Salt marshes and tidal flats are ecosystems providing considerable value to human beings and widely distributed in the world temperate coastal zone. However, they have been seriously damaged due to human activities over the past half century. Previous studies often used area-based statistic measures to track tidal wetlands loss while neglected to extract other important information such as changes in the composition of tidal wetlands, transformation between salt marsh and tidal flat as well as the quality of them. Little is known about the overlap degree among the spatial patterns of tidal wetlands loss and degradation levels. Shanghai’s tidal wetlands in the Yangtze River Estuary of China was used as a cased study. Changes in the area and composition of Shanghai’s natural tidal wetlands was measured by building a continuous land use/land cover (LULC) map database (1990, 2000, 2008 and 2018). LULC transformation analysis was used to detect the impact extent of different loss and degradation driving forces (land reclamation, coastal erosion and biological invasion) within different periods. Quality of the remaining tidal wetlands was determined by modeling ecosystem health index (EHI) on 661 assessment units of the study area. Spatial relationship between the EHI and area-based tidal wetland loss indicators were tested with bivariate Moran’s I. The results showed that net loss rate (0.6%/yr) of Shanghai’s tidal wetlands was relatively low between 1990 and 2018, although its composition changed dramatically. The extent of native ecosystems reduced significantly. For example, 42% of the natural tidal marshes were occupied by S. alterniflora in 2018. Actual degradation and loss of Shanghai’s coastal wetlands caused by the three driving forces between 1990 and 2018 was much higher than the maximum area of tidal wetlands of the four time periods. Average EHI value of all units was 0.58, which suggested that Shanghai’s remaining tidal wetlands were in the middle health status. Only a few units on the Jiuduan shoal showed net gain or low actual loss in area, and with good health. This study confirmed that changes in tidal wetlands area cannot be independently used to represent the conservation status at large scale because even ‘natural’ tidal wetlands can become degraded at varying levels due to multiple stressors. This study provides valuable insights in integrating comprehensive information of coastal wetlands loss and degradation into conservation and restoration planning.