Evaluating the Effectiveness of Earplugs in Preventing Noise-Induced Hearing Loss in an Auto Parts Factory in China
, Liangliang Zhao 1, Ling Li 1, Thais C. Morata 2, Wei Qiu 3, Huiling Amy Feng 2 and Baoli Zhu 4,*
Round 1
Reviewer 1 Report
This is an interesting and informative study presenting the incorrect wearing of HPD resulting in no detectable effect on the prevention of hearing loss at the auto parts factory. The manuscript is well-written; however, several criticisms need to be clarified in its revised version:
Criticisms:
1) In the Materials and Methods, why chose different noise dosimeters to measure no noise-exposed and noise-exposed participants?
2) To calculate the PAR, 2 dB C minus A corrected SNR was applied. The authors should confirm whether the value should be reduced by 2.5 or 3 dB but not 2 dB, citing references containing ANSI or CSA standard.
3) In general, a 50% NRR for earplugs was recognized to calculate the NRR in the real world. Therefore, the 3M earplug for use with dBA through de-rating should be 25 x (50%) - 3 = 9.5 dBA, not the value of 15 dBA as mentioned in your manuscript. The difference between SNR and PAR (9.5 vs. 7) exists but not as great as your calculated data (15 vs. 7).
4) In the collected data, is the personal noise exposure dosage calculated from each participant exposed to noise?
5) How to calculate the PAR? How long does it take for data acquisition?
Is there any microphone inserting into the ear canal, then followed by 3M earplug insertion?
6) Please check if the CNE formula CNE= ALAeq,8h +10 log T should be corrected as
CNE= LAeq,8h +10 log T ?
Author Response
Reply to Reviewer #1
Reviewer #1 Comment 1: In the Materials and Methods, why chose different noise dosimeters to measure no noise-exposed and noise-exposed participants?
Response Reviewer #1 Comment 1: Thanks for the comment.
For the non-noise-exposed employees, we used the Casella dBadge noise dosimeters because they were not overexposed to noise. This dosimeter can allow us to extract LAeq,8h directly.
While for the noise-exposed participant, we wanted to address the temporal characteristics of noise in this study, so we used the ASV5910-R digital recorder (same recorder used in reference 12). To address this comment, we changed the sentence in lines 131-133 into “Immediately after recordings were completed, the data were transferred from the recorder to a computer for subsequent LAeq,8h and kurtosis calculation.”
We also added to the second paragraph of 2.3 Personal noise exposure levels in lines 125 - 128 the sentences: “This instrument was selected because of our interest in the temporal characteristics of the noise exposures. The recorders comply with IEC 61672 and IEC 61252 standards, and used in previous studies [12,13].
Reviewer #1 Comment 2: To calculate the PAR, 2 dB C minus A corrected SNR was applied. The authors should confirm whether the value should be reduced by 2.5 or 3 dB but not 2 dB, citing references containing ANSI or CSA standard.
Response Reviewer #1 Comment 2: Thanks for the comment.
In the ANSI/ASA S12.68-2007, 5.3 application, it says, “The NRSG should be used instead of the NRSA for noises that have a rumbling, thunderous, or heavy sound to them, for sounds that are from air-moving equipment, for passengers in moving vehicles, for sound levels exceeding 100 dBA, or if the measured difference between C- and A-weighted sound levels exceeds 5 dB. At that LC − LA difference, the NRSA overestimates performance by 3 dB on average.” The average LAeq,8h of noise-exposed workers in this study was 87 dB(A). Unfortunately, we did not measure the Lc level. To check the LC − LA difference in this study, we extract around 300 pieces of noise data from all machinery manufacturing industries used in the classical report” Karplus HB and Bonvallet GL. A noise survey of manufacturing industries. Am Ind Hyg Assoc Q. 1953 ec;14(4):235-63. doi: 10.1080/00968205309343944. PMID: 13114194.”. The median and average LC − LA differences were 1.8 and 2.6 dB, respectively, less than 5 dB.
