Introduction
Hearing loss (HL) is one of the most common sensory deficits worldwide, affecting millions of individuals and significantly diminishing quality of life. In India alone, approximately 63 million individuals experience substantial auditory impairment. This burden is expected to increase, as the global prevalence of HL is expected to reach 630 million by 2030 [1,2]. The consequences of HL go beyond auditory function, influencing communication, education, employment, mental health, and social integration [3,4]. The burden of HL is especially high among older adults, with more than 25% of individuals over 60 years affected [5], contributing significantly to years lived with disability [6].
Multiple risk factors contribute to hearing impairment, including aging, chronic otitis media, noise exposure, diabetes, hypertension, and lifestyle behaviors such as smoking and alcohol use [7–9]. Sensineural hearing loss (SNHL) is often associated with aging and systemic comorbidities, whereas conductive hearing loss (CHL) is frequently linked to middle ear pathologies, such as chronic suppurative otitis media [10]. Less recognized causes include autism [11], central auditory processing disorders [12], and tumors [13]. Noise remains a significant and preventable cause of HL, especially in occupational settings. It is considered a pervasive industrial pollutant, often leading to noise-induced hearing loss (NIHL), which typically manifests as bilateral, symmetrical, high-frequency SNHL [14–17]. A plethora of studies suggest that genetics and immune response also play a vital role in the causation of HL [18].
Despite the prevalence of HL, particularly in industrial regions of India, epidemiological data exploring its multifactorial etiology remain sparse. This study aims to examine the degree, pattern, and magnitude of risk factors associated with HL among adults attending a tertiary care hospital in East Singhbhum, an industrial district of eastern India. By identifying the relative contribution of various demographic, occupational, lifestyle, and medical factors, this research seeks to support targeted prevention and early intervention strategies.
Subjects and Methods
Study design and ethical clearance
An analytical cross-sectional study was conducted from February 2024 to July 2024 in the ENT Department of Tata Main Hospital (TMH), a tertiary care center attached as a teaching hospital to Manipal Tata Medical College, Jamshedpur, East Singhbhum, India. The study was started after obtaining ethics approval (No. TMH/IEC/MAR/142/2024) from the Institutional Review Board of TMH and Manipal Tata Medical College. Consent to participate was obtained from all participants involved in the study.
Study population and sampling
A hospital-based cross-sectional observational study with analytical components, using a convenience sampling method, was employed, and 295 patients presenting with complaints of HL were included from the Department of ENT. Patients aged 20–80 years presenting with complaints of HL who consented were included in the study. Patients unable to respond to pure tone audiometry (PTA) or who did not provide informed consent were excluded from the study.
Clinical examination and audiometry
Each patient underwent a detailed medical history review and a comprehensive examination of the ear, nose, and throat. The history of hypertension and diabetes mellitus was retrieved from patients’ medical history and further confirmed by clinical records.
An otoscopic examination was first performed, followed by microscopic evaluation of the ear to assist in diagnosis. Hearing assessment was conducted using PTA in a soundproof room meeting ISO 8253–1:2010 standards, with the MAICO MA 53 clinical audiometer (MAICO Diagnostics GmbH) and standard headphones. The audiometer was calibrated according to ANSI S3.6–2018 standards, and examinations were conducted by certified audiologists with 10 years of clinical experience. The TDH39 headphones were used to measure air conduction hearing thresholds, while the B71 headphones were used for bone conduction hearing thresholds for each ear in a two-room soundproof setup. PTA was conducted from 250 Hz to 8 kHz for air conduction and from 250 Hz to 4 kHz for bone conduction thresholds. The Hughson-Westlake procedure was employed to determine the hearing thresholds for each ear. The World Health Organization (WHO) classification criteria (mild: 26–40 dB; moderate: 41–60 dB; severe: 61–80 dB; and profound: >80 dB) for HL were used to classify the clinical cases. HL was further classified as conductive, sensorineural, or mixed [2,19].
