Occupational Medicine 2009;59:483–486

Published online 3 July 2009 doi:10.1093/occmed/kqp091

Noise-induced hearing loss in French police officers

Francxois-Xavier Lesage1, Nicolas Jovenin2, Frederic Deschamps1 and Samuel Vincent1
...................................................................................................................................................................................

Background There is a lack of data about police officers’ hearing thresholds and the risk of noise-induced hearing loss
(NIHL) associated with this occupation. In France, 129 000 national police officers, 96 000 state police
force members and 16000 municipal police officers may be affected by occupational noise exposure.
...................................................................................................................................................................................
Aims To evaluate the association between police employment and NIHL.

Downloaded from https://academic.oup.com/occmed/article/59/7/483/2751295 by guest on 18 October 2022

...................................................................................................................................................................................

Methods We undertook a cross-sectional study using review of medical records. Audiometric and otological
data and information on potential confounders were extracted from medical records. Global hearing
loss and selective 4000 Hz hearing loss were analysed.
...................................................................................................................................................................................

Results Of total, 1692 subjects (887 policemen and 805 civil servants) participated in the study. After adjust-
ing for potential cofounders, police officers were 1.4 times more likely to have a selective 4000 Hz
hearing loss than civil servants (95% CI 1.1–1.9). This difference was greater between motorcycle
police officers and civil servants (OR 5 3; 95% CI 1.4–6.3).
...................................................................................................................................................................................
Conclusions These data suggest that occupational noise exposure in police work, particularly in motorcycle police
officers, may induce hearing loss. Noise sources need to be more accurately defined to confirm high-
level noise exposures, to better define significant sources of noise and to identify effective solutions.
...................................................................................................................................................................................

Key words Motorcycle police officers; noise-induced hearing loss; occupational hearing loss; occupational noise
exposure; police officers.
...................................................................................................................................................................................

Introduction LEq,d 80 dB(A) is exceeded. To our knowledge, there

According to the World Health Organization, 250 million
people in the world have disabling hearing impairment of
moderate or greater severity [1]. The leading causes of
hearing loss for deaf and hard-of-hearing adults in the
USA are noise, age and ear infection [2]. Although occu-
pational hearing loss is a well-recognized hazard in some
industries or occupations with exposure to high noise lev-
els (e.g. airport ground crews), it has not been evaluated
as fully in occupations for which the risk is not so overt,
such as police officers. Hearing impairment caused by the
noise of gunfire has been studied, mainly in soldiers [3,4].
Increased hearing thresholds have been identified in some
specialized police forces, such as traffic police officers in
a large Egyptian city [5] and police dog handlers [6], but
data are unavailable for Western European police officers
other than motorcycle officers. Since 2006, French legis-
lation has introduced a limit value of equivalent con-
tinuous noise level exposure at work (LEq,d) of 87
dB(A). Actions to control exposure are required where
1
UF de Pathologie Professionnelle et Santé au Travail (Occupational Health
Department), Hôpital Sébastopol, Reims, France.
2
Institut Jean Godinot, Reims, France.
Correspondence to: Francxois-Xavier Lesage, UF de Pathologie Professionnelle et
Santé au Travail (Occupational Health Department), Hôpital Sébastopol, 48 rue
de Sébastopol, 51092 Reims Cedex, France. Tel: 103 26 78 89 33;
fax: +03 26 78 43 56; e-mail: fxlesage@chu-reims.fr
has to date been no noise level exposure assessment in
French police forces. Because our department provides
an occupational health service for the police force of
Reims, France, we sought to verify whether the risk of
noise-induced hearing loss (NIHL) was present. We pres-
ent here the results of a comparative cross-sectional study
of the association between police work and hearing loss.
Methods
We undertook a comparative cross-sectional study involv-
ing police officers (including motorcycle police officers)
who worked from 1999 to 2005 in Reims, France (pop-
ulation 200 000). The comparison group comprised
other municipal or state civil servants from the same city,
specifically administrative staff, research workers, techni-
cians and janitors, selected because they were free from
occupational noise exposure.
All the subjects in each group had been in their job for
at least 6 months and were seen at our clinic for routine
health surveillance. This included systematic standard-
ized audiometric testing and an interview about past
and current noise exposure. Subjects with current exter-
nal or middle ear disease (i.e. otitis or wax) were not in-
cluded in the study.
Standardized medical records for all subjects attending
our clinic include information regarding medical history,

