A cross-sectional study of the relationship between background noise and hearing threshold among Indian mine workers: formulating regression equations to reduce the false- positive hearing impairment cases

doi:10.21203/rs.3.rs-2878565/v1

Background noise is believed to affect audiometry results conducted particularly in non-soundproof booth in various occupational settings. The present study first of its kind investigated the effect of background noise on the results of audiometry among mine workers. Furthermore, the study formulated regression equations to rationalise audiometry conducted in a non-soundproof audiometry booth. In this cross-sectional study, audiometry was performed on mineworkers (n=388) divided into four groups with variable background noise outside the soundproof booth and then within the soundproof booth. The hearing levels of audiograms obtained outside and inside the soundproof booth were compared. A graph was plotted based on the differences in hearing level and background noise outside the audiometry booth. The relationship between variations in hearing level and background noise was explored using a coefficient of correlation. When the outside background noise level increased from 26 dB to 62 dB, the difference in hearing level increased from 2 dB to 27 dB across all groups. The mean hearing levels were found to be more affected at lower frequencies than at higher frequencies. Based on background noise and the variance in hearing level for each frequency, the study proposed regression equations. The study concluded that background noise had a greater impact on audiometry results at lower frequencies. Formulated regression equations may have the advantage of reducing the number of false positive hearing impairment cases that would have been identified in non-soundproof booths. The findings of the study are important particularly among mine workers in developing countries like India.

Health sciences/Health occupations

Health sciences/Medical research

audiometry

background noise

hearing level

mine worker

Noise is one of the most common occupational hazards in the mining and manufacturing industries ¹. Continuous exposure to noise at workplace can cause cumulative permanent loss of hearing over the years. Audiometry is the standard test for detecting and evaluating hearing impairment and determining an individual's hearing threshold to pure tones ranging from 0.5K to 8K Hz ².

Background noise is believed to affect audiometry results during the procedure, but excluding all ambient noise from an audiometric test room is neither feasible nor practical due to structural and cost considerations. There are prescribed international guidelines which allow acceptable background noise for conducting audiometry like International Standard (ISO: 8253) and American National Standard (ANSI: S3.1-1999) ^3,4 Soundproof booth is believed an ideal set-up for conducting audiometry wherein the background noise is below the standards.

Audiometry conducted outside a soundproof booth exceeds these standards due to the presence of background noise above the maximum permissible ambient noise level (MPANL). The audiometry conducted outside a soundproof booth show an increased hearing threshold at lower and higher frequencies, resulting in higher cases of hearing impairment.

Due to the lack of an ideal soundproof booth, audiometry is mostly conducted in a small closed room with minimal background noise during routine medical examinations in an industrial setting, wherein audiograms of the majority of the workers show hearing impairment at both lower and higher frequencies ^5,6. It is possible that excessive background noise resulted in false positive cases.

According to literature, audiometry performed outside of a soundproof booth primarily impacts lower frequencies, with little to no impact on higher frequencies. Bromwich, et al (2008) concluded that audiometry conducted in presence of background noise affected mostly in the low-frequency range ⁷. Lo and McPherson, (2013) observed that the audiometry conducted without soundproof booth masks the test tones at all frequencies wherein the detection levels of the hearing threshold got affected resulting in a false positive diagnosis. As a result, the individual may be referred for further diagnostic tests ⁸. Wong, et al (2003) reported that testing environments with moderate to substantial ambient noise may result in an overestimation of the shift in hearing threshold more at both lower and less at higher frequencies ⁹. However, none of the earlier studies proposed a regression equation to rationalise audiometry conducted booth outside the soundproof booth to inside soundproof booth.

The study was taken up with the objective to ascertain differences in hearing levels due to the presence of background noise in audiometry conducted in a non-soundproof booth as compared to soundproof booth. Further, the study formulated regression equations to rationalise audiometry conducted in a non-soundproof audiometry booth.

Study design, setting, and population:

The study was carried out among 388 mine workers of 11 mines from April 2018 to June 2019 in Nagpur, Maharashtra State and Bharatpur, Rajasthan State. The study subjects were divided into four groups, the study of group 1 and group 2 was conducted at Nagpur while the study of group 3 and group 4 was conducted at Bharatpur. The workers available on the day of the examination who were ready to be part of the study were included. The study was carried out in accordance with the appropriate regulations and guidelines after receiving approval from the Institutional Ethics Committee. Informed consent was obtained from each study subject in their vernacular language, which contained a clause concerning the possibility of publishing their identifying information or images. The workers who were suffering from neurological disorders, with hearing aids were excluded from the study.

