Occupational noise-induced hearing loss (ONIHL) is a complex, but preventable, health problem for South African miners. Meticulously collected data should be made use of to design interventions to address this health issue.
A single mine’s electronic data were reviewed in a secondary data review to determine, from the records, factors that hearing conservation practitioners deemed useful for identifying ‘at risk’ miners and to establish factors that would pave the way for the integration of the 2014 hearing conservation programme (HCP) milestones into the mine’s current proactive data management system (PDMS). The objectives of this article were to establish how miners with published risk factors associated with ONIHL were managed by the mine’s hearing conservation practitioners as part of the HCP; to determine if the mine’s hearing conservation practitioners could estimate miners’ risk of ONIHL using baseline percentage loss of hearing (PLH) as a hearing conservation measure; and to estimate the contribution of noise exposure to ONIHL risk.
In a secondary data review design, records in a platinum mine’s two electronic data sets were reviewed: the first contained diagnostic audiometry records (
The mean age of the miners (all male candidates) was 47 ± 8.5 years; more than 80% had worked for longer than 10 years. Valid baseline audiometry records were available for only 34% (
These findings indicate significant gaps in the mine’s PDMS, requiring attention. Nonetheless, the mine’s current data capturing may be used to identify miners at risk of developing ONIHL. The PLH referral cut-off point (≥2.5%) used by the mine’s hearing conservation practitioners, when used in conjunction with baseline PLH shifts, was the major factor in early identification of ONIHL in miners exposed to ≥85 dBA noise. An inclusive integrative data management programme that includes the medical surveillance data set of the miners’ noise exposure levels, occupations, ages and medical treatments for tuberculosis and human immunodeficiency syndrome is recommended, as these are important risk indicators for developing ONIHL, particularly within the South African context.
The World Health Organization (WHO) reported a 16% prevalence of occupational noise-induced hearing loss (ONIHL) in 2010 (WHO,
In the diagnosis and management of ONIHL, it is important to include tracking and monitoring all the aforementioned risk factors to ensure efficient implementation of any hearing conservation programme (HCP). Occupational diseases in the mining industry, including pulmonary tuberculosis and silicosis, for example, have been well documented (Hermanus,
The aforementioned comorbid health conditions have not been considered in HCPs (Nelson, Nelson, Concha-Barrientos, & Fingerhut,
A number of studies on HCPs in South African mines have reported factors associated with the failure of HCPs to prevent ONIHL. These include excessive noise levels emitted by equipment (higher than the legislated level of 85 dBA), unsatisfactory uptake on the use of hearing protection devices (HPDs) by miners (Hansia & Dickinson,
The ADBA procedures were developed by the American National Standards Institute Working Group (ANSI S12/WGS12) (ANSI,
Functional structure of a risk management system for occupational noise.
This technique requires the use of a data management system to ‘identify occupations, workplaces and activities where hearing loss is progressing most, particularly in the case of a large workforce…..’ (Franz & Phillips,
We analysed data from the mine’s medical surveillance and audiometry records to describe the mine’s ONIHL prevention efforts (
A proactive data management system (PDMS), as a single point of contact for organisations, has been used for many years to understand complex organisational trends. The PDMS follows best practice principles, allows for efficient monitoring of key factors identified by the company and provides intervention measures where required (Jantti & Cater-Steel,
Previously reported high incidences of ONIHL in the South African mining industry indicate the miners’ hearing loss as a function of percentage loss of hearing (PLH) at 10% or greater since 2001 when the
This led to the review of HCP milestones by the Minerals Council South Africa, focussing on the mining equipment noise emissions and the tracking of the miners’ hearing thresholds (high-frequency standard thresholds) (Inspectorate, 2017). The new milestones set out in 2014 by the mining industry stakeholders stated that no individual miner’s standard hearing threshold shift (STS) should exceed 25 dB from baseline, in both ears, and that the total operational or process noise emitted by any equipment must not exceed 107 dBA, by the year 2024 (Mine Health and Safety Council – MHSC, 2015). The use of the STS as a sensitive measure for early detection of hearing loss was deemed a proactive practice of managing miners at risk for ONIHL. This initiative prompted the mines to integrate the miners’ newly recorded standard thresholds and threshold shifts (2014–2016) into their current reporting systems in order to improve their efficiency in identifying miners ‘at risk’ of developing ONIHL and to be in line with WHO targets for ONIHL (WHO,
The objectives of this article were to establish how miners with published risk factors associated with ONIHL were managed by the mine’s hearing conservation practitioners as part of the HCP; to determine if the mine’s hearing conservation practitioners could estimate miners’ risk of ONIHL using baseline PLH as a hearing conservation measure; and to estimate the contribution of noise exposure to ONIHL risk.
