Exposure to greenspace and cancer incidence, prevalence, and mortality: A systematic review and meta-analyses

We conducted a systematic review and meta-analysis of the available literature on the association between greenspace exposure and all-sites and site-specific cancer incidence, prevalence, and mortality in adults. We searched PubMed, Scopus, and Web of Science for original articles published, without language restriction until September 2021. We assessed the risk of bias in each study and the overall quality of evidence for exposure-outcome pairs that were reported in two or more studies. Out of the 18 included studies, cross-sectional studies were the most common study design (n = 8), and most of the studies were conducted in Europe (n = 8). In terms of risk of bias, the majority of cohorts (four out of six) and case-control studies (three out of four) were of good or very good quality, and cross-sectional studies were mostly (five out of eight) of poor quality. Outcomes (incidence, prevalence, mortality) on different cancer sites were reported: lung cancer (n = 9), prostate cancer (n = 4), breast cancer (n = 4), skin cancer (n = 3), colorectal cancer (n = 2), all-sites cancer (n = 2), brain cancer (n = 1), mouth and throat cancer (n = 1), and esophageal cancer (n = 1). The meta-analyses for the breast, lung, and prostate cancer incidence did not show statistically significant associations (for example for breast cancer: hazard ratio = 0.83; 95% confidence interval: 0.47-1.48). For skin cancer, the available evidence suggests that greenspace could be a potential risk factor. For the other cancers, the evidence was non-conclusive. The overall quality of evidence of all of the exposure-outcome pairs was very low. Given the wide confidence interval of the pooled estimates and very low quality of evidence, the findings should be interpreted with caution. Future large and longitudinal studies are needed to assess the potential association of greenspace exposure with cancers, considering types and quality of greenspace, evaluation of cancer sub-types, and adjustment for a sufficient set of covariates.