Residential greenness, air pollution and visual impairment: a prospective cohort study Authors: Jia-min Yan, Min-zhe Zhang, Hong-jie Yu, Qi-qiang He Published: 2024 ABSTRACT ObjectiveThe aim of this study was to investigate the associations between residential greenness, air pollution (PM2.5, PM10, and NO2), and the risk of visual impairment in Chinese middle-aged and elderly adults.MethodsThe data were used from the China Health and Retirement Longitudinal Study (CHARLS), focusing on adults aged 45 years and older. Exposure to residential greenness was assessed using the Normalized Difference Vegetation Index (NDVI), while air pollution exposure (PM2.5, PM10, and NO2) was estimated from the ChinaHighAirPollutants dataset. Visual impairment was self-reported and categorized as distance visual impairment (DVI) and near visual impairment (NVI). Cox proportional hazards models were applied to investigate the associations between residential greenness, air pollution, and visual impairment. Furthermore, we investigated the potential modification and mediation relationship between residential greenness, air pollution and visual impairment.ResultsA total of 9,591 participants were included in this study. During a 7-year period, 2,650 cases of DVI and 3,130 cases of NVI were identified. In the single exposure model, a higher level of residential greenness was associated with a significantly reduced risk of both DVI and NVI. Specifically, for every 0.1 unit increase in NDVI, the hazard ratio (HR) was 0.68 (95%CI: 0.65,0.71) for DVI, and 0.67 (95%CI: 0.64,0.69) for NVI. Conversely, exposure to air pollution was positively associated with increased risks for both DVI and NVI. For each 10 µg/m3 increase in PM2.5, the HR was 1.60 (95%CI: 1.56,1.65) for DVI and 1.64 (95%CI: 1.60,1.69) for NVI. Similarly, PM10 and NO2 exposure showed elevated risks with HR of 1.28 (95%CI: 1.27,1.30) and 1.73 (95%CI: 1.62,1.85) for DVI, and 1.30 (95%CI: 1.28,1.32) and 1.82 (95%CI: 1.72,1.94) for NVI, respectively. The multiple exposure model also showed significant association between residential greenness, air pollution and visual impairment risk. Additionally, the adverse effects of PM2.5, PM10, and NO2 were attenuated in areas with the highest levels of residential greenness. Mediation analysis revealed that air pollution mediated a significant portion of the association between residential greenness and visual impairment.ConclusionThis study suggests that increased residential greenness may reduce the risk of visual impairment, partially by mitigating air pollution exposure. These findings highlight the environmental benefits to improve air quality and enhance green spaces, thereby promoting visual health in aging populations.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12889-024-20853-7. Introduction Visual impairment, including distance visual impairment (DVI) and near visual impairment (NVI), is a significant health problem worldwide. According to the World Health Organization (WHO), at least 2.2 billion people worldwide suffer from DVI or NVI, which represents a huge financial burden with an estimated annual global productivity cost of 411 billion US dollars [1]. In China, the estimated number of people with visual impairment is almost 60 million [2]. Visual impairment significantly affects the quality of life of those affected and has far-reaching consequences that go beyond the loss of sight. Visual impairment can lead to social isolation and higher rates of depression and anxiety in adults, as well as an increased risk of accidents, injuries, and the need for long-term care in older adults [3, 4]. The etiology of visual impairment is multifactorial, involving genetic predispositions, aging, and environmental factors. In recent years, environmental factors, particularly exposure to air pollution and access to green spaces, have gained increasing attention. Air pollution, primarily characterized by the presence of particulate matter (PM2.5, PM10), nitrogen dioxide (NO2), and ozone (O3), is known to have deleterious effects on human health, especially the respiratory and cardiovascular systems [5, 6]. Emerging evidence suggests that air pollution may also be linked to the development and progression of visual impairment [7]. In contrast, residential greenness could have a positive impact on health outcomes by improving mental health, reducing stress, and enhancing physical activity [8]. However, there are limited studies on the relationship between green spaces and visual health, and most focused on children and adolescents with cross-sectional in design [9–11]. It remains unclear regarding the longitudinal effects of air pollution, green space on visual function, particularly in adults. Thus, we investigated the association between residential greenness, air pollution and visual impairment among Chinese middle-aged and elderly people in a prospective cohort