This was a secondary data review of miners’ electronic records from a platinum mine in Limpopo province, South Africa. Secondary data are called as such for various reasons, including the fact that the data were obtained by somebody else; the data had already undergone one layer of analysis prior to the secondary analysis; and that the data were collected for a focus or objective different to the one these are currently used for (Sorensen, Sabroe, & Olsen,
We reviewed all miners’ electronic medical surveillance and audiometry records at the research site. This comprised records including miners’ age, sex, years of experience, occupations, noise exposure levels, medical surveillance and audiometric test results.
The records reviewed in this secondary data review were stored in two databases, and comprised data from 2014 to 2017. The main database contained 1963 medical surveillance and diagnostic audiometry records of miners, all of which had a baseline PLH shift of ≥2.5% (value set by the mine to refer miners for diagnostic audiometry), and had been referred to an occupational medical practitioner for further intervention, and to an audiologist for diagnostic audiometry. The second data set was a subset of records of 73 miners who had calculated PLH shifts of ≥10% from baseline. Following the diagnostic audiometry assessment, a diagnosis of ONIHL was confirmed by the audiologist, and some miners had been presented to the Rand Mutual Assurance company for ONIHL compensation. These two sets of data were handled separately.
All data underwent data cleaning processes that included the removal of duplicate records (
It should be noted that because this is a secondary data review, PLH values were what was recorded and therefore reviewed, but it is acknowledged that the South African mines had started moving towards the new HCP milestones (2014) of using STS.
Extracted data were captured into Microsoft Excel, after which they were imported into and analysed using Stata (version 15.1). The variables, such as age, baseline PLH, diagnostic audiometry PLH, years of noise exposure, noise exposure levels and recorded risk factors, were analysed using descriptive statistical analysis, logistic regression analysis, and a two-way sample proportions test (z-test).
To establish how the mine’s hearing conservation practitioners managed miners presenting with risk factors associated with ONIHL, risk factors stated in the functional risk management structure (
To determine if the mine’s hearing conservation practitioners could estimate miners’ risk of ONIHL at baseline, PLH margins from 0% to 40% (5% increments) were used to estimate and interpret adjusted predictions, using logistic regression analysis. Finally, the contribution of noise exposure to the risk for ONIHL was estimated using a two-way sample proportion test (
Ethical clearance to conduct the study was obtained from the University of the Witwatersrand’s Human Ethics Committee (M180273) on 11 April 2018, and permission to access the records was obtained from the management team of the mining company. The study adhered to the Declaration of Helsinki 1975, as revised in 2008, as far as ethical considerations were concerned.
The workforce comprised only male mine workers.
Demographic characteristics of the miners in the main subset (
Characteristic | % | |
---|---|---|
Age (years) |
||
20–30 | 40 | 2.1 |
31–40 | 273 | 14.1 |
41–50 | 594 | 30.6 |
51–60 | 1023 | 52.8 |
≥ 61 | 8 | 0.4 |
Working experience (years) |
||
1–5 | 162 | 0.8 |
6–10 | 1010 | 52.3 |
11–15 | 636 | 32.9 |
≥ 16 | 22 | 0.1 |
,
,
Demographic characteristics of the miners (
Characteristic | % | |
---|---|---|
Age (years) |
||
20–30 | 0 | 0 |
31–40 | 11 | 15.1 |
41–50 | 38 | 52.1 |
51–60 | 24 | 33 |
≥ 61 | 0 | 0 |
Working experience (years) |
||
1–5 | 5 | 6.8 |
6–10 | 29 | 39.7 |
11–15 | 32 | 43.8 |
≥ 16 | 7 | 9.6 |
,
,
The data seem to indicate that hearing conservation factors included in the mine’s PDMS to manage noise exposure and miners at risk for ONIHL were PLH scores for screening and diagnostic audiometry, with categorised referral points for the miners, as ≤2.5%, 5%, 7% and ≥10%, to monitor those miners at risk for ONIHL and for compensation purposes.
