Prevalence, risk factors and knowledge of noise-induced hearing loss among metal fabricators in Mbarara city, Uganda: a cross-sectional study Authors: Brian Agaba, Doreen Nakku, Esther Nakasagga, Daniel Atwine Published: 2025 ABSTRACT AbstractBackground and objectivesNoise-induced hearing loss (NIHL) is a common permanent health condition and yet preventable. Individuals who work in noisy environments such as metal fabricators in the informal sector are at an especially high risk of developing NIHL. We set out to determine the prevalence, risk factors and knowledge about NIHL among metal fabricators in Mbarara City (Uganda).MethodsWe conducted a descriptive cross-sectional study in Mbarara city among metal fabricators for a period of 6 months. 178 participants were recruited using a consecutive sampling method, data was collected using a data collection tool which captured their responses and examination findings, workshop noise levels were measured using a hand-held sound level meter and the hearing threshold by a pure tone audiometer. Then analysed the data using STATA V.15.0.ResultsThe majority were young (18–24 years) males (96.1%). The prevalence of NIHL was 19.7% and we noted this increased with advancing age and work experience in the metal fabrication industry. Only 23.2% of metal fabricators had adequate knowledge levels about NIHL. After adjusting for confounding, we found that only ‘history of employment in a noisy environment’ truly increased the odds (OR 3.3 95% CI 1.1 to 9.0 and a p value of 0.028) of developing NIHL in this population. Advancing age, having a ‘relative with hearing loss’, increasing work experience in metal fabrication (>1 year), a positive history of smoking and recent use of ototoxic drugs also demonstrated increased odds of association with NIHL, that is 3.4 (0.7–16.8, 0.132), 2.0 (0.8–5.1, 0.137), 9.9 (1.0–16.8, 0.055), 3.4 (0.5–24.2, 0.061) and 1.7 (0.6–4.9, 0.310) respectively, with their corresponding 95% CIs and p values, though these were not statistically significant.ConclusionThe prevalence of NIHL among metal fabricators in Mbarara city is low. Knowledge about NIHL was inadequate among most fabricators. NIHL prevalence increased with work experience, more so among those with a history of work in a noisy environment. Other contributory factors were advancing age, a family history of hearing loss, smoking and use of ototoxic drugs.Keywords: Community Health, Environmental Medicine, Noise, Prevalence WHAT IS ALREADY KNOWN ON THIS TOPIC Globally, 5% of people live with disabling hearing loss, disproportionately impacting developing countries due to inadequate advocacy and preventive measures. In Uganda, 12% of adults experience hearing loss, with occupational noise exposure common in industries like metal fabrication being a major preventable cause. Globally, 5% of people live with disabling hearing loss, disproportionately impacting developing countries due to inadequate advocacy and preventive measures. In Uganda, 12% of adults experience hearing loss, with occupational noise exposure common in industries like metal fabrication being a major preventable cause. WHAT THIS STUDY ADDS Among 178 metal fabricators in Mbarara City, Uganda, this study identifies a 19.7% prevalence of noise-induced hearing loss (NIHL, strongly linked to age, prolonged work duration (>5 years) and poor knowledge (only 23.2% demonstrated awareness of NIHL prevention). It highlights the critical gap in workplace safety practices and education in informal sectors. Among 178 metal fabricators in Mbarara City, Uganda, this study identifies a 19.7% prevalence of noise-induced hearing loss (NIHL, strongly linked to age, prolonged work duration (>5 years) and poor knowledge (only 23.2% demonstrated awareness of NIHL prevention). It highlights the critical gap in workplace safety practices and education in informal sectors. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY These findings call for urgent health education on NIHL, enforcement of hearing protection policies and regular audiometric screenings in Uganda’s informal industries. Future research should evaluate causal pathways between noise exposure and NIHL and test interventions like subsidised protective gear to curb projected increases in hearing loss among at-risk workers. These findings call for urgent health education on NIHL, enforcement of hearing protection policies and regular audiometric screenings in Uganda’s informal industries. Future research should evaluate causal pathways between noise exposure and NIHL and test interventions like subsidised protective gear to curb projected increases in hearing loss among at-risk workers. Background According to the WHO, disabling hearing loss (HL) affected 360 million people worldwide, approximately 5% of the global population in 2011.1 In Uganda alone, a 2008 study found the prevalence of disabling HL in the general adult population at 12%.2 If not addressed, HL contributes to reduced social, psychological and cognitive function and consequently to conditions such as depression and somatisation.3 WHO estimates that one-third of all global cases of HL can be attributed to noise exposure, of which the majority is occupation-related.4 Occupational noise is a more frequent cause of HL as compared with recreational noise because of its association with individuals’ livelihoods and survival.5 According to a study conducted in the USA, about a quarter of adults were reported to have noise-induced