Noise-induced hearing loss in construction workers being assessed for Hand-arm Vibration Syndrome.
House, Ronald A. ; Sauve, John T. ; Jiang, Depeng 等
Many workers are exposed to high noise levels and may develop
noise-induced hearing loss (NIHL). (1,2) Tak et al. (3) estimated that,
after the manufacturing sector, the construction sector had the greatest
number of workers occupationally exposed to noise in the US.
Construction workers in Canada are similarly exposed to high noise
levels (4) and at risk of NIHL. (5,6) In construction work, the use of
hand-held vibrating tools is an important source of noise exposure. (4)
Exposure to hand-arm vibration may result in Hand-Arm Vibration
Syndrome (HAVS) which is comprised of vascular, neurological and
musculoskeletal components. (7,8) The main vascular component consists
of cold-induced blanching of the fingers, a form of secondary
Raynaud's phenomenon which is often referred to as Vibration White
Finger (VWF). (9) Several studies have found that in workers exposed to
both noise and hand-arm vibration, those workers with VWF have greater
hearing loss than workers without VWF, after controlling for noise
exposure. (10-15) However a well-designed study by Pyykko et al. (16
found that the permanent hearing loss in workers exposed to noise and
vibration did not exceed the permanent hearing loss from exposure to
noise only. Therefore it appears that VWF might be a marker for
increased individual susceptibility to noise-induced hearing loss
without the vibration conferring increased risk of hearing loss from
noise exposure. (17,18)
The Occupational Health Clinic at St. Michael's Hospital in
Toronto, Ontario has a comprehensive program to assess workers for HAVS.
Most of the construction workers assessed for HAVS have not had
audiometry at work and therefore the clinic offers audiometry to these
workers as a means of case finding for NIHL. This study was carried out
to assess the extent of NIHL in these construction workers and to
examine possible predictors, in particular the duration of work in
construction and the severity of VWF.
METHODS
A total of 191 construction workers were assessed for HAVS between
January 1, 2006 and December 31, 2007. All had been exposed to noise
from their vibrating tools such as grinders, drills and impact tools.
Twenty-two workers decided not to have the audiometric test that was
offered and therefore the study group with new audiometric data
comprised 169 subjects.
The data abstracted from these workers' medical charts in
2008-9 included the age at assessment, years worked in construction, job
title, audiometric test results and the HAVS Stockholm vascular scale
results in each hand. This scale is based on the frequency and severity
of cold-induced blanching in the fingers and ranges from 0 to 4 with
stage 0 indicating no blanching and stage 4 the most severe
abnormalities. (19) Audiometry had been carried out using a calibrated
Tracor Instruments RA 410S micro-automated audiometer in a soundproof
booth with measurement of hearing levels at 0.5, 1, 2, 3, 4, 6 and 8 kHz
in each ear. The Workplace Safety and Insurance Board (WSIB) in Ontario
uses a presbycusis-corrected average hearing level in the speech range
(0.5 to 3 kHz) of 26.25 dB as the threshold for granting a pension for
NIHL20 which is based on the American Medical Association Guides to the
Evaluation of Permanent Impairment. (21) The presbycusis correction is
(age in years - 60) x 0.5 for those over age 60 which is deducted from
the average hearing level in the speech range to obtain the
presbycusis-corrected average hearing level. Therefore this level of
hearing loss was used in our study as the basis for identifying
clinically significant hearing loss.
Statistical analyses were performed using SAS Version 9.2 (SAS
Institute, Cary, NC). The continuous variables age, years worked in
construction and many of the hearing levels at the specific audiometric
frequencies were not normally distributed and therefore, in the
bivariate analysis, Spearman rank correlations were used. The
relationship between the number (%) of workers with a
presbycusis-corrected hearing level [greater than or equal to]26.25 dB
averaged over the speech range and years worked in construction in the
categories <25, 25-40, >40 years was evaluated using the
Cochran-Mantel-Haenszel test and treating the data as ordinal. To
examine the effect of the HAVS Stockholm vascular scale on hearing loss,
after controlling for years worked in construction, we first conducted
multivariate (more than one dependent variable) linear regression with
the hearing levels at all of the audiometric frequencies (0.5 to 8 kHz)
as dependent variables and the HAVS Stockholm vascular scale and years
worked in construction as independent variables. Follow-up univariate
(single dependent variable) linear regressions were then conducted to
examine whether hearing loss at each audiometric frequency was
associated with the HAVS Stockholm vascular scale after controlling for
years worked in construction. All of these regressions were carried out
for each ear using the Stockholm vascular scale variable from the same
side. Age could not be included in these models because it was highly
correlated with years worked in construction. All p-values were
two-tailed with p<0.05 being statistically significant.
