Enhancing the spatial accessibility to medical facilities is an important public health strategy in many countries. However, it is not clear whether enhancing spatial accessibility is capable of addressing medical care issues due to the complexity of individuals' health-seeking behavior. This study draws from Andersen's behavioral model to identify enabling factors in the associations of spatial accessibility to medical facilities and socio-institutional factors with individuals' health-seeking behavior. Logistic regression and mixed geographically weighted regression techniques are used to analyze large-scale health survey data collected in Shanghai, China. The results show that spatial accessibility to medical facilities plays a significant role in health-seeking behaviors, but the mechanism at play is complicated, involving people's preference for high-grade hospitals and the demand for easy access to public transit. Moreover, there is spatial heterogeneity in the relationship between spatial accessibility to medical facilities and health-seeking behaviors. Especially, people who live in the inner suburb of Shanghai have worse access to these facilities and thus are more likely to perform self-treatment compared with urban residents. Moreover, localized contexts formed by the interweaving of spatial accessibility and socio-institutional factors, such as registered residence status (hukou), complicate the association between spatial accessibility and health-seeking behaviors.

With great concern over the health-promoting environment worldwide, there is a growing body of research into the neighborhood effects on health beyond the sole focus on individual socioeconomic disadvantages and lifestyle risks. Our study contributes to neighborhood health research by investigating the combined effects of multi-dimensional neighborhood environmental characteristics and recreational physical activity under different geographic contexts on residents' self-rated health. Drawing upon a health survey conducted in suburban Shanghai in 2017, we employ a series of multilevel models to examine how the multi-scale environmental and behavioral factors are related to residents' self-rated physical and mental health, respectively. The results show that the greening rate of the community, rather than accessibility to large-scale urban parks, is a significant indicator of self-rated health for suburban residents. Subjective evaluations on neighborhood safety and air pollution exposure are significantly associated with residents' physical and mental health, while neighborhood attachment matters more for mental health. Outdoor recreational exercises, especially in the environment of neighborhood green space, are conducive to better physical health, while indoor physical activity shows weaker and insignificant health benefits. These findings offer a promising way for public health policymakers and urban planners to implement place-based health interventions and develop health-supportive neighborhoods.