hearing loss (NIHL) and a third reported noise at their workplaces.3 NIHL is permanent sensorineural HL that results from repeated exposure to moderate or high-intensity noise levels.6 Sensorineural HL stems from issues within the inner ear or auditory nerves, which can be congenital or acquired due to various causes, such as ageing, ototoxic drugs, infections and trauma, including noise exposure.1 Populations at considerable risk of developing NIHL include those working in noisy workplaces such as factories, mines, military and metal works.7 On a historical note, NIHL was first recognised among ‘boiler makers’ during the Industrial Revolution.8 NIHL typically begins with a temporary threshold shift (TTS), a reversible change in hearing threshold after acute noise exposure, lasting up to 48 hours. If noise exposure continues, TTS can progress to a permanent threshold shift (PTS).9 NIHL is characterised by HL at high frequencies, especially between 3000–6000 Hz, often forming a notch at 4000 Hz. Continued exposure deepens this notch and can spread to adjacent frequencies.4 Pure tone audiometry is the standard diagnostic tool for NIHL, distinguishing among SNHL, conductive HL and mixed HL, by measuring air and bone conduction thresholds. The test requires a soundproof environment and calibrated equipment, performed by trained personnel. Proper patient preparation, including avoiding loud noise and treating infections beforehand, is crucial for accurate results.10 The prevalence of NIHL is reported to be higher in developing countries, including Uganda, compared with the developed world, and this is attributed to the lack of advocacy, proper hearing hygiene practices and targeted education regarding the preventive measures for NIHL.1 11 Several studies have been conducted to determine the burden of NIHL among some high-risk skilled workers worldwide, and prevalence ranging between 25% and 59.7% has been documented.12,16 These are way above the average global prevalence for NIHL of 16%.6 Other factors associated with NIHL development include male gender, advancing age, genetic predisposition, duration of working in a noisy environment, smoking and use of ototoxic drugs, among others.312 14,17 Individuals with NIHL usually present with difficulty hearing and tinnitus; however, imbalance and hyperacusis have also been reported.4 12 Ignorance about NIHL among individuals involved in these risky occupations is listed as one of the major factors potentiating its development.13 18 Also of note, sustainable use of hearing protection gear is akin to good knowledge about NIHL. Studies have shown that workers with adequate knowledge about NIHL have a lower prevalence than their counterparts, just as is the case in Europe and North America.7 18 19 It is therefore imperative to assess the knowledge about NIHL among the most at-risk individuals, such that the gaps can be appropriately addressed. Once established, NIHL has no cure, yet it is preventable.17 20,22 It has increased significantly, especially in developing countries. To halt this trend, WHO has recommended that hearing protective measures be prioritised as well as regular audiometric tests for all individuals who work in environments whose noise levels are above 85 dB.1 20 23 In Uganda, the Occupational Safety and Health (OSH) Act of 2007 clearly states this requirement. NIHL progression is influenced by noise intensity and exposure duration.24 Regular audiometric testing in noisy workplaces can monitor hearing deterioration. Hypertension, diabetes, hypercholesterolaemia and concurrent use of ototoxic drugs also impact its development. Awareness and proper implementation of preventive measures are key to mitigating this burden.6 There had not been studies on the prevalence of NIHL in Mbarara City (a city in western Uganda). However, reviewing anecdotal data from the Mbarara Regional Referral Hospital’s (MRRH) ear, nose and throat (ENT) department records showed that clients with hearing complaints comprised more than one-third of the total attendees (4750 total clients in calendar year 2019). Also, out of the total Pure Tone Audiograms (PTAs) done (about 40 per month in 2019/2020), approximately 50% of them were suggestive of SNHL and NIHL was among the most likely cause. Unfortunately, the details of patient occupations and lifestyles were not routinely captured in these records, thus presenting a knowledge gap for the role played by noise at workplaces for HL development among individuals visiting this department. Numerous informal metal fabrication workshops (MFWs) were among the workplaces where noise was generated within this city. Metal fabricators from these were at risk of repeated and prolonged exposure to noise with subsequent NIHL development. There were no data about the burden of NIHL among metal fabricators in Mbarara city. Information about knowledge about NIHL among these metal fabricators and other associated factors such as the intensity of noise generated, daily hours of exposure and duration of involvement in metal fabrication, among others, was also lacking. This made it imperative to determine the prevalence, risk factors and knowledge of NIHL among metal fabricators in Mbarara City to help devise the most appropriate means of preventing or stalling its development and progression. The objectives of this study were: To determine the prevalence of NIHL among Metal Fabricators in Mbarara City. To determine the prevalence of NIHL among