene in a study of 32 boat builders (Hoffmann et al. 2006) nor in a European population exposed to solvents (Fransen et al. 2008) or in styrene-exposed German ship workers (Triebig et al. 2009).

In a study of 412 Taiwanese steel mill workers, an association between lead in blood and hearing loss was reported, but the study revealed no effect on hearing of exposure to copper, zinc, arsenic and cadmium (Hwang et al. 2009).

Choi et al. (2012) reported an association between cadmium in blood and hearing loss, but not for lead in blood.

Crawford et al. reported a weak correlation between exposure to organophosphates and self-reported hearing loss in a survey of 14,229 pesticide-exposed workers (OR 1.17) (Crawford et al. 2008).

There are numerous studies on humans showing that various chemicals may affect hearing. The exposure data on noise and chemicals are of low quality in most studies, and it is therefore difficult to conclude on a dose–response relationship.

In a review article, Morata and Johnson (2012) have described a number of animal studies showing dose–response relationship. Several solvents, such as styrene, toluene, xylene, ethylbenzene, trichloroethylene, n-hexane, jet fuel, white spirit and other solvent mixtures, have ototoxic properties in humans and affect both cochlear and more central nervous structures (Johnson and Morata 2010). Ototoxic properties are also evidenced by exposure to lead, carbon monoxide and pesticides, such as organophosphates (Burgess and Williams 2006). There is experimental evidence from animal studies that the effects of noise and solvents are synergistic with regard to hearing loss (Morata and Johnson 2012), and this may also be the case in human studies (Jacobsen et al. 1993; Morata et al. 1993; Kim et al. 2005; Fuente and McPherson 2007a, b; Botelho et al. 2009; Fuente et al. 2009; Fuente and Hickson 2011). The damage to the cochlea of ototoxic substances is similar to that from exposure to noise and may explain the synergistic effect.

In addition, it appears that age may modify the ototoxic effect. Young laboratory animals appear to be more vulnerable than older ones with respect to hearing loss from both noise and chemicals. Whether this also applies to exposure to chemicals or noise in humans is more uncertain (Johnson and Morata 2010).

In summary, the chemicals best documented to have an effect on hearing in humans at exposures around the occupational exposure limit (OEL) are styrene, toluene, lead, mercury, carbon disulfide and carbon monoxide (Johnson and Morata 2010).

Several medications have ototoxic properties. Most notable is the cancer drug, cisplatin, which has a fairly strong ototoxic effect. Another group of ototoxic drug is aminoglycosides. Aspirin may also cause hearing loss (Johnson and Morata 2010).

The US National Institute of Occupational Safety and Health (NIOSH) and the American College of Occupational and Environmental Medicine (ACOEM) have recommended that exposure to ototoxic drugs and chemicals must be taken into account when risk assessments and hearing conservation programs are being made (Johnson and Morata 2010; Kirchner et al. 2012). The US Occupational Safety and Health Administration (OSHA) recommends regular hearing testing of workers exposed to ototoxic substances. In Australia and New Zealand, compensation for occupational hearing loss caused by ototoxic drugs and chemicals may be granted (Johnson and Morata 2010).

The relationship between ototoxic drugs and chemicals and hearing loss is far from fully examined, but such exposures must be included in the assessment of possible occupational hearing loss in workers and in the design of preventive measures (Johnson and Morata 2010; Kirchner et al. 2012).