In the ANSI S12.68 standard, the NIOSH 100 Noises, the Air Force 50 Noises, and an additional 20 Civilian Aviation noises were used to better understand the change in the noise reduction as a function of the spectral balance, LC − LA. The selected noises were meant to provide a uniform distribution, not a normal distribution like what Gauger and Berger(2004) evaluated, which combined 6 databases from classic reports. By quoting the reference we cited in this study Berger 2008 “Based on Gauger and Berger (2004), the mean and median C – A value for industrial noises are 2.5 and 1.9 dB respectively. Thus, a 2-dB C – A correction was subtracted from the NRR to compare to PAR.”, we have confirmed that 2 dB C minus A corrected SNR was applied in this study.
Reviewer #1 Comment 3: In general, a 50% NRR for earplugs was recognized to calculate the NRR in the real world. Therefore, the 3M earplug for use with dBA through derating should be 25 x (50%) - 3 = 9.5 dBA, not the value of 15 dBA as mentioned in your manuscript. The difference between SNR and PAR (9.5 vs. 7) exists but not as great as your calculated data (15 vs. 7).
Response Reviewer #1 Comment 3: Thanks for the comment. As the reviewer pointed out, in general, a 50% NRR for earplugs is used to calculate the NRR in the United States. However, China requires a 40% derating of the SNR. So, the derating value should be 25 x 60% = 15 dB.
If we use both the derating and correction together, similar as OSHA dictates that 7 dB (C – A correction)is subtracted from the NRR prior to derating by 50% as in (NRR-7)/2, the derated value would be (25-2) x 60% = 13.8 dB. Thus, the difference between SNR and PAR would be 13.8 vs. 7.
Reviewer #1 Comment 4: In the collected data, is the personal noise exposure dosage calculated from each participant exposed to noise?
Response Reviewer #1 Comment 4: Thanks for the comment. Yes, the personal noise exposure dosage was calculated from each participant exposed to noise. To address this comment, we have changed the sentence in line 123 to “Each noise-exposed worker from the auto parts factory wore an ASV5910-R digital recorder.’
Reviewer #1 Comment 5: How to calculate the PAR? How long does it take for data acquisition? Is there any microphone inserting into the ear canal, then followed by 3M earplug insertion?
Response Reviewer #1 Comment 5: Thanks for the comment. The field attenuation estimation system used in this study was the 3MTM E-A-RfitTM Dual Ear Validation System (https://multimedia.3m.com/mws/media/1585320O/3m-value-story-of-hearing-fit-testing.pdf) and we cited a study [9] that gives further information on the test. The dual-element microphone assembly attaches to a specially probed hearing protectors to allow measurement of the sound level inside the subject’s ear canal while the hearing protector is worn. The “in-ear” sound level is simultaneously compared to the sound level outside the hearing protector, as measured by the external microphone. In this study, the workers were asked to wear their earplugs in their normal manner, then the experimenter attached the dual-element microphone array to the corresponding left and the right probed test earplugs. The subject sat facing the test speaker at a proper distance. The experimenter presented the test stimulus, and the equipment calculated the PARs for each ear as well as binaurally. All PAR values were reported as PAR minus the uncertainty value, which is effectively a PAR84 (ANSI S12.71-2018). As the test procedures are the same as reference 9 cited in this paper, we did not include it in the current manuscript.
Reviewer #1 Comment 6: Please check if the CNE formula CNE= ALAeq,8h +10 log T should be corrected as CNE= LAeq,8h +10 log T ?
Response Reviewer #1 Comment 6: Thanks for pointing out our mistake. The CNE formula should be CNE= LAeq,8h +10 log T. We have corrected it as suggested.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Excellent paper! It was a pleasure reviewing it!
Author Response
Reply to Reviewer #2
Reviewer #2 Comment 1: Excellent paper! It was a pleasure reviewing it!