Metal Fabricators in Mbarara City. To assess the level of knowledge about NIHL among metal fabricators in Mbarara City. To assess the level of knowledge about NIHL among metal fabricators in Mbarara City. To identify the factors associated with the risk of developing NIHL among metal fabricators in Mbarara City. To identify the factors associated with the risk of developing NIHL among metal fabricators in Mbarara City. Materials and methods A descriptive cross-sectional study was conducted in Mbarara city among metal fabricators from August 2021 to March 2022. The study was conducted in Mbarara city, the largest urban centre in Southwestern Uganda (located 270 km west of Kampala), was elevated to city status in May 2019 and became operational in July 2020 with an estimated population of 250 000 residents. The city hosts established industries such as Century Bottling Company, Roofings and Nile Breweries, alongside numerous small-scale MFWs. Over 30 informal MFWs employed approximately 300 workers, most operating without adherence to occupational safety guidelines or purpose-built infrastructure. These workshops often clustered together, with smaller units (employing fewer than five workers) sharing equipment or outsourcing specialised tasks. The study focused on MFWs due to their rising prevalence driven by increasing demand for metal products amid the city’s rapid development. Fabrication activities occurred in open-air environments, generating significant noise pollution from hammering/banging and metal-cutting machinery. For logistical efficiency, we recruited participants from neighbouring MFWs at a particular site that we termed a ‘study station’. These stations were only used for recruitment and ease of access purposes for our study participants. We created a total of 10 such stations for this study from which we referred participants to MRRH, the region’s largest public healthcare facility, offering ENT services including audiometry. Our inclusion criteria targeted all individuals employed in MFWs for ≥1 month (to account for the high turnover in Uganda’s informal sector) who agreed to avoid loud noise exposure for ≥12 hours prior to audiometric testing at MRRH and provided written informed consent. This ‘12-hour’ precaution minimised confounding effects of TTS on results. The exclusion criteria excluded participants with pre-existing documented HL prior to employment in metal fabrication and those with a history of ear surgery. We calculated the sample size using a formula for single population proportion with correction for a finite population,25 based on an estimated 48% prevalence of NIHL among metal fabricators,16 a population size of 300, and a 5% margin of error. The calculated sample size was 169, adjusted to 178 for attrition. A consecutive sampling method was employed to recruit eligible metal fabricators until the target sample size of 178 participants was achieved from a screened pool of 196 workers. Of the 18 excluded individuals, 12 declined participation, and the remaining six had pre-existing HL documented prior to their employment in metal fabrication. No stratification was applied in the sampling process. The research team visited various metal fabrication workshops in Mbarara City to introduce the study and recruit participants, who were then instructed to visit MRRH for study procedures, including hearing testing, questionnaire administration, blood pressure measurement, ear examinations and counselling. Participation was voluntary; workshops were notified one day prior to data collection, and participants were instructed to avoid noise exposure for ≥12 hours before their scheduled assessments. Each participant completed all study activities in approximately 40 minutes (see figure 1 for the study flow). Study organogram. Noise levels at each ‘study station’ were measured twice daily: once during peak operational hours (with machinery active) and once during breaks (quieter periods). Measurements were conducted using a calibrated BENETECH GM1352 Sound Level Meter (range: 30–130 dBA; frequency response: 31.5 Hz–8 kHz), operated by a certified ENT clinical officer from MRRH. Data were recorded on standardised forms for subsequent analysis. A pretested data collection tool with three sections was used to assess the different variables. It contained three sections, the first had semistructured interviewer-administered questions in both English and Runyankore (the local language) on sociodemographic factors, occupational history, hearing symptoms, knowledge about NIHL and comorbidities such as diabetes, hypertension and HIV infection. A knowledge assessment tool from Malaysia18 was adapted to assess knowledge about NIHL among metal fabricators for this study. It was translated, pretested and modified and then back translated into English for local use. The same tool had also been used in a Tanzanian study.26 This tool evaluated five domains: ‘general aspects of noise,’ ‘causes,’ ‘signs and symptoms,’ ‘treatment,’ and ‘prevention.’ Responses from this tool were scored, with 75% and above indicating ‘adequate knowledge’ and those with below as ‘inadequate knowledge’ about NIHL. A pretested data collection tool with three sections was used to assess the different variables. It contained three sections, the first had semistructured interviewer-administered questions in both English and Runyankore (the local language) on sociodemographic factors, occupational history, hearing symptoms, knowledge about NIHL and comorbidities such as diabetes, hypertension and HIV infection. A knowledge assessment tool from Malaysia18 was adapted to assess knowledge about NIHL among metal fabricators for this study. It was translated, pretested and modified and then back translated into English for local use. The same tool had also been used in a Tanzanian study.26 This tool evaluated five domains: ‘general aspects of noise,’ ‘causes,' 'signs and symptoms,’ ‘treatment,’ and ‘prevention.’ Responses from this tool were scored, with 75% and above indicating ‘adequate knowledge’ and those with below as ‘inadequate knowledge’ about NIHL. The second section recorded general and ear examination findings that were performed by the principal investigator. Wax was removed from 41 participants before the hearing test. 15 participants with active acute upper respiratory tract infections were treated and had their hearing test rescheduled. Six of them had other ear conditions (including three with eardrum perforations) that also required management and advice. The last section contained an audiogram which captured the hearing test data. Participants were asked to avoid loud noise for at least 12 hours prior to the test to minimise contribution of TTS on the results. For the same reason, the tests were performed in the morning hours. The hearing tests, PTA, were conducted in an audiometry booth at MRRH, while observing the ISO 8253-1:2010 guidelines. Background noise was monitored, with a maximum of 31 dBA using a sound level meter. Audiometry was performed using a BELL PLUS DIAGNOSTIC AUDIOMETER, with frequencies from 250–8000 Hz tested including the 3000 and 6000 Hz interoctaves. We used the modified Hughson-Westlake method to determine hearing thresholds.10 We first screened participants for normal hearing (<25 dB thresholds). Those failing the screening (>25 dB thresholds) underwent diagnostic PTA to characterise and assess the severity of HL. In this study, we defined NIHL as an SNHL with thresholds above 25 dB at frequencies of 3000 Hz, 4000 Hz or 6000 Hz in any or both ears. Those with audiograms suggesting SNHL with a configuration of NIHL that is a notch or trough at 3000–6000 Hz were thought to have followed noise exposure.8 Participants received their results and further assessment from ENT clinicians for those that needed it. Pure tone average was calculated from the average frequencies of 3000 Hz, 4000 Hz and 6000 Hz. In this study, the severity of HL was grouped into mild (26–40), moderate (41–60), severe (61–80) and profound (above 80).24 Study organogram We checked all study tools for completeness before data entry. The data from these tools was then entered into a database designed using EPI Info software V.7.2. The completed dataset was imported into STATA V.15.0 (Stata, College Station, Texas, USA) from EPI Info for analysis. We then generated descriptive statistics for all study participants’ characteristics using means and medians for continuous variables. Also used frequencies and proportions for categorical variables. The first objective of our study was to determine the prevalence of NIHL among metal fabricators in Mbarara City. We achieved this by calculating the proportion of metal fabricators with NIHL and expressing it as a percentage with a 95% CI. Then we explored the relationship between NIHL prevalence and factors such as age and gender using the χ2 test, applying a significance level of 5% to determine statistical relevance. The second objective focused on assessing the knowledge of NIHL among these metal fabricators. We used a knowledge assessment questionnaire for this, each correct response earning a mark and the total score expressed as a percentage. Then categorised scores as either adequate (75% and above) or inadequate (below 75%) knowledge. The study compared these knowledge levels against age, education level and duration of involvement in metal fabrication, using the χ2 test to identify statistically significant differences, with a p value threshold of <0.05. Lastly, the study sought to identify factors associated with NIHL by analysing various independent variables from the conceptual framework against the binary outcome of NIHL. Bivariate analysis using χ2 and logistic regression generated unadjusted ORs. We used factors that we found to be statistically significant or biologically plausible in multivariate analysis to control confounding factors. We then developed a final multivariate model using a backward stepwise selection method, with adjusted ORs and their 95% CIs reported for significant factors. Then, assessed the model’s quality using the goodness-of-fit test. Public and patient involvement Metal fabricators (patients) and their workshop supervisors were first engaged during the study design phase (around June 2021) to codefine research priorities and ensure feasibility. Research questions and outcome measures including knowledge gaps, barriers to hearing protection and other NIHL risk factors were refined through two focus groups (n=10 metal fabricators) to align with the workers’ lived occupational experiences and health concerns. Their input directly shaped the study design, particularly regarding practical logistics: workshop visit schedules, locally appropriate consent processes, simplified questionnaire language and audiometry arrangements at MRRH to minimise work disruption. During recruitment and conduct, workshop supervisors acted as community liaisons to endorse