Statistical analysis
The demographic data, including age, sex, religion, and location, were recorded for each patient. The collected data were organized and tabulated using Microsoft Excel 2016, and statistical analysis was performed using the Statistical Package for Social Sciences (SPSS) version 27.0 (IBM Corp.) and RStudio Version 2025.05.1+513 (Posit Software, PBC). The data were analyzed using appropriate statistical methods, with the chi-square test applied to examine associations. To determine a significant linear trend in the binomial proportion across ordinal distribution of the variables across increasing levels of HL severity, the Cochran–Armitage Trend test was used. A multiple linear regression analysis was conducted with the average PTA threshold (dB HL) as the dependent (continuous) variable, representing the severity of HL. The estimated effects (β coefficients) with a 95% confidence interval (CI) for HL were calculated. The model included predictors such as age group, sex, occupation, personal history (e.g., smoking, noise exposure), type of HL, and diagnosis, to examine factors associated with the degree of hearing impairment in patients already diagnosed with HL. A p-value less than 0.05 was considered statistically significant.
Results
Sociodemographic characteristics
The demographic characteristics of the study population are shown in Table 1. The study sample (n=295) consisted of slightly more females (52.54%) than males (47.46%). Sex distribution did not show a statistically significant difference (χ2(1)=0.763, p=0.382), suggesting sex parity in HL presentation. On the other hand, the age distribution of participants was statistically significant (χ2(4)=65.288, p<0.001), indicating that HL was not evenly distributed across age groups. The highest proportion of participants was aged between 46 and 60 years (33.22%), followed by those aged 61–75 years (26.44%) and 30–45 years (21.36%). Participants aged above 75 years accounted for 8.14% of the sample. Occupational status also showed a significant association with the distribution of HL (χ2(2)=73.329, p<0.001), with factory workers and non-worker individuals each accounting for 45.08% of the study population.
This substantial representation of factory workers underscores the relevance of occupational noise exposure as a potential risk factor, although a definitive diagnosis of NIHL was limited by the lack of detailed exposure histories. These findings confirm the heterogeneity of the study sample. The statistically significant variation across age and occupational categories justifies further subgroup analysis and supports the inclusion of these variables in regression models for HL prediction. It also indicates that the risk of HL is multifactorial and varies significantly based on demographic and occupational factors, reinforcing the need for stratified public health interventions.
Pattern and causative factors of HL
Table 2 summarizes the pattern of HL among 295 participants based on personal history, predisposing factors, diagnosis, type, severity, and PTA findings. Most participants (82.37%) reported no personal history, while smaller proportions reported smoking (6.44%), alcohol use (3.73%), or both (2.71%), with a significant association between personal history and HL pattern (χ2(7)=1,322.708, p<0.001). Nearly half (47.80%) had no known predisposing factors, while 13.22% had diabetes, 16.95% hypertension, and 20.68% both, also showing a significant association (χ2(4)=173.051, p<0.001). Chronic otitis media (34.20%) and presbycusis (28.10%) were the most common diagnoses, followed by others such as NIHL (7.80%), BPPV (5.10%), acute suppurative otitis media (ASOM) (4.70%), and sudden SNHL (3.70%), with a significant distribution (χ2(6)=190.610 p<0.001). Most participants had bilateral HL (94.92%) versus unilateral (5.08%) (χ2(1)= 238.051, p<0.001), and the majority had moderate (50.51%) or mild (31.86%) loss, with fewer showing severe (11.86%), profound (3.05%), or normal hearing (2.71%) (χ2(4)=254.271, p<0.001). PTA results showed 45.76% had SNHL, 26.10% CHL, 25.42% mixed, and 2.71% normal hearing, with a significant association (χ2(3)=109.65, p<0.001).
Association between the level of HL and various variables
The association between the level of HL and various demographic and clinical variables is presented in Table 3. There was a statistically significant association between age group and level of HL (p<0.001). Participants aged 46–60 and 61–75 years exhibited a higher prevalence of moderate to severe HL, while younger age groups (15–30 and 31–45 years) were more commonly in the normal to mild HL categories.
