The present study aimed to explore the combined impact of noise and dust exposure on hearing loss and extra-auditory effects, such as cardiovascular health, glucose metabolism, and obesity risks, in occupationally exposed populations. The study also sought to identify associated risk factors using machine learning algorithms. We found that combined exposure to noise and dust significantly increases the risk of noise-induced hearing loss (NIHL), as well as diastolic blood pressure, BMI, fasting blood glucose, and abnormal liver function, compared to exposure to noise alone.
Furthermore, male gender, age, combined exposure, elevated diastolic blood pressure, elevated fasting blood glucose, and smoking were identified as independent positive predictors of ONIHL. The results of multivariate logistic regression were validated using machine learning algorithms, with the logistic regression model confirmed as the optimal predictive model. This finding was consistent with the multivariate analysis. Additionally, men without underlying diseases, aged 30–39 years, and with less than 10 years of work experience, were found to be at higher risk of developing NIHL.
Several cross-sectional studies have investigated the relationship between noise exposure, age, and hearing loss. Somma et al.12 observed 5 dB and 20 dB hearing losses in younger and older cement workers (exposed to > 85 dB daily noise) compared to controls. Golmohammadi et al.13found that hearing loss increased with age and years of exposure in 1,062 tractor workers (Leq > 85 dB(A)), with specific regression coefficients. Hederstierna and Rosenhall14 reported an additive relationship between noise-induced hearing loss (NIHL) and aging in individuals aged 70–75 years.
Overall, the literature supports an additive model of NIHL and aging, which is consistent with our findings. Specifically, our study demonstrated that the detection rate of NIHL increases with both years of exposure and age in workers exposed to noise.
The detection rate of noise-induced hearing loss (NIHL) was higher in males than in females, and this difference was statistically significant, which is consistent with previous studies15,16.This may be attributed to differences in hormone levels, as well as the higher prevalence of harmful habits such as smoking and alcohol consumption in males compared to female17,Our results also align with these findings, showing higher detection rates of NIHL in populations with smoking and alcohol consumption habits.
Additionally, our study indicated that individuals with hypertension, high fasting blood glucose, and a high body mass index (BMI), particularly those who are obese, are more vulnerable to NIHL. The underlying mechanisms are as follows: In cases of hyperglycemia, persistent high blood sugar levels damage the inner ear by thickening the microvessel basement membranes, leading to microvascular lesions, impaired nerve fiber conduction, and disrupted metabolism, all of which ultimately impair hearing[18].
Hypertension similarly disrupts cochlear microcirculation, increasing the risk of hemorrhage and subsequent reperfusion injury. Damaged hair cells and reduced blood flow further lower noise tolerance, enhancing the risk of hearing loss. Systolic and diastolic blood pressure have different physiological mechanisms. Therefore, in this study, systolic and diastolic pressures were analyzed separately. Systolic pressure reflects the volume of blood ejected by the heart during contraction and the tension in the arterial walls, while diastolic pressure reflects the pressure in the arteries when the heart is relaxed. Their effects on blood vessels and organs are distinct, so analyzing them separately provides a clearer understanding of their different roles in health. Systolic pressure likely has a more significant impact on hemodynamics and cochlear blood supply, whereas diastolic pressure may play a more crucial role in long-term vascular health. Additionally, clinical treatment and interventions typically address systolic and diastolic pressures with different strategies. Thus, separate analysis allows for a more accurate assessment of their independent effects and helps avoid confounding their influence.
Obesity also contributes to inner ear damage. Vascularly, it increases capillary pressure and promotes atherosclerosis, while metabolically, it reduces adiponectin levels and induces oxidative stress. These combined effects contribute to damage in the inner ear[19].
Furthermore, a novel finding in our study is that combined exposure to noise and dust significantly increases the risk of NIHL in workers compared to single noise exposure. This suggests a synergistic effect of noise and dust on the development of NIHL.
Previous research has shown that noise exposure acutely affects the autonomic nervous and endocrine systems3,, influencing blood pressure (BP) levels and altering blood lipid and glucose concentrations, which are key factors in the development of cardiovascular disease18.Occupational noise exposure has also increasingly been identified as a risk factor for obesity. The potential mechanisms underlying this association may involve noise exposure inducing dysregulation of the endocrine system through activation of the hypothalamus-pituitary-adrenal axis and elevated cortisol levels. This, in turn, can alter metabolism, promote central fat deposition, and contribute to obesity5,19.Additionally, some studies indicate that long-term dust exposure leads to lung fibrosis and the entry of dust particles into the bloodstream, causing stress reactions in other systems. This can lead to the development of pulmonary hypertension, accelerate atherosclerosis, and further affect the function of cardiovascular and other systems[23]. In our study, we grouped participants based on the presence or absence of combined exposure and found significant differences between the two groups in terms of diastolic blood pressure (DBP), BMI (especially obesity), fasting blood glucose, and liver function. However, we did not observe significant changes in the prevalence of DBP or abnormal electrocardiogram (ECG) in exposed participants, although there was an increase in systolic blood pressure (SBP) and abnormal liver function.Our findings are largely consistent with those from a controlled exposure study, which showed an increase in DBP but no change in SBP or heart rate in 18 healthy human volunteers exposed to occupational noise at 95 dBA for 20Â min20Additionally, exposure to noise exacerbated liver damage21 .
