The association between living environmental factors and hearing loss in Chinese middle-aged and older adults: Results from a cross-sectional and longitudinal study Authors: Wanyi Huang, Songjian Wang, Yi Liu, Yuan Wang, Liting Zheng, Younuo Chen, Tong Liu, Junyi Liu, Shuo Wang Published: 2026 ABSTRACT AbstractBackgroundHearing loss (HL) substantially impairs quality of life and functional independence among middle-aged and older adults. Although individual environmental exposures have been associated with hearing health, the relationship between overall living environment quality and HL remains insufficiently understood. This study aimed to examine the association between a composite living environment score and HL in a nationally representative sample of Chinese adults.MethodsData were obtained from the China Health and Retirement Longitudinal Study (CHARLS). Cross-sectional analyses included 4,404 adults aged ≥ 45 years, and longitudinal analyses followed 3,603 participants without HL at baseline for up to 7 years. A composite living environment score (range: 0–6) was constructed based on ambient PM₂.₅ exposure, household fuel type, drinking water source, building type, and indoor temperature. Cross-sectional associations were assessed using logistic regression models, while longitudinal associations were assessed using Cox proportional hazards models.ResultsIn cross-sectional analyses, higher living environment scores were inversely associated with prevalent HL (all P < 0.05). During longitudinal follow-up, 645 incident HL cases were identified. Each one-point increase in the living environment score was associated with a 5.7% lower risk of incident HL (fully adjusted hazard ratio [HR] = 0.943, 95% confidence interval [CI]: 0.889–0.999). Compared with participants in high-risk environments, those in low-risk environments showed a lower risk of HL (HR = 0.801, 95% CI: 0.638–1.005), with evidence of a dose–response relationship (P for trend = 0.038). Restricted cubic spline analyses indicated a linear association (P for nonlinearity > 0.05).ConclusionBetter overall living environment quality was associated with a lower risk of hearing loss among middle-aged and older adults, suggesting that improvements in living conditions may contribute to healthier auditory aging.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28279-z.Keywords: Hearing loss, Living environment score, Middle-aged and older adults, Epidemiological studies Introduction Hearing loss is a common chronic condition and a major contributor to disability among older adults worldwide. According to estimates from the World Health Organization, approximately 430 million individuals live with disabling hearing loss, with prevalence increasing markedly with age [1]. In China, rapid population aging has been accompanied by a growing burden of hearing loss, posing an important public health concern [2]. This condition is associated with substantial functional, psychological, and social consequences, underscoring the need to identify modifiable risk factors to support healthy aging. In the context of rapid urbanization and population aging, the living environment has increasingly been recognized as an important determinant of chronic health outcomes. Several environmental exposures have been linked to auditory dysfunction. Ambient air pollution, particularly fine particulate matter (PM₂.₅), has been hypothesized to impair cochlear function through oxidative stress and microvascular injury [3]. Household air pollution resulting from solid fuel combustion—still widely used in many regions—has also been identified as a modifiable risk factor for hearing loss [4, 5]. In addition, emerging evidence suggests that drinking water quality [6, 7] and characteristics of the built environment [8] may influence auditory health. However, existing studies have largely focused on individual exposures in isolation, limiting their ability to capture the combined effects of real-world environmental conditions. This single-exposure approach largely overlooks the interrelatedness of environmental factors and their potential synergistic effects. A composite measure that aggregates multiple exposures into a holistic living environment score offers a more ecologically valid framework for assessing overall health impact—a methodology successfully employed in cardiovascular epidemiology [9]. Leveraging data from the China Health and Retirement Longitudinal Study (CHARLS), this study therefore aims to investigate the association between a comprehensive living environment score—encompassing ambient PM₂.