Greenness and its interaction with air pollution in relation to postmenopausal breast cancer risk in UK Biobank

Carmen Smotherman; Brian L. Sprague; Dejana Braithwaite; Lusine Yaghjyan

doi:10.1371/journal.pone.0334744

Abstract

It is unclear if natural vegetation (greenness) is associated with breast cancer breast cancer risk and if it interacts with air pollution. We investigated the associations of greenness with breast cancer risk in postmenopausal women, and the extent to which the association of the air pollutant PM₁₀ with breast cancer differs by greenness level. Of the 154,804 postmenopausal women included in the study, 6,131 developed breast cancer after the baseline assessment for enrollment into the UK Biobank. Data on greenness measures, and established breast cancer risk factors were available at baseline assessment (2006−2010), while data for PM₁₀ were available from 2007 and 2010. We used Cox proportional hazards models to assess adjusted associations between greenness and breast cancer risk. We also examined associations of PM₁₀ with breast cancer risk by greenness levels and their interactions. For each 0.1 unit increase in NDVI, we found a 2.6% increase in breast cancer risk (Hazard ratio [HR]=1.03, 95%CI 1.00–1.05). We also found evidence of a significant interaction, with stronger association between cumulative average PM₁₀ and breast cancer risk at lower NDVI levels (1^{^st} quartile NDVI: HR per 10 µg/m^³ PM₁₀ = 3.03, 95%CI 1.98–4.64) versus higher NDVI levels (4^{^th} quartile: HR = 1.52, 95%CI 1.16–1.98). We did not find interactions with other greenness measures. Our results suggest that various measures of greenness differ in their association with breast cancer and that there may be an interaction between greenness and PM₁₀ exposure in relation to breast cancer risk.

Citation: Smotherman C, Sprague BL, Braithwaite D, Yaghjyan L (2025) Greenness and its interaction with air pollution in relation to postmenopausal breast cancer risk in UK Biobank. PLoS One 20(11): e0334744. https://doi.org/10.1371/journal.pone.0334744

Editor: Mattias Gaglio, University of Ferrara, ITALY

Received: June 9, 2025; Accepted: October 1, 2025; Published: November 12, 2025

Copyright: © 2025 Smotherman et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data used for this project are available for interested investigators from UK Biobank per its standard data access policy. UK Biobank data are available for use by eligible researchers from academic, charity, government and commercial organizations – across the world – for health-related research that is in the public interest. More information on access procedure and application can be found at https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access.

Funding: BS effort was supported in part by grant P20GM103644 from the National Institute of General Medical Sciences. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Natural green environments (greenness), commonly assessed using remote sensing indices such as the Normalized Difference Vegetation Index (NDVI), may reduce harmful environmental exposures (such as air pollution, extreme heat), increase physical activity and reduce stress levels, which subsequently could impact health outcomes [1–5]. Importantly, use of greenness in epidemiologic studies captures multiple interrelated effects rather than isolating individual health benefits. Multiple lines of evidence suggest that greenness improves health, as well as social and environmental outcomes especially in populations with lower socioeconomic status [6]. Previous studies have linked exposure to urban greenness with various health endpoints [7–13], demonstrating in some instances interactions with air pollution [14–16], and in others, an independent effect [17,18]. However, the associations of greenness with cancer risk remain poorly understood. A recent report from Nurses’ Health Study found an inverse association of greenness with overall cancer mortality, independent of air pollution [19]. Two studies suggested an inverse association between greenness and prostate cancer risk, independently of air pollution, social and demographic characteristics, and lifestyle factors [20,21]. A few recent reports found an inverse association of greenness with breast cancer risk [20,22,23], while other reported a positive association [24]. These contrasting findings may be due to differences in study designs, geographic setting, exposure assessment methods, and population characteristics. Some studies suggested that these associations were independent of air pollution, physical activity, and obesity [22,24]. Recent studies also demonstrated associations of greenness with several biomarkers implicated in breast carcinogenesis, including lower insulin resistance and less obesity [2,25,26], lower levels of pro-inflammatory markers [27,28], improved cortisol profiles [12], and longer telomeres [29], thus further supporting a potential link to breast carcinogenesis.

For development and implementation of efficient public health interventions aimed to reduce breast cancer incidence, the potential complex relationship between greenness and air pollution in relation to breast cancer should be comprehensively explored to determine if greenness is indeed associated with breast cancer risk and whether the association is independent of air pollution [30]. Failing to account for the confounding or modifying effects of greenness could misrepresent the true impact of air pollution on health. Understanding these independent and joint effects is critical for public health and urban planning, as it allows for more targeted interventions, e.g., designing green spaces that maximize health benefits while minimizing harmful exposures. It also helps in identifying population subgroups who may benefit most from greenness or be most vulnerable to pollution.

The goal of this study was to examine the association of greenness with postmenopausal breast cancer risk, as well as explore whether association of air pollution with breast cancer differs by the level of greenness for postmenopausal women enrolled in the UK Biobank. We focused on pollution measures that were previously linked to breast cancer risk in our analyses using data from 155,235 postmenopausal women from UK Biobank, specifically PM₁₀ annual averages from 2007 and cumulative average PM₁₀ (calculated as the average across measures from 2007 and 2010) [31].

Methods

Study population

This study included postmenopausal women selected from participants in UK Biobank. The data received from the UK Biobank were accessed for research purposes on January 20, 2021. The authors did not have access to information that could identify individual participants during or after data collection. The recruitment, enrollment and follow-up processes were described in details elsewhere [32]. In short, about 500,000 volunteers, 40–69 years old from the UK were recruited during 2006–2010, and about 20% of the participants completed a follow-up first assessment during 2012–2013.

In this study, we used the same subset of postmenopausal women from our previous study [31]. Briefly, exclusion criteria encompassed a history of any cancer except non-melanoma skin cancer, premenopausal or unknown menopausal status, and missing data on all air pollution measures, all greenness measures, and on any of the following covariates: race, BMI, parity, alcohol, and smoking. National Health Service (NHS) Central Registers records from England (through February 29, 2020) and from Scotland (through August 31, 2021) were used to assess participants’ outcomes. The International Classification of Diseases (ICD) codes used to identify a breast cancer diagnosis were 174 and C50 for invasive breast cancer, and 2330 and D05 for in-situ breast cancer. This study included 154,804 women (representing 97% of all eligible women in UK Biobank) and 6,131 developed breast cancer during the follow (Fig 1).

Download:

Fig 1. Study sample selection diagram.

Flow diagram illustrating the selection process of participants for the study.

https://doi.org/10.1371/journal.pone.0334744.g001

UK Biobank was approved by the North West Multi-centre Research Ethics Committee (MREC), which covers the UK. All participants of UK Biobank provided written consent at recruitment. The current analyses were approved by the University of Florida Institutional Review Board (protocol #202002317).

Greenness assessment

The following greenness measures were assessed in UK Biobank: Normalized Difference Vegetation Index (NDVI), Natural Environment Percentage (NEP), and Greenspace Percentage (GP). All measures were recorded during the assessment period at baseline (2006–2010), while GP and NEP were also available at the first repeat assessment (2012–2013).