study. We also examined the mediation and interaction effects of air pollutants on the association between green space exposure and visual impairment. Method Study population The data utilized in this study were derived from the China Health and Retirement Longitudinal Study (CHARLS), a comprehensive, nationally representative survey focusing on the aging population in China. CHARLS was initiated in 2011 by the National School of Development at Peking University, aiming to evaluate the health, economic status, and social well-being of individuals aged 45 and older [12]. The baseline survey recruited over 17,000 participants from 150 counties across China, and the follow-up surveys were conducted in 2013, 2015, 2018, and 2020, respectively. CHARLS received ethical approval from the Biomedical Ethics Committee of Peking University (IRB00001052–11015), and informed consent was obtained from all participants. The present study included participants who were 45 years or older and had no DVI/NVI at baseline (2011), and participated at least one of the subsequent follow-up surveys (2013, 2015 and 2018). Participants with missing data on exposure measurements were excluded from the study. Finally, a total of 9,591 participants were included in the data analysis (Figure S1). Assessment of exposure Residential greenness was assessed using the Normalized Difference Vegetation Index (NDVI), a satellite-derived measure of vegetation density. NDVI values range from − 1 to 1, with higher values indicating greater vegetation density. For each participant, we calculated the average NDVI within a 250-meter buffer around their residential address. The NDVI data were obtained from global MODIS product MOD13Q1 version 5, a 16-day composite with a high spatial resolution (250 m × 250 m), using annual average NDVI estimates during the study period to minimize seasonal variations. We calculated the annual average NDVI for the year over 1 year prior to the occurrence of visual impairment, loss to follow-up, or the end of the study period, whichever came first. The exposure to air pollution including particulate matter (PM2.5 and PM10) and nitrogen dioxide (NO2) was obtained from the ChinaHighAirPollutants (CHAP) dataset [13–17]. The CHAP dataset provides long-term, full-coverage, high-resolution estimates of ground-level air pollutants across China, leveraging artificial intelligence to account for the spatiotemporal heterogeneity of air pollution. The cross-validation coefficient of determination (CV-R2) for PM2.5, PM10 and NO2 were 0.92, 0.90, and 0.80, respectively, demonstrating the high accuracy of these estimates. Similarly, we calculated participants’ long-term air pollution exposure using annual average values, applying the same method used for NDVI. Assessment of visual impairment Visual impairment was categorized into DVI and NVI. Distance vision was assessed by the question- “How good is your eyesight for seeing things at a distance, like recognizing a friend from across the street (with glasses or corrective lenses if you wear them)?” Near vision was evaluated by the question- “How good is your eyesight for seeing things up close, like reading ordinary newspaper print (with glasses or corrective lenses if you wear them)?”. The answers were “excellent, very good, good, fair, or poor”. A response of “poor” was defined as having either DVI or NVI. Covariates Data on covariates included sex, age (< 65 years and ≥ 65 years), body mass index (BMI, calculated as weight in kilograms divided by height in meters squared), education level (illiterate, primary school and middle school or above), marital status (married and single or other), residence (rural and urban), smoking status (smoking and non- smoking), alcohol consumption (> 1 times per month, < 1 times per month and non-consumer), and household income (low and high). Additionally, a history of hypertension, dyslipidemia, diabetes, heart disease, and stroke were collected through face-to-face interviews. Environmental factors such as cooking fuel (solid fuel or clean fuel), average annual temperature, and relative humidity were also included, as they have been identified in previous studies having potential influences on visual impairment [18–20]. Temperature and humidity data were obtained from the National Earth System Science Data Center, National Science & Technology Infrastructure of China (http://www.geodata.cn), using the same method employed for calculating the NDVI to determine average annual temperature and relative humidity. Statistical analysis Descriptive statistics for baseline characteristics were presented as mean ± standard deviation or median with interquartile range for continuous variables, and as number with proportion for categorical variables, respectively. Group differences were assessed using t-test or Mann-Whitney U test for continuous variables, and chi-squared test or Fisher’s exact test