Data collection procedures:

Detailed history and general examination:

Participants’ data on personal demography, work, addiction, etc. were recorded in the data collection form. A physical examination including a general medical check-up and measurements of height (in centimetres), body weight (in kilograms), and blood pressure conducted and recorded.

An otoscopic examination was done to ascertain the presence of wax, foreign body, inflammation of the outer ear, eczema of the pinna etc. to rule out conductive deafness before audiometry ¹⁰.

Pure tone audiometry:

The audiometric evaluation of these workers was carried out using a 2 Channel clinical portable PC-based audiometer (Labat Asia) with TDH-39 supra-aural earphones which were calibrated before the study. The study subjects were not exposed to noise for more than 12 hours to avoid the risk of temporary threshold shift (TTS) which would have influenced the test results ¹¹.

The audiometry was conducted for sound frequencies ranging from 0.5K Hz to 8K Hz and the minimum intensity for measurement was from 10 dB(A) and the responses of the subject were noted. Audiograms of study subjects were evaluated for classification based on pure tone audiogram evaluating the thresholds of hearing for frequencies of 0.5K, 1K, 2K, 3K, 4K, 6K and 8K Hz. Trials were repeated at least 3 times to determine the lowest signal intensity, which served as the final threshold value for each ear. The mean threshold values at 500, 1000, and 2000 Hz were used to determine low-frequency hearing status, while the mean threshold values at 3000, 4000, 6000 and 8000 Hz were used to determine high-frequency hearing status ¹².

The test was conducted for the same person in a room with background noise [fig. 1 (A)] and was repeated after an interval of about two hours in a soundproof audiometry booth [fig. 1 (B)]. Both the audiometry booths used in the study complied with ANSI standards for ears covered with supra-aural earphones. Audiometry of all study subjects inside and outside the soundproof audiometry booth was conducted by the same technician.

Measurement of the background noise

Type 1 Sound Level meter (make Casella) was used for measuring the background noise in the soundproof booth and outside the booth for seven studied frequencies. Sound level meter was placed about one feet distance from the subject and background sound was measured. To assess the effect of background noise on audiometry results, each study group was assigned to a different location with a different background sound ¹³.

Statistical Analysis: Mean ± SD calculated for all studied parameters. The decimals being rounded off to the nearest number. The correlation between background noise and differences in hearing level was studied by using a coefficient of correlation (r). To predict the difference in hearing level i.e. dependent variable (y-axis) by known background noise i.e. independent variable (x-axis) coefficient of regression was used and to determine the accuracy of the coefficient of regression R-squared values were obtained by using Excel.

The study was conducted among 388 mine workers from the 11 mines divided into 4 groups with varying background noise (Figure 2).

General characteristics of study subjects:

The study participants were male, with a mean (±SD) age of 37 (±7) years (Table 1). The mean (±SD) work experience was 9 (±5) years. The workers had a mean (±SD) BMI of 22 (±4) kg/m². Eighteen, twenty-four and five percent of study participants admitted to tobacco smoking, tobacco/gutka chewing, and alcohol drinking, respectively. The mean (±SD) systolic and diastolic blood pressure was 114 (±12) and 76 (±7) mmHg.

Table 1: General Characteristics of mine workers from eleven mines of Maharashtra and Rajasthan between April 2018 to June 2019

Characteristics	Group 1 (n= 62)	Group 2 (n= 35)	Group 3 (n= 164)	Group 4 (n= 127)	Total (n=388)
Age (yrs.)	40 ± 6	45 ± 8	30 ± 8	33 ± 10	34 (±10)
Work experience (yrs)	13 ± 4	19 ± 11	5 ± 4	6 ± 6	8 (±7)
Height (cm)	166 ± 7	162 ± 7	166 ± 6	166 ± 7	165 ± 6
Weight (kg)	57 ± 10	60 ± 10	64 ± 12	66 ± 11	66 ± 10
BMI (kg/m²)	20 ± 4	23 ± 3	23 ± 4	24 ± 4	23 (±4)
Systolic BP (mm hg)	112 ± 15	134 ± 19	121 ± 13	120 ± 11	120 ± 12
Diastolic BP (mm hg)	77 ± 9	86 ± 13	81 ± 8	79 ± 7	80 ± 7

