LAUSR

Introduction As the world becomes increasingly urbanized and human-nature contact declines, urban greenspace's impact on human health has garnered growing interest across academic disciplines. Various definitions and multiple indicators of greenspace have been utilized, with most studies finding an overall positive association between greenspace and health. Nevertheless, studies directly comparing how different greenspace indicators impact different disease types have been limited. Moreover, to verify the robustness of conclusions drawn, studies should compare multiple measures of greenspace across various spatial scales. Thus, a more comprehensive analysis is necessary to help inform future study design, especially in determining which greenspace indicators would be most useful in data-limited areas. Methods Chengdu, the capital city of Sichuan Province, is West China's largest and most urban city, being typical of other large cities in lower to middle-income countries (LMICs). With twenty county-level jurisdictions spanning various degrees of urbanization, Chengdu's landscape heterogeneity and large population make it ideal for studying greenspace's impact on public health. This study took Chengdu as a case study to assess the association and potential impact of three traditional measures of greenspace (Normalized Difference Vegetation Index, Enhanced Vegetation Index, and Fractional Vegetation Cover) and urban ratio (% of population being urban) on hospitalization rates and medical expenses paid for three major disease categories (circulatory system diseases, neoplasms, and respiratory system diseases). Results and discussion We found greenspace did have a significant impact on public health, but this relationship differed by disease type. Greenspace exhibited significant positive association with respiratory diseases, but insignificant negative associations with the other disease categories. Urban ratio showed significant negative association with greenspace abundance. The higher the urban ratio (e.g., less greenspace), the more money was paid on medical expenses. This relationship was found not only in terms of urban ratio being positively correlated with medical expenses, but also in that all three greenspace indicators were negatively correlated with medical expenses. Consequently, in future health outcome studies, urban ratio could be an acceptable negative indicator of greenness in LMICs where urban ratio is likely to imply less greenness.