Greenspace percentage was modelled using 2005 data from the Generalized Land Use Database (GLUD) for England, which shows the areas of land use at lower-layer super output areas (LSOAs) for 2001 Census Output Areas in England [33]. GP was estimated as the percentage of LSOAs covered by green space, including gardens. Each residential address was allocated circular distance buffers of 300m and 1000m, respectively. An area-weighted mean of GP coverage was allocated to participants from England, at the baseline and at first repeat assessment. GLUD data did not include addresses from outside of England; therefore, greenness measurements were not assigned by UK Biobank for addresses outside England, resulting in missing data for these participants. GP measurements from baseline assessment and the first repeat assessment were highly correlated (correlation coefficient r = 0.971 [p < 0.001, n = 6,417] for GP, buffer 1000m, and r = 0.968 [p < 0.001, n = 6,417] for GP, buffer 300m); thus, only baseline greenness measurements were used in the analyses.

Natural Environment Percentage (NEP) was derived using the Centre for Ecology and Hydrology (CEH) Land Cover Map (LCM) 2007 data to extend coverage to Wales and Scotland [34].The percentage coverage of “natural environment” were calculated for two buffer sizes (300m and 1000m) around the participant residential address. Only baseline NEP measurements were used in analyses since NEP values from baseline and the first repeat assessment were highly correlated: r = 0.969 (p < 0.001, n = 6,417) for NEP, buffer 1000m, and r = 0.965 (p < 0.001, n = 6,417) for NEP, buffer 300m.

The Normalized Difference Vegetation Index (NDVI) is a unit-less index that indicates a relative overall green healthy vegetation, described previously in details [35]. Residential greenness was expressed as mean deviation in the NDVI values within 500m radial area of the participant home address. The NDVI ranges from −1 (indicating water) to +1 (indicating dense green vegetation). Values of 0.2–0.3 indicate shrub and grassland, while 0–0.1 indicate barren rock, sand, or snow. NDVI measure was not available for participants from northern England or Scotland (total of 45,851 or 28.8% of eligible women).

Air pollution assessment

Exposures to air pollution were assessed using annual averages of particulate matter PM₁₀ derived from Land Use Regression (LUR) models, described in details elsewhere [36–38]. Briefly, the estimates for 2007 were based on PM₁₀ concentrations obtained from EuroAirnet, the regulatory air pollution monitoring network in Europe [36]. The predictors in the final model were Satellite-derived surface PM_2.5 from 2001–2006, the Y coordinate of participant location, length of major (200–2500 meters) and minor (200 meters) roads, altitude, tree canopy (100 meters). The model for 2007 explains about 47% of the variation in measured concentrations (about 50% of the variation in log-transformed PM₁₀). The estimates for 2010 were based on the monitoring done by the European Study of Cohorts for Air Pollution Effects (ESCAPE), including two sites in UK, Manchester and London/Oxford [37,38]. Predictors in the best model to predict the annual average of PM₁₀ for 2010 differed by region. For Manchester, the final model included surface area of urban green (100 meters), total heavy-duty traffic load of major roads (1000 meters), surface area of semi-natural and forested (300 meters), surface area of high-density residential area (100 meters). The model for London/Oxford included inverse distance and inverse squared distance to the nearest major road, and heavy-duty traffic intensity on nearest major road. These regional models were developed and applied separately to account for differences in urban structure and emission patterns; hence for differences in availability of comparable predictor variables across all regions; a single harmonized national model was not used. Due to poor performance of the model for further locations from the initial ESCAPE study area, participants living beyond 400 km from Greater London (northern England and Scotland), were not assigned PM₁₀ concentrations for 2010 to prevent exposure misclassification (n = 11,499 or 7.4% or the study sample).

Covariates information

Data on established breast cancer risk factors (age at menarche, age at menopause, Body Mass Index (BMI), race/ethnicity, postmenopausal hormone therapy, parity, age at first birth, smoking status, alcohol use, and family history for breast cancer were available from baseline and, for a subset of women, from follow-up assessments. We included only baseline risk factor data in the analyses as we previously found high correlations between baseline and repeat assessments for these risk factors, as reported previously [31].

Statistical analyses

Associations between each greenness measure and breast cancer risk were assessed using Cox proportional hazards regression models, as previously described [39,40]. Multivariate models also included age at recruitment (continuous), age at menarche (continuous), age at menopause (<46, 46 to <50, 50 to <55, 55 or more years, unknown), BMI (kg/m², continuous), race (Caucasian vs Other races), combined parity and age at first birth (nulliparous, parous and age at first birth 25 years or less, parous and age at first birth more than 25 years, parous and unknown age at first birth), postmenopausal hormone use, family history of breast cancer in first degree relatives, smoking and alcohol consumption. “Other races” included: Mixed (White and Black Caribbean, White and Black African, White and Asian, Any other Mixed background), Asian or Asian British (Indian, Pakistani, Bangladeshi, Any other Asian background), Black or Black British (Caribbean, African, Any other Black background, Chinese or Other ethnic group. Follow-up started at enrollment and ended at the time of breast cancer diagnosis, another cancer type diagnosis, death, or at the last linkage to the cancer registries. Greenness and air pollution measures were modeled as continuous variables and as quartiles, respectively. Risk estimates were described using hazard ratios (HRs) and their 95% confidence intervals (CIs). The regression models included the interaction term between age at recruitment and time due to the fact that age at recruitment violated the proportional hazards assumption [40]. Tests for trend were performed, modeling each greenness measure as an ordinal variable, and using the median greenness level in each category.

The regression models were developed for 2007 PM₁₀, 2010 PM₁₀, and cumulative average PM₁₀ (calculated as the average across available PM₁₀ measures) and included the interaction between greenness and PM₁₀ (modeled as continuous per 10 µg/m³ and as quartiles). The interaction was tested in two ways: with greenness and PM₁₀ as continuous exposure, and using the medians for each quartile. Next, the associations between PM₁₀ and breast cancer risk were examined across the greenness levels. The analyses of 2007 PM₁₀ exposure and 2010 PM₁₀ exposure, respectively, had a follow-up time starting with the respective year in which the PM₁₀ was available. For analyses of cumulative average exposure to PM₁₀, the follow-up start time was defined as the latest available PM₁₀ measure.

Additional analyses were performed to (1) allow a 2-year lag in air pollution as previously described [31] and (2) include only diagnoses of invasive breast cancer. In the lagged analysis, we first applied the lag to the individual annual PM₁₀ exposure values and then computed the cumulative PM₁₀ exposure using these lagged values. Significance of effects was assessed at 0.05 level. Statistical analyses were done using SAS (SAS Institute Inc. version 9.4).

Results

Of the 154,804 postmenopausal women included in the analytical sample, 6,131 (4%) were diagnosed with breast cancer during follow-up. The women included in the study were, on average, 60.1 years old (ranging from 40 to 71 years). Those who developed breast cancer were followed-up for 5.4 years, on average (standard deviation [SD]=3.2 years), while cancer-free women were followed for 10.7 years (SD = 2.0 years). Table 1 presents the distributions of patients’ characteristics at baseline by cancer diagnosis, and by greenspace percentage, buffer 1000m, respectively. As compared to women without breast cancer, women with breast cancer were on average older (60.6 vs 60.1 years, p for difference < 0.0001), had higher BMI (27.6 vs 27.2 kg/m², p for difference < 0.0001), had higher proportion of nulliparous (70.1% vs 16.2%) or lower proportion of having a child before 25 years of age (38.1% vs 40.1% p = 0.001), were more likely to have reported currently using postmenopausal hormone therapy at enrollment (11.6% vs. 8.9%, p < 0.001), and having a family history of breast cancer (11.1% vs. 7.2%, p < 0.001) (Table 1).