Risk rankings for noise exposure levels were not available, from lowest to highest (1–4), for all the miners. Furthermore, only 34% of the miners (
The miners’ estimated risk for ONIHL was based on the baseline PLH shift of 2.5% used by the mine. Baseline PLH is core to tracking miner’s hearing deterioration and is required for the calculation of the PLH shift for compensation; to determine a PLH shift of >10%.
Baseline percentage loss of hearing adjusted predictions with 95% confidence intervals for occupational noise-induced hearing loss.
The miners’ noise exposure levels were recorded as continuous data and were aligned with their occupations.
Noise exposure risk rating categories of exposed miners (
Noise exposure level (dBA) | Noise exposure risk rating category | % | |
---|---|---|---|
≤ 82 | 0 (low) | 81 | 4.2 |
83–85 | 1 (medium) | 279 | 14.4 |
86–90 | 2 (significant) | 328 | 16.9 |
91–95 | 3 | 941 | 48.6 |
96–105 | 4 | 309 | 15.9 |
> 105 | 5 (high) | 0 | 0 |
Fifty-nine (80.8%) of the 73 miners diagnosed with ONIHL were exposed to noise levels of 85 dBA – 104 dBA; 14 (19.2%) were exposed to noise levels of <85 dBA. Although both groups (those exposed to <85 dBA and those exposed to ≥85 dBA) were exposed to high noise levels, the results indicated that the risk for ONIHL increased with the noise exposure category. The
Our study was guided by Franz and Phillips’ (
With regard to establishing how the mine’s hearing conservation practitioners proactively managed miners presenting with risk factors associated with ONIHL findings showed that miners older than 40 years and those who had worked in mining for longer than 6 years were at higher risk of developing ONIHL. The fact that they were referred for diagnostic audiometry with a screening audiometry PLH score of ≥2.5% meant they were suspected of being at risk of developing ONIHL. In other studies, conducted in South African mines (Khoza-Shangase,
The ADBA technique used by the mines requires more sensitive measures to identify miners at risk of ONIHL. The South African mines consider annual and diagnostic audiometry procedures as sensitive measures for identification of hearing loss; hence, both audiometry procedures were conducted on miners working in areas identified as high risk for ONIHL. Although Franz and Phillips (
In a review of gold miners’ hearing results, researchers argued for the importance of adding an objective diagnostic tool – the distortion-product otoacoustic emissions (DPOAEs) tool – for early detection of ONIHL (Moepeng et al.,
The contextual implications for TB and HIV as contributory factors towards miners’ hearing loss cannot be ignored. Our review of the records showed that these data were not included in the miners’ hearing conservation data, in spite of several studies in the South African mining industry indicating that TB and HIV are associated with hearing loss (Hermanus,
In a recent record review of hearing function among gold miners, Khoza-Shangase (
Our findings indicated that baseline audiometry referral scores used by the mines’ hearing conservation practitioners delay the early identification process and intervention of miners at risk of ONIHL. Even at 0% baseline PLH, the miners were at a 20% risk of developing ONIHL. In addition, the PLH scores provide low-frequency thresholds at baseline for the miners. The 2.5% PLH referral used by the mine was seen as proactive when compared to the legislated referral of 3.2% (SANS,
There is a dearth of literature on the use of baseline PLH as part of HCPs in South African industries where noise is a problem. The only other study that reviewed baseline PLH scores was conducted by Bronkorst and Schutte (
The miners who were exposed to excessive noise levels (≥85 dBA) had a greater chance of being diagnosed with ONIHL. Of the 73 miners diagnosed with ONIHL, 14 (19.1%) were exposed to noise levels <85 dBA. This implies that the OEL of ≥85 dBA used by the mines did not protect miners from developing ONIHL, or that there could be other risk factors that led to the miners’ ONIHL diagnosis. Our findings and those of Edwards et al. (2011) are in agreement as far as noise exposure levels in South African mines are concerned. However, we questioned the association of other risk factors with ONIHL, specific to miners exposed to <85 dBA noise levels. Thus, a recommendation is made for the hearing conservation practitioners in the mining industry to review the legislated OEL (≥85 dBA), and to consider comorbid factors associated with occupational hearing loss, in order to prevent ONIHL.