the study’s credibility and assist in participant enrolment. Fabricators codeveloped recruitment materials (eg, vernacular posters) and identified optimal times for hospital visits to reduce income loss. Participants were explicitly consulted about the burden of participation; they reviewed the 40 min assessment duration and 12-hour noise-avoidance requirement, leading to the implementation of transport reimbursement to offset financial barriers. For dissemination, fabricators co-lead community workshops using visual summaries, participated in local radio dialogues (in Runyankole, their local language) and codesigned plain-language brochures highlighting actionable prevention steps. These strategies were to ensure results reached both participants and the related occupational communities in accessible, relevant formats. Results This study was conducted among metal fabricators in Mbarara City, Uganda, between August 2021 and March 2022. Of the 10 study stations, 178 participants were recruited. During peak working hours, the mean noise level across all stations was 106±3 dBA (range: 102–108 dBA). In contrast, during quieter periods, the mean noise level measured 67±4 dBA (range: 62–71 dBA). Overall, the median age of the participants was 22 years (IQR 19–27). Participants were mostly male, that is, 171 (96.1%) of them and only seven (3.9%) were female. Most participants (74.0%) reported spending more than 10 hours a day involved in metal fabrication activities. It is important to note that the median duration of involvement in metal fabrication among participants was 3 years (IQR 1–5). More details are in table 1 below. In exploring other population factors, 48 of 178 (27%) of the participants had relatives with HL, seven of 178 (3.4%) had ever smoked, 52 of 178 (29.2%) were using alcohol then, whereas the majority (59.0%) reported exposure to recreational noise which included listening to music with headsets/earphones and visiting ‘disco’ hall. Also, to note, only one (0.6%) participant reported using a hearing protection device and these were earmuffs as he conducted metal fabrication activities. 80 (44.9%) of them reported tinnitus, whereas only 47 (26.6%) reported difficulty hearing. 31 (17.4%) participants reported undocumented HL before joining metal fabrication. 16 (9.0%) and 29 (16.3%) of the participants reported having hypertension and a recent use of ototoxic drugs, respectively. Table 1Patient clinical and non-clinical characteristics including NIHL prevalenceVariableParticipants (%)NIHL (%)Age categories (in years) 18–24115 (64.6)15 (13.0) 25–3444 (24.7)10 (22.7) >3519 (10.7)10 (52.6)Gender Male175 (98.3)1 (33.3) Female3 (1.7)34 (19.4)Education level No formal education/primary92 (51.7)20 (21.7) Secondary/tertiary86 (48.3)15 (17.4)Work in noisy environment before metal fabMetal Fabrication (MF)31 (17.4)13 (41.9)Ever smoked7 (3.4)5 (71.4)Current alcohol user52 (29.2)11 (21.2)Previous alcohol user44 (24.7)9 (20.5)Exposed to recreational noise105 (59.0)17 (16.2)Recent use of ototoxic drugs29 (16.3)10 (34.5)Duration of involvement in MF in years <130 (16.9)1 (3.3) 1–492 (51.7)16 (17.4) >556 (31.5)18 (32.1)Wears hearing protection device1 (0.6)Number of hours spent in MFWs per day <54 (2.3)1 (25.0) 5–937 (20.8)12 (32.4) >10132 (74.2)21 (15.9)Any difficulty hearing47 (26.4)11 (23.4)Tinnitus80 (44.9)18 (22.5)Hearing loss before joining MF31 (17.4)10 (32.3)Relative with hearing loss48 (27.0)13 (27.1)Hypertension16 (9.0)7 (43.8)Diabetes mellitus2 (1.1)HIV infection3 (1.7)Wax present and removed41 (23.0)10 (24.4)Perforated eardrum3 (1.7)Other ear abnormalities3 (1.7)Knowledge about NIHL Inadequate136 (76.8)25 (18.4) Adequate41 (23.2)9 (22.0)MF, metal fabrication; MFW, metal fabrication workshop; NIHL, noise-induced hearing loss. Patient clinical and non-clinical characteristics including NIHL prevalence MF, metal fabrication; MFW, metal fabrication workshop; NIHL, noise-induced hearing loss. The most common finding during physical examination was wax impaction and that was 41 (23.0%) of the total participants. Only three (1.7%) of our participants had a perforated ear drum. The overall prevalence of HL among the 178 metal fabricators in Mbarara City was 29.2% (52/178). Of these cases, we classified 67.3% (35/52) as having SNHL, corresponding to a prevalence of 19.7% in the study population. And according to our study criteria, all these 35 SNHL cases met the definition of NIHL (see table 2 for details). Thus, the prevalence of NIHL among metal fabricators was 19.7%. Table 2HL typesPrevalenceHL (%)NIHL (%)Overall52 (29.2)35 (19.7)Type Conductive14 (26.9) SNHL35 (67.3) Mixed3 (5.8)Laterality Right18 (34.6)14 (40.0) Left12 (23.1)4 (11.4) Both22 (42.3)17 (48.6)Severity Mild32 (61.5)21 (60.0) Moderate15 (28.9)12 (34.3 Severe4 (7.7)2 (5.7) Profound1 (1.9)0 (0.0)HL, hearing loss; NIHL, noise-induced hearing loss; SNHL, sensorineural hearing loss. HL types HL, hearing loss; NIHL, noise-induced hearing loss; SNHL, sensorineural hearing loss. Notably, the prevalence of NIHL increased with age, that is 13.0% for 18–24 years, 22.7% for 25–34 years and 52.6% for those aged 35 years. Among the three female participants, one (33.3%) had NIHL although the association between gender and NIHL was not statistically significant. Regarding laterality, 48.6% (17/35) of those affected by NIHL experienced bilateral HL. For unilateral cases, the right ear was more significantly affected, with 14 cases (40.0%) compared with four