Download:

Table 1. Characteristics of study participants at baseline, by the quartiles of greenspace percentage, buffer 1000m and by breast cancer status.

https://doi.org/10.1371/journal.pone.0334744.t001

The average greenness indicators were 45.68% (range 4.96–99.18%) for greenspace percentage, buffer 1000m; 35.77% (range 0.23–99.18%) for greenspace percentage, buffer 300m; 41.46% (range 0–100%) for natural environment percentage, buffer 1000m; 26.72% (range 0–100%) for natural environment percentage, buffer 300m (S1 Table).

Associations of greenness with breast cancer risk

For each 0.1 unit increase in NDVI, we found a 2.6% increase in BCa risk (Hazard ratio [HR]=1.03, 95%CI 1.00–1.05, Table 2). Women with higher exposure levels had greater risk of breast cancer (HR for 2^{^nd} vs 1^{^st} quartile = 1.10, 95% CI 1.01–1.21, HR for 3^{^rd} vs 1^{^st} quartile = 1.10, 95% CI 1.00–1.99, HR for 4^{^th} vs 1^{^st} quartile = 1.13, 95% CI 1.04–1.24, p for trend = 0.012) (Table 2). Our findings indicated no other associations of any other greenness measures with breast cancer risk.

Download:

Table 2. Associations of greenness measures with breast cancer risk.

https://doi.org/10.1371/journal.pone.0334744.t002

Interactions of greenness with PM₁₀ in relation to breast cancer risk

As reported previously in this sample [31], PM₁₀ was positively associated with breast cancer risk. We found a significant interaction effect between NDVI and cumulative average PM₁₀ exposure, modeled as continuous variable, indicating a stronger association of cumulative average PM₁₀ with breast cancer risk at lower levels of NDVI (1^{^st} NDVI quartile: HR per 10 µg/m^³ = 3.03, 95% CI 1.98–4.64) compared to higher levels of NDVI (4^{^th} NDVI quartile: HR per 10 µg/m^³ = 1.52, 95% CI 1.16–1.98) (p < 0.001, Fig 2). While the interactions between cumulative average PM₁₀ and several greenness metrics (greenspace percentage within 1000m, greenspace percentage within 300m, and natural environment percentage within 300m were statistically significant, there was no consistent or dose-dependent trend in the association of cumulative average PM₁₀ and breast cancer risk across greenness levels (Fig 2). No other significant interactions between air pollution (on a continuous scale) and greenness measures were found in these analyses (Figs 3 and 4).

Download:

Fig 2. Association of cumulative average PM₁₀ (continuous measure, per 10 µg/m³) with breast cancer risk, by the quartiles of the greenness measures, (hazard ratios and 95% confidence intervals)^a.

(A) without and (B) with 2-year air pollution exposure lag. NDVI – normalized difference vegetation index; PM₁₀ – particulate matter≤10 µm in diameter_; Q – quartiles; ^{^a}Risk estimates adjusted for age, body mass index, race, age at menopause, age at menarche, parity/age at first birth, postmenopausal hormone use, family history of breast cancer, alcohol consumption, and smoking; p-interaction: P for interaction between air pollutant measure and greenness measure, with both variables modeled as continuous.

https://doi.org/10.1371/journal.pone.0334744.g002

Download:

Fig 3. Association of 2007 PM₁₀ (continuous measure, per 10 µg/m³) with breast cancer risk, by the quartiles of the greenness measures, (hazard ratios and 95% confidence intervals)^a.

(A) without and (B) with 2-year air pollution exposure lag. NDVI – normalized difference vegetation index; PM₁₀ – particulate matter≤10 µm in diameter_; Q – quartiles; ^{^a}Risk estimates adjusted for age, body mass index, race, age at menopause, age at menarche, parity/age at first birth, postmenopausal hormone use, family history of breast cancer, alcohol consumption, and smoking; p-interaction: P for interaction between air pollutant measure and greenness measure, with both variables modeled as continuous.

https://doi.org/10.1371/journal.pone.0334744.g003

Download:

Fig 4. Association of 2010 PM₁₀ (continuous measure, per 10 µg/m³) with breast cancer risk, by the quartiles of the greenness measures, (hazard ratios and 95% confidence intervals)^a.

(A) without and (B) with 2-year air pollution exposure lag. NDVI – normalized difference vegetation index; PM₁₀ – particulate matter≤10 µm in diameter_; Q – quartiles; ^{^a}Risk estimates adjusted for age, body mass index, race, age at menopause, age at menarche, parity/age at first birth, postmenopausal hormone use, family history of breast cancer, alcohol consumption, and smoking; p-interaction: P for interaction between air pollutant measure and greenness measure, with both variables modeled as continuous.

https://doi.org/10.1371/journal.pone.0334744.g004

When PM₁₀ was modeled as quartiles, we found no associations of 2007 PM₁₀ or 2010 PM₁₀ with breast cancer risk at any of the greenness levels (Tables 3 and 4, respectively). However, women who were in the 4^{^th} quartile for cumulative average PM₁₀ exposure had significantly higher risk of breast cancer as compared to those in the lowest quartile within almost all strata of greenness measures (Table 5). For example, for women in the 1^{^st}, 2^{^nd}, and 3^{^rd} quartiles of mean NDVI we observed 32%, 37%, and 19% increase in breast cancer risk, for women who were in the 4^{^th} PM₁₀ quartile as compared to those who were in the 1^{^st} PM₁₀ quartile (1^{^st} NDVI quartile: HR for 4^{^th} vs. 1^{^st} PM₁₀ quartile = 1.32, 95% CI 1.09–1.59; 2^{^nd} NDVI quartile: HR = 1.37, 95% CI 1.13–1.66; 3^{^rd} NDVI quartile: HR = 1.19, 95% CI 1.00–1.67) (Table 5). A marginally significant increase in the risk of breast cancer with increased air pollution exposure was seen among women in the 4^{^th} quartile of mean NDVI exposure (HR for 4^{^th} vs 1^{^st} PM₁₀ quartile = 1.13, 95% 0.95–1.36) (Table 5).

Download:

Table 3. Association of quartiles of 2007 PM₁₀ with breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals)^a.

https://doi.org/10.1371/journal.pone.0334744.t003

Download:

Table 4. Association of quartiles of 2010 PM₁₀ with breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals)^a.

https://doi.org/10.1371/journal.pone.0334744.t004

Download:

Table 5. Association of quartiles of cumulative average PM₁₀ with breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals)^a.

https://doi.org/10.1371/journal.pone.0334744.t005

Secondary analyses

Similar to findings from the main analysis, in the lagged analysis we found that the association of cumulative average PM₁₀ with breast cancer risk remained stronger at lower NDVI levels as compared to higher NDVI levels (p = 0.002) (Fig 2), and that the cumulative average PM₁₀ was positively associated with breast cancer risk within most of the greenness strata (Table 5).