Response Reviewer #2 Comment 1: Thanks for the comment and encouragement.
Author Response File: Author Response.docx
Reviewer 3 Report
I thank the editors for the opportunity to review this article titled “Evaluating the effectiveness of earplugs in preventing noise-induced hearing loss in an auto parts factory in China”. In this article, the authors report a challenging and well-designed study into the impact of earplug use on the high-frequency pure tone audiometric thresholds of workers from an auto parts factory in comparison to a control group of health department employees. The article reviews the handling of occupational noise exposure prevention in China and introduces the concepts involved in evaluating the achieved compared to predicted effectiveness of hearing protection. The authors address the need for assessing what realistic expectations of attenuation are, and concludes that in two groups of relatively-well age matched individuals, those who encountered occupational noise exposure gained little if any benefit from mandatory hearing protection, and as such had suffered increased audiometric thresholds.
Below are some comments that the authors may wish to integrate into future revisions of their work.
I do not have major concerns about the article. The analyses and results appear to be appropriate and represented reproducibly and transparently. The work is very interesting and a timely reminder of the impact of noise exposure and of not taking hearing protection seriously.
I have concerns around the use of different dosimeters for each cohort of study participants. Why was this done? Was anything done to try and establish that the data captured were equivalent?
Can the authors comment on how realistic the fit testing procedure felt in comparison to just putting in earplugs normally? Were participants encouraged to use their usual method and/or asked if the earplugs felt normal for them? Given that 37% of participants in that group received no benefit from the earplugs, does that seem consistent with the participant’s own description of what wearing earplugs feels like to them?
Line 176: why were better ear thresholds used, rather than the mean across ears or (if there is any hypothesis for unilateral noise exposure) each ear separately?
Lines 210-213: please could you reword the paragraph so that ages are shown as figures in the format mean ± standard deviation, which is easier to pull out from the text? It doesn’t look like the ages were significantly different across the groups but it would be good to confirm this (i.e. a p value).
Can the authors comment on whether there were any gender differences in earplug fit or proportion/duration of use?
Could the sub-axes on Figure 5 be rearranged so that a visual comparison can be made between male and female participants of each age group? i.e. just have two rows of graphs and show each male age group on the top row and each female age group along the bottom row to allow easier visual inspection of the data?
Author Response
Reply to Reviewer #3
Below are some comments that the authors may wish to integrate into future revisions of their work.
Reviewer #3 Comment 1: I do not have major concerns about the article. The analyses and results appear to be appropriate and represented reproducibly and transparently. The work is very interesting and a timely reminder of the impact of noise exposure and of not taking hearing protection seriously.
Response Reviewer #3 Comment 1: Thanks for the comment.
Reviewer #3 Comment 2: I have concerns around the use of different dosimeters for each cohort of study participants. Why was this done? Was anything done to try and establish that the data captured were equivalent?
Response Reviewer #3 Comment 2: Thanks for the comment. Yes, you are right. We used different dosimeters to measure no noise-exposed and noise-exposed participants. We added the sentence to 2.3.
“This instrument was selected because of our interest in the temporal characteristics of the noise exposures. The recorders comply with IEC 61672 and IEC 61252 standards, and used in previous studies [12,13].”
For the non-noise-exposed employees, we used the Casella dBadge noise dosimeters because they were not overexposed to noise. This dosimeter allows us to extract LAeq,8h directly.
While for the noise-exposed participant, we wanted to address the temporal characteristics of noise in this study, so we used the ASV5910-R digital recorder (same recorder used in reference 12). To address this comment, we changed the sentence in lines 131-133 into “Immediately after recordings were completed, the data were transferred from the recorder to a computer for subsequent LAeq,8h and kurtosis calculation.”
Because the dosimeter and digital recorder both used the same protocol which was consistent with NIOSH recommendations, and they were calibrated before and after the measurement, we believe that the data captured were equivalent.