The study was approved by the Research Ethics Board of St.
Michael's Hospital, a teaching hospital affiliated with the
University of Toronto.
RESULTS
All 169 workers in this study were men, median age of 57 (range:
28-75) and median years worked in construction of 35 (range: 4-52). One
hundred and thirty-eight (81.7%) of the workers were pipefitters with
the others working in similar trades. Of the 169 workers, 22 could not
provide an accurate history of the frequency and distribution of finger
blanching to allow the Stockholm vascular classification to be done. Of
the 147 subjects who could be classified, the number (%) of subjects in
the Stockholm scale stages is indicated in Table 1.
The audiometric hearing levels at each frequency in each ear are
provided in Table 2. The highest mean hearing levels were obtained at 6
kHz in each ear and the highest median value of 40 dB was obtained at 4
kHz in the left ear and 6 kHz bilaterally. As indicated in Table 2,
years worked in construction had statistically significant correlations
(p<0.001) with the hearing levels at every audiometric frequency in
each ear. The highest correlations were obtained at the audiometric
frequencies of 6 kHz (r=0.47) for the right ear and 4 kHz (r=0.50) for
the left ear.
The number (%) of subjects with a presbycusis-corrected average
hearing level in the speech range >26.25 dB was 35 (20.7%) in the
right ear, 48 (28.4%) in the left ear and 31 (18.3%) in both ears. As
indicated in Table 3, there was a statistically significant increase
(p<0.001) in the percentage of workers with this level of impairment
for each ear and for both ears combined as the number of years in
construction increased.
The results from the multivariate linear regression (Table 4)
indicated that, after controlling for years worked in construction and
hence noise exposure, the Stockholm vascular scale had a statistically
significant effect on the audiometric hearing levels at all of the
audiometric frequencies (0.5 to 8 kHz) for both the right side (F
(7,138)=3.20, p=0.004) and the left side (F (7,138)=3.11, p=0.005).
However, follow-up univariate linear regression analysis indicated that,
after controlling for years worked in construction, the effect of the
Stockholm vascular scale on the hearing level at each of the specific
audiometric frequencies did not reach statistical significance except
for 0.5 Hz on the left side.
DISCUSSION
In our study, there were statistically significant correlations
between the hearing levels at each audiometric frequency and years
worked in construction. The correlations were highest at 4 and 6 kHz and
tapered off at 8 kHz, which is a typical pattern for NIHL. (1) Therefore
the number of years worked in construction was probably a good metric
for cumulative noise exposure in our study group. However, because of
the high correlation between years worked in construction and age, it is
likely that age also contributed to the hearing loss in some of the
workers in our study.
This study found a high prevalence of clinically significant
hearing loss in construction workers being assessed for HAVS. Thirty-one
(18.3%) subjects had a presbycusis-corrected average hearing level in
the speech frequencies of >26.25 dB in both ears, which is the level
at which a pension would be granted for NIHL by the WSIB in Ontario. The
proportion of subjects with this level of hearing loss increased as the
number of years worked in construction increased, with the trend being
statistically significant.
Other studies have found an increased risk of hearing loss in
construction workers. Tak and Calvert (22) recently reported that the
construction sector had the largest number of workers of any industrial
sector in the US with hearing difficulty attributable to employment. The
estimated prevalence (SE) of hearing difficulty in workers in the
construction industry was 15.1% (0.5%) and the prevalence ratio (95% CI)
in comparison to workers assumed not to be exposed to noise was 1.43
(1.31-1.57). Arndt et al. found a prevalence ratio of hearing loss of
1.5 (1.29-1.82) for construction workers in Germany in comparison to
non-exposed controls. (23) Workers in Canada likely face similar risks
of NIHL. Sinclair and Haflidson carried out a noise survey of workers in
27 construction projects and contractors' facilities in Ontario and
found that the average noise exposure was 98.8 dB(A) with 91.3% of the
samples exceeding 85 dB(A).4
Our study also found a statistically significant multivariate
effect of the HAVS Stockholm vascular scale on the hearing levels at the
audiometric frequencies 0.5 to 8 kHz after controlling for years worked
in construction. However, the effect of HAVS on hearing appeared to be
small because we could only identify a single audiometric frequency of
the 14 specific frequencies tested that had a statistically significant
association with the Stockholm vascular scale. Other studies have
reported an association between VWF and hearing loss with the assumed
mechanism being an increased susceptibility to generalized vasospasm,
including vasospasm in the cochlea, in workers who develop HAVS.10-15
Interestingly, primary Raynaud's phenomenon unrelated to vibration
exposure also has been found to be associated with increased
susceptibility to NIHL. (10)
Zhu et al. reported greater temporary threshold shift in hearing
from acute noise and hand-arm vibration exposure in comparison to just
noise exposure in a small group of healthy subjects. (24) However,
permanent hearing loss is the most important auditory outcome and this
does not appear to be increased from combined noise and hand-arm
vibration exposure in comparison to just noise exposure. (16) Our study
included only workers exposed to hand-arm vibration and therefore could
not further address this question.