₅, household fuel type, drinking water source, housing type, and indoor temperature—and incident hearing loss in a national cohort of middle-aged and older adults. Methods Study population The China Health and Retirement Longitudinal Study (CHARLS) is a nationally representative survey of middle-aged and older adults in China. The baseline survey was conducted in 2011–2012, with follow-up visits every two years. During each wave, physical measurements were collected, and blood samples were obtained every two survey cycles. To ensure national representativeness, the baseline survey covered 450 villages or communities within 150 counties across China, successfully interviewing 17,708 individuals from 10,257 households, representative of the middle-aged and older population in China. The CHARLS study protocol was approved by the Ethical Review Committee at Peking University, and written informed consent was obtained from all participants. This study analyzed CHARLS data from 2011 to 2018. Environmental exposure variables were assessed at baseline, while hearing loss outcomes were followed up through 2018. The 2018 wave was selected as the endpoint because it was the most recent survey cycle with the most complete data on hearing loss (HL) outcomes available at the time of analysis. A combination of cross-sectional and longitudinal analyses was employed to examine the association between living environmental factors and HL in middle-aged and older adults. The inclusion and exclusion criteria for the study are detailed in Fig. 1. Initially, 777 participants with missing age data or aged below 45 years were excluded. Next, 141 individuals without hearing loss information and 10,559 participants with missing environmental data were removed. Thereafter, 1,827 participants with incomplete covariate data were excluded. After applying these criteria, 4,404 participants were included in the cross-sectional analysis. For the longitudinal analysis, an additional 548 participants with prevalent HL at baseline and 253 with no follow-up data were excluded, resulting in a final analytical sample of 3,603 participants. Visual flowchart for population screening Assessment of living environmental factors Living environmental factors were assessed at baseline using a structured questionnaire capturing five key domains: ambient fine particulate matter (PM₂.₅) levels, household energy type, water source, building type, and indoor temperature. Annual average PM₂.₅ concentrations for the baseline year 2011 were obtained at the city level from the National Aeronautics and Space Administration (NASA) Earth Observing System Distributed Information System. Specifically, satellite-derived aerosol optical depth data were integrated with the Goddard Earth Observing System chemical transport model and a geographically weighted regression model to estimate ground-level PM₂.₅ concentrations [10]. City-level PM₂.₅ concentrations were then calculated by averaging all grid cells falling within each city’s administrative boundary, using official Chinese city shapefiles. Individual participants were assigned the city-level mean PM₂.₅ value according to their self-reported city of residence at baseline. Based on the Ambient Air Quality Standards of China, an annual PM₂.₅ concentration of 35 µg/m³ was used as the threshold to define high exposure [9, 11]. Previous epidemiological studies have demonstrated that long-term exposure to elevated PM₂.₅ concentrations is associated with an increased risk of hearing impairment and other adverse health outcomes [3, 12, 13]. Accordingly, participants with PM₂.₅ concentrations below 35 µg/m³ were assigned a score of 1, whereas those exposed to concentrations of 35 µg/m³ or higher were assigned a score of 0. Household energy type was collected at the individual level via the baseline questionnaire. It was categorized as clean or solid fuels based on sources used for cooking and heating. Clean fuels included natural gas, biogas, liquefied petroleum gas (LPG), or electricity for cooking; and natural gas, LPG, solar energy, electricity, or municipal heating for heating. Solid fuels included coal, crop residues, wood, or charcoal for either purpose. Household energy was scored 0–2 to reflect graded exposure from cooking and heating: 2 for clean fuels for both purposes, 1 for clean fuels for one purpose only, and 0 for solid fuels for both. This three-level classification captures the major sources of indoor air pollution and the gradient of household air pollution exposure. Water source was assessed by the question, “Does your residence have tap water?” A “yes” response scored 1, and “no” scored 0. Building type was determined by asking whether the residence was a single-story or multi-story structure, with multi-story buildings assigned a score of 1 and single-story a score of 0. Indoor temperature was evaluated by the interviewer’s assessment of “How is the temperature in this household?” A rating of “comfortable” scored 1; otherwise, the score was 0. Scores for ambient PM₂.₅, household energy, water source, building type, and indoor temperature were summed into a composite living environment score ranging from 0 to 6. Participants were subsequently classified into low-risk (≥ 4 points), intermediate-risk (3 points), and high-risk (0–2 points) groups. A higher composite score indicated better overall living environment quality. Scores for ambient PM₂.