for categorical variables. Cox proportional hazards models were used to estimate hazard ratio (HR) and 95% confidence intervals (CI) for the association between residential greenness (as measured by the NDVI) and air pollution (PM2.5, PM10, and NO2) with the risk of DVI and NVI. We performed single exposure model by including residential greenness and air pollution separately and two-exposure models by including residential greenness and air pollution simultaneously. When considered as categorical variables, the results were presented as HR (95%CI) for the quintiles of exposure variables (Quintile 1: lowest, Quintile 5: highest), with Quintile 1 as the reference. For continuous variables, the HR (95%CI) was reported for each increase of 0.1 unit for NDVI and 10 µg/m3 for air pollution. Nonlinear relationship between single exposure with risk of DVI and NVI was assessed by restricted cubic spline (RCS) with three degrees of freedom [21]. All models were adjusted for sex, age, BMI, education level, marital status, residence, smoking status, alcohol consumption, household income, history of hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel type, and environmental factors, such as average annual temperature and humidity. To assess whether the potential association between air pollution and the risk of DVI and NVI was modified by residential greenness, we divided the cohort into quintiles based on residential greenness and conducted separate models for air pollution within each quintile. Mediation analysis was performed to quantify the proportion of the effect of residential greenness on visual impairment that could be explained by air pollution, following the method of Huang et al. [22]. First, we fitted a linear regression model to calculate the influence of residential greenness on air pollution. A Cox proportional hazards model was then applied to examine the relationship between the mediator variable (air pollution), residential greenness, and visual impairment after adjusting for various covariates. The direct and indirect effects were calculated, and the proportion mediated was determined for each pollutant (PM2.5, PM10, and NO2). To ensure robust statistical inference, a resampling method (Monte Carlo simulation) was used to estimate the CIs and P values for these effects. The mediation proportion was calculated as follows: Direct Effect (DE) = ln(HRDE), Indirect Effect (IE) = ln(HRIE), and Total Effect (TE) = ln(HRTE). Here, DE + IE = TE, and the mediation proportion was computed as IE/TE. Subgroup analyses were conducted to explore potential effect by age (< 65 years, ≥ 65 years), sex (male, female), education level (illiterate, primary school, middle school or above), residence (rural, urban), smoking status (no, yes), hypertension status (no, yes), and household income (low, high). Interaction P values between exposures and covariates were obtained using likelihood ratio tests. To assess the robustness of the findings, sensitivity analyses were performed by recalculating the average values of NDVI and air pollution from baseline until the occurrence of visual impairment, loss to follow-up, or the end of the study period. All statistical analyses were performed using R version 4.3.2, with a two-sided P value of < 0.05 considered statistically significant. Results Table 1 shows the baseline characteristics of the participants stratified by DVI and NVI. A total of 9,591 participants were included in the study, with an average age of 61.75 ± 9.39 years. During the seven-year follow-up period, 2,650 DVI cases and 3,130 NVI cases were identified. Participants with DVI or NVI were more likely to be female, have lower educational attainment, live in rural areas, use solid fuels for cooking, have lower household income, suffer from hypertension and heart disease. An inverse correlation was observed between NDVI and air pollutants, while PM2.5, PM10, and NO2 were positively correlated with each other (Figure S2). There was a strong correlation between PM2.5 and PM10, whereas weak correlations were found between PM2.5 and NO2, as well as PM10 and NO2. Table 1Characteristics of study participantsVariableAll participants (N = 9591)Without DVI (N = 6941)With DVI (N = 2650)P valueWithout NVI (N = 6461)With NVI (N = 3130)P value Age (years) 61.75 (9.39)61.01 (9.10)63.69 (9.84)< 0.00161.88 (9.39)61.48 (9.38)0.049 BMI (kg/m 2 ) 23.57 (3.75)23.64 (3.70)23.39 (3.87)0.00723.62 (3.75)23.48 (3.76)0.114Sex, n (%)< 0.001< 0.001Male4,789 (50.0)3,703 (53.4)1,086 (41.0)3,369 (52.2)1,420 (45.4)Female4,794 (50.0)3,232 (46.6)1,562 (59.0)3,086 (47.8)1,708 (54.6)Education level, n (%)< 0.001< 0.001Illiterate2,230 (23.3)1,330 (19.2)900 (34.0)1,405 (21.7)825 (26.4)Primary school3,763 (39.2)2,688 (38.7)1,075 (40.6)2,462 (38.1)1,301 (41.6)Middle school or above3,595 (37.5)2,920 (42.1)675 (25.5)2,593 (40.1)1,002 (32.0)Marital status, n (%)< 0.0010.801Married8,583 (89.5)6,293 (90.7)2,290 (86.4)5,786 (89.6)2,797 (89.4)Single or other1,008 (10.5)648 (9.3)360 (13.6)675 (10.4)333 (10.6)Residence, n (%)< 0.001< 0.001Rural5,687 (59.3)3,893 (56.1)1,794 (67.7)3,642 (56.4)2,045 (65.3)Urban3,904 (40.7)3,048 (43.9)856 (32.3)2,819 (43.6)1,085 (34.7)Smoking, n (%)3,816 (39.8)2,887 (41.6)929 (35.1)< 0.0012,644 (40.9)1,172 (37.5)0.001Alcohol consumption, n (%)< 0.001< 0.001> 1 times per month2,519 (26.3)1,952 (28.1)567 (21.4)1,768 (27.4)751 (24.0)< 1 times per month791 (8.2)620 (8.9)171 (6.5)552 (8.5)239 (7.6)Non consumer6,278 (65.5)4,366 (62.9)1,912 (72.2)4,138 (64.1)2,140 (68.4)Cooking fuel, n (%)< 0.001< 0.001Solid fuel4,908 (51.5)3,333 (48.3)1,575 (59.7)3,158 (49.2)1,750 (56.2)Clean fuel4,627 (48.5)3,565 (51.7)1,062 (40.3)3,263 (50.8)1,364 (43.8)Household income, n (%)< 0.001< 0.001Low4,442 (46.8)3,012 (43.8)1,430 (54.7)2,851 (44.6)1,591 (51.4)High5,046 (53.2)3,861 (56.2)1,185 (45.3)3,543 (55.4)1,503 (48.6)Hypertension, n (%)2,102 (22.0)1,447 (20.9)655 (24.8)< 0.0011,376 (21.4)726 (23.3)0.037Dyslipidemia, n (%)832 (8.8)565 (8.3)267 (10.3)0.003559 (8.8)273 (8.9)0.876Diabetes, n (%)439 (4.6)275 (4.0)164 (6.2)< 0.001279 (4.4)160 (5.2)0.086Heart disease, n (%)998 (10.5)651 (9.4)347 (13.2)< 0.001616 (9.6)382 (12.3)< 0.001Stroke, n (%)152 (1.6)102 (1.5)50 (1.9)0.170106 (1.6)46 (1.5)0.596 Table 1Characteristics of study participantsVariableAll participants (N = 9591)Without DVI (N = 6941)With DVI (N = 2650)P valueWithout NVI (N = 6461)With NVI (N = 3130)P value Age (years) 61.75 (9.39)61.01 (9.10)63.69 (9.84)< 0.00161.88 (9.39)61.48 (9.38)0.049 BMI (kg/m 2 ) 23.57 (3.75)23.64 (3.70)23.39 (3.87)0.00723.62 (3.75)23.48 (3.76)0.114Sex, n (%)< 0.001< 0.001Male4,789 (50.0)3,703 (53.4)1,086 (41.0)3,369 (52.2)1,420 (45.4)Female4,794 (50.0)3,232 (46.6)1,562 (59.0)3,086 (47.8)1,708 (54.6)Education level, n (%)< 0.001< 0.001Illiterate2,230 (23.3)1,330 (19.2)900 (34.0)1,405 (21.7)825 (26.4)Primary school3,763 (39.2)2,688 (38.7)1,075 (40.6)2,462 (38.1)1,301 (41.6)Middle school or above3,595 (37.5)2,920 (42.1)675 (25.5)2,593 (40.1)1,002 (32.0)Marital status, n (%)< 0.0010.801Married8,583 (89.5)6,293 (90.7)2,290 (86.4)5,786 (89.6)2,797 (89.4)Single or other1,008 (10.5)648 (9.3)360 (13.6)675 (10.4)333 (10.6)Residence, n (%)< 0.001< 0.001Rural5,687 (59.3)3,893 (56.1)1,794 (67.7)3,642 (56.4)2,045 (65.3)Urban3,904 (40.7)3,048 (43.9)856 (32.3)2,819 (43.6)1,085 (34.7)Smoking, n (%)3,816 (39.8)2,887 (41.6)929 (35.1)< 0.0012,644 (40.9)1,172 (37.5)0.001Alcohol consumption, n (%)< 0.001< 0.001> 1 times per month2,519 (26.3)1,952 (28.1)567 (21.4)1,768 (27.4)751 (24.0)< 1 times per month791 (8.2)620 (8.9)171 (6.5)552 (8.5)239 (7.6)Non consumer6,278 (65.5)4,366 (62.9)1,912 (72.2)4,138 (64.1)2,140 (68.4)Cooking fuel, n (%)< 0.001< 0.001Solid fuel4,908 (51.5)3,333 (48.3)1,575 (59.7)3,158 (49.2)1,750 (56.2)Clean fuel4,627 (48.5)3,565 (51.7)1,062 (40.3)3,263 (50.8)1,364 (43.8)Household income, n (%)< 0.001< 0.001Low4,442 (46.8)3,012 (43.8)1,430 (54.7)2,851 (44.6)1,591 (51.4)High5,046 (53.2)3,861 (56.2)1,185 (45.3)3,543 (55.4)1,503 (48.6)Hypertension, n (%)2,102 (22.0)1,447 (20.9)655 (24.8)< 0.0011,376 (21.4)726 (23.3)0.037Dyslipidemia, n (%)832 (8.8)565 (8.3)267 (10.3)0.003559 (8.8)273 (8.9)0.876Diabetes, n (%)439 (4.6)275 (4.0)164 (6.2)< 0.001279 (4.4)160 (5.2)0.086Heart disease, n (%)998 (10.5)651 (9.4)347 (13.2)< 0.001616 (9.6)382 (12.3)< 0.001Stroke, n (%)152 (1.6)102 (1.5)50 (1.9)0.170106 (1.6)46 (1.5)0.596 Characteristics of study participants Table 2 shows the association between residential greenness, air pollution, and the risk of visual impairment. In the single-exposure model, each 0.1 unit increase in residential greenness was associated with a lower risk of both DVI (HR: 0.68, 95%CI: 0.65,0.71) and NVI (HR: 0.67, 95%CI: 0.64,0.69). Compared to the lowest NDVI quintile, the HR for DVI and NVI in the highest quintile were 0.30 (95%CI: 0.26,0.34) and 0.29 (95%CI: 0.25,0.33), respectively. In contrast, exposure to air pollution (PM2.5, PM10, and NO2) was positively associated with an increased risk of both DVI and NVI. The HR for each 10 µg/m³ increase in PM2.5, PM10, and NO2 was 1.60 (95% CI: 1.56,1.65), 1.28 (95%CI: 1.27,1.30), and 1.73 (95%CI: 1.62,1.85) for DVI and 1.64 (95%CI: 1.60,1.69), 1.30 (95%CI: 1.28,1.32), and 1.82 (95%CI: 1.72,1.94) for NVI, respectively. Moreover, the respective HR for DVI in the highest quintile of PM2.5, PM10, and NO2 was 6.04 (95%CI: 5.21,7.01), 8.39 (95%CI: 7.12,9.88), and 3.49 (95%CI: 2.93,4.16), compared with the lowest quintile. For NVI, the respective HR in the highest quintile of PM2.5, PM10, and NO2 was 7.10 (95%CI: 6.16,8.17), 9.39 (95%CI: 8.05,10.97), and 4.43 (95%CI: 3.77,5.21), compared with the lowest quintile. Table 2Associations of residential greenness and air pollution with risk of DVI and NVI in single-exposure modelExposureDVINVINumber of cases/TotalAdjusted HR (95%CI)Number of cases/TotalAdjusted