Table 2 depicts background noise in the non-soundproof audiometry booth, hearing thresholds inside and outside the soundproof booth, and hearing level differences. Group 1 had minimum background noise and group 4 had maximum background noise outside the soundproof booth in all frequencies. In all groups, maximum background noise was seen at 0.5K Hz., while it was minimal at 8K Hz.

It was observed that the mean hearing levels were affected more at lower frequencies than at higher frequencies in all groups depicting that background noise impacted lower frequencies as compared to higher. The highest impact was seen in group 4 at 0.5K of 25 dB while the least was seen in group 1 at higher frequencies of 2 dB.

Table - 2: Background noise, hearing levels outside and inside the soundproof booth and mean difference in hearing levels

Frequency ( Hz)		0.5K	1K	2K	3K	4K	6K	8K
Frequency ( Hz)		Lower frequency			Higher frequency
Group 1 (n= 62)	Background noise outside booth	35 ± 1	33 ± 1	41 ± 2	32 ± 1	33 ± 1	27 ± 1	26 ± 1
	Hearing level outside booth	25 ± 6	24 ± 6	23 ± 6	28 ± 10	29 ± 13	33 ± 13	29 ± 11
	Hearing level inside booth	20 ± 6	19 ± 5	18 ± 6	26 ± 10	27 ± 14	31 ± 12	27 ± 11
	Difference in hearing level	5 ± 3	5 ± 3	5 ± 2	2 ± 2	2 ± 2	2 ± 2	2 ± 1
Group 2 (n= 35)	Background noise outside booth	42 ± 2	43 ± 2	42 ± 2	40 ± 3	39 ± 2	39 ± 1	35 ± 1
	Hearing level outside booth	36 ± 7	35 ± 8	33 ± 8	37 ± 11	36 ± 11	41 ± 13	36 ± 17
	Hearing level inside booth	25 ± 7	26 ± 8	24 ± 9	30 ± 12	30 ± 11	37 ± 14	33 ± 17
	Difference in hearing level	11 ± 2	09 ± 3	09 ± 3	7 ± 3	06 ± 3	04 ± 2	03 ± 2
Group 3 (n= 164)	Background noise outside booth	57 ± 2	56 ± 3	56 ± 3	52 ± 2	51 ± 4	50 ± 3	49 ± 3
	Hearing level outside booth	48 ± 8	42 ± 9	38 ± 11	42 ± 13	40 ± 15	46 ± 16	44 ± 15
	Hearing level inside booth	27 ± 8	25 ± 9	23 ± 11	29 ± 13	32 ± 16	35 ± 16	32 ± 15
	Difference in hearing level	21 ± 5	17 ± 5	15 ± 5	13 ± 4	08 ± 5	11 ± 5	12 ± 5
Group 4 (n= 127)	Background noise outside booth	62 ± 4	58 ± 3	56 ± 4	54 ± 3	53 ± 5	52 ± 3	52 ± 4
	Hearing level outside booth	48 ± 10	41 ± 9	34 ± 9	35 ± 10	35 ± 10	38 ± 11	35 ± 10
	Hearing level inside booth	23 ± 5	21 ± 5	18 ± 6	21 ± 7	24 ± 9	27 ± 10	23 ± 9
	Difference in hearing level	25 ± 10	20 ± 8	16 ± 8	14 ± 7	11 ± 8	11 ± 9	12 ± 9

It was observed that at 0.5 K HZ background noise outside the booth increased from 35 dB in group 1 to 62 dB in group 4, there was a gradual increase in the difference in hearing level from 5 dB to 25 dB. Similar to this, at 1K Hz and 2K Hz with increase in background noise of 33 dB to 58 dB and 41 dB to 56 dB correspondingly, the difference in hearing level gradually increased from 5 dB to 20 dB and 5 dB to 16 dB.