The Author 2009. Published by Oxford University Press on behalf of the Society of Occupational Medicine.
All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
484 OCCUPATIONAL MEDICINE
including ear and hearing-related problems, noise expo-
sure during leisure activity, military service and work his-
tory, including prior occupations, specific job exposure to
occupational risks and number of years in their occupa-
tion. These data were extracted anonymously from the
medical records for our study. Cigarette smoking status
at the time of the medical and hearing examination was
also recorded.
Audiometric testing was conducted using a single audi-
ometer (Audioscan, Essilor) calibrated to the manufac-
turer’s instructions according to the ISO 389 standard.
This audiometer also complied with the French CEI 645
and ISO 6189 norms. Trained occupational physicians
conducted audiometric testing in a dedicated room that
met the audiometer manufacturer’s specifications and
the French ISO 8253-1 norm. A senior occupational
health specialist was responsible for audiometry quality
control. Pure-tone air-conduction thresholds were deter-
mined for each ear with chromatic progression from 250
to 8000 Hz.
Hearing damage was assessed in terms of either global
hearing loss or selective 4000 Hz hearing loss. Global
hearing loss was defined as a pure-tone average of thresh-
olds at 500, 1000, 2000 and 4000 Hz .20 dB. In cases of
asymmetric hearing loss, in accordance with the Bureau
International d’AudioPhonology [7] classification sys-
tem, the average loss in the better ear was multiplied
by 7 and in the worse ear by 3, and the sum divided by 10.
Selective 4000 Hz hearing loss is characteristic of the
onset of NIHL [9], defined here as a hearing loss .30 dB
(uni- or bilateral) at 4000 Hz without loss at 8000 Hz.
Independent ethical approval is not needed for this
type of study design in France.
Statistical analyses were conducted using SAS Version
8.2 Software (SAS Institute Inc., Cary, NC, USA). For
univariate analyses, we used the x2 test for categorical var-
iables and Student’s t-tests of differences in means for
continuous data. Logistic regression was used to evaluate
the odds of having a hearing loss (global hearing loss and
NIHL) associated with being a police officer, adjusting
for age, sex, smoking and other potential confounders
[8]. Potential interaction effects were tested but elimi-
nated from our presented models as they were not statis-
tically significant. Differences were considered significant
if the two-sided P-value was 0.05 or less.
Results
Of the 1880 people eligible for the study, 1692 people
participated, including 887 police officers (of whom 33
were motorcycle police officers) and 805 civil servants.
Of total, 188 were not included because of the missing
data.
The descriptive characteristics of participants in the
study are presented in Table 1.
In all, 23% of police department staff and 21.5% of
civil servant had global hearing loss. In all, 28% of police
department staff had NIHL versus 16% of civil servants.
Univariate analysis showed that police officers were sig-
nificantly younger that controls and more predominantly
male (Table 1). They also were more likely to have a prior
history of noise exposure and to be cigarette smokers.
In the multivariate analysis, after adjusting for age, sex,
smoking status, medical and ear, nose and throat (ENT)
history and previous or leisure-time noise exposures,
working in the police department was not associated with
the global hearing loss (OR 5 1.04; 95% CI 0.7–1.4)
(Table 2). However using the same model, selective 4
Downloaded from https://academic.oup.com/occmed/article/59/7/483/2751295 by guest on 18 October 2022
kHz hearing loss was more likely in police officers than
among controls (OR 5 1.4; 95% CI 1.1–1.9). It appears
that this increased risk of NIHL in all police officers is not
entirely attributable to motorcycle officers as the NIHL
risk for non-motorcycle officers was also weakly increased
(OR 5 1.37; 95% CI 1.0–1.8), although the risk was great-
est in motorcycle officers (OR 5 3; 95% CI 1.46.6).
Discussion
Our study found an increased risk of NIHL in motorcycle
officers and to a lesser extent in other police officers.
Some technical limitations could not be avoided in this
study. First, the timing of the audiometry assessment in
relation to when subjects were last exposed to noise could
not be controlled. In addition, we did not identify partic-
ular sources of noise exposure among the different occu-
pational groups. The French norm recommends testing
hearing 3 day after the last noise exposure, but it was
not possible to achieve this in this study. Therefore, it
is possible that the effect of temporary threshold shift
has led to an overestimate of the real risk of NIHL.
We did not use a soundproof booth in our study, but
noise levels in the dedicated room used for testing were
,30 dB(A). Moreover, the testing conditions were the
same for policemen and administrative staff. Therefore,
the lack of a soundproof booth has probably had a weak
influence on our results.
The main strengths of this study are the size of the
study groups and the fact that we took into consideration
potential sources of bias, notably past noise exposures.
In a logistic regression model adjusted for age, sex,
current smoking, previous occupational noise exposure,
leisure-time noise exposure and a history of acute noise
exposures, being a police officer was associated with se-
lective 4000 Hz hearing loss, but not with global hearing
loss. This selective hearing loss suggests that there is an
occupational risk of NIHL in police officers.
In the literature, global hearing loss is associated with
factors such as presbycusis and ear diseases. Only very
substantial noise exposure results in global hearing loss
and an association between global hearing loss and occu-
pational noise exposure is therefore found only if
 F.-X. LESAGE ET AL.: NIHL IN FRENCH POLICE OFFICERS 485