The association between NDVI and invasive breast cancer became stronger as compared to the main analysis (HR per 0.1-unit NDVI change = 1.04, 95% CI 1.02–1.06) (S2 Table). There were no other associations found between greenness measures and invasive breast cancer risk.

Lower levels of NDVI had stronger association of cumulative average PM₁₀ with invasive breast cancer risk (for 1^{^st} NDVI quartile: HR per 10 µg/m^³ = 3.10, 95% CI 1.94–4.95) compared to higher levels of NDVI (for 4^{^th} NDVI quartile: HR per 10 µg/m^³ = 1.66, 95% CI 1.24–2.21) (S3 Table). Similar to the overall analyses, there was not a clear trend for the association of cumulative average PM₁₀ with invasive breast cancer risk for the other greenness measures. No other interactions were significant.

When PM₁₀ was modeled as quartiles (S4, S5 and S6 Tables), women who were in the 4^{^th} quartile for cumulative average PM₁₀ exposure remained at higher risk of invasive breast cancer as compared to those who were in the 1^{^st} PM₁₀ quartile within almost all strata of greenness measures (S6 Table).

In the lagged analysis, the association between cumulative average PM₁₀ and invasive breast cancer remained stronger for lower levels of NDVI (1^{^st} NDVI quartile: HR per 10 µg/m^³ = 2.17, 95% CI 1.27–3.70) compared to higher NDVI levels (4^{^th} NDVI quartile: HR per 10 µg/m^³ = 1.58, 95% CI 1.14–2.18). There were no other interactions between any other PM₁₀ and greenness measures (S3 Table). When PM₁₀ was modeled as quartiles, cumulative average PM₁₀ was positively associated with invasive breast cancer risk in most of the greenness strata (S6 Table). A trend of higher breast cancer risk for women in the 4^{^th} vs 1^{^st} quartile for cumulative average PM₁₀ exposure was observed within each of the greenness levels; however, not all were statistically significant (S6 Table).

Discussion

Our findings indicated that NDVI was positively associated with postmenopausal breast cancer risk. The association between cumulative average PM₁₀ and breast cancer risk was stronger among individuals residing in areas with lower NDVI levels compared to those in areas with higher NDVI. Greenness percentage and natural environment percentage were not associated with breast cancer risk.

Positive associations of NDVI with breast cancer were found in some studies [24], but not all studies [20,22,23] that investigated the association of greenness with breast cancer risk. These inconsistencies between reported associations between greenness and breast cancer risk, including from our study, may also be due to the differences in study designs, geographic context, exposure assessment methods, and participants characteristics. Consistent with our findings, a study conducted by O’Callaghan-Gordo et al. [24] found that on average NDVI within the 300m-500m buffer was associated with increased risk of breast cancer (Odds ratio [OR] per 1-Interquartile range [IQR]=1.17, 95% CI 1.02–1.34) [24]. In a study including 6,500 women with breast cancer [22], the authors found that the risk of breast cancer was reduced by 4% for a 1-IQR increase of NDVI (HR per 1-IQR = 0.96, 95% CI 0.92-0.99); however, they did not account for risk factors for breast cancer nor examined the associations separately by menopausal status. Hart et al. [23] found that women living in the 5^{^th} quintile of cumulative average NDVI had lower risk of breast cancer compared to those in the 1^{^st} quintile, adjusting for a wide range of breast cancer risk factors (HR = 0.87, 95% CI 0.75–1.01, p for trend = 0.02). However, the cohort was limited to specific sub-population of women (female nurses), which may limit the generalizability to more diverse populations. In addition, Villeneuve et al. [22] and James et al. [23] used prospective cohort designs in large populations in Canada and the United States, respectively, assessing long-term residential greenness using satellite-derived NDVI. In contrast, O’Callaghan et al. [20] and O’Callaghan-Gordo et al. [24] conducted case-control studies within Spanish populations, with the former assessing proximity to green spaces rather than NDVI.

The use of different units for the greenness measures (e.g., percent of greenspace and natural environment vs. NDVI measured using an index measure), which might explain why NDVI might be associated with breast cancer risk in contrast to greenspace percentage and natural environment percentage. In addition, the choice of buffer distance might represent different exposure pathways, which a shorter distance capturing nearby and frequent contacts with greenness [33]. Finally, NDVI focuses on the health and quantity of vegetation while the other greenness measures included in these analyses are reflective of one area covered by green space.

To the best of our knowledge, no other study looked at associations of PM₁₀ with postmenopausal breast cancer by levels of NDVI. We found that the association of cumulative average PM₁₀ with breast cancer risk was stronger at lower levels of NDVI, suggesting the potential modifying effect of greenness on the association of PM₁₀ and breast cancer risk. Possible mechanisms through which greenness might modify this association include reduction in the level of pollutants, especially PM, and increase in pollutants disperse [41–43].

Among postmenopausal women, our study is the largest to date to examine the interaction between greenness and air pollution in relation to breast cancer risk. UK Biobank is a large population-based prospective cohort with rigorous protocols, regular updates of health outcomes, and valid methods for estimating the exposure measures, such as air pollution and greenness. Several limitations need to be considered. Only data from the baseline assessment for greenspace percentage, natural environment percentage, and covariates were used in analyses; however, repeated measures from baseline and follow-up assessments were highly correlated. Greenspace percentage was not available for participants outside England, and NDVI was not available for northern England and Scotland, which might affect the generalizability of the observed associations. As the participant’s residential address at the time of enrollment in the UK Biobank was used to calculate the concentrations of air, we could have been potentially misclassifying the participants based on their exposure due to other activities away from their residential address. However, other studies suggest that similar effects were observed when analyses included both time at home and work as compared to analyses including only time at home [44,45]. Next, despite the difference in the average follow-up time between breast cancer cases (5.4 years) and women without breast cancer (10.7 years), the proportion of women with available data on each of the air pollution measures was similar between the two groups (2007 PM₁₀: 99.5% for both cases and controls; 2010 PM₁₀: 92.5% for cases and 94.6% for controls; cumulative average PM₁₀: 99.7% for both cases and controls). Additionally, 92.5% of the controls and 94.7% of the cases had the cumulative average PM₁₀ estimate based on both 2007 PM₁₀ and 2010 PM₁₀ measures. Finally, we could not differentiate among breast cancer subtypes because the data was not available in the UK Biobank. As breast cancer subtypes may have distinct etiologies and environmental risk factors, future studies with access to richer clinical data are needed to explore subtype-specific associations with air pollution and greenness exposure.

Conclusion

In this study, we found a positive association between NDVI, a common measure of greenness, and postmenopausal breast cancer risk. Additionally, we observed a significant interaction between cumulative average PM₁₀ and NDVI, where the association between PM₁₀ exposure and breast cancer risk was stronger among women living in areas with lower greenness compared to those in greener environments. These findings suggest that greenness may not have a uniformly protective effect and that its role in modifying the health impacts of air pollution may be more complex than previously thought.