The analysis revealed that the severity of HL was significantly associated with several factors, including age (p<0.001), occupation (p=0.005), personal history (such as smoking and noise exposure) (p<0.001), predisposing medical conditions (diabetes and hypertension) (p<0.001), type of HL (p<0.001), PTA classification (p<0.001), and diagnosis (p<0.001). Older age groups, factory workers, and individuals with combined risk factors (e.g., smoking, noise exposure, or comorbidities) were more likely to have moderate to severe HL. Bilateral HL and sensorineural or mixed types were more common in higher severity categories. Chronic otitis media and presbycusis were the most frequent diagnoses in moderate to severe cases.
The Cochran-Armitage test for trend showed significant linear associations between HL severity and key binary variables. Male participants had a higher grade (severe) of HL (Z=2.08, p=0.038), and bilateral HL became more common with increasing severity (Z=−2.92, p=0.004). Factory workers also showed a significant trend toward more severe HL (Z= 2.84, p=0.005), indicating a potential occupational risk. A weak but statistically significant positive correlation was observed between age and average HL (r=0.202, p<0.001), suggesting that HL tends to increase with age (Fig. 1).
Risk factors associated with HL
To predict HL (dB HL) based on age group, sex, occupation, personal history, predisposing factors, type of HL, PTA final classification, and diagnosis, a multiple linear regression was conducted (Table 4). The overall regression model was statistically significant, F(28, 266)=8.84, p<0.001, explaining approximately 48.2% of the variance in HL (R2=0.482, adjusted R2=0.427). This suggests that the included variables collectively provided a good fit for predicting HL severity. The dependent variable in this model is the average hearing threshold in decibels (dB HL). Age was a strong predictor of HL, with participants aged 15–75 years showing significantly lower hearing thresholds compared to those over 75 (all p<0.05). Sex and occupational status were not significant predictors. Participants with both smoking and alcohol use (β=−18.58, p<0.001) or alcohol use alone (β=−13.02, p=0.003) had significantly greater HL. Combined noise and smoking exposure (β=23.27, p=0.004) was also significant, while noise or smoking alone was not. Among comorbidities, hypertension (β= 4.84, p=0.037), combined diabetes and hypertension (β=4.70, p=0.045), and other conditions (β=16.00, p=0.018) were significantly associated with increased HL. Unilateral HL was linked to higher thresholds than bilateral (β=13.45, p=0.002). All types of PTA-confirmed HL, i.e., CHL (β=23.44, p<0.001), SNHL (β=22.41, p<0.001), and mixed (β=38.38, p<0.001), were significantly associated with higher hearing threshold levels (dB). Clinical diagnoses such as presbycusis, chronic otitis media, noise-induced HL, sudden SNHL, and BPPV were also significantly associated with greater HL (all p<0.01), while acute suppurative otitis media was not (p=0.372).
Frequency-wise average pure tone thresholds by occupation and personal history (in dB HL) are shown in Table 5. Factory workers show significantly elevated thresholds at 4 kHz and 8 kHz, consistent with patterns seen in NIHL. The noise exposure + smoking group demonstrates a steep drop at 4 kHz and 8 kHz, typical of acoustic trauma or cumulative NIHL. Smoking-only individuals also show elevated thresholds compared to those without exposure. Among participants reporting occupational noise exposure, the self-reported mean (±SD) duration of exposure was 9.2 (±3.5) years, primarily among factory workers.
Discussion
This study investigated the degree, pattern, and risk factors of HL among adult individuals in East Singhbhum, an industrial district of eastern India. The results showed that there is a significant association between HL severity and factors such as age, industrial noise exposure, smoking, and chronic medical conditions such as diabetes and hypertension. Moreover, the present study highlights the synergistic impact of noise and smoking on auditory outcomes. The most common audiological diagnosis observed was SNHL (45.76%), followed by CHL (26.10%) among older adults and factory workers.