HR (95%CI)NDVIContinuous (per 0.1 unit increase)0.68 (0.65,0.71)0.67 (0.64,0.69)Quintile 1729/1919Ref815/1919RefQuintile 2472/19180.51 (0.44,0.58)568/19180.55 (0.49,0.62)Quintile 3631/19180.65 (0.57,0.74)794/19180.76 (0.67,0.86)Quintile 4425/19180.41 (0.36,0.47)522/19180.42 (0.37,0.47)Quintile 5393/19180.30 (0.26,0.34)431/19180.29 (0.25,0.33)P for trend< 0.001< 0.001PM2.5Continuous (per 10 µg/m³ increase)1.60 (1.56,1.65)1.64 (1.60,1.69)Quintile 1461/1919Ref475/1919RefQuintile 2405/19181.10 (0.95,1.29)498/19181.27 (1.10,1.47)Quintile 3431/19181.51 (1.29,1.76)602/19182.00 (1.74,2.31)Quintile 4569/19182.72 (2.35,3.16)680/19183.45 (2.99,3.97)Quintile 5784/19186.04 (5.21,7.01)875/19187.10 (6.16,8.17)P for trend< 0.001< 0.001PM10Continuous (per 10 µg/m³ increase)1.28 (1.27,1.30)1.30 (1.28,1.32)Quintile 1372/1919Ref389/1919RefQuintile 2551/19181.72 (1.47,2.00)644/19181.85 (1.60,2.15)Quintile 3358/19181.54 (1.30,1.81)518/19182.12 (1.82,2.46)Quintile 4658/19184.10 (3.54,4.76)796/19185.42 (4.71,6.24)Quintile 5711/19188.39 (7.12,9.88)783/19189.39 (8.05,10.97)P for trend< 0.001< 0.001NO2Continuous (per 10 µg/m³ increase)1.73 (1.62,1.85)1.82 (1.72,1.94)Quintile 1561/1919Ref586/1919RefQuintile 2502/19181.33 (1.16,1.53)664/19181.70 (1.49,1.93)Quintile 3470/19181.25 (1.08,1.45)612/19181.66 (1.45,1.90)Quintile 4556/19182.32 (2.00,2.70)600/19182.54 (2.20,2.93)Quintile 5561/19183.49 (2.93,4.16)668/19184.43 (3.77,5.21)P for trend< 0.001< 0.001Adjusted for age, sex, education level, marital status, residence, smoking status, alcohol consumption, household income, hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel, BMI, temperature, and humidity Table 2Associations of residential greenness and air pollution with risk of DVI and NVI in single-exposure modelExposureDVINVINumber of cases/TotalAdjusted HR (95%CI)Number of cases/TotalAdjusted HR (95%CI)NDVIContinuous (per 0.1 unit increase)0.68 (0.65,0.71)0.67 (0.64,0.69)Quintile 1729/1919Ref815/1919RefQuintile 2472/19180.51 (0.44,0.58)568/19180.55 (0.49,0.62)Quintile 3631/19180.65 (0.57,0.74)794/19180.76 (0.67,0.86)Quintile 4425/19180.41 (0.36,0.47)522/19180.42 (0.37,0.47)Quintile 5393/19180.30 (0.26,0.34)431/19180.29 (0.25,0.33)P for trend< 0.001< 0.001PM2.5Continuous (per 10 µg/m³ increase)1.60 (1.56,1.65)1.64 (1.60,1.69)Quintile 1461/1919Ref475/1919RefQuintile 2405/19181.10 (0.95,1.29)498/19181.27 (1.10,1.47)Quintile 3431/19181.51 (1.29,1.76)602/19182.00 (1.74,2.31)Quintile 4569/19182.72 (2.35,3.16)680/19183.45 (2.99,3.97)Quintile 5784/19186.04 (5.21,7.01)875/19187.10 (6.16,8.17)P for trend< 0.001< 0.001PM10Continuous (per 10 µg/m³ increase)1.28 (1.27,1.30)1.30 (1.28,1.32)Quintile 1372/1919Ref389/1919RefQuintile 2551/19181.72 (1.47,2.00)644/19181.85 (1.60,2.15)Quintile 3358/19181.54 (1.30,1.81)518/19182.12 (1.82,2.46)Quintile 4658/19184.10 (3.54,4.76)796/19185.42 (4.71,6.24)Quintile 5711/19188.39 (7.12,9.88)783/19189.39 (8.05,10.97)P for trend< 0.001< 0.001NO2Continuous (per 10 µg/m³ increase)1.73 (1.62,1.85)1.82 (1.72,1.94)Quintile 1561/1919Ref586/1919RefQuintile 2502/19181.33 (1.16,1.53)664/19181.70 (1.49,1.93)Quintile 3470/19181.25 (1.08,1.45)612/19181.66 (1.45,1.90)Quintile 4556/19182.32 (2.00,2.70)600/19182.54 (2.20,2.93)Quintile 5561/19183.49 (2.93,4.16)668/19184.43 (3.77,5.21)P for trend< 0.001< 0.001Adjusted for age, sex, education level, marital status, residence, smoking status, alcohol consumption, household income, hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel, BMI, temperature, and humidity Associations of residential greenness and air pollution with risk of DVI and NVI in single-exposure model Adjusted for age, sex, education level, marital status, residence, smoking status, alcohol consumption, household income, hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel, BMI, temperature, and humidity Table 3 presents the results of two-exposure models that simultaneously assessed the effects of residential greenness and air pollution on the risk of DVI and NVI. In these models, the protective effect of NDVI on visual impairment was attenuated compared to the single-exposure model. For example, in the NDVI + PM2.5 model, the HR for DVI was 0.85 (95%CI: 0.81,0.89) and for NVI was 0.84 (95%CI: 0.80,0.88), slightly weaker than in the single-exposure model. Similarly, the harmful effects of air pollution were also attenuated when NDVI was included in the model. For instance, in the same NDVI + PM2.5 model, the HR for DVI associated with PM2.5 was reduced to 1.53 (95%CI: 1.49,1.58), and for NVI to 1.57 (95%CI: 1.53,1.62). Similar trends were observed in the NDVI + PM10 and NDVI + NO2 model. In the NDVI + PM10 model, the HR for NVDI were 0.88 (95%CI: 0.84,0.92) for DVI and 0.87 (95%CI: 0.84,0.91) for NVI, while the HR for PM10 was 1.25 (95%CI: 1.23,1.27) for DVI and 1.27 (95%CI: 1.25,1.29) for NVI, both showing a reduction compared to the single-exposure models. In the NDVI + NO2 model, the HR