Importantly, our results contribute to the very limited body of evidence examining how environmental exposures such as greenness and air pollution interact in relation to breast cancer risk, particularly among postmenopausal women. The differential associations observed across various greenness metrics also underscore the need for careful selection and interpretation of environmental exposure indicators in future research.

Further studies are warranted to replicate these findings in diverse populations and geographic settings, and to explore underlying biological mechanisms. A better understanding of how greenness and air pollution jointly influence breast cancer risk could inform public health strategies aimed at reducing environmental risk factors and promoting healthier living environments for women.

Supporting information

S1 Table. Distribution of greenness and PM₁₀ measures in the study sample.

https://doi.org/10.1371/journal.pone.0334744.s001

(PDF)

S2 Table. Associations of greenness measures with invasive breast cancer risk.

https://doi.org/10.1371/journal.pone.0334744.s002

(PDF)

S3 Table. Association of PM₁₀ (continuous measure, per 10 µg/m³) with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

https://doi.org/10.1371/journal.pone.0334744.s003

(PDF)

S4 Table. Association of quartiles of 2007 PM₁₀ with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

https://doi.org/10.1371/journal.pone.0334744.s004

(PDF)

S5 Table. Association of quartiles of 2010 PM₁₀ with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

https://doi.org/10.1371/journal.pone.0334744.s005

(PDF)

S6 Table. Association of quartiles of cumulative average PM₁₀ with breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

https://doi.org/10.1371/journal.pone.0334744.s006

(PDF)

