Abstract: The multi-variable assessment method is a current trend for studying the health of river ecosystems. A system of candidate indices, including physical-chemical parameters for water quality, aquatic life index and physical habitatquality index, were set based on monitoring data collected from the Liao River basin in 2005. Principal component analysis (PCA) and correlation analysis were used to select ten parameters from candidate ones, such as BOD₅, DO, conductivity, TN, TP, COD_Mn, FCC, DAA, B-IBI and HQI, which were then incorporated into an integrated assessment index system for assessing the ecosystem health of the Liao River basin. The health level at each sampling site was analyzed using modified grey correlation degree method. The results showed that three of 25 sites were“healthy" level, while 14 sites were “sub-healthy" level. These sites were located upstream of distributaries and headstream of the Liao River, where little disturbance by human activities occurred. The other 32% of all sites were designated as “sub-sick" or “sick" level, and were mainly located at the mid-stream and downstream of the Liao River. Discharge of urban and industrial pollution was the primary factor impacting the river health.