TY - JOUR
T1 - Greenspace exposure and cancer incidence: A 27-year follow-up of the French GAZEL cohort
AU - Zare Sakhvidi, Mohammad Javad
AU - Yang, Jun
AU - Siemiatycki, Jack
AU - Dadvand, Payam
AU - de Hoogh, Kees
AU - Vienneau, Danielle
AU - Goldberg, Marcel
AU - Zins, Marie
AU - Lequy, Emeline
AU - Jacquemin, Bénédicte
N1 - Publisher Copyright:
© 2021 Elsevier B.V.
PY - 2021
Y1 - 2021
N2 - Background Greenspace exposure has been suggested to be associated with a range of health outcomes. The available evidence on the association of this exposure with cancer is still very scarce and inconsistent. Objectives We aimed to study the association between greenspace exposure and all-site and site-specific (prostate, breast, colorectal, bladder, lung, and malignant melanoma of skin) cancer incidence in the GAZEL cohort. Methods This study was based on over 27 years of follow-up (1989–2016) of 19,408 participants across France. We assessed the residential greenspace exposure within several buffers as well as residential proximity to green spaces (agricultural, urban, and forests) in each follow-up. We used time-dependent Cox models, controlling for time-varying personal and area-level variables, with different lags between exposure and outcome. Additional analysis was conducted according to the urban-rural residence of the participants' over follow-up. Results Over the 294,645 person-years of follow-up, we registered 4075 incident cases of cancer. We found an increase in the risk for all-sites cancer with an inter-quartile range increase of Normalized Difference in Vegetation Index across different buffers (hazard ratio (HR) of 1.08; 95% CI: 1.02, 1.14 for the 100 m buffer). We found a positive association of all-sites cancer with proximity to agricultural lands (HR: 1.03; 95% CI: 1.00, 1.05), and forests (HR:1.04; 95% CI: 1.00, 1.07), but not with urban green spaces. The cancer site-specific analyses suggested a protective role of greenspace for breast, lung, and colorectal cancers (e.g. breast cancer HR at 100 m buffer: 0.82; 95% CI: 0.69, 0.99). Non-significant associations were observed for prostate, bladder, and skin cancer. Stratified analyses based on urban, semi-urban, and rural classification did not suggest any differential pattern. Conclusion We identified an increased risk of all-site cancer with increased greenspace and proximity to agricultural lands and forests; whereas potential protective role of greenspace for breast cancer.
AB - Background Greenspace exposure has been suggested to be associated with a range of health outcomes. The available evidence on the association of this exposure with cancer is still very scarce and inconsistent. Objectives We aimed to study the association between greenspace exposure and all-site and site-specific (prostate, breast, colorectal, bladder, lung, and malignant melanoma of skin) cancer incidence in the GAZEL cohort. Methods This study was based on over 27 years of follow-up (1989–2016) of 19,408 participants across France. We assessed the residential greenspace exposure within several buffers as well as residential proximity to green spaces (agricultural, urban, and forests) in each follow-up. We used time-dependent Cox models, controlling for time-varying personal and area-level variables, with different lags between exposure and outcome. Additional analysis was conducted according to the urban-rural residence of the participants' over follow-up. Results Over the 294,645 person-years of follow-up, we registered 4075 incident cases of cancer. We found an increase in the risk for all-sites cancer with an inter-quartile range increase of Normalized Difference in Vegetation Index across different buffers (hazard ratio (HR) of 1.08; 95% CI: 1.02, 1.14 for the 100 m buffer). We found a positive association of all-sites cancer with proximity to agricultural lands (HR: 1.03; 95% CI: 1.00, 1.05), and forests (HR:1.04; 95% CI: 1.00, 1.07), but not with urban green spaces. The cancer site-specific analyses suggested a protective role of greenspace for breast, lung, and colorectal cancers (e.g. breast cancer HR at 100 m buffer: 0.82; 95% CI: 0.69, 0.99). Non-significant associations were observed for prostate, bladder, and skin cancer. Stratified analyses based on urban, semi-urban, and rural classification did not suggest any differential pattern. Conclusion We identified an increased risk of all-site cancer with increased greenspace and proximity to agricultural lands and forests; whereas potential protective role of greenspace for breast cancer.
KW - Cancer incidence Greenspace Environmental health Longitudinal study
U2 - 10.1016/j.scitotenv.2021.147553
DO - 10.1016/j.scitotenv.2021.147553
M3 - Article
SN - 0048-9697
VL - 787
JO - Science of the Total Environment
JF - Science of the Total Environment
M1 - 147553
ER -