References

1. Villeneuve PJ, Jerrett M, Su JG, Weichenthal S, Sandler DP. Association of residential greenness with obesity and physical activity in a US cohort of women. Environ Res. 2018;160:372–84. pmid:29059619
- View Article
- PubMed/NCBI
- Google Scholar
2. Sarkar C. Residential greenness and adiposity: findings from the UK Biobank. Environ Int. 2017;106:1–10. pmid:28551493
- View Article
- PubMed/NCBI
- Google Scholar
3. James P, Banay RF, Hart JE, Laden F. A review of the health benefits of greenness. Curr Epidemiol Rep. 2015;2(2):131–42. pmid:26185745
- View Article
- PubMed/NCBI
- Google Scholar
4. Ai H, Zhang X, Zhou Z. The impact of greenspace on air pollution: empirical evidence from China. Ecol Indic. 2023;146:109881.
- View Article
- Google Scholar
5. Akopov AS, Beklaryan LA, Saghatelyan AK. Agent-based modelling of interactions between air pollutants and greenery using a case study of Yerevan, Armenia. Environ Modell Softw. 2019;116:7–25.
- View Article
- Google Scholar
6. World Health Organization. Regional Office for E. Urban green space interventions and health: a review of impacts and effectiveness. Copenhagen: World Health Organization; 2017.
7. Urban green spaces and health. Copenhagen: WHO Regional Office for Europe AO. 2019. https://www.who.int/europe/publications/i/item/WHO-EURO-2016-3352-43111-60341
- View Article
- Google Scholar
8. Donovan GH, Michael YL, Butry DT, Sullivan AD, Chase JM. Urban trees and the risk of poor birth outcomes. Health Place. 2011;17(1):390–3. pmid:21106432
- View Article
- PubMed/NCBI
- Google Scholar
9. Takano T, Nakamura K, Watanabe M. Urban residential environments and senior citizens’ longevity in megacity areas: the importance of walkable green spaces. J Epidemiol Community Health. 2002;56(12):913–8. pmid:12461111
- View Article
- PubMed/NCBI
- Google Scholar
10. Sugiyama T, Leslie E, Giles-Corti B, Owen N. Associations of neighbourhood greenness with physical and mental health: do walking, social coherence and local social interaction explain the relationships? J Epidemiol Community Health. 2008;62(5):e9. pmid:18431834
- View Article
- PubMed/NCBI
- Google Scholar
11. Ward Thompson C, Roe J, Aspinall P, Mitchell R, Clow A, Miller D. More green space is linked to less stress in deprived communities: evidence from salivary cortisol patterns. Landsc Urban Plan. 2012;105(3):221–9.
- View Article
- Google Scholar
12. Roe JJ, Thompson CW, Aspinall PA, Brewer MJ, Duff EI, Miller D, et al. Green space and stress: evidence from cortisol measures in deprived urban communities. Int J Environ Res Public Health. 2013;10(9):4086–103. pmid:24002726
- View Article
- PubMed/NCBI
- Google Scholar
13. Iyer HS, James P, Valeri L, Hart JE, Pernar CH, Mucci LA, et al. The association between neighborhood greenness and incidence of lethal prostate cancer: a prospective cohort study. Environ Epidemiol. 2020;4(2):e091. pmid:32656487
- View Article
- PubMed/NCBI
- Google Scholar
14. Kim S, Kim H, Lee J-T. Interactions between ambient air particles and greenness on cause-specific mortality in seven Korean Metropolitan Cities, 2008-2016. Int J Environ Res Public Health. 2019;16(10):1866. pmid:31137847
- View Article
- PubMed/NCBI
- Google Scholar
15. Ji JS, Zhu A, Lv Y, Shi X. Interaction between residential greenness and air pollution mortality: analysis of the Chinese Longitudinal Healthy Longevity Survey. Lancet Planet Health. 2020;4(3):e107–15. pmid:32220672
- View Article
- PubMed/NCBI
- Google Scholar
16. Crouse DL, Pinault L, Balram A, Brauer M, Burnett RT, Martin RV, et al. Complex relationships between greenness, air pollution, and mortality in a population-based Canadian cohort. Environ Int. 2019;128:292–300. pmid:31075749
- View Article
- PubMed/NCBI
- Google Scholar
17. Vienneau D, de Hoogh K, Faeh D, Kaufmann M, Wunderli JM, Röösli M, et al. More than clean air and tranquillity: residential green is independently associated with decreasing mortality. Environ Int. 2017;108:176–84. pmid:28863390
- View Article
- PubMed/NCBI
- Google Scholar
18. Yeager R, Riggs DW, DeJarnett N, Tollerud DJ, Wilson J, Conklin DJ, et al. Association between residential greenness and cardiovascular disease risk. J Am Heart Assoc. 2018;7(24):e009117. pmid:30561265
- View Article
- PubMed/NCBI
- Google Scholar
19. James P, Hart JE, Banay RF, Laden F. Exposure to greenness and mortality in a Nationwide Prospective Cohort Study of Women. Environ Health Perspect. 2016;124(9):1344–52. pmid:27074702
- View Article
- PubMed/NCBI
- Google Scholar
20. O’Callaghan C, Valentin A, Cirach M, Castaño-Vinyals G, Gràcia-Lavedan E, Echezarreta N, et al. Natural outdoors environments and prostate and breast cancer risk: a case-control in Spain. Annual meeting International Society for Environmental Epidemiology. Environmental Health Perspectives; 2016.
21. Demoury C, Thierry B, Richard H, Sigler B, Kestens Y, Parent M-E. Residential greenness and risk of prostate cancer: a case-control study in Montreal, Canada. Environ Int. 2017;98:129–36. pmid:27823799
- View Article
- PubMed/NCBI
- Google Scholar
22. Villeneuve P, Mansouri R, Crouse D, To T, Wall C, Miller A. Greenness, obesity and incident breast cancer: evidence from the Canadian National Breast Screening Study. Environ Epidemiol. 2019;3:416.
- View Article
- Google Scholar
23. James P, Hart JE, Bertrand K, Bezold C, VoPham T, Tamimi R, et al. Greenness and breast cancer in a US-Based Nationwide Prospective Cohort Study. ISEE Conference Abstracts. 2018;2017(1).
- View Article
- Google Scholar
24. O’Callaghan-Gordo C, Kogevinas M, Cirach M, Castaño-Vinyals G, Aragonés N, Delfrade J, et al. Residential proximity to green spaces and breast cancer risk: The multicase-control study in Spain (MCC-Spain). Int J Hyg Environ Health. 2018;221(8):1097–106. pmid:30076044
- View Article
- PubMed/NCBI
- Google Scholar
25. Fong KC, Hart JE, James P. A review of epidemiologic studies on greenness and health: updated literature through 2017. Curr Environ Health Rep. 2018;5(1):77–87. pmid:29392643
- View Article
- PubMed/NCBI
- Google Scholar
26. Thiering E, Markevych I, Brüske I, Fuertes E, Kratzsch J, Sugiri D, et al. Associations of residential long-term air pollution exposures and satellite-derived greenness with insulin resistance in German Adolescents. Environ Health Perspect. 2016;124(8):1291–8. pmid:26863688
- View Article
- PubMed/NCBI
- Google Scholar
27. Im SG, Choi H, Jeon Y-H, Song M-K, Kim W, Woo J-M. Comparison of effect of two-hour exposure to forest and urban environments on cytokine, anti-oxidant, and stress levels in young adults. Int J Environ Res Public Health. 2016;13(7):625. pmid:27347982
- View Article
- PubMed/NCBI
- Google Scholar
28. Li Q, Kobayashi M, Kumeda S, Ochiai T, Miura T, Kagawa T, et al. Effects of forest bathing on cardiovascular and metabolic parameters in middle-aged males. Evid Based Complement Alternat Med. 2016;2016:2587381. pmid:27493670
- View Article
- PubMed/NCBI
- Google Scholar
29. Martens DS, Nawrot TS. Ageing at the level of telomeres in association to residential landscape and air pollution at home and work: a review of the current evidence. Toxicol Lett. 2018;298:42–52. pmid:29944903
- View Article
- PubMed/NCBI
- Google Scholar
30. Berrigan D, Berger NA. Geospatial approaches to energy balance and breast cancer. Springer International Publishing; 2019.
31. Smotherman C, Sprague B, Datta S, Braithwaite D, Qin H, Yaghjyan L. Association of air pollution with postmenopausal breast cancer risk in UK Biobank. Breast Cancer Res. 2023;25(1):83. pmid:37443054
- View Article
- PubMed/NCBI
- Google Scholar
32. UK Biobank. Protocol for a large-scale prospective epidemiological resource. Date: March 21 AS, 2021. https://www.ukbiobank.ac.uk/media/gnkeyh2q/study-rationale.pdf
- View Article
- Google Scholar
33. Department for Communities and Local Government. Generalised land use database statistics for England 2005 (enhanced basemap). London: DCLG.
34. Morton D, Rowland C, Wood C, Meek L, Marston C, Smith G, et al. Final Report for LCM2007 - the new UK land cover map. Countryside Survey Technical Report No 11/07. 2011.
35. Sarkar C, Gallacher J, Webster CJ. Morphometric analysis of the built environment in UL Biobank: Data analyses and specification manual. UK: UK Biobank, CTSU, Oxford University; 2014.
36. Vienneau D, de Hoogh K, Bechle MJ, Beelen R, van Donkelaar A, Martin RV, et al. Western European land use regression incorporating satellite- and ground-based measurements of NO2 and PM10. Environ Sci Technol. 2013;47(23):13555–64. pmid:24156783
- View Article
- PubMed/NCBI
- Google Scholar
37. Eeftens M, Beelen R, de Hoogh K, Bellander T, Cesaroni G, Cirach M, et al. Development of Land Use Regression models for PM(2.5), PM(2.5) absorbance, PM(10) and PM(coarse) in 20 European study areas; results of the ESCAPE project. Environ Sci Technol. 2012;46(20):11195–205. pmid:22963366
- View Article
- PubMed/NCBI
- Google Scholar
38. Beelen R, Hoek G, Vienneau D, Eeftens M, Dimakopoulou K, Pedeli X, et al. Development of NO2 and NOx land use regression models for estimating air pollution exposure in 36 study areas in Europe – The ESCAPE project. Atmospheric Environ. 2013;72:10–23.
- View Article
- Google Scholar
39. Cox DR. Regression Models and Life-Tables. J R Stat Soc Series B: Stat Methodol. 1972;34(2):187–202.
- View Article
- Google Scholar
40. Therneau T, Grambsch P. Modeling survival data: extending the cox model. 2000.
41. Nowak DJ, Crane DE, Stevens JC. Air pollution removal by urban trees and shrubs in the United States. Urban Forestry Urban Greening. 2006;4(3–4):115–23.
- View Article
- Google Scholar
42. Nieuwenhuijsen MJ, Kruize H, Gidlow C, Andrusaityte S, Antó JM, Basagaña X, et al. Positive health effects of the natural outdoor environment in typical populations in different regions in Europe (PHENOTYPE): a study programme protocol. BMJ Open. 2014;4(4):e004951. pmid:24740979
- View Article
- PubMed/NCBI
- Google Scholar
43. Markevych I, Schoierer J, Hartig T, Chudnovsky A, Hystad P, Dzhambov AM, et al. Exploring pathways linking greenspace to health: Theoretical and methodological guidance. Environ Res. 2017;158:301–17. pmid:28672128
- View Article
- PubMed/NCBI
- Google Scholar
44. Ragettli MS, Phuleria HC, Tsai M-Y, Schindler C, de Nazelle A, Ducret-Stich RE, et al. The relevance of commuter and work/school exposure in an epidemiological study on traffic-related air pollution. J Expo Sci Environ Epidemiol. 2015;25(5):474–81. pmid:25492241
- View Article
- PubMed/NCBI
- Google Scholar
45. Dibben C, Clemens T. Place of work and residential exposure to ambient air pollution and birth outcomes in Scotland, using geographically fine pollution climate mapping estimates. Environ Res. 2015;140:535–41. pmid:26005952
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Villeneuve PJ, Jerrett M, Su JG, Weichenthal S, Sandler DP. Association of residential greenness with obesity and physical activity in a US cohort of women. Environ Res. 2018;160:372–84. pmid:29059619
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Sarkar C. Residential greenness and adiposity: findings from the UK Biobank. Environ Int. 2017;106:1–10. pmid:28551493
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. James P, Banay RF, Hart JE, Laden F. A review of the health benefits of greenness. Curr Epidemiol Rep. 2015;2(2):131–42. pmid:26185745
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Ai H, Zhang X, Zhou Z. The impact of greenspace on air pollution: empirical evidence from China. Ecol Indic. 2023;146:109881.
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref5] 5. Akopov AS, Beklaryan LA, Saghatelyan AK. Agent-based modelling of interactions between air pollutants and greenery using a case study of Yerevan, Armenia. Environ Modell Softw. 2019;116:7–25.
View Article
Google Scholar

[17] View Article

[18] Google Scholar

[ref6] 6. World Health Organization. Regional Office for E. Urban green space interventions and health: a review of impacts and effectiveness. Copenhagen: World Health Organization; 2017.