for NDVI was 0.74 (95%CI: 0.71,0.77) for DVI and 0.73 (95%CI: 0.70,0.76) for NVI, with NO2 showing reduced harmful effects (HR for DVI: 1.45, 95%CI: 1.35,1.55; HR for NVI: 1.51, 95%CI: 1.42,1.61). Table 3Associations of residential greenness and air pollution with risk of DVI and NVI in two-exposure modelModelExposureDVINVIAdjusted HR (95%CI)Adjusted HR (95%CI)NDVI + PM2.5NDVI0.85 (0.81,0.89)0.84 (0.80,0.88)PM2.51.53 (1.49,1.58)1.57 (1.53,1.62)NDVI + PM10NDVI0.88 (0.84,0.92)0.87 (0.84,0.91)PM101.25 (1.23,1.27)1.27 (1.25,1.29)NDVI + NO2NDVI0.74 (0.71,0.77)0.73 (0.70,0.76)NO21.45 (1.35,1.55)1.51 (1.42,1.61)Adjusted for age, sex, education level, marital status, residence, smoking status, alcohol consumption, household income, hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel, BMI, temperature, and humidity. HR was presented for per 0.1 unit increase for NDVI and for per 10 µg/m³ increase for PM2.5, PM10, and NO2 Table 3Associations of residential greenness and air pollution with risk of DVI and NVI in two-exposure modelModelExposureDVINVIAdjusted HR (95%CI)Adjusted HR (95%CI)NDVI + PM2.5NDVI0.85 (0.81,0.89)0.84 (0.80,0.88)PM2.51.53 (1.49,1.58)1.57 (1.53,1.62)NDVI + PM10NDVI0.88 (0.84,0.92)0.87 (0.84,0.91)PM101.25 (1.23,1.27)1.27 (1.25,1.29)NDVI + NO2NDVI0.74 (0.71,0.77)0.73 (0.70,0.76)NO21.45 (1.35,1.55)1.51 (1.42,1.61)Adjusted for age, sex, education level, marital status, residence, smoking status, alcohol consumption, household income, hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel, BMI, temperature, and humidity. HR was presented for per 0.1 unit increase for NDVI and for per 10 µg/m³ increase for PM2.5, PM10, and NO2 Associations of residential greenness and air pollution with risk of DVI and NVI in two-exposure model Adjusted for age, sex, education level, marital status, residence, smoking status, alcohol consumption, household income, hypertension, dyslipidemia, diabetes, heart disease, stroke, cooking fuel, BMI, temperature, and humidity. HR was presented for per 0.1 unit increase for NDVI and for per 10 µg/m³ increase for PM2.5, PM10, and NO2 Table 4 shows the mediation analysis of the indirect effects of air pollution on the association between residential greenness and visual impairment. For DVI, PM2.5 mediated 49.73% of the association between greenness and visual impairment (Direct effect HR: 0.89, 95%CI: 0.86,0.91; Indirect effect HR: 0.89, 95%CI: 0.88,0.90). Similar mediation effects were observed for PM10 (53.96%) and NO2 (17.04%). For NVI, PM2.5 mediated 49.95% of the association, with mediation proportions of 53.79% for PM10 and 17.81% for NO2. Table 4Mediation effects of air pollution on the association between residential greenness and risk of distance visual impairment and near visual impairmentOutcomeMediatorDirect effect (HR 95%CI)Indirect effect (HR 95%CI)Total effect (HR 95%CI)Proportion mediatedDVIPM2.50.89 (0.86,0.91)0.89 (0.88,0.90)0.79 (0.76,0.81)49.73%PM100.89 (0.87,0.92)0.88 (0.86,0.89)0.79 (0.76,0.81)53.96%NO20.84 (0.81,0.86)0.96 (0.95,0.97)0.81 (0.78,0.83)17.04%NVIPM2.50.87 (0.84,0.89)0.87 (0.85,0.88)0.75 (0.73,0.77)49.95%PM100.88 (0.85,0.90)0.86 (0.84,0.87)0.75 (0.73,0.77)53.79%NO20.81 (0.78,0.83)0.95 (0.94,0.96)0.77 (0.75,0.79)17.81% Table 4Mediation effects of air pollution on the association between residential greenness and risk of distance visual impairment and near visual impairmentOutcomeMediatorDirect effect (HR 95%CI)Indirect effect (HR 95%CI)Total effect (HR 95%CI)Proportion mediatedDVIPM2.50.89 (0.86,0.91)0.89 (0.88,0.90)0.79 (0.76,0.81)49.73%PM100.89 (0.87,0.92)0.88 (0.86,0.89)0.79 (0.76,0.81)53.96%NO20.84 (0.81,0.86)0.96 (0.95,0.97)0.81 (0.78,0.83)17.04%NVIPM2.50.87 (0.84,0.89)0.87 (0.85,0.88)0.75 (0.73,0.77)49.95%PM100.88 (0.85,0.90)0.86 (0.84,0.87)0.75 (0.73,0.77)53.79%NO20.81 (0.78,0.83)0.95 (0.94,0.96)0.77 (0.75,0.79)17.81% Mediation effects of air pollution on the association between residential greenness and risk of distance visual impairment and near visual impairment Figures 1 and 2 show the associations between air pollution and the risks of DVI and NVI, across different quintiles of residential greenness. The detrimental effects of air pollution on DVI and NVI were pronounced in participants living in areas with lower levels of residential greenness (quintiles 1–4 of the NDVI), with insignificant association in the highest level of greenness areas (quintile 5 of the NDVI). Fig. 1Associations between air pollution and DVI across quintiles of greenness Associations between air pollution and DVI across quintiles of greenness Fig. 2Associations between air pollution and NVI across quintiles of greenness Associations between air pollution and NVI across quintiles of greenness The concentration response curves (Figures S3 and Figure S4) illustrate the dose-dependent associations of residential greenness and air pollution with the risk of DVI and NVI. These curves suggest protective