A striking 94.92% of participants exhibited bilateral HL, with moderate HL being most common (50.51%) in this study. These findings are consistent with existing literature, indicating that bilateral, moderate-level hearing impairments are typically observed in age-related and noise-induced cases [20]. Age was a strong predictor (p<0.001), with individuals above 60 years showing the highest prevalence of moderate to severe HL, consistent with presbycusis-related degeneration [21]. Younger age groups (15–45 years) had significantly lower hearing thresholds.
The linear trend association between participant characteristics and the severity of HL was analyzed by employing the Cochran-Armitage test. Male participants (Z=2.08, p= 0.038), those with bilateral HL (Z=−2.92, p=0.004), and factory workers (Z=2.84, p=0.005) showed increasing trends with greater hearing impairment. These findings underscore the need for targeted prevention, especially among high-risk groups like male industrial workers.
Further, factory workers significantly exhibited disproportionately higher levels of moderate and severe HL than non-factory workers and unemployed individuals (p=0.005). Although occupation alone was insignificant in multivariate regression, univariate analysis suggests a probable link between industrial noise exposure and hearing impairment. This is supported by prior studies demonstrating occupational noise as a major contributor to HL [14].
Personal history variables such as noise exposure with smoking history (β=23.27, p=0.004), smoking and alcohol (β=−18.58, p<0.001), and alcohol (β=−13.02, p=0.003) were found as significant risk factors for HL, echoing findings that it can impair cochlear blood flow and intensify the effects of noise [22].
Diabetes with hypertension was significantly associated with HL in univariate analysis (p<0.001), consistent with studies linking these conditions to cochlear microvascular damage [23,24]. However, they were not retained as independent predictors in regression analysis, suggesting an indirect role or interaction with dominant variables like age or PTA classification.
Factors associated with HL severity were identified using a regression model applied among patients with confirmed auditory impairment. As all individuals in the sample had some degree of HL, the model was not intended to assess risk for the development of HL per se. Instead, it helped to identify demographic and clinical correlates of more severe auditory dysfunction. Unequal distribution of etiologies or patient-level factors may have influenced the estimates. The PTA classification was significantly associated with hearing severity (p<0.001). Mixed HL predicted worse outcomes (β=38.38, p<0.001), while a CHL PTA result indicated higher thresholds (β=23.44, p<0.001), reinforcing PTA’s diagnostic utility [25]. Diagnoses such as SNHL (β=29.06, p<0.001), followed by BPPV (β=13.72, p=0.001), NIHL (β=10.85, p=0.007), presbycusis (β=9.66, p<0.001), and chronic otitis media (β=8.65, p=0.003) were all strongly linked to greater HL severity [26]. Moreover, it was observed that sudden SNHL was significantly associated with worse hearing outcomes. This aligns with prior studies that highlight the abrupt and often severe impact of sudden SNHL on auditory thresholds [27].
No significant sex differences were observed (p=0.038), consisting of slightly more females (52.54%) than males (47.46%), diverging from some studies that report higher male susceptibility due to occupational noise exposure [28]. This sex parity may reflect similar environmental and behavioral exposures in this study population. The weak positive correlation between age and average HL (r=0.202) aligns with existing literature, suggesting age as a contributing but not sole factor in hearing decline. The scatter plot (Fig. 1) illustrates the heterogeneity in HL among individuals of the same age. This variability may be attributed to differences in occupational exposure to noise, environmental factors, lifestyle habits, genetic predisposition, or the presence of chronic diseases such as diabetes or hypertension.
Despite limitations such as its single-center hospital setting, limited resources, lack of patient follow-up, societal barriers, and the use of convenience sampling (which may introduce selection bias and limit generalizability), this study provides valuable insight into regional patterns and risk factors of HL in an industrial setting in eastern India. A larger, community-based study is recommended to validate these findings.
This study highlights age, industrial noise exposure, smoking, and comorbidities such as hypertension with diabetes as significant risk factors for HL in the eastern Indian population. Chronic otitis media emerged as the most common diagnosis; however, the high prevalence of sensorineural and mixed HL, particularly among factory workers, underscores the impact of occupational and systemic factors. These findings call for early detection programs, stricter noise regulation, and integrated preventive strategies to mitigate HL in high-risk groups.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