Reviewer #3 Comment 3: Can the authors comment on how realistic the fit testing procedure felt in comparison to just putting in earplugs normally? Were participants encouraged to use their usual method and/or asked if the earplugs felt normal for them? Given that 37% of participants in that group received no benefit from the earplugs, does that seem consistent with the participant’s own description of what wearing earplugs feels like to them?
Response Reviewer #3 Comment 3: Thanks for the comment.
To get the real attenuation from the earplugs as they were worn in the normal manner, we fit tested the workers when they inserted the surrogate probed test earplugs into their ears the way they normally wore them during their work shift. The researcher observed the worker’s fitting technique without coaching or intervening. The researcher then fit test them and the PARs were measured which were used in the study data analysis.
If a worker achieved the attenuation less than 5 dB, or was observed to fit their earplugs improperly, the worker was instructed on how to wear earplug correctly, and if time allowed and the worker was interested, he/she could wear the earplugs again and was retest to make sure he/she achieve the efficient attenuation. The worker was encouraged to wear the earplugs correctly when he/she was exposed to hazardous noise. However, to avoid confusing the readers, this attenuation value was not reported in this study.
In this study, we did not collect the data about what wearing earplugs feel like to the workers. So, it is impossible to compare the PARs with the worker’s description.
We have revised the sentences in lines 171-172 as “workers’ PARs were measured by asking the workers to fit the surrogate probed test earplugs by themselves the way they normally wore them during their work shift. No coaching or demonstration was provided.”
Reviewer #3 Comment 4: Line 176: why were better ear thresholds used, rather than the mean across ears or (if there is any hypothesis for unilateral noise exposure) each ear separately?
Response Reviewer #3 Comment 4: Thanks for the comment. The better ear was used because this was the criteria for establishing median hearing threshold levels of the control group (Hoffman HJ, Dobie RA, Ko CW, Themann CL, Murphy WJ. Americans hear as well or better today compared with 40 years ago: hearing threshold levels in the unscreened adult population of the United States, 1959-1962 and 1999-2004. Ear Hear. 2010 Dec;31(6):725-34. doi: 10.1097/AUD.0b013e3181e9770e. PMID: 20683190.
We have added this sentence in line 186 as “The better ear was used because this was the criteria for establishing median hearing threshold levels of the control group [16].”
Reviewer #3 Comment 5: Lines 210-213: please could you reword the paragraph so that ages are shown as figures in the format mean ± standard deviation, which is easier to pull out from the text? It doesn’t look like the ages were significantly different across the groups, but it would be good to confirm this (i.e. a p value).
Response Reviewer #3 Comment 5: Thanks for the comment. As you suggested, we reworded the sentences in lines 219-229 as “A total of 385 noise-exposed workers and 1268 non-noise-exposed employees participated in this study. Among noise-exposed workers, 70% were male and 30% were female; their median duration of noise exposure was eight years (25th - 75th percentile: 4 - 11 years); their median age was 37 years with a standard deviation (SD) of seven years. The male workers (35 ± 7.5 years) were younger than the female workers (40 ± 5.5 years) (F = 40.04, p < 0.001). For non-noise-exposed employees, 36% were male and 64% were female; their median tenure was 11 years (25th - 75th percentile: 5 - 23 years); their mean age was 38 years with a SD of 11 years. The male employees (41 ± 10.6 years) were older than the female employees (37 ± 10.2 years) (F = 38.23, p < 0.001)”. Because the ages were significantly different between gender, the comparison of the hearing threshold between noise-exposed workers, non-noise-exposed employees, and ISO-estimated hearing threshold was conducted exclusively within gender.
Reviewer #3 Comment 6: Can the authors comment on whether there were any gender differences in earplug fit or proportion/duration of use?
Response Reviewer #3 Comment 6: Thanks for the comment. As you suggested, we have added “Male workers achieved higher attenuation than female workers (F = 4.342, p = 0.038)” in lines 271-272 in Section 3.3. Personal attenuation rating section.
