Hand, foot, and mouth disease (HFMD) has brought millions of attacks and a substantial burden in the Asia-Pacific region. Previous studies assessed disease risks around the world, which demonstrated great heterogeneity, and few determined the modification effect of social factors on temperature-disease relationship. We conducted a time-series study to evaluate the temperature-associated HFMD morbidity risk using daily data (from 2011 to 2017) and to identify potential modifiers relating to urban-rural status and aggregation mode of children. By applying a distributed lag nonlinear model (DLNM) and controlling for time-varying factors and other meteorological factors, we found that the relationship between daily mean temperature and the cumulative risk of HFMD was an approximately M-shaped curve. The effects of higher temperature appeared to be greater and more persistent than those of lower temperature. With the reference of -6°C, the cumulative relative risk (RR) values of high temperature (95 percentile) and low temperature (5 percentile) were 3.74 (95% CI: 2.50-5.61) and 1.72 (95% CI: 1.24-2.37) at lag 4-7, respectively. Temperature-associated HFMD morbidity risks were more pronounced among rural children and those attending kindergartens or schools at specific lags and temperatures. Relative risk values for temperature-disease association was highest among the 3- to 6-year group, whereas no gender difference was observed. Studying effect estimates and their modifications using the DLNM on a daily scale helps to identify susceptible groups and guide policy-making and resource allocation according to specific local conditions.