Our results demonstrated that several factors were significantly associated with the occurrence of noise-induced inner ear hearing loss (ONIHL). Male gender emerged as a strong independent predictor, which is consistent with previous research suggesting that males may be more susceptible to noise-induced hearing loss, possibly due to differences in occupational distribution and exposure levels between genders2. Combined exposure to noise and dust was found to have a substantial impact, with an odds ratio of 11.27. This indicates a synergistic effect of these two pollutants on the auditory system, underscoring the importance of considering their combined exposure in occupational health assessments.
Elevated diastolic blood pressure was positively associated with ONIHL, aligning with the growing body of evidence linking cardiovascular health and hearing loss. The underlying mechanism may involve vascular changes in the inner ear, as increased blood pressure could affect the delicate blood supply to the cochlea3.Smoking and elevated fasting blood glucose were also identified as independent risk factors. Smoking can cause vasoconstriction and reduce oxygen supply to the cochlea, while hyperglycemia may lead to oxidative stress and damage to the auditory nerve fibers4,5. Interestingly, systolic blood pressure was found to be a negative predictor in the multivariate analysis. Consider the possible causes, moderate increases in systolic pressure may improve cochlear blood supply, reducing inner ear damage caused by hypoxia or abnormal blood flow. The labyrinth, which supplies the inner ear, is a terminal artery, making it susceptible to the influence of overall circulation. Stable blood supply ensures its normal function and prevents ischemia-induced hearing loss. Furthermore, elevated systolic pressure often prompts individuals to pay more attention to cardiovascular health, leading to more proactive health management, which may indirectly contribute to hearing protection and reduce the risk of ONIHL. Further research is needed to clarify this relationship.
The application of machine learning (ML) algorithms in this study demonstrated their potential superiority over traditional regression methods in handling the complex relationships within our dataset, particularly when dealing with a large number of samples. The resampling validation mechanism employed was essential in providing a comprehensive assessment of the models’ performance across multiple training runs. We developed six types of ML algorithms to model our data: Logistic Regression, Random Forest Classifier, Decision Tree Classifier, XGB Classifier, AdaBoost Classifier, and Gaussian Naive Bayes (NB). After optimizing, adjusting, and training the ML models, we found that the logistic model achieved the highest predictive performance for ONIHL. The results of multivariate logistic regression were further validated by the machine learning algorithms, confirming that logistic regression was the best model and consistent with the prior results of multivariate logistic regression.
To explore the relationship between dust exposure and ONIHL in specific populations, stratified analyses were conducted across different subgroups. The results revealed that men without pre-existing conditions, aged 30–39 years, and with less than 10 years of work experience are at an elevated risk of developing ONIHL. A possible explanation is that certain subgroups may be more vulnerable to the combined effects of noise and dust, and may lack the protective effects of estrogen. Additionally, these individuals are more likely to be in occupations with high noise exposure. This suggests the need for targeted preventive strategies within these populations. For instance, in male-dominated occupations with high noise and dust exposure, stricter hearing protection measures and regular health monitoring should be implemented. Younger workers and those with less work experience may require additional training on the importance of occupational safety and the potential risks of exposure.
The AUC value of 0.714 indicates a moderate discriminatory capacity of the model, although there is room for improvement. Future research could explore more advanced machine learning techniques or incorporate additional relevant variables to enhance predictive precision. The comparison of different algorithm performances provides valuable insights, assisting researchers in selecting appropriate models based on data characteristics and research objectives.
The results of this study have significant implications for occupational health policies and practices. Employers must recognize the substantial risks associated with combined noise and dust exposure and take proactive measures to mitigate these risks. Engineering controls, the provision of personal protective equipment, and proper workplace ventilation are crucial components of these efforts. Regular health check-ups should include comprehensive assessments of hearing, cardiovascular health, and metabolic markers to identify potential health issues at an early stage. Additionally, health promotion initiatives should encourage workers to adopt healthier lifestyles, such as quitting smoking, maintaining a balanced diet, and engaging in regular physical activity, which could potentially counteract the harmful effects of occupational exposures.