Despite the fact that the hazardous auditory effects of noise have
been known for many years, our study has indicated that construction
workers continue to develop NIHL. An increased focus on prevention is
needed for workers in this sector and should involve input from workers
and employers and focus on all levels of prevention. (25) This would be
aided by legislation restricting noise exposure, but in Ontario, the
legislation which limits noise exposure to 85 dB(A) for an eight-hour
daily exposure does not apply to con struction workers and there is no
requirement for these workers to have routine audiometry. The situation
is similar in the US in that the OSHA standard for noise exposure for
construction workers is less rigorous than the standard for general
industrial workers. (3,26) In addition, construction workers often move
from one job site to another to carry out serial construction projects
and this pattern of work is not conducive to having employers provide
comprehensive surveillance as well as other program elements to prevent
hearing loss. (3,26) It also creates difficulty for monitoring of
programs by government inspectors. Construction contractors, workers,
unions, health and safety associations and Ministries of Labour in
Canada need to consider improved methods for prevention of NIHL in
construction workers.
Received: October 21, 2009
Accepted: January 23, 2010
REFERENCES
(1.) Kurmis AP, Apps SA. Occupationally-acquired noise-induced
hearing loss: A senseless workplace hazard. Int J Occup Med Environ
Health 2007;20(2):127-36.
(2.) Nelson DI, Nelson RY, Concha-Barrientos M, Fingerhut M. The
global burden of occupational noise-induced hearing loss. Am J Ind Med
2005;48(6):446 58.
(3.) Tak S, Davis RR, Calvert M. Exposure to hazardous workplace
noise and use of hearing protection devices among US workers-NHANES,
1999-2004. Am J Ind Med 2009;52(5):358-71.
(4.) Sinclair JDN, Haflidson WO. Construction noise in Ontario.
Appl Occup Environ Hyg 1995;10(5):457-60.
(5.) Hessel PA. Hearing loss among construction workers in
Edmonton, Alberta, Canada. J Occup Environ Med 2000;42(1):57-63.
(6.) House RA, Pasut G. Evaluation of the audioscope in an
industrial setting. J Occup Med 1992;34(5):539-45.
(7.) Griffin MJ, Bovenzi M. The diagnosis of disorders caused by
hand-transmitted vibration: Southhampton Workshop 2000. Int Arch Occup
Environ Health 2002;75(1-2):1-5.
(8.) Bovenzi M. Health risks from occupational exposures to
mechanical vibration. Med Lav 2006;97(3):535-41.
(9.) Noel B. Pathophysiology and classification of the vibration
white finger. Int Arch Occup Environ Health 2000;73(3):150-55.
(10.) Palmer KT, Griffin MJ, Syddall HE, Pannett B, Cooper C,
Coggon D. Raynaud's phenomenon, vibration induced white finger, and
difficulties in hearing. Occup Environ Med 2002;59(9):640-42.
(11.) Pyykko I, Starck J, Farkkila M, Hoikkala M, Korhonen O,
Nurminen M. Handarm vibration in the aetiology of hearing loss in
lumberjacks. Br J Ind Med 1981;38(3):281-89.
(12.) Pyykko I, Koskimies K, Starck J, Pekkarinen J, Farkkila M,
Inaba R. Risk factors in the genesis of sensorineural hearing loss in
forestry workers. Br J Ind Med 1989;46(7):439-46.
(13.) Iki M, Kurumatani N, Hirata K, Moriyama T. An association
between Raynaud's phenomenon and hearing loss in forestry workers.
Am Ind Hyg Assoc J 1985;46(9):509-13.
(14.) Miyakita T, Miura H, Futatsuka M. Noise-induced hearing loss
in relation to vibration-induced white finger in chain-saw workers.
Scand J Work Environ Health 1987;13(1):32-36.
(15.) Szanto C, Ligia S. Correlation between vibration induced
white finger and hearing loss in miners. J Occup Health
1999;41(4):232-37.