effects of increasing residential greenness and detrimental effects of increasing air pollution on DVI and NVI. In particular, there was a linear relationship between residential greening and the risk of DVI and NVI, and a non-linear relationship between air pollution (PM2.5, PM10 and NO2) and the risk of DVI and NVI. Figure S5 and S6 show stratified analyses for the associations of residential greenness and air pollution with risk of DVI and NVI. The inverse association between NDVI and the risk of DVI was more pronounced among participants in urban areas, while the inverse association with NVI was stronger among those aged < 65 years. Furthermore, stronger associations of PM2.5 and PM10 with the risk of both DVI and NVI were observed among participants aged < 65 years and with higher education levels. The association between PM10 and the risk of DVI and NVI was more significant among participants with higher household income. The results of the sensitivity analyses were generally consistent with the main findings, indicating that the associations observed between environmental exposures and visual impairment risk were robust (Supplementary tables and Figure S7-S8). Discussion This study investigated the associations between residential greenness, air pollution, and the risk of visual impairment among Chinese middle-aged and older adults. Our results suggest that high levels of air pollution were associated with an increased risk of both DVI and NVI, while residential greenness could reduce DVI and NVI risk. Furthermore, the adverse effects of air pollution on DVI and NVI were more pronounced in individuals exposed to low levels of residential greenness. The mediation analysis suggested that the positive effects of residential greenness on DVI and NVI were partly mediated by the reduction in exposure to PM2.5, PM10 and NO2. Although several studies have explored the association between air pollution exposure and visual impairment, the findings remain inconsistent. Similar to our study, a cross-sectional study across six low- and middle-income countries involving 36,620 participants aged 50 years and older found that each 10 µg/m³ increase in PM2.5 above 15 µg/m³ was associated with a higher risk of presbyopia (OR: 1.37, 95%CI: 1.23,1.54) [23]. Another cross-sectional study with 61,995 children from 7 provinces/municipalities across China found that an interquartile range increase in PM1, PM2.5, PM10, and NO2 was associated with a 1.13- (95%CI, 1.04,1.24), 1.27- (95%CI, 1.08,1.48), 1.14- (95%CI, 1.02,1.28), and 1.28-fold (95%CI, 1.17,1.39) increased odds of visual impairment [24], respectively. In contrast, a cross-sectional study including 18,622 Korean adults reported that high NO2 had protective effect on nuclear cataracts [25]. The discrepancy may be due to variations in population characteristics, exposure levels, and the specific types of visual impairment studied. Thus, further investigations are warranted to clarify the relationship between air pollution and eye health. To date, there is only limited evidence of a link between residential greenness and visual impairment. In a survey of 13,380 school–aged adolescents from Beijing, a 0.1 increase in NDVI in the 500 m buffer may reduce myopia risk at school level by 15% (OR: 0.85, 95%CI: 0.74,0.98) [9]. Another cross-sectional study of 61,995 Chinese children and adolescents showed that an interquartile range increase in NDVI in the 500 m buffer was associated with a lower likelihood of visual impairment (OR: 0.95, 95%CI: 0.93,0.97) [10]. In a cohort study of 53,575 preschool children, higher neighborhood greenness in a 100 m buffer around the home address was associated with a lower risk of myopia, and higher neighborhood greenness in 100, 250, and 500 m reduced the risk of astigmatism [11]. Although it may be hard to directly compare our results and previous findings due to the variations in study populations and the definitions of visual impairment, the inverse association between NDVI in the 250 m buffer and visual impairment in our study provides further evidence that reinforce the potential benefits of residential greenness on visual health among middle-aged and older adults. The associations of air pollution and residential greenness with risk of DVI and NVI observed in our study may be explained by several mechanisms. Exposure to air pollutants can induce oxidative stress, resulting in cellular damage to ocular tissues [26, 27]. Furthermore, chronic exposure to air pollutants has been linked to systemic inflammation [28, 29], and inflammatory cytokines may disrupt the blood-retinal barrier, leading to retinal damage and visual impairment [30, 31]. The potential reasons underlying the beneficial effects of residential greenness may be owe to increased outdoor activities against myopia, particularly in children [32]. Moreover, exposure to natural environments has been linked to