₅, household energy, water source, building type, and indoor temperature were summed into a composite living environment score using the formula: Composite score = Score_PM₂.₅ + Score_energy + Score_water + Score_building + Score_temperature. Each domain contributed equally, yielding a total range of 0 to 6. Participants were subsequently classified into low-risk (≥ 4 points), intermediate-risk (3 points), and high-risk (0–2 points) groups. A higher composite score indicated better overall living environment quality. This additive scoring approach has been used in previous studies to create multi-domain environmental indices [9, 14]. Assessment of hearing loss and covariates Hearing loss was assessed via self-reported questionnaires at baseline (2011) and subsequent follow-up surveys (2013, 2015 and 2018). Assessment was based on three baseline questions: (1) “Are you deaf or semi-deaf?” (2) “Have you ever used a hearing aid?” (3) “How would you rate your hearing: excellent, very good, good, fair, or poor?” Participants were classified as having hearing loss if they met any of the following criteria: (1) being deaf or semi-deaf, (2) having hearing aid use experience, or (3) self-rating their hearing as “poor.” Those not meeting any criterion were classified as having no hearing loss. This classification approach for HL has been widely adopted in previous relevant studies [15–17]. For the cross-sectional analysis, baseline hearing loss status was used. For the longitudinal analysis, participants with prevalent hearing loss at baseline were excluded, and incident hearing loss was identified based on follow-up assessments. All covariates were assessed at baseline. Covariates were categorized into three domains. Sociodemographic variables included age, gender, education level (junior high school and above vs. primary school and below), region (Central, East, West), residence location, and marital status (married vs. other). Lifestyle behaviors included smoking status and alcohol use. Health-related factors included hypertension, diabetes, vision impairment, pain, depression, and cognitive impairment. Depressive symptoms were assessed using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10), with a total score ≥ 10 indicating clinically significant depressive symptoms [18]. Cognitive function was evaluated as the sum of episodic memory (0–10 points) and executive function (0–11 points), with higher scores indicating better cognitive performance [19]. A total score < 6 defined cognitive impairment based on established criteria [20]. Statistical analysis Continuous variables are presented as mean ± standard deviation (SD) and categorical variables as frequency (percentage). Group differences in baseline characteristics were examined using independent samples t-tests for continuous variables and chi-square tests for categorical variables. The association between living environment risk scores and hearing loss was investigated using logistic regression models for cross-sectional analyses and Cox proportional hazards regression models for prospective analyses. Model adjustments included age, gender, region, residence, marital status, smoking, drinking, education level, hypertension, diabetes, visual impairment, pain, depression, and cognitive impairment. Subgroup and interaction analyses were performed for age, gender, region, residence, marital status, smoking, drinking, education level, hypertension, diabetes, visual impairment, pain, depression, and cognitive impairment to assess the robustness of associations. Mediation analysis was conducted to examine potential mediating effects of age, gender, region, residence, marital status, smoking, drinking, education level, hypertension, diabetes, visual impairment, pain, depression, and cognitive impairment in the relationship between living environment factors and hearing loss. A regression-based approach was used to estimate the indirect effect, with the proportion mediated calculated accordingly. All statistical analyses were performed using R version 4.4.3, with two-sided p-values < 0.05 considered statistically significant. Results Baseline characteristics of study participants Table 1 presents the baseline characteristics of the study population. Overall, 4,404 participants were included, with a mean age of 57.5 years (SD: 8.93). Among them, 548 (12.4%) had prevalent hearing loss (HL) at baseline. In the longitudinal study involving 3,603 participants, 645 incident HL cases occurred during follow-up, corresponding to an incidence of 17.9%. Compared to participants in the high-risk living environment group, those in the low-risk group were more likely to reside in the Eastern region and urban areas, have higher education levels, and a lower prevalence of