Table 1. Characteristics of participants and univariate analyses (police officers versus civil servants) of age, sex, smoking history, noise
exposure, medical history and hearing loss

Characteristics No. (%) of participants Univariate

analyses
All participants, Policemen, Motorcycle policemen, Others civil servants, P
N 5 1692 n (%) n 5 887, n (%) n 5 33, n (%) n 5 805, n (%)

Age (mean, years) 37.6 years 41.8 years ***

,25 108 (6) 89 (10) 0 19 (2) ***
25–34 504 (30) 317 (36) 13 (39) 187 (23) ***
35–44 473 (28) 186 (21) 9 (27) 287 (36) ***
45–54 515 (30) 292 (33) 11 (33) 223 (28) *
$55 92 (5) 3 (0) 0 89 (11) ***

Downloaded from https://academic.oup.com/occmed/article/59/7/483/2751295 by guest on 18 October 2022

Male 1119 (66) 776 (87) 33 (100) 343 (43) ***
Current smokers 516 (30) 303 (34) 10 (30) 213 (26) ***
Exposure and medical history
Leisure-time noise exposure 79 (5) 66 (7) 3 (9) 13 (2) ***
Acute noise exposure history 19 (1) 16 (2) 1 (3) 3 (0) **
Previous occupational 225 (13) 165 (19) 7 (21) 60 (7) ***
noise exposure
Medical historya 262 (15) 141 (16) 6 (18) 121 (15) ns
ENT historyb 67 (4) 17 (2) 1 (3) 50 (6) ***
Hearing lossc
,21 dB 1310 (77) 678 (76) 25 (76) 632 (79) ns
21–40 dB 250 (5) 191 (21) 8 (24) 159 (20) ns
41–54 dB 28 (2) 17 (2) 0 11 (1) ns
55–79 dB 4 (0) 1 (0) 0 3 (0) nsd
Selective hearing loss at 4000 Hz 377 (22) 249 (28) 15 (45) 128 (16) ***

ns, no significant difference; *P , 0.05, **P , 0.01, ***P , 0.001.

a
History of non-ENT disorder such as toxic, iatrogenic, genetic or neurological hearing loss.
b
History of ENT disorder.
c
According the International Bureau for AudioPhonology (BIAP) classification.
d
Tested with Fisher’s exact test.