cases (11.4%) whose left ear was affected. The severity of HL was in the mild and moderate categories, with only two cases classified as severe. Of the 178 participants, only 41 (23.2%) had good knowledge of NIHL. The majority (76.8%) had a total score of below 75% in the knowledge assessment tool indicating inadequate knowledge about NIHL. When we did stratification with age, education levels and duration in metal fabrication, there was no evidence for a statistically significant relationship. The most correctly answered items were in the domain of ‘general aspects on noise’, which is by an average of 93.6% of the participants. That of ‘causes’ at 89.4%, ‘signs and symptoms’ at 58.4%, ‘prevention’ at 39.1% followed and the least performed being ‘treatment’ with an average of 29.0% of the participants. The item ‘loud noise in metal fabrication workshops can cause HL’ was the most correctly answered, that is, by 172 of 178 (96.6%) participants, and the least correctly answered item was ‘HL due to noise can be treated by taking medicine’ by only 19 of 178 (10.7%) participants. We then ran a bivariate analysis for all the factors which included the noise exposure levels, occupational parameters and the other biologically plausible factors in table 1 that we thought to be associated with the development of NIHL. Metal fabricators aged 35 years and above had 7.4 times the odds of developing NIHL than those aged 18–24 years (95% CI 2.6 to 21.2, p=0.000). A positive family history of HL increased the odds of developing NIHL to two times that of their counterparts (95% CI –0.9 to 4.4, p=0.095). Among the behavioural factors, a history of smoking posed 11.8 times the odds of developing NIHL (95% CI 2.2 to 63.5, p=0.004). Similarly, a history of recent use of ototoxic drugs had 2.6 times the odds of developing NIHL (95% CI 1.1 to 6.3, p=0.019). Participants who had spent 5+ years in metal fabrication had odds of 13.7 times developing NIHL. More details follow in table 3. Table 3Bivariate and multivariate analysis results examining the factors associated with NIHL among metal fabricators in Mbarara CityVariableBivariate analysisMultivariate analysisUnadjusted OR (95% CI)P valueAdjusted OR (95% CI)P valueAge category in years 18–241.01.0 25–342.0 (0.8 to 4.8)0.1381.3 (0.4 to 4.1)0.680 >357.4 (2.6 to 21.2)0.000*3.4 (0.7 to 16.8)0.132Previous work in a noisy environment No1.01.0 Yes4.1 (1.8 to 9.6)0.001*3.3 (1.1 to 9.0)0.028*Duration in MF <1 year1.01.0 1–4 years6.1 (0.8 to 48.1)0.0867.6 (0.9 to 62.8)0.061 >5 years13.7 (1.7 to 109.0)0.013*9.9 (1.0 to 16.8)0.055Hearing loss prior to MF No1.01.0 Yes2.3 (1.0 to 5.5)0.0572.2 (0.7 to 6.5)0.173Relative with hearing loss No1.01.0 Yes2.0 (0.9 to 4.4)0.0952.0 (0.8 to 5.1)0.137 I do not know5.3 (0.7 to 39.8)0.1054.5 (0.4 to 46.2)0.210Ever smoked No1.01.0 Yes11.8 (2.2 to 63.5)0.004*3.4 (0.5 to 24.2)0.061Recent use of ototoxic drugs No1.01.0 Yes2.6 (1.1 to 6.3)0.032*1.7 (0.6 to 4.9)0.310*p value<0.05.MF, metal fabrication; NIHL, noise-induced hearing loss. Bivariate and multivariate analysis results examining the factors associated with NIHL among metal fabricators in Mbarara City p value<0.05. MF, metal fabrication; NIHL, noise-induced hearing loss. After adjusting for statistically and biologically significant variables (age, previous work in a noisy environment, duration in MF, HL before joining MF, having a relative with HL, smoking and recent use of ototoxic drugs) from the multivariate analysis, we found that ‘previous work in a noisy environment’ was the only factor which truly increased the odds of developing NIHL in this population with an OR of 3.3 (CI 1.1 to 9.0) and a p value of 0.028. However, it is important to note that ‘duration in metal fabrication’ of 5 years or more had odds of 9.9 (CI 1.0 to 16.8) times of developing NIHL and a p value of 0.055. Advancing ‘age’, having a ‘relative with HL’, increased ‘duration of involvement in metal fabrication’ of 1–4 years, having ‘ever smoked’ and a ‘recent use of ototoxic drugs’ also demonstrated increased odds with wide CIs for the development of NIHL among metal fabricators in Mbarara City, but these were not statistically significant. Refer to table 3 for details. Discussion Metal fabricators were exposed to elevated levels of occupational noise in Mbarara city, with measurements reaching up to 108 dBA during peak working hours as noted across all the study stations. These levels significantly exceed the recommended occupational exposure limit of 85 dBA above which we expect damage to the inner ear sensory hair cells.4 Most participants in this study were male (98.3%), which aligns with the prevailing cultural norms in Uganda. Here, we traditionally view metal fabrication as a physically demanding occupation suited for young men. The few women involved in the industry typically occupied roles in administration and accounting which require less physical exertion and fewer hours of exposure to hazardous noise levels. While male predominance in this field has been reported in other studies,12 27 our findings reveal an even higher proportion than theirs. One study in Nigeria reported an equal distribution of male and female participants, contrasting sharply with our findings.28 Most study participants reported working for more than 8 hours a day for at least 5 days a week indicating prolonged exposure to high noise levels, a key factor contributing to the development of NIHL. The most common symptoms that were reported by