smoking. They also exhibited significantly lower rates of vision impairment, pain, depressive symptoms, cognitive impairment, and hearing loss (all P < 0.05). As shown in Table S1, participants with HL were significantly more likely to use solid fuels for multiple purposes, lack tap water access, and live in single-story buildings compared to those without HL (all P < 0.001). Table 1Baseline characteristics of the study populationTotalLiving environmental score P High riskMiddle riskLow riskN = 4404N = 1339N = 1852N = 1213Age57.5 (8.93)57.9 (9.11)57.3 (8.84)57.3 (8.86)0.165Gender0.738 Female2157 (49.0%)644 (48.1%)913 (49.3%)600 (49.5%) Male2247 (51.0%)695 (51.9%)939 (50.7%)613 (50.5%)Region< 0.001 Central1947 (44.2%)685 (51.2%)804 (43.4%)458 (37.8%) East1443 (32.8%)248 (18.5%)703 (38.0%)492 (40.6%) West1014 (23.0%)406 (30.3%)345 (18.6%)263 (21.7%)Residence< 0.001 Rural2918 (66.3%)1043 (77.9%)1319 (71.2%)556 (45.8%) Urban1486 (33.7%)296 (22.1%)533 (28.8%)657 (54.2%)Marital status0.153 Married3757 (85.3%)1136 (84.8%)1566 (84.6%)1055 (87.0%) Other647 (14.7%)203 (15.2%)286 (15.4%)158 (13.0%)Smoking< 0.001 Yes1785 (40.5%)585 (43.7%)771 (41.6%)429 (35.4%) No2619 (59.5%)754 (56.3%)1081 (58.4%)784 (64.6%)Drinking0.110 Yes1785 (40.5%)512 (38.2%)763 (41.2%)510 (42.0%) No2619 (59.5%)827 (61.8%)1089 (58.8%)703 (58.0%)Education< 0.001 Junior high school and above2004 (45.5%)553 (41.3%)792 (42.8%)659 (54.3%) Primary school and below2400 (54.5%)786 (58.7%)1060 (57.2%)554 (45.7%)Hypertension0.217 Yes1179 (26.8%)379 (28.3%)473 (25.5%)327 (27.0%) No3225 (73.2%)960 (71.7%)1379 (74.5%)886 (73.0%)Diabetes0.782 Yes320 (7.27%)101 (7.54%)136 (7.34%)83 (6.84%) No4084 (92.7%)1238 (92.5%)1716 (92.7%)1130 (93.2%)Visual impairment0.001 Yes1278 (29.0%)418 (31.2%)558 (30.1%)302 (24.9%) No3126 (71.0%)921 (68.8%)1294 (69.9%)911 (75.1%)Pain< 0.001 Yes1177 (26.7%)431 (32.2%)483 (26.1%)263 (21.7%) No3227 (73.3%)908 (67.8%)1369 (73.9%)950 (78.3%)Depression< 0.001 Yes1318 (29.9%)489 (36.5%)555 (30.0%)274 (22.6%) No3086 (70.1%)850 (63.5%)1297 (70.0%)939 (77.4%)Cognitive impairment< 0.001 Yes179 (4.06%)76 (5.68%)71 (3.83%)32 (2.64%) No4225 (95.9%)1263 (94.3%)1781 (96.2%)1181 (97.4%)Hearing loss< 0.001 Yes548 (12.4%)217 (16.2%)212 (11.4%)119 (9.81%) No3856 (87.6%)1122 (83.8%)1640 (88.6%)1094 (90.2%) Baseline characteristics of the study population Associations between living environmental factors and hearing loss in the cross-sectional study The associations between living environmental factors and HL in the cross-sectional analysis are shown in Table 2. In the model treating the living environment score as a continuous variable, each unit increase in the living environment score was associated with an unadjusted odds ratio (OR) of 0.827 (95% CI: 0.777, 0.881) for HL. After full adjustment for all covariates, the OR was 0.910 (95% CI: 0.849, 0.976). When the score was analyzed categorically, a significant monotonic decreasing trend in HL risk was observed across increasing living environment quality levels (all P for trend < 0.05). Specifically, compared to the high-risk group, the low- and intermediate-risk groups showed significant inverse associations with HL, with unadjusted ORs of 0.562 (95% CI: 0.443, 0.714) and 0.668 (95% CI: 0.545, 0.820), respectively. These inverse associations persisted after full adjustment, with adjusted ORs of 0.780 (95% CI: 0.599, 1.016) for the low-risk group and 0.781 (95% CI: 0.628, 0.970) for the intermediate-risk group. Table 2Cross-sectional association between living environment score and hearing losscontinuous variableLow riskMiddle riskHigh riskP for trendCrude model0.827(0.777,0.881)*0.562(0.443,0.714)*0.668(0.545,0.820)*Reference< 0.001Model 10.838(0.785,0.893)*0.573(0.449,0.730)*0.683(0.555,0.841)*Reference< 0.001Model 20.880(0.822,0.942)*0.698(0.538,0.906)*0.738(0.596,0.914)*Reference0.004Model 30.910(0.849,0.976)*0.780(0.599,1.016)*0.781(0.628,0.970)*Reference0.041Crude model: without any adjustmentModel 1 adjust for: age; genderModel 2 adjust for: age; gender; region; residence; marital status; smoking; drinking; educationModel 3 adjust for: age; gender; region; residence; marital status; smoking; drinking; education; hypertension; diabetes; visual impairment; pain; depression; cognitive impairment* indicates p < 0.05 Cross-sectional association between living environment score and hearing loss Crude model: without any adjustment Model 1 adjust for: age; gender Model 2 adjust for: age; gender; region; residence; marital status; smoking; drinking; education Model 3 adjust for: age; gender; region; residence; marital status; smoking; drinking; education; hypertension; diabetes; visual impairment; pain; depression; cognitive impairment * indicates p < 0.05 Associations between living environmental factors and hearing loss in the longitudinal study The Kaplan-Meier curves showed the cumulative incidence of HL in the overall study population (Figure S1). Among