The background noise levels at higher frequencies of 3K, 4K, 6K, and 8K Hz varied from 32 dB to 54 dB, 33 dB to 53 dB, 27 dB to 52 dB, and 26 dB to 52 dB, respectively, with an increase in the difference in hearing level from 2 dB to 14 dB, 2 dB to 11 dB. 2dB to 11dB and 2dB to 12dB.

A graph is plotted for background noise on the y-axis and the difference in hearing level on the x-axis for the individual frequency of four groups. [Fig. 3 (A) to (G)]. It was observed that the relation between background noise and the hearing level was linear, i.e. as the background sound increased there was an increase in hearing level.

The study showed differences in a mine worker's hearing level when audiometry was performed inside a soundproof booth versus outside a soundproof booth at frequencies ranging from 0.5K Hz to 8K Hz. Background noise had an effect on audiometry results across all frequency ranges and groups, with a greater impact on lower frequencies than on higher frequencies. Our study is in agreement with earlier studies that have been conducted to determine the influence of background noise on audiometry findings ^7–9. These studies revealed that the audiometry conducted outside the soundproof booth with moderate to substantial background noise result in an overestimation of hearing threshold as compared to the audiometry conducted inside soundproof booth. These studies also noted that the impact was more apparent at lower frequencies compared to higher frequencies, and our study confirmed this trend.

Further, our study attempted to formulate regression equations first of its kind between background noise and the difference in hearing level for each frequency. For that, correlation between background noise and differences in hearing level was studied by using a coefficient of correlation (r) and to determine the accuracy of the coefficient of regression R-squared values were obtained as shown in table 3.

Table 3: Regression equations for individual frequency

Frequency (Hz)	Equation	R²
0.5K	y = 0.7263x - 20.251	0.9975
1K	y = 0.5873x - 15.03	0.9681
2K	y = 0.5943x - 17.722	0.9218
3K	y = 0.5387x - 14.972	0.9972
4K	y = 0.3732x - 9.6703	0.8991
6K	y = 0.392x - 9.4623	0.9265
8K	y = 0.4371x - 10.452	0.9368

R² is the proportion of the variance in the dependent variable that can be predicted by the independent variable (s).The value of R squared is between 0 and 1 where values closer to 0 represent a poor fit while values closer to 1 represent a perfect fit ¹⁴. The R squared values in our regression equations were found to range from 0.8991 to 0.9975, which were close to 1, and the model may thus be considered satisfactory. The regression equations may be used to predict the value of the dependent variable i.e. the difference in hearing level for the independent variables i.e. background noise outside the booth. To use these regression equations, substitute the value of background noise for x, and the resulting y value must be subtracted from the audiometry result for each frequency. Using these equations, audiometry performed outside of a soundproof booth may be used to assess audiograms and determine hearing loss.

These equations have the benefit of reducing the number of cases of hearing impairment that would have been diagnosed more frequently in non-soundproof booths. Hence, these equations would be used for screening and those found to have hearing loss would require audiometry in a suitable soundproof booth to confirm diagnosis.

In developing countries such as India, where soundproof booths are not available in for screening purposes, in primary health/community centres, rural areas, these equations can be used to reduce the effect of background noise and produce results comparable to audiometry performed in a soundproof booth.

Strengths and limitations

Our study is one of the first studies, proposing regression equations to adjust hearing levels for audiometry conducted in a non-soundproof booth. Four groups were studied in order to increase variability, however not all four groups' studies took place in the same location.

Despite the fact that the study was conducted in two different locations with two different soundproof booths, both of them met prescribed international standard. Further, the sample size was different in each of the four groups, but the same samples were used for inside and outside the audiometry booth, so the difference in sample size may be insignificant.

Authors also accepted that factors such as learning by study subjects as a result of repeated audiometry and technician performance as a result of continuous tasks might have confounded the audiometry results.

The findings of the study showed that the hearing threshold significantly increased outside the soundproof audiometry booth among mine workers. The study devised regression equations for individual frequencies from 0.5 K Hz to 8 K HZ. These equations may be useful as a screening tool to determine the difference in hearing level when audiometry is conducted outside the soundproof booth. However, validation of these equations in large sample sizes needs to be performed.

Acknowledgements: The authors thank Mrs Shilpa Ingole, Technical officer at National Institute of Miners’ Health to assist in field studies.