Reviewer #3 Comment 7: Could the sub-axes on Figure 5 be rearranged so that a visual comparison can be made between male and female participants of each age group? i.e. just have two rows of graphs and show each male age group on the top row and each female age group along the bottom row to allow easier visual inspection of the data?
Response Reviewer #3 Comment 7: Thanks for the comment. The original Figure 4 and Figure 5 present the ISO estimated median HT of the noise-exposed workers and the average HT of the non-noise -exposed workers, at each frequency, for male workers and female workers. For easier visual inspection of the data between males and females, we have added Figure 4 in the “3.4 Hearing threshold analysis“ section.
Figure 4. Visualized comparison of the average HTs between noise-exposed male workers (left panel) and female workers (right panel) by frequency and age group.
Author Response File: Author Response.docx
Reviewer 4 Report
attached
Comments for author File: 
Comments.pdf
Author Response
Reply to Reviewer #4
The title is descriptive. Methods are inadequate in that
Reviewer #4 Comment 1: Although non-noise factors affecting hearing may have been assessed in the employee survey, the authors do not describe how they controlled for these confounders.
Response Reviewer #4 Comment 1: Thanks for the comment.
Usually, non-noise factors that can affect hearing in the manufacturing industry consider ototoxicants, such as solvents, asphyxiants, nitriles, metal and compounds (https://www.cdc.gov/niosh/docs/2018-124/pdfs/2018-124.pdf). In addition to noise exposure, some workers used metalworking fluid (MWF) to cool, lubricate, and remove metal chips from tools and parts during machining of metal stock. MWFs are complex mixtures and often contain other substances, including biocides, corrosion inhibitors, metal fines, tramp oils, and biological contaminants. Inhalation of MWF aerosols may irritate the throat, nose, and lung and has been associated with chronic bronchitis, asthma, hypersensitivity pneumonitis, and worsening of pre-existing respiratory problems (https://www.cdc.gov/niosh/hhe/reports/pdfs/2015-0070-3304.pdf). For now, there is no evidence that MWF could affect hearing. To clarify the non-noise confounder issue, we have added one sentence in line 92 “Workers were not exposed to ototoxicants at this factory”.
Another non-noise factor related to hearing loss is age. To control age as a confounder, we calculated the prevalence of hearing loss at each group, as well as compared the hearing between noise-exposed workers, non-noise-exposed employees, and ISO-estimated hearing threshold exclusively within gender by age group. We also controlled age in the person correlation analysis and the linear model. To address the age as the confounder, we have added the following phase in the data analysis section as follows “the prevalence of high-frequency hearing loss of each age group were calculated using Microsoft Excel” in line 192, and “The comparison of the hearing threshold between noise-exposed workers, non-noise-exposed employees, and ISO-estimated hearing threshold was conducted exclusively within gender by age group” in lines 205 to 207
Smoking and alcohol consumption are considering related to hearing loss in some literatures. We pre-checked smoke and alcohol by using a general linear model. The results showed that none of them were associated with hearing loss in this study. To address smoking and alcohol as the potential confounders, we have added the phase in the data analysis section as follows “Second, after confirmed that smoking (F = 3.40, p = 0.066) and alcohol consumption (F = 0.28, p = 0.971) were not confounders in this study, a linear model “log (HFHT) = a + b(CNE) + c(CNE-ACNE) + age + gender” was conducted to check the association between HFHT with CNE, earplug protection, age, and gender. ”in lines 212 to 213.
In summary, we pre-checked that ototoxicants, smoke, and alcohol were not the confounders in this study, and we controlled age as a confounder through the data analysis. Thus, we believe that no limitation of non-noise factors to draw our conclusion. We have added this sentence to the second paragraph of the Discussion in lines 358 to 361: “Many occupational and non-occupational factors can have an impact on hearing thresholds. In this investigation we controlled for other work-related factors (ototoxicants, tenure) and examined the contribution of other factors such as age, gender, smoking and alcohol consumption.”