participants included difficulty hearing and tinnitus, findings consistent with similar studies conducted in Nigeria and Tanzania.12 27 Imbalance and hyperacusis are the other symptoms reported in the literature; however, they were not assessed in our study. We found the prevalence of NIHL among metal fabricators in Mbarara city to be 19.7%. This figure is lower than the global prevalence estimates of NIHL among workers exposed to noisy environments which range between 37% and 59.7%.7 It is also lower compared with findings from similar occupational settings in various African countries such as that among textile workers in Tanzania12 at 58.5%, employees of a mineral industry in Zimbabwe17 at 53%, quarry workers in Ghana29 at 44%, plus iron and steel workers in Tanzania16 at 48%. Several factors may explain the comparatively lower prevalence observed in our study. One explanation is the difference in the scale and mechanisation of operations. The study sites in Mbarara consisted of small-scale workshops with limited mechanisation and less work volumes, resulting in reduced noise exposure. In contrast, participants in most of the comparison studies were drawn from larger, more formalised industries, where exposure to high noise levels from heavy machinery is more common. Also, to note, most of the metal fabrication activities in Mbarara City were conducted in open-air environments as observed during our visits to the workshops. This contrasts with the enclosed or underground workspaces reported in some of the comparison studies. For example, in the mining sector, it has been documented that underground mining is more likely to be associated with a higher prevalence of NIHL compared with open-air counterparts due to the confined spaces that amplify noise levels.15 Therefore, the open-air work setting in Mbarara could have served as a protective factor by allowing noise to dissipate more effectively. Overall, these contextual differences in work environments and operational scales may account for the lower NIHL prevalence observed among metal fabricators in Mbarara City. Most of the affected individuals in our study had bilateral HL, and the right ear was found to be more significantly affected. This asymmetry may be attributed to the predominance of right-handedness among workers, leading to greater noise exposure on the right side as they perform their duties, though, from existing literature, NIHL is usually bilateral and symmetrical.4 5 Another example, a study involving 42 777 army personnel found more pronounced NIHL in the left ear, regardless of handedness, suggesting that factors beyond handedness, such as cortical pathway asymmetries, may influence susceptibility to cochlear damage.30 Additionally, our findings indicated that all cases of NIHL were categorised as mild to moderate in severity. This agrees with existing literature, which suggests that NIHL seldom results in severe HL.4 12 31 Among metal fabricators, only 23.2% of the participants demonstrated adequate knowledge about NIHL, a finding similar to observations from a Tanzanian study.26 This could be due to similarities in geographical context and education curricula within East Africa. In contrast, a Malaysian study reported that only 11% of participants had good knowledge of NIHL.13 In our study, the domains of treatment and prevention were the poorest performing, a trend also observed among Tanzanian iron and steel workers plus Malaysian sawmill workers.18 26 These findings suggest a large knowledge gap among our artisans. Additionally, we found a significant association between NIHL knowledge and education level, reflecting the varying emphasis on health education across different schooling levels. Notably, studies from Ghana and Nigeria reported higher knowledge levels among workers (57% and 93%, respectively),14 32 which could be because they used a different assessment tool from ours. Previous employment in a noisy environment emerged as the only factor found independently associated with NIHL among metal fabricators in our study, a finding that is consistent with a 2019 study conducted in Tanzania16 and other literature.24 Additionally, a duration of involvement in metal fabrication of 5 years or more was associated with up to a 10-fold increased risk of developing NIHL, demonstrating a meaningful relationship attributable to inner ear damage that occurs following prolonged noise exposure. Other factors, including increasing age, having a family history of HL, smoking and recent use of ototoxic drugs, showed relative associations with NIHL, though these did not reach statistical significance, potentially due to the limited sample size in our study. The other observed associations were with advancing age, family history of HL, smoking and ototoxic drugs and are biologically plausible. For example, ageing is linked to diminished regenerative capacity of sensory epithelia,5 a genetic predisposition may render individuals more susceptible to noise-induced damage4 and smoking induces vascular changes that impair the repair of damaged auditory cells.5 33 34 These findings align with reports from other studies13 16 and underscore the multifactorial nature of NIHL development. Recent use of ototoxic drugs was similarly significantly associated with NIHL development. However, in this study, these data weres self-reported and may not represent the true picture of ototoxicity. The the most common ototoxic drug that our participants reported in