The fact that the participating mine did not record noise exposure risk rankings is problematic, and it limits efforts for the prevention of ONIHL. It is important for the mines to provide accurate noise exposure rankings in the miners’ audiometry records, specific to tasks and occupations, in order to correctly identify miners at risk of ONIHL. In addition, the noise measurement technologies and recording techniques used by the mines’ occupational health practitioners should be effective towards the prevention of ONIHL. Although the duration of exposure was not recorded in the current study, we acknowledge the importance of this aspect in ONIHL, over and above other personal factors such as genetic predisposition.
Although significant, the findings should be interpreted carefully, taking cognisance of methodological limitations. The data that we reviewed were exclusively drawn from one large-scale mine, from the miners’ diagnostic audiometry records, and revealed three critical limitations. Firstly, the shortcoming associated with the use of secondary data is that some interpretations of our findings are based on assumptions. Secondly, the occupational hygienist’s data were not included in the records that we reviewed; thus, additional exposures and duration thereof were not included in the data that we reviewed. Our results therefore reflect only a narrow view of the diagnostic audiometry data set. Thirdly, the miners’ medical surveillance records regarding TB and HIV status could not be accessed. This compromised our review, as we could not comment on the broader spectrum of the hearing health risks associated with the miners’ ONIHL. This exclusion of data from the medical surveillance database indicates that the mine did not necessarily view TB, HIV and their medications as risk factors for hearing loss and therefore did not consider them in their HCP strategies. Finally, the non-classification of noise exposure levels according to risk rankings (Franz & Phillips,
Our findings, guided by the ADBA technique, which was recommended by Franz and Phillips in 2000, indicated that the mine was using an adaptation of a technique that was recommended more than 15 years ago. This resulted in the mine’s hearing conservation practitioners considering noise exposure as the sole occupational hazard for ONIHL. Other occupational exposures associated with ONIHL were excluded. Nonetheless, the effort towards keeping records for the miners at risk of developing ONIHL was commendable.
A lower baseline referral PLH score should be used to identify miners at risk of ONIHL early, also allowing for early intervention to be instituted. The mining company is encouraged to add the miner’s data on burden of diseases such as TB and HIV as part of the miners’ audiometry monitoring programmes. This would facilitate proactive management of miners at risk of developing any type of hearing loss, including ONIHL. The current data management system used by the mine has the potential to be integrated into, and used together with, the new 2014 HCP milestones, to monitor the miners’ hearing function across all frequencies. It is probable that the mines are not achieving the desired outcomes of tracking miners’ hearing loss accurately, because they are using the PLH levels to report miners at risk for ONIHL, rather than STS. We recommend that the mines implement the new milestones (use of STS) across all stages of HCP reporting, to ensure that all new baselines are based on STS levels. This will ensure early identification of ONIHL instead of focussing on PLH, which is concerned with compensation after a disability has occurred.
The authors are grateful to Dr Innocent Maposa for assistance with biostatistical analysis in this study.
The authors declare that they have no financial or personal relationships which may have inappropriately influenced them in writing this article.
L.N. conceptualised the idea for the research, as well as the design and methodology adopted, with assistance from G.N. and K.K.-S. L.N. collected the data for the study and analysed these with supervision from G.N. and K.K.-S. L.N. was the lead author of the manuscript, which was critically reviewed by G.N. and K.K.-S. All three authors read and approved the final manuscript.
The authors wish to thank the National Institute for the Humanities and Social Sciences (NIHSS) and the Consortium for Advanced Research Training in Africa (CARTA) for financial assistance towards the publication of this manuscript.
Data sharing is not applicable to this article as no new data were created or analysed in this study.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.