the 3,603 participants included in the longitudinal analysis, 645 incident HL cases occurred during follow-up, yielding an incidence rate of 17.9%. The HL incidence varied by living environment category: 22.13% in the high-risk, 18.06% in the intermediate-risk, and 13.07% in the low-risk group (Table 3). As summarized in Table 3, the longitudinal findings were consistent with the cross-sectional results. When analyzed as a continuous variable, each unit increase in the living environment score was associated with an unadjusted hazard ratio (HR) of 0.865 (95% CI: 0.819, 0.914) for incident HL. After full adjustment, the HR was 0.943 (95% CI: 0.889, 0.999). Categorical analysis also revealed a significant monotonic decrease in HL risk with improving living environment quality in both unadjusted and fully adjusted models (all P for trend < 0.05), indicating that participants in low- and intermediate-risk environments had a lower risk of developing HL compared to those in high-risk environments. Furthermore, restricted cubic spline analysis confirmed a linear inverse association between the living environment score and HL risk (Fig. 2; all P for nonlinearity > 0.05). Table 3Longitudinal association between living environment and HL in different modelsLiving environment scoreTotalIncidence of HL, n%Crude modelModel 1Model 2Model 3Continuous variable3603645 (17.90%)0.865(0.819,0.914)*0.870(0.824,0.919)*0.927(0.875,0.982)*0.943(0.889,0.999)*High risk1071237 (22.13%)ReferenceReferenceReferenceReferenceMiddle risk1545279 (18.06%)0.802(0.674,0.953)*0.815(0.685,0.969)0.909(0.762,1.085)0.940(0.787,1.122)Low risk987129 (13.07%)0.584(0.472,0.724)*0.589(0.476,0.730)*0.756(0.602,0.948)*0.801(0.638,1.005)*P for trend< 0.001< 0.0010.0170.038Outcome: HL, HR(95%CI)Crude model: without any adjustmentModel 1 adjust for: age; genderModel 2 adjust for: age; gender; region; residence; marital status; smoking; drinking; educationModel 3 adjust for: age; gender; region; residence; marital status; smoking; drinking; education; hypertension; diabetes; visual impairment; pain; depression; cognitive impairment* indicates p < 0.05 Longitudinal association between living environment and HL in different models Outcome: HL, HR(95%CI) Crude model: without any adjustment Model 1 adjust for: age; gender Model 2 adjust for: age; gender; region; residence; marital status; smoking; drinking; education Model 3 adjust for: age; gender; region; residence; marital status; smoking; drinking; education; hypertension; diabetes; visual impairment; pain; depression; cognitive impairment * indicates p < 0.05 Restricted cubic spline curves for the association between the living environment score and hearing loss. A without any adjustment. B adjusted for: age; gender. C adjusted for: age; gender; region; residence; marital status; smoking; drinking; education. D adjusted for: age; gender; region; residence; marital status; smoking; drinking; education; hypertension; diabetes; visual impairment; pain; depression; cognitive impairment Subgroup and mediation analyses Stratified analyses by age, gender, region, residence location, marital status, smoking, alcohol use, education, and health conditions are presented in Table S2. Results were largely consistent with the primary analysis across most subgroups. A significant interaction was observed only for residence location (P for interaction = 0.013), wherein the association between living environment quality and HL was statistically significant only among urban residents. Mediation analysis examining the potential mediating role of all covariates in the association between living environment factors and HL identified depressive symptoms as a significant mediator. In the fully adjusted model, the proportion mediated by depressive symptoms was 7.37% (Figure S2; P < 0.001), suggesting that depression may partially explain the pathway from living environment factors to HL. Discussion In this large national cohort of middle-aged and older adults in China, we found that higher overall living environment quality, as assessed by a composite index incorporating air quality, household fuel use, water source, building type, and indoor temperature, was associated with a lower risk of hearing loss. This association was observed in both cross-sectional and longitudinal analyses and demonstrated a dose–response pattern. Importantly, the relationship remained consistent after adjustment for a wide range of sociodemographic characteristics, lifestyle factors, and health conditions, suggesting that living environment quality may play a role in auditory health among aging populations. This study extends existing evidence linking specific environmental exposures, such as ambient PM₂.