Author’s contributions:

Dr. Sarang Dhatrak and Dr. Subroto Nandi contributed to the conception, design, and definition of intellectual content. Dr. Sarang Dhatrak, Dr. Nitishkumar Tank and Dr. Subroto Nandi contributed to the literature search, data analysis, manuscript preparation, manuscript editing, and manuscript review. Dr. Sarang Dhatrak drafted the first draft of the manuscript. Dr. Sarang Dhatrak, Dr. Nitishkumar Tank and Dr. Subroto Nandi reviewed and edited the manuscript for corrections. Dr. Sarang Dhatrak, Dr. Nitishkumar Tank and Dr. Subroto Nandi approve of the final version of the article. Dr. Sarang Dhatrak and Dr. Subroto Nandi will act as the guarantor of the research.

Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding: This work was funded by the Ministry of Mines, Government of India (vide letter no. F.No. 14/25/2015 –Met.IV dated 29.12.0215)

Conflict of interest: The authors declare that they have no potential of interests.

Ethics approval: The study was approved by the Institutional Human Ethics Committee of National Institute of Miners’ Health [approval number: NIMH/IEC/2017/01, date: May 16, 2017.

Consent to participate: Written informed consent was obtained from all study participants before their enrolment in the study.

Consent for publication: Not applicable

Authors Affiliation declaration: The first and third authors of the paper were employed with the National Institute of Miners' Health in Nagpur during the study's execution. Further, The National Institute of Occupational Health, Ahmedabad, amalgamated with the National Institute of Miners' Health. Authors' current designations and affiliations are mentioned in author details.

Smith, A. W. The World Health Organisation and the prevention of deafness and hearing impairment caused by noise. Noise Health 1, 6 (1998).
Nandi, S. S. & Dhatrak, S. V. Occupational noise-induced hearing loss in India. Indian J. Occup. Environ. Med. 12, 53–56 (2008).
EN ISO 8253-1:2010 - Acoustics - Audiometric test methods - Part 1: Pure-tone air and bone conduction audiometry (ISO 8253-1:2010). iTeh Standards Store https://standards.iteh.ai/catalog/standards/cen/b5d6089a-99c2-4a78-9086-be7da495e999/en-iso-8253-1-2010.
Frank, T. ANSI update: maximum permissible ambient noise levels for audiometric test rooms. Am. J. Audiol. 9, 3–8 (2000).
Behar, A. Audiometric Tests without Booths. Int. J. Environ. Res. Public. Health 18, 3073 (2021).
Frank, T. & Williams, D. L. Ambient noise levels in industrial audiometric test rooms. Am. Ind. Hyg. Assoc. J. 55, 433–437 (1994).
Bromwich, M. A., Parsa, V., Lanthier, N., Yoo, J. & Parnes, L. S. Active noise reduction audiometry: a prospective analysis of a new approach to noise management in audiometric testing. The Laryngoscope 118, 104–109 (2008).
Lo, A. H. C. & McPherson, B. Hearing screening for school children: utility of noise-cancelling headphones. BMC Ear Nose Throat Disord. 13, 6 (2013).
Wong, T. W. et al. Agreement between hearing thresholds measured in non-soundproof work environments and a soundproof booth. Occup. Environ. Med. 60, 667–671 (2003).
Mankowski, N. L. & Raggio, B. S. Otoscope Exam. in StatPearls (StatPearls Publishing, 2022).
Quaranta, A., Portalatini, P. & Henderson, D. Temporary and permanent threshold shift: an overview. Scand. Audiol. Suppl. 48, 75–86 (1998).
Hughson, W. & Westlake, H. D. Manual for program outline for rehabilitation of aural casualties both military and civilian, sponsored by the American Academy of Ophthalmology and Otolaryngology. (Douglas Print. Co., 1944).
Darbyshire, J. L. & Young, J. D. An investigation of sound levels on intensive care units with reference to the WHO guidelines. Crit. Care 17, R187 (2013).
Hamilton, D. F., Ghert, M. & Simpson, A. H. R. W. Interpreting regression models in clinical outcome studies. Bone Jt. Res. 4, 152–153 (2015).

No competing interests reported.

A cross-sectional study of the relationship between background noise and hearing threshold among Indian mine workers: formulating regression equations to reduce the false- positive hearing impairment cases

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1