The spatial difference between regional tourism efficiency and tourism scale is evident. It is of great theoretical and practical value to dynamically grasp the spatial-temporal evolution characteristics and the coupling coordination relationship of them to promote the high-quality sustainable development of tourism. This paper measures the tourism scale of 126 city units in the Yangtze River Economic Belt from 2001 to 2018. The DEA-MI model was introduced to measure and decompose tourism efficiency, while the exploratory spatio-temporal data analysis method was used to explore the spatial and temporal characteristics of regional tourism differences and spatial structure. In addition, the coupling coordination degree model of tourism efficiency and scale was established to examine the coupling superiority and synergy consistency of the two. The results demonstrated the following: (1) the spatial difference in comprehensive tourism efficiency of the study area was recognizable. The average situation exhibited the spatial distribution characteristics of "high in the east and west and low in the middle", and inter-annual changes decline in fluctuations. Scale efficiency played a supporting role in overall efficiency, and technical efficiency played a restrictive role. (2) The fluctuation range with the local spatial structure of tourism efficiency and tourism scale was relatively small, and the direction of dependence was relatively stable. The volatility of the latter is slightly stronger than that of the former, and the change of spatial dependence direction is comparable. The competitive situation of local spatial structure of tourism efficiency was stronger than that of cooperation, and the integration of tourism scale cooperation was strong. (3) The local spatial structure of tourism scale was relatively stable as it was was difficult to change the relative position of the city unit. However, the local spatial structure of tourism efficiency was still unstable, and it was likely to change the possibility within the city unit. (4) The overall coupling degree and coupling coordination degree of tourism efficiency and scale were gradually improved with similar spatial and temporal differentiation characteristics. There was spatial heterogeneity and volatility in local evolution, and the high value region of the degree of coupling coordination generated a broader range and slower diffusion.