₅ [3] and household solid fuel use [4, 21] to hearing impairment by adopting an integrated environmental perspective. Previous research has primarily examined environmental risk factors individually, which may not adequately reflect the complex and cumulative nature of real-world exposures [9]. Although studies directly examining broad residential environmental composite indices in relation to hearing loss remain limited, emerging evidence has begun to apply summary environmental indicators to hearing outcomes. For example, Wu et al. examined the association of the Air Quality Index and individual air pollutants with sudden sensorineural hearing loss, providing evidence that composite air pollution indicators may be relevant to hearing-related outcomes [22]. In an occupational cohort, Li et al. further used an Air Quality Comprehensive Index and found that air pollution indicators, particularly when combined with noise exposure, were associated with increased occupational hearing loss risk [23]. Compared with these studies, which focused mainly on air pollution indices or occupational environments, our study extends the composite-exposure approach to a broader residential living environment context among middle-aged and older adults. By constructing a composite living environment score that incorporates multiple indoor and outdoor environmental domains, our findings suggest that a more holistic assessment of environmental conditions may better characterize overall environmental risk related to hearing loss. This integrated approach provides a complementary framework for understanding environmental influences on auditory aging. Several biological and psychosocial pathways may help explain the observed association between living environment quality and hearing loss. Exposure to ambient PM₂.₅ and household air pollutants from solid fuel combustion has been linked to systemic oxidative stress and chronic inflammation, which may contribute to cochlear hair cell damage and microvascular dysfunction [24–26]. In addition, characteristics of the built environment, such as building structure and sound insulation, may influence long-term exposure to environmental noise, a well-established risk factor for hearing impairment [27, 28]. Indoor thermal discomfort may also act as a chronic physiological stressor, potentially affecting auditory function through stress-related neuroendocrine pathways [27–30]. Furthermore, our mediation analysis suggests that depressive symptoms may partially mediate the relationship between living environment quality and hearing loss. Adverse living conditions may increase psychological distress, which in turn has been associated with inflammatory processes and neurobiological changes relevant to auditory health [31–33]. Beyond these specific biological and psychosocial mechanisms, our findings may also be interpreted from a broader life-course and functional health perspective. The WHO healthy ageing framework emphasizes that functional ability in older age is shaped by intrinsic capacity, environmental characteristics, and their interaction [34]. Hearing is an important sensory component of intrinsic capacity, and auditory decline may therefore reflect not only isolated sensory impairment but also the cumulative influence of environmental, psychosocial, and functional vulnerabilities across the life course [11, 34]. Consistent with this perspective, Yang et al. recently reported in CHARLS that functional dependence was associated with cardiovascular disease among middle-aged and older adults, illustrating how aggregated functional deficits can contribute to ageing-related chronic outcomes [35]. Although cardiovascular disease and hearing loss are distinct outcomes, this framework provides a useful parallel: cumulative disadvantage and multidimensional exposures may jointly shape ageing-related health trajectories. In this context, the composite living environment score used in our study may capture broader environmental vulnerability rather than a single exposure pathway, thereby offering a more holistic perspective on auditory ageing. Although these pathways remain speculative and cannot be confirmed in an observational study, they provide plausible explanations for the associations observed in our analyses. The subgroup analysis showed a statistically significant interaction by residence location (P for interaction = 0.013), with the protective association between better living environment quality and hearing loss observed only among urban residents. This finding is important and warrants further consideration. One possible explanation is that the dominant environmental exposures captured by the composite living environment score may differ between urban and rural settings. In urban areas, variations in the composite score may more strongly reflect differences in ambient environmental exposures, particularly outdoor PM₂.