(16.) Pyykko I, Pekkarinen J, Stark J. Sensory-neural hearing loss
during combined noise and vibration exposure. Int Arch Occup Environ
Health 1987;59(5):439 54.
(17.) Pekkarinen J. Noise, impulse noise, and other physical
factors: Combined effects on hearing. Occup Med 1995;10(3):545-59.
(18.) House R. The effect of combined noise and vibration exposure
on hearing. Occup Health Ont 1988;9(2):72-85.
(19.) Gemne G, Pyykko I, Taylor W, Pelmear PL. The Stockholm
Workshop scale for the classification of cold-induced Raynaud's
phenomenon in the hand-arm vibration syndrome (revision of the
Taylor-Pelmear scale). Scand J Work Environ Health 1987;13(4):275-78.
(20.) Noise-induced Hearing Loss, On / After January 2, 1990.
Document No. 1601-04. Toronto, ON: Workplace Safety and Insurance Board,
2008.
(21.) Guides to the Evaluation of Permanent Impairment, Third
Edition (Revised). Milwaukee, WI: American Medical Association, 1990.
(22.) Tak S, Calvert GM. Hearing difficulty attributable to
employment by industry and occupation: An analysis of the National
Health Interview Survey United States, 1997 to 2003. J Occup Environ
Health 2008;50(1):46-56.
(23.) Arndt V, Rothenbacher D, Brenner H, Fraisse E, Zschenderlein
B, Daniel U, et al. Older workers in the construction industry: Results
of routine health examination and a five year follow-up. Occup Environ
Med 1996;53(10):686-91.
(24.) Zhu S, Sakakibara H, Yamada S. Combined effects of hand-arm
vibration and noise on temporary threshold shifts of hearing in healthy
subjects. Int Arch Occup Environ Health 1997;69(6):433-36.
(25.) Malchaire J, Piette A. A comprehensive strategy for the
assessment of noise exposure and risk of hearing impairment. Ann Occup
Hyg 1997;41(4):467-84.
(26.) Shuter AH. Construction noise: Exposure, effects and the
potential for remediation; a review and analysis. Am Ind Hyg Assoc J
2002;63(6):768-89.
Ronald A. House, MD,CM, MSc, FRCPC, [1,2] John T. Sauve, BScH, [3]
Depeng Jiang, PhD [4]
Author Affiliations
[1.] Division of Occupational and Environmental Health, Dalla Lana
School of Public Health, University of Toronto, Toronto, ON
[2.] Department of Occupational and Environmental Health, St.
Michael's Hospital, Toronto, ON
[3.] Medical student, University of Toronto, Toronto, ON
[4.] Keenan Research Centre, Li Ka Shing Knowledge Institute, St.
Michael's Hospital, Toronto, ON
Correspondence: Dr. Ronald A. House, Department of Occupational and
Environmental Health, St. Michael's Hospital, 30 Bond St.,
Toronto, ON M5B 1W8, Tel: 416-864-5074, Fax: 416-864-5421, E-mail:
houser@smh.toronto.on.ca
Conflict of Interest: None to declare.
Table 1. Number (%) of Subjects in the Stockholm Vascular
Scale Stages
Hand Stockholm Vascular Scale Stage *
0 1 2 3 4
Right 28 (19.0%) 14 (9.5%) 51 (34.7%) 54 (36.7%) 0 (0%)
Left 28 (19.0%) 17 (11.6%) 48 (32.7%) 54 (36.7%) 0 (0%)
* The Stockholm scale stages as described by Gemne et al.19
are as follows:
Stage 0: No attacks of finger blanching
Stage 1: Occasional attacks affecting only the tips of one or
more fingers
Stage 2: Occasional attacks affecting distal and middle
(rarely also proximal) phalanges of one or more fingers
Stage 3: Frequent attacks affecting all phalanges of
most fingers
Stage 4: As in Stage 3, with trophic skin changes in
the fingertips
Table 2. Audiometric Hearing Levels and Spearman
Correlations with Years Worked in Construction
Ear Frequency Hearing Level Correlation
(kHz) Mean (SD) * Median (Range) ([dagger])
(95% CI)
Right 0.5 18.4 (13.7) 15 (-10, 80) 0.16 (0.01-0.30)
([double dagger])
1 17.1 (13.9) 15 (0, 75) 0.24 (0.09-0.38)
2 17.6 (16.2) 15 (-10, 90) 0.36 (0.22-0.48)
3 27.4 (20.9) 25 (-5, 90) 0.43 (0.29-0.54)
4 36.8 (21.5) 35 (-5, 90) 0.42 (0.28-0.53)
6 40.9 (20.6) 40 (5, 90) 0.47 (0.34-0.58)
8 35.5 (23.3) 30 (-5, 90) 0.42 (0.28-0.53)
Left 0.5 19.4 (15.1) 15 (0, 70) 0.22 (0.07-0.36)
1 17.5 (15.6) 15 (-5, 80) 0.25 (0.10-0.38)
2 20.1 (16.7) 20 (-5, 80) 0.34 (0.19-0.46)
3 32.1 (20.8) 30 (-5, 90) 0.48 (0.35-0.59)
4 40.4 (21.0) 40 (0, 90) 0.50 (0.37-0.60)
6 43.7 (21.7) 40 (5, 90) 0.48 (0.36-0.59)
8 37.8 (23.8) 35 (-10, 90) 0.45 (0.32-0.57)
* SD=Standard deviation.