Table 2. Adjusted ORsa for job and global or NIHL hearing impairment for this reason, making it an appro-
priate measurement for our study [9].
Variables Global hearing loss NIHL We therefore believe that police officers may have mod-
OR (95% CI) OR (95% CI) erate but deleterious occupational noise exposures. As
this has not to date been reported in ordinary police offi-
Civil servants 1.00 (referent) 1.00 (referent) cers, other studies are needed to confirm our results.
Active policemen 1.04 (0.72–1.40) 1.41 (1.06–1.9) Police officers are potentially exposed to multiple sour-
Non-motorcycle 1.03 (0.76–1.40) 1.37 (1.02–1.84) ces of noise, including vehicle horns, gunfire, dog barking
policemen and traffic noise, including, for motorcycle officers, mo-
Motorcycle policemen 1.06 (0.42–2.64) 3.00 (1.36–6.63)
torcycle noise.
These may involve both impulse and chronic noise
OR, odds ratio; CI, confidence interval. exposures, the latter being defined as .85 dB LEq,d
a
Adjusted for age, sex, smoking status, leisure-time noise exposure, acute noise (equivalent daily continuous sound level).
exposure and previous occupational noise exposure.
Reims police officers are required to undertake shooting
practice (80 cartridges per year). However, safety measures
occupational noise exposure is sufficiently intense and are strict and their use seems to be effective [10]. Hearing
prolonged [9]. Occupational noise exposure in police loss associated with gunfire is more likely to come from
work is not known to induce such extensive hearing loss. inadvertent unexpected exposure than from planned
Therefore, the absence of an association between global shooting range activity where hearing protection is used.
hearing loss and police work is consistent with no more Chronic exposure to traffic noise could be an impor-
than moderate occupational noise exposure. tant source of occupational hearing loss, especially in mo-
By contrast, selective 4000 Hz hearing loss is thought torcycle police officers, as previously observed [11,12].
to indicate early or moderate NIHL and to be specific for Noise exposure in motorcyclists is known to be
486 OCCUPATIONAL MEDICINE
hazardous. According to Ross, during town driving equiv-
alent continuous noise levels ranged from 63 to 90 dB(A),
intercoms giving the highest peak levels. On the open
road, levels were up to 105 dB(A) LEq [13]. Our findings
are consistent with this previous data.
It would be interesting to examine any correlation be-
tween the duration of employment as a motorcycle police
officer and the degree of NIHL. However, the use of Pear-
son’scoefficient ofcorrelation wasnotappropriate,asafun-
damental requirement of its use is that the two variables be
normally distributed, which was not the case with our data.
Traffic noise is very variable within and between cities
[4,14,15] and at different times. Unfortunately, we could
not quantify the impact of such variables in our study.
Moreover, there is a variety of police tasks, each likely
to involve differing noise exposures, such as patrol work,
traffic police duties, work in motorcycle squads and dog-
handling units and riot control activities. Our study was
only able to single out motorcycle police officers (Table
2), and it would be useful to further refine the analysis
taking into account the noise exposure levels specific to
different police tasks.
This study was conducted on convenience samples,
namely Reims civil servants and police officers, rather than
randomly selected ones. France has 129 000 national
police officers, 96 000 state police officers and 16 000 mu-
nicipal police officers. Although extrapolation of our con-
clusions to other French police forces requires caution, the
size of the study, the great uniformity of police training and
practice in France and the homogeneity of French cities
suggest these findings may be widely applicable, not only
in France but also possibly in other European police forces
as well. It is important to replicate and extend our obser-
vations, because if confirmed, effective preventive meas-
ures are available and could be implemented.
Key points
• Occupational noise exposure in non-motorcycle
police officers may induce hearing loss.
• Motorcycle police officers appear to be at greater risk
of hearingloss causedbyoccupationalnoise exposure.
• If these observations are confirmed by other studies,
effective preventive measures should be implemented,
both in France and in police forces elsewhere.
Conflicts of interest
None declared.
References
1. Facts about Deafness. Bull WHO. http://www.who.int/pbd/
deafness/facts/en/ (15 November 2007, date last accessed).
2. Ries PW. Prevalence and characteristics of persons with
hearing trouble: United States, 1990–1991. Vital Health
Stat 1994;10:1–75.
3. Kiukaanniemi H, Lopponen H, Sorri M. Noise-induced
Downloaded from https://academic.oup.com/occmed/article/59/7/483/2751295 by guest on 18 October 2022
low- and high-frequency hearing losses in Finnish con-
scripts. Mil Med 1992;157:480–482.
4. Ylikoski J. Acute acoustic trauma in Finnish conscripts. Eti-
ological factors and characteristics of hearing impairments.
Scand Audiol 1989;18:161–165.
5. Kamal AA, Eldamati SE, Faris R. Hearing threshold of
Cairo traffic policemen. Int Arch Occup Environ Health
1989;61:543–545.
6. Reid A, Dick F, Semple S. Dog noise as a risk factor for
hearing loss among police dog handlers. Occup Med (Lond)
2004;54:535–539.
7. BIAP Recommendation 02/1 bis. Audiometric Classification
of Hearing Impairments. http://www.biap.org/biapanglais/
rec021eng.htm (15 November 2007, date last accessed).
8. Hosmer DW, Lemeshow S. Applied Logistic Regression. New
York, NY: John Wiley and sons Inc, 1989.
9. McBride DI, William S. Audiometric notch as a sign of
noise induced hearing loss. Occup Environ Med 2001;58:
46–51.
10. Ylikoski M, Pekkarinen JO, Starck JP, Paakkonen RJ,
Ylikoski JS. Physical characteristics of gunfire impulse noise
and its attenuation by hearing protectors. Scand Audiol
1995;24:3–11.
11. Mc Combe AW, Binnington J, Davis A, Spencer H. Hear-
ing loss and motorcyclists. J Laryngol Otol 1995;109:
599–604.
12. Pierson WR, Mahe JE. Noise and the highway patrolman.
J Occup Med (Lond) 1973;15:892–893.
13. Ross BC. Noise exposure of motorcyclists. Ann Occup Hyg
1989;33:123–127.
14. Seto EY, Holt A, Rivard T, Bhatia R. Spatial distribution of
traffic induced noise exposures in a US city: an analytic tool
for assessing the health impacts of urban planning decisions.
Int J Health Geogr 2007;6:24.
15. Orlando P, Perselli F, Cristina ML, Piromalli W. Environ-
mental and personal monitoring of exposure to urban noise
and community response. Eur J Epidemiol 1994;10:
549–554.