₅ and traffic- or industry-related pollution [36]. By contrast, in rural areas, household solid fuel use for cooking and heating remains relatively common and may contribute substantially to indoor air pollution exposure [37]. Because indoor household air pollution, occupational exposures, and lifestyle-related factors may not be fully captured by an area-level composite living environment score, the exposure gradient represented by the score may be less distinct among rural residents. Another plausible explanation is that urban–rural differences in baseline socioeconomic and health profiles may have modified the association between living environment quality and hearing loss. In our sample, urban residents had higher educational attainment and lower prevalence of comorbid conditions, which may influence both environmental exposure patterns and susceptibility to hearing impairment. Previous studies in China have also reported persistent urban–rural disparities in socioeconomic resources and health status among middle-aged and older adults [38]. Therefore, the observed urban–rural heterogeneity may reflect not only differences in environmental exposure composition, but also differences in vulnerability, healthcare access, occupational histories, and cumulative life-course socioeconomic conditions. The findings of this study may have implications for both clinical practice and public health. From a geriatric care perspective, consideration of living environment characteristics may complement existing risk assessments for hearing loss, particularly among older adults with multiple vulnerabilities. Incorporating brief, structured inquiries regarding household fuel use, indoor conditions, and access to basic infrastructure into routine assessments could help identify individuals who may benefit from closer auditory monitoring [39]. At the population level, our results support broader public health efforts aimed at improving residential environmental conditions, including the promotion of clean household energy, improved water infrastructure, and healthy housing standards. Such environmental interventions, although not specific to hearing health, have been widely recognized as important components of strategies to promote healthy aging and reduce chronic disease burden [40, 41]. Importantly, these implications should be interpreted cautiously given the observational nature of the study. This study has several strengths. The use of a large, nationally representative cohort with both cross-sectional and longitudinal data enhances the robustness of the findings. The construction of a composite living environment score allowed for a more comprehensive assessment of environmental conditions than single-exposure approaches, better reflecting real-world exposure patterns. In addition, the consistency of results across analytic methods and subgroups lends support to the stability of the observed associations. Nevertheless, several limitations should be acknowledged. First, hearing loss was assessed using self-reported measures rather than objective audiometric testing, which may have introduced misclassification. Although self-reported hearing status has been validated in epidemiological studies [15–17], it remains less precise than objective assessments of hearing status. In addition, the accuracy of self-reported hearing loss may have varied across participants. Participants with greater health literacy or environmental awareness may have been more likely to notice, report, or attribute hearing difficulties, whereas those with lower awareness may have underreported hearing problems. Such differential reporting could have biased the observed associations, although the direction and magnitude of this bias are difficult to determine [42]. Future studies using standardized audiometric assessments are needed to confirm our findings. Second, and importantly, we lacked information on lifetime occupational noise exposure, a recognised and well-documented risk factor for hearing loss [1, 43]. Occupational noise may also confound associations between environmental exposures and hearing outcomes, because socioeconomic disadvantage can be associated with both higher exposure to adverse residential environments and greater likelihood of employment in noisy occupations [44, 45]. According to the WHO World Report on Hearing (2021) [1], occupational noise exposure is among the leading preventable causes of hearing loss globally. Although we adjusted for education and region as proxies for socioeconomic position and occupational type, these variables cannot fully capture the complex relationship between occupational exposures and hearing health. The observed urban-rural interaction may partly reflect differential occupational noise exposure patterns, as rural and urban workers in China typically engage in different economic sectors with distinct noise exposure profiles. Future studies should consider incorporating detailed occupational histories and personal noise dosimetry measurements to better characterise lifetime occupational noise exposure and further strengthen the evidence base. Third, residual confounding