Reviewer #4 Comment 2: The study involved workers at a single noisy work setting.
Response Reviewer #4 Comment 2: Thanks for the comment. Although the involved workers are from the same factory, they were exposed to 8 different types of work (cutting, machining, heat treatment, forklift, inspection, etc.). We have added the following sentence in the section of “2.1. Selection of participants and background information about the surveyed factory” (in lines 89-91) to address the eight job titles: “Each building had multiple production lines and eight job titles (e.g., cutting, grinding, numerical control machine, heat treatment, sorting, packing, forker driver, and inspection)”.
We analyzed the temporal characteristics of each worker's noise and found that the noise they were exposed to ranged from continuous noise to complex noise with a kurtosis value between 3 and 75. Therefore, the actual noise these workers exposed to were relatively complex and varied. However, as the reviewer pointed out, the noise settings from a single factory are limited. We plan to conduct this study in more industrial noise environments in the future. We have added the following sentences in the section of “Limitation of the study” (in lines 479-482) to address the limitation of all the data collected from a single factory: “Although the workers were exposed to eight different types of work noise, the generality of the findings in this study may be limited by the fact that the data came from a single factory. Similar studies in different industries will be helpful to further evaluate the actual pro-tection capability of HPDs in the real world.”
Reviewer #4 Comment 3: Authors justifyiably conclude that the hearing conservation program at the production facility is not effective, but this is not clearly stated in the abstract;
Response Reviewer #4 Comment 3: Thanks for the comment. We have added the following sentence in the abstract and conclusion section, as you and another reviewer suggested:“We have added the sentence, “The hearing conservation program at the surveyed factory was not effective.” in line 179.
Reviewer #4 Comment 4: The study would make a good quality improvement study, should the authors choose to focus on hearing conservation program factors that are amenable to improvement.
Response Reviewer #4 Comment 4: Thanks for the constructive comment. We have added the following sentences in the abstract and conclusion section, as you and another reviewer suggested: “The hearing conservation program at the surveyed factory was not effective. Periodic hearing tests , earplug fit testing, expanding the offer of different types of hearing protection, and employee education about the importance of protecting their hearing were recommended to the surveyed factory’s occupational health and safety program.”
Author Response File: Author Response.docx
Reviewer 5 Report
With noise-induced hearing loss a growing problem it is essential that ways to reduce hearing loss be widely disseminated. To learn that wearing hearing protection in a working environment did not protect workers is shocking because one would have assumed the opposite. However, just providing workers with ear plugs is not sufficient in protecting workers; training on use of ear plugs must be part of the program as noted by the authors. Thus, this study provided most useful information.
I would also suggest to the authors that periodic checks on the hearing of employees by employers be part of a hearing conservation program as well as lectures on the importance of protecting one's hearing.
Author Response
Reply to Reviewer #5
Reviewer #5 Comment 1: With noise-induced hearing loss a growing problem it is essential that ways to reduce hearing loss be widely disseminated. To learn that wearing hearing protection in a working environment did not protect workers is shocking because one would have assumed the opposite. However, just providing workers with ear plugs is not sufficient in protecting workers; training on use of ear plugs must be part of the program as noted by the authors. Thus, this study provided most useful information.
Response Reviewer #5 Comment 1: Thanks for the comment and encouragement.
Reviewer #5 Comment 2: I would also suggest to the authors that periodic checks on the hearing of employees by employers be part of a hearing conservation program as well as lectures on the importance of protecting one's hearing.
