Spatial access to healthy foods has drawn much attention due to its potential to facilitate healthy eating habits, and subsequently reduce the risk of chronic diseases. However, the associations of healthy food accessibility with grocery shopping patterns depending on various spatial contextual units remain little known. This study aimed to (1) compare healthy food accessibility among residential neighborhoods with varying buffer sizes and activity space; (2) examine the associations of grocery shopping patterns and healthy food accessibility in residential neighborhoods and activity space; (3) explore the differences in healthy food environment and the associations between healthy food accessibility and grocery shopping patterns by populations density depending on residential neighborhoods and activity space. Data came from GPS-based Household Interview Survey for the Cincinnati Ohio Region conducted during 2009–2010, of which final sample consisted of 1625 participants. The results illustrated that (1) healthy food accessibility in activity space had greater variability than in residential neighborhoods; (2) healthy food accessibility had significant associations with grocery shopping patterns across various spatial contextual units; and (3) healthy food accessibility and its associations with grocery shopping patterns differed significantly by population density in residential neighborhoods but not in activity space. This study provides a deeper insight into the relationships between grocery shopping patterns and healthy food accessibility by accounting for various spatial contextual units. It also improves nuanced understanding of the spatial heterogeneity in healthy food environment and its associations with grocery shopping patterns between more urbanized areas and more rural areas depending on residential neighborhoods and activity space. One implication is that improving healthy food exposures to people's residence and other routine activity destinations may encourage them to visit the healthy food stores for food purchase. Another implication is that future research should carefully consider the place effects in evaluating people's food environment in residential neighborhoods and its associations with grocery shopping behaviors.