reductions in stress, improvements in mental health, and a strengthening of the immune system, which may in turn indirectly contribute to better eye health by promoting general well-being [33]​. Additionally, residential greenness could act as natural buffers against air pollution by reducing residents’ exposure to harmful pollutants [34]. In stratified analysis, the association between residential greenness and incident DVI was modified by residence, with a stronger effect among participants living in urban areas. This may be attributed to the typically higher levels of air pollution in urban areas compared to rural areas. In such environments, residential greenness may provide greater relative benefits by mitigating the harmful effects of pollutants on visual health [35]. Furthermore, we found stronger associations of PM2.5 and PM10 with incident DVI and NVI among participants age < 65 years and with high education level. This might because that individuals aged < 65 years may have higher exposure levels to outdoor air pollution due to more active lifestyles, including greater participation in outdoor activities and longer commuting times [36], and individuals with high education/household income usually live in urban areas with greater air pollution [37]. Notably, we also found a stronger association between PM10 exposure and the incidence of DVI and NVI among participants with higher household incomes, who were more likely to live in urban areas, where PM10 concentrations tend to be higher due to increased traffic and industrial activities. The strengths of our study include the large, nationally representative cohort and the robust statistical models to assess the effects of multiple exposures. However, several limitations should be acknowledged. First, the assessment of visual impairment was based on self-reported measures, which may be subject to reporting bias. Future studies should incorporate objective measures of visual function, such as standardized visual acuity tests, to improve accuracy. Second, although we controlled for numerous confounders, residual confounding such as genetic predisposition, dietary habits, and occupational exposures, cannot be entirely ruled out. Third, the most detailed address information for each participant was only available at city level, this may introduce some degree of misclassification. In conclusion, our findings suggest that increasing residential greenness and reducing air pollution could be effective strategies for mitigating the risk of visual impairment in aging populations. Future studies should explore the underlying biological mechanisms and assess the long-term impact of environmental interventions on visual health. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1: Figure S1. Selection framework of study participants Supplementary Material 1: Figure S1. Selection framework of study participants Supplementary Material 2: Figure S2. (A-B) Spearman’s correlation coefficients between residential greenness and air pollution for DVI and NVI Supplementary Material 2: Figure S2. (A-B) Spearman’s correlation coefficients between residential greenness and air pollution for DVI and NVI Supplementary Material 3: Figure S3. Concentration-response curves for the association of residential greenness and air pollution with risk of DVI using restrict cubic splines Supplementary Material 3: Figure S3. Concentration-response curves for the association of residential greenness and air pollution with risk of DVI using restrict cubic splines Supplementary Material 4: Figure S4. Concentration-response curves for the association of residential greenness and air pollution with risk of NVI using restrict cubic splines Supplementary Material 4: Figure S4. Concentration-response curves for the association of residential greenness and air pollution with risk of NVI using restrict cubic splines Supplementary Material 5: Figure S5. Subgroup analysis for the associations of NDVI, PM2.5, PM10, and NO2 with DVI Supplementary Material 5: Figure S5. Subgroup analysis for the associations of NDVI, PM2.5, PM10, and NO2 with DVI Supplementary Material 6: Figure S6. Subgroup analysis for the associations of NDVI, PM2.5, PM10, and NO2 with NVI Supplementary Material 6: Figure S6. Subgroup analysis for the associations of NDVI, PM2.5, PM10, and NO2 with NVI Supplementary Material 7: Figure S7. Associations between air pollution and DVI across quintiles of greenness (Sensitivity analysis) Supplementary Material 7: Figure S7. Associations between air pollution and DVI across quintiles of greenness (Sensitivity analysis) Supplementary Material 8: Figure S8. Associations between air pollution and NVI across quintiles of greenness (Sensitivity analysis) Supplementary Material 8: Figure S8. Associations between air pollution and NVI across quintiles of greenness (Sensitivity analysis) Supplementary Material 9 Supplementary Material 9