Based on Moran index, hot spot analysis and geographically weighted regression model, this paper analyzed the spatial distribution characteristics and influencing factors of per capita hospitalization costs in 31 provinces of mainland China from 2017 to 2019. The results showed that the Moran's I index of per capita hospitalization cost was greater than 0 from 2017 to 2019 (P < 0.01). Hot spot analysis shows that there are "hot spot" and "cold spot" area of hospitalization cost per capita. The number of licensed assistant physicians per thousand population and urbanization rate have a positive impact on the per capita hospitalization cost, while the number of beds per thousand population has a negative impact on the per capita hospitalization cost. The per capita hospitalization cost in China is on the rise as a whole, and there is aggregation in geographical distribution, and there are differences in the strength of influencing factors in different regions. Therefore, policies should be formulated according to the actual situation.