[ref7] 7. Urban green spaces and health. Copenhagen: WHO Regional Office for Europe AO. 2019. https://www.who.int/europe/publications/i/item/WHO-EURO-2016-3352-43111-60341
View Article
Google Scholar

[21] View Article

[22] Google Scholar

[ref8] 8. Donovan GH, Michael YL, Butry DT, Sullivan AD, Chase JM. Urban trees and the risk of poor birth outcomes. Health Place. 2011;17(1):390–3. pmid:21106432
View Article
PubMed/NCBI
Google Scholar

[24] View Article

[25] PubMed/NCBI

[26] Google Scholar

[ref9] 9. Takano T, Nakamura K, Watanabe M. Urban residential environments and senior citizens’ longevity in megacity areas: the importance of walkable green spaces. J Epidemiol Community Health. 2002;56(12):913–8. pmid:12461111
View Article
PubMed/NCBI
Google Scholar

[28] View Article

[29] PubMed/NCBI

[30] Google Scholar

[ref10] 10. Sugiyama T, Leslie E, Giles-Corti B, Owen N. Associations of neighbourhood greenness with physical and mental health: do walking, social coherence and local social interaction explain the relationships? J Epidemiol Community Health. 2008;62(5):e9. pmid:18431834
View Article
PubMed/NCBI
Google Scholar

[32] View Article

[33] PubMed/NCBI

[34] Google Scholar

[ref11] 11. Ward Thompson C, Roe J, Aspinall P, Mitchell R, Clow A, Miller D. More green space is linked to less stress in deprived communities: evidence from salivary cortisol patterns. Landsc Urban Plan. 2012;105(3):221–9.
View Article
Google Scholar

[36] View Article

[37] Google Scholar

[ref12] 12. Roe JJ, Thompson CW, Aspinall PA, Brewer MJ, Duff EI, Miller D, et al. Green space and stress: evidence from cortisol measures in deprived urban communities. Int J Environ Res Public Health. 2013;10(9):4086–103. pmid:24002726
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref13] 13. Iyer HS, James P, Valeri L, Hart JE, Pernar CH, Mucci LA, et al. The association between neighborhood greenness and incidence of lethal prostate cancer: a prospective cohort study. Environ Epidemiol. 2020;4(2):e091. pmid:32656487
View Article
PubMed/NCBI
Google Scholar

[43] View Article

[44] PubMed/NCBI

[45] Google Scholar

[ref14] 14. Kim S, Kim H, Lee J-T. Interactions between ambient air particles and greenness on cause-specific mortality in seven Korean Metropolitan Cities, 2008-2016. Int J Environ Res Public Health. 2019;16(10):1866. pmid:31137847
View Article
PubMed/NCBI
Google Scholar

[47] View Article

[48] PubMed/NCBI

[49] Google Scholar

[ref15] 15. Ji JS, Zhu A, Lv Y, Shi X. Interaction between residential greenness and air pollution mortality: analysis of the Chinese Longitudinal Healthy Longevity Survey. Lancet Planet Health. 2020;4(3):e107–15. pmid:32220672
View Article
PubMed/NCBI
Google Scholar

[51] View Article

[52] PubMed/NCBI

[53] Google Scholar

[ref16] 16. Crouse DL, Pinault L, Balram A, Brauer M, Burnett RT, Martin RV, et al. Complex relationships between greenness, air pollution, and mortality in a population-based Canadian cohort. Environ Int. 2019;128:292–300. pmid:31075749
View Article
PubMed/NCBI
Google Scholar

[55] View Article

[56] PubMed/NCBI

[57] Google Scholar

[ref17] 17. Vienneau D, de Hoogh K, Faeh D, Kaufmann M, Wunderli JM, Röösli M, et al. More than clean air and tranquillity: residential green is independently associated with decreasing mortality. Environ Int. 2017;108:176–84. pmid:28863390
View Article
PubMed/NCBI
Google Scholar

[59] View Article

[60] PubMed/NCBI

[61] Google Scholar

[ref18] 18. Yeager R, Riggs DW, DeJarnett N, Tollerud DJ, Wilson J, Conklin DJ, et al. Association between residential greenness and cardiovascular disease risk. J Am Heart Assoc. 2018;7(24):e009117. pmid:30561265
View Article
PubMed/NCBI
Google Scholar

[63] View Article

[64] PubMed/NCBI

[65] Google Scholar

[ref19] 19. James P, Hart JE, Banay RF, Laden F. Exposure to greenness and mortality in a Nationwide Prospective Cohort Study of Women. Environ Health Perspect. 2016;124(9):1344–52. pmid:27074702
View Article
PubMed/NCBI
Google Scholar

[67] View Article

[68] PubMed/NCBI

[69] Google Scholar

[ref20] 20. O’Callaghan C, Valentin A, Cirach M, Castaño-Vinyals G, Gràcia-Lavedan E, Echezarreta N, et al. Natural outdoors environments and prostate and breast cancer risk: a case-control in Spain. Annual meeting International Society for Environmental Epidemiology. Environmental Health Perspectives; 2016.

[ref21] 21. Demoury C, Thierry B, Richard H, Sigler B, Kestens Y, Parent M-E. Residential greenness and risk of prostate cancer: a case-control study in Montreal, Canada. Environ Int. 2017;98:129–36. pmid:27823799
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref22] 22. Villeneuve P, Mansouri R, Crouse D, To T, Wall C, Miller A. Greenness, obesity and incident breast cancer: evidence from the Canadian National Breast Screening Study. Environ Epidemiol. 2019;3:416.
View Article
Google Scholar

[76] View Article

[77] Google Scholar

[ref23] 23. James P, Hart JE, Bertrand K, Bezold C, VoPham T, Tamimi R, et al. Greenness and breast cancer in a US-Based Nationwide Prospective Cohort Study. ISEE Conference Abstracts. 2018;2017(1).
View Article
Google Scholar

[79] View Article

[80] Google Scholar

[ref24] 24. O’Callaghan-Gordo C, Kogevinas M, Cirach M, Castaño-Vinyals G, Aragonés N, Delfrade J, et al. Residential proximity to green spaces and breast cancer risk: The multicase-control study in Spain (MCC-Spain). Int J Hyg Environ Health. 2018;221(8):1097–106. pmid:30076044
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref25] 25. Fong KC, Hart JE, James P. A review of epidemiologic studies on greenness and health: updated literature through 2017. Curr Environ Health Rep. 2018;5(1):77–87. pmid:29392643
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref26] 26. Thiering E, Markevych I, Brüske I, Fuertes E, Kratzsch J, Sugiri D, et al. Associations of residential long-term air pollution exposures and satellite-derived greenness with insulin resistance in German Adolescents. Environ Health Perspect. 2016;124(8):1291–8. pmid:26863688
View Article
PubMed/NCBI
Google Scholar