([dagger]) Spearman correlation between audiometric hearing level
and years worked in construction.
([double dagger]) p<0.001 for all correlations.
Table 3. The Number (Percentage) of Workers with a
Presbycusis-corrected Average Hearing Level in the
Speech Range >26.25 dB According to Years Worked in
Construction
Ear Mean HL * Years Worked in Construction
([greater than
or equal to] <25 25-40 >40
>26.25
Right Yes 1 (3.7%) 13 (14.1%) 21 (42.0%)
No 26 (96.3%) 79 (85.9%) 29 (58.0%)
Left Yes 3 (11.1%) 19 (20.6%) 26 (52.0%)
No 24 (88.9%) 73 (789.4%) 24 (48.0%)
Both Yes 1 (3.7%) 11 (12.0%) 19 (38.0%)
No 26 (96.3%) 81 (88.09%) 31 (62.0.0%)
Ear p-value
([dagger])
Right <0.001
Left <0.001
Both <0.001
* Mean HL=hearing level averaged over the speech range (0.5-3 kHz)
after subtracting a presbycusis correction for subjects aged over 60
years as follows: [(age in years - 60) x 0.5].
([dagger]) The p-value was obtained from the Cochran-Mantel-Haenszel
test treating the data as ordinal (i.e., indicative of trend).
Table 4. Linear Regression Analysis of Audiometric Hearing Levels
Ear Audiometric Predictor Estimate
Frequency (SE)
Right Overall Years in construction *
Stockholm vascular scale
([dagger])
0.5 Years in construction 0.37(0.12)
Stockholm vascular scale 1.08(1.07)
1 Years in construction 0.47(0.12)
Stockholm vascular scale 0.80(1.07)
2 Years in construction 0.58(0.14)
Stockholm vascular scale 2.03(0.19)
3 Years in construction 1.03(0.17)
Stockholm vascular scale 1.18(1.47)
4 Years in construction 1.02(0.17)
Stockholm vascular scale 1.27(1.54)
6 Years in construction 1.05(0.17)
Stockholm vascular scale 1.50(1.46)
8 Years in construction 1.02(0.19)
Stockholm vascular scale 0.71(1.70)
Left Overall Years in construction
Stockholm vascular scale
0.5 Years in construction 0.49(0.13)
Stockholm vascular scale 2.32(1.14)
1 Years in construction 0.55(0.13)
Stockholm vascular scale 1.12(1.17)
2 Years in construction 0.72(0.14)
Stockholm vascular scale 0.58(1.20)
3 Years in construction 1.16(0.16)
Stockholm vascular scale 0.17(1.38)
4 Years in construction 1.27(0.16)
Stockholm vascular scale 1.97(1.37)
6 Years in construction 1.11(0.16)
Stockholm vascular scale 0.50(1.44)
8 Years in construction 1.11(0.19)
Stockholm vascular scale 1.68(1.62)
Ear Audiometric F-value p-value
Frequency
Right Overall 6.81 <0.001
3.20 0.004
0.5 0.003
0.31
1 <0.001
0.46
2 <0.001
0.09
3 <0.001
0.42
4 <0.001
0.41
6 <0.001
0.30
8 <0.001
0.68
Left Overall 9.83 <0.001
3.11 0.005
0.5 <0.001
0.04
1 <0.001
0.34
2 <0.001
0.63
3 <0.001
0.90
4 <0.001
0.15
6 <0.001
0.73
8 <0.001
0.30
* Years worked in construction is a continuous variable. The estimates
indicate the effect of a single year worked in construction.
([dagger]) Stockholm scale is treated as a continuous variable. The
estimates indicate the effect of one stage of the Stockholm scale.
