Associations between Residential Greenness and Influenza Virus Infection in China: An Individual-Level, National, Cross-Sectional Study, 2010–2017

Background Residential greenness, a fundamental component of urban design, could contribute to the prevention of respiratory infections via several potential mechanisms.. However, the health benefits of greenness on influenza epidemics in real world are not as clear. Therefore, in the present study, we investigated the association between residential greenness exposure and influenza virus infection risk using a large and diverse cross-sectional dataset from the Chinese influenza surveillance system. Methods In this cross-sectional, associational study, we used information from influenza-like illness (ILI) patients who were tested for influenza from 2010 to 2017 from the Chinese influenza surveillance system. Residential greenness was assessed with the normalized difference vegetation index (NDVI) within a 250 m radius of the ILI residential addresses. Other environmental metrics included the mean air temperature; relative humidity; precipitation; wind speed; sunshine duration; and O ₃ and PM _2.5 concentrations. A series of logistic models were constructed to examine the associations between residential greenness exposure and the odds of influenza virus infection after adjusting for covariates such as individual age, gender, climate, air pollution and seasonality. Findings From 2010–2017, 3,131,881 ILI cases were tested for influenza, and 1,012,430 (32.3%) participants with detailed building-level residential addresses were included in this study. Overall, a protective effect of residential greenness on the risk of influenza virus infection was observed, with 2.6% lower odds of influenza virus infection per one-quartile increase in the NDVI (odds ratio (OR) = 0.974, 95% confidence interval (CI): 0.963–0.985, p < 0.001). However, the impact varied across the different subgroups. Stratified analyses indicated that the protective effects of residential greenness were strongest among adults aged ≥ 60 years (OR = 0.853, 95% CI: 0.814–0.894, p < 0.001), but among children aged 7–17 years (i.e., school-aged children), the association was positive (OR = 1.104, 95% CI: 1.079–1.129, p < 0.001). There were no protective effects at other city scales except in megacities (OR = 0.907, 95% CI: 0.886–0.930, p < 0.001). Similarly, the protective effects of residential greenness against the development of influenza were observed only during the influenza season, i.e., in spring (OR = 0.914, 95% CI: 0.893–0.936, p < 0.001) and winter (OR = 0.933, 95% CI: 0.915–0.952, p < 0.001), and in southern China (OR = 0.975, 95% CI: 0.963–0.988, p < 0.001). Residential greenness had protective effects against influenza B infection (OR = 0.888, 95% CI: 0.872–0.906, p < 0.001), but no such effect was observed for influenza A infection (OR = 1.022, 95% CI: 1.008–1.036). The results from the interaction analysis between covariates were consistent with the results from the stratified analyses, except when the age group interacted with geographic regions. The mediating effects of PM _2.5 and O ₃ exposure on the association between residential greenness exposure and the risk of influenza virus infection were 33.90% and 16.38%, respectively. Discussion The results highlight the benefits of well-designed green environments for influenza prevention. Given the rapid ageing and urbanization process in China, policies aimed at optimizing the allocation and design of green spaces might help reduce respiratory infection transmission.