[90] View Article

[91] PubMed/NCBI

[92] Google Scholar

[ref27] 27. Im SG, Choi H, Jeon Y-H, Song M-K, Kim W, Woo J-M. Comparison of effect of two-hour exposure to forest and urban environments on cytokine, anti-oxidant, and stress levels in young adults. Int J Environ Res Public Health. 2016;13(7):625. pmid:27347982
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref28] 28. Li Q, Kobayashi M, Kumeda S, Ochiai T, Miura T, Kagawa T, et al. Effects of forest bathing on cardiovascular and metabolic parameters in middle-aged males. Evid Based Complement Alternat Med. 2016;2016:2587381. pmid:27493670
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref29] 29. Martens DS, Nawrot TS. Ageing at the level of telomeres in association to residential landscape and air pollution at home and work: a review of the current evidence. Toxicol Lett. 2018;298:42–52. pmid:29944903
View Article
PubMed/NCBI
Google Scholar

[102] View Article

[103] PubMed/NCBI

[104] Google Scholar

[ref30] 30. Berrigan D, Berger NA. Geospatial approaches to energy balance and breast cancer. Springer International Publishing; 2019.

[ref31] 31. Smotherman C, Sprague B, Datta S, Braithwaite D, Qin H, Yaghjyan L. Association of air pollution with postmenopausal breast cancer risk in UK Biobank. Breast Cancer Res. 2023;25(1):83. pmid:37443054
View Article
PubMed/NCBI
Google Scholar

[107] View Article

[108] PubMed/NCBI

[109] Google Scholar

[ref32] 32. UK Biobank. Protocol for a large-scale prospective epidemiological resource. Date: March 21 AS, 2021. https://www.ukbiobank.ac.uk/media/gnkeyh2q/study-rationale.pdf
View Article
Google Scholar

[111] View Article

[112] Google Scholar

[ref33] 33. Department for Communities and Local Government. Generalised land use database statistics for England 2005 (enhanced basemap). London: DCLG.

[ref34] 34. Morton D, Rowland C, Wood C, Meek L, Marston C, Smith G, et al. Final Report for LCM2007 - the new UK land cover map. Countryside Survey Technical Report No 11/07. 2011.

[ref35] 35. Sarkar C, Gallacher J, Webster CJ. Morphometric analysis of the built environment in UL Biobank: Data analyses and specification manual. UK: UK Biobank, CTSU, Oxford University; 2014.

[ref36] 36. Vienneau D, de Hoogh K, Bechle MJ, Beelen R, van Donkelaar A, Martin RV, et al. Western European land use regression incorporating satellite- and ground-based measurements of NO2 and PM10. Environ Sci Technol. 2013;47(23):13555–64. pmid:24156783
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref37] 37. Eeftens M, Beelen R, de Hoogh K, Bellander T, Cesaroni G, Cirach M, et al. Development of Land Use Regression models for PM(2.5), PM(2.5) absorbance, PM(10) and PM(coarse) in 20 European study areas; results of the ESCAPE project. Environ Sci Technol. 2012;46(20):11195–205. pmid:22963366
View Article
PubMed/NCBI
Google Scholar

[121] View Article

[122] PubMed/NCBI

[123] Google Scholar

[ref38] 38. Beelen R, Hoek G, Vienneau D, Eeftens M, Dimakopoulou K, Pedeli X, et al. Development of NO2 and NOx land use regression models for estimating air pollution exposure in 36 study areas in Europe – The ESCAPE project. Atmospheric Environ. 2013;72:10–23.
View Article
Google Scholar

[125] View Article

[126] Google Scholar

[ref39] 39. Cox DR. Regression Models and Life-Tables. J R Stat Soc Series B: Stat Methodol. 1972;34(2):187–202.
View Article
Google Scholar

[128] View Article

[129] Google Scholar

[ref40] 40. Therneau T, Grambsch P. Modeling survival data: extending the cox model. 2000.

[ref41] 41. Nowak DJ, Crane DE, Stevens JC. Air pollution removal by urban trees and shrubs in the United States. Urban Forestry Urban Greening. 2006;4(3–4):115–23.
View Article
Google Scholar

[132] View Article

[133] Google Scholar

[ref42] 42. Nieuwenhuijsen MJ, Kruize H, Gidlow C, Andrusaityte S, Antó JM, Basagaña X, et al. Positive health effects of the natural outdoor environment in typical populations in different regions in Europe (PHENOTYPE): a study programme protocol. BMJ Open. 2014;4(4):e004951. pmid:24740979
View Article
PubMed/NCBI
Google Scholar

[135] View Article

[136] PubMed/NCBI

[137] Google Scholar

[ref43] 43. Markevych I, Schoierer J, Hartig T, Chudnovsky A, Hystad P, Dzhambov AM, et al. Exploring pathways linking greenspace to health: Theoretical and methodological guidance. Environ Res. 2017;158:301–17. pmid:28672128
View Article
PubMed/NCBI
Google Scholar

[139] View Article

[140] PubMed/NCBI

[141] Google Scholar

[ref44] 44. Ragettli MS, Phuleria HC, Tsai M-Y, Schindler C, de Nazelle A, Ducret-Stich RE, et al. The relevance of commuter and work/school exposure in an epidemiological study on traffic-related air pollution. J Expo Sci Environ Epidemiol. 2015;25(5):474–81. pmid:25492241
View Article
PubMed/NCBI
Google Scholar

[143] View Article

[144] PubMed/NCBI

[145] Google Scholar

[ref45] 45. Dibben C, Clemens T. Place of work and residential exposure to ambient air pollution and birth outcomes in Scotland, using geographically fine pollution climate mapping estimates. Environ Res. 2015;140:535–41. pmid:26005952
View Article
PubMed/NCBI
Google Scholar

[147] View Article

[148] PubMed/NCBI

[149] Google Scholar

Figures

Abstract

Introduction

Methods

Study population

Greenness assessment

Air pollution assessment

Covariates information

Statistical analyses

Results

Associations of greenness with breast cancer risk

Interactions of greenness with PM10 in relation to breast cancer risk

Secondary analyses

Discussion

Conclusion

Supporting information

S1 Table. Distribution of greenness and PM10 measures in the study sample.

S2 Table. Associations of greenness measures with invasive breast cancer risk.

S3 Table. Association of PM10 (continuous measure, per 10 µg/m3) with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

S4 Table. Association of quartiles of 2007 PM10 with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

S5 Table. Association of quartiles of 2010 PM10 with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

S6 Table. Association of quartiles of cumulative average PM10 with breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

References

Interactions of greenness with PM₁₀ in relation to breast cancer risk

S1 Table. Distribution of greenness and PM₁₀ measures in the study sample.

S3 Table. Association of PM₁₀ (continuous measure, per 10 µg/m³) with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

S4 Table. Association of quartiles of 2007 PM₁₀ with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

S5 Table. Association of quartiles of 2010 PM₁₀ with invasive breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).

S6 Table. Association of quartiles of cumulative average PM₁₀ with breast cancer risk, by the quartiles of the greenness measures, without and with 2-year air pollution exposure lag (hazard ratios and 95% confidence intervals).