Response Reviewer #5 Comment 2: Thank you for your constructive suggestion.It is important to check the employee's hearing periodically and to educate the employees on the importance of protecting their hearing. Actually, the employer in this study did conduct an annual hearing screen for their employees and had annual training on hearing protection. After the survey, we recommended the employees conduct a hearing threshold test instead of a hearing screen and conduct yearly earplugs fit testing. We have added the following sentence in the conclusion section as suggested: “Periodic hearing tests, earplug fit testing, expanding the offer of different types of hearing protection, and employee education about the importance of protecting their hearing were recommended to the surveyed factory’s occupational health and safety program.”
Author Response File: Author Response.docx
Round 2
Reviewer 4 Report
attached
Comments for author File: Comments.pdf
Author Response
Reply to Reviewer #4
Evaluating the effectiveness of earplugs in preventing noise-induced hearing loss in an auto parts factory in China (revision #1)
International Journal of Environmental Research and Public Health
Reviewer #4 Comment 1:This is a review of a revision of the first manuscript submission with this title. The study addresses a topic of potential interest to readers: noise-induced hearing loss. The stated aims of the study were “to evaluate the attenuation offered by earplugs and the effectiveness of earplug use in the prevention of noise-induced hearing loss (NIHL).“ There is discrepancy in aims. In line 62, the authors suggest that one aim of the study was to determine what factors can contribute to the success of the hearing conservation program. However, (line 242) the study does not address this goal well.
Response Reviewer #4 Comment 1: Thanks for the comment. We deleted the last part of the sentence (now in lines 64-65) that caused the confusion. This study at an auto parts factory in China was designed to investigate the attenuation offered by earplugs and their effectiveness in preventing NIHL. (line 65 - 67).
Reviewer #4 Comment 2:The manuscript reports results of a case-control study. The report is well organized, and the writing is clear. The study has low potential impact, in that only a single noise-exposed setting is sampled. The topic of the report is appropriate to the Journal in that it focuses on an issue highly relevant to environmental and public health. Although the authors identify the study design as cross-sectional (L 72), they go on to describe it as a case-control.
Response Reviewer #4 Comment 2: Thanks for the comment. We have changed into “case-control study” in lines 74 and 473, as suggested.
Reviewer #4 Comment 3:The manuscript is improved in that it acknowledges the limitations of being conducted with workers at a single noisy work setting, and the abstract has been revised to state that the hearing conservation program at the production facility is not effective. Although confounders of smoking and alcohol consumption were controlled (Line 207), authors do not describe how confounders were controlled.
Response Reviewer #4 Comment 3: Thanks for the comment. We put smoking and alcohol consumption in a generalized linear model “log (HFHT) = a + b(CNE) + c(CNE-ACNE) + age + gender + smoking + alcohol consumption”. The results confirmed that smoking (F = 3.40, p = 0.066) and alcohol consumption (F = 0.28, p = 0.971) were not confounders in this study. Then a final liner model as described in line 214-216 was conducted and reported. We have reworded the sentence in lines 212-214 as following “after confirmed that smoking (F = 3.40, p = 0.066) and alcohol consumption (F = 0.28, p = 0.971) were not confounders using a generalized linear model, a final linear model “log (HFHT) = a + b(CNE) + c(CNE-ACNE) + age + gender” was conducted ……”
Reviewer #4 Comment 4:Some obvious recommendations are missing from the discussion, e.g., supervision of workers and modification of workplace policy.
Response Reviewer #4 Comment 4: Thanks for the comment. We have added the following sentence to the discussion section (lines 440-442) as suggested: “Recommendations included supervision of workers on consistent use of earplugs and modification of workplace policy to the occupational health and safety program.”
Reviewer #4 Comment 5:This reviewer maintains her position that the study would make a good quality improvement study, should the authors choose to focus on hearing conservation program factors that are amenable to improvement. However, the editor may choose to publish in its present form.
Response Reviewer #4 Comment 5: Thanks for the comment. This study aimed to investigate the attenuation offered by earplugs and their effectiveness in preventing NIHL. A systematic review of the effectiveness of hearing conservation program factors will be considered in future studies.
Author Response File: Author Response.docx