by other unmeasured factors—such as early-life socioeconomic conditions, dietary patterns, ototoxic medication use, recreational noise exposure, and genetic susceptibility—cannot be excluded. Fourth, environmental exposures were assessed only at baseline, which limited our ability to account for temporal changes in exposure levels during follow-up and may have introduced non-differential exposure misclassification. Moreover, constructing the environmental exposure score using an equal-weighting approach may not fully reflect the varying relative importance of different environmental domains. Fifth, the composite living environment score was developed based on data available in CHARLS, and its applicability to other populations or settings requires further validation. Finally, the relatively short follow-up duration of 7 years limited our ability to examine longer-term hearing trajectories. Given that hearing loss is a slowly progressive condition, more extended observation periods may be needed to fully characterise the contribution of environmental exposures to auditory ageing. Conclusion In conclusion, this study found that better overall living environment quality was associated with a lower risk of hearing loss among middle-aged and older adults, with evidence of a dose–response relationship. By applying a composite environmental assessment in both cross-sectional and longitudinal analyses, our findings suggest that multiple aspects of the living environment may jointly influence auditory health. These results highlight the potential relevance of environmental conditions in the context of healthy aging, while emphasizing the need for future studies using objective hearing assessments and refined exposure measures to further clarify underlying mechanisms. Supplementary Information Supplementary Material 1. Fig. S1 Kaplan-Meier analysis of cumulative hearing loss incidence by living environmental score groups. Fig.S2 Mediation effect analysis of covariates in the relationship between living environmental factors and hearing loss, adjusted for age, gender, region, residence, marital status, smoking, drinking, education, hypertension, diabetes, visual impairment, pain, depression and cognitive impairment. Supplementary Material 2. Table S1 Baseline distribution of living environment risk factors in the study population. Table S2 Subgroup analyses of the association between living environment factors and hearing loss in the cohort. Acknowledgements We sincerely thank those who participated in the data collection and management of the China Health and Retirement Longitudinal Study. Authors’ contributions Huang Wanyi: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing - original draft, Visualization. Wang Songjian: Methodology, Software, Validation, Formal analysis. Yi Liu, Yuan Wang, Liting Zheng, Younuo Chen, Tong Liu, and Junyi Liu: Data extraction. Wang Shuo: Conceptualization, Resources, Writing - review & editing, Supervision, Project administration, Funding acquisition. All authors have read and approved the final version of the manuscript. Funding The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was funded in part by grants from the National Key Research and Development Program of China (2023YFC3604200-project 2023YFC3604203). Data availability Data supporting the results of this study are available from official websites of ChinaHealth and Retirement Longitudinal Study (http://charls.pku.edu.cn). Declarations Ethics approval and consent to participate This study was conducted using data from the China Health and Retirement Longitudinal Study (CHARLS), which was approved by the Ethical Review Committee of Peking University. All procedures involving human participants were performed in accordance with the ethical standards of the institutional research committee and with the Declaration of Helsinki. Written informed consent was obtained from all participants at the time of enrollment in the CHARLS study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Wanyi Huang and Songjian Wang contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1. Fig. S1 Kaplan-Meier analysis of cumulative hearing loss incidence by living environmental score groups. Fig.S2 Mediation effect analysis of covariates in the relationship between living environmental factors and hearing loss, adjusted for age, gender, region, residence, marital status, smoking, drinking, education, hypertension, diabetes, visual impairment, pain, depression and cognitive impairment. Supplementary Material 2. Table S1 Baseline distribution of living environment risk factors in the study population. Table S2 Subgroup analyses of the association between living environment factors and hearing loss in the cohort. Data Availability Statement Data supporting the results of this study are available from official websites of ChinaHealth and Retirement Longitudinal Study (http://charls.pku.edu.cn).