our study was quinine use in the past 1 year. This is due to the synergistic effect that ototoxic drugs offer with noise exposure for inner ear sensory epithelia damage.5 105 of our participants reported exposure to recreational noise; however, this was not statistically associated with NIHL in our study as in the literature.5 This could also have been due to our low sample size. Conclusions The prevalence of NIHL among metal fabricators in Mbarara City was 19.7%, with most participants demonstrating limited knowledge of NIHL causes and prevention. A history of ‘prior employment in noisy environments’ before entering metal fabrication emerged as the sole factor independently associated with NIHL development. Additionally, prolonged occupational exposure (≥5 years in metal fabrication) showed a significant association with increased risk. Contributing factors included advancing age, family history of HL, smoking and ototoxic drug use. Recommendations This study demonstrated that MFWs in Mbarara City generate high levels of noise to which the metal fabricators are exposed with no hearing protection and thus at risk of development of NIHL among them. To mitigate this, we recommend The implementation of targeted health education programmes either at individual workshops or training designated representatives to raise awareness about NIHL. The implementation of targeted health education programmes either at individual workshops or training designated representatives to raise awareness about NIHL. Enforcing robust safety policies, including the use of ear protective devices, routine hearing screenings and monitoring of work schedules, is crucial. Enforcing robust safety policies, including the use of ear protective devices, routine hearing screenings and monitoring of work schedules, is crucial. Further research is needed to explore the long-term effects of noise exposure, assess the effectiveness of interventions and assess the contribution of additional noise sources, like traffic, on NIHL development in Mbarara. Further research is needed to explore the long-term effects of noise exposure, assess the effectiveness of interventions and assess the contribution of additional noise sources, like traffic, on NIHL development in Mbarara. Study limitations The absence of documented noise exposure profiles in Uganda complicates the accurate determination of NIHL risk among metal fabricators. The absence of documented noise exposure profiles in Uganda complicates the accurate determination of NIHL risk among metal fabricators. The lack of personal dosimeters and reliance on self-reported abstinence from loud noise before testing may have influenced the study’s findings. The lack of personal dosimeters and reliance on self-reported abstinence from loud noise before testing may have influenced the study’s findings. Our study did not capture the distinct roles of the study participants in these MFWs and thus we couldn’t assess the effect of NIHL on these subpopulations. Our study did not capture the distinct roles of the study participants in these MFWs and thus we couldn’t assess the effect of NIHL on these subpopulations. Additionally, variations in work environments might mean the results are not fully representative of the entire population of metal fabricators in Mbarara. Additionally, variations in work environments might mean the results are not fully representative of the entire population of metal fabricators in Mbarara. The cross-sectional design also introduces potential recall bias on factors such as use of ototoxic drugs, despite efforts to minimise it. The cross-sectional design also introduces potential recall bias on factors such as use of ototoxic drugs, despite efforts to minimise it. A cross-sectional study could not establish a causal relationship between noise exposure and NIHL, suggesting a need for a longitudinal study for more definitive conclusions. A cross-sectional study could not establish a causal relationship between noise exposure and NIHL, suggesting a need for a longitudinal study for more definitive conclusions. Acknowledgements I deeply thank my supervisors for their invaluable guidance, support and insights throughout this research. Your mentorship was crucial in shaping the study. I also thank the study participants, whose willingness to respond and be examined enabled the exploration of noise-induced hearing loss among metal fabricators in Mbarara City. Their contributions provided essential insights for advocating better occupational health practices. Footnotes Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Patient consent for publication: Not applicable. Ethics approval: This study received ethical approval from the Research Ethics Committee of Mbarara University of Science and Technology (MUST-REC; Reference: MUST-2021-61) and all relevant local administrative bodies. Written informed consent was obtained from all participants after explaining the study’s purpose, risks and benefits. Confidentiality was ensured through anonymised data collection and secure storage. Participants gave informed consent to participate in the study before taking part. Provenance and peer review: Not commissioned; externally peer reviewed. Patient and public involvement: Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research. Data availability statement Data are available upon reasonable request. References Associated Data Data Availability Statement Data are available upon reasonable request.