Biological community composition in rivers is driven jointly by multiple environmental factors, however, the mechanism is largely unknown. For this, an eDNA-based technology was used to investigate macroinvertebrate community composition and biodiversity at 10 sampling sites from 4 mountainous rivers (Jinxiuchuan, Jinyangchuan, Jinyunchuan and rivers after their confluence) at lower Yellow River Basin. Results showed that the water quality varied in different rivers. For example, the concentrations of dissolved salt and nutrients were higher in Jinyunchuan and antibiotics was higher in Jinyangchuan. Compared with the other two rivers, the Jinxiuchuan had better water quality. The community composition of macroinvertebrates in the mountainous rivers exhibited significant difference. Jinxiuchuan was dominated by aquatic insects with small proportion of molluscs. Among these organisms, Insects (Tipula abdominalis and Uracanthella punctisetae) and Oligochaeta (Aeolosoma sp.) were the key species that contributed greatly to the differences in community composition of macroinvertebrates between Jinxiuchuan and other rivers. Redundancy analysis (RDA) and variance partitioning analysis (VPA) revealed that salinization, nutrients and antibiotic significantly affected the changes of macroinvertebrate community composition. Of these factors, electrical conductivity (EC) were the most important factors for driving the macroinvertebrate community composition changes with a contribution rate of 22.86%, followed by TP (20.12%), NH₃-N (13.25%) and TN (7.81%). In addition, the contribution rates from the coupling effects of salinization with nutrient and antibiotics was 21.60% and 19.60%, respectively, and nutrients and antibiotics was 16.20%. Stepwise regression models showed that Margalef index (d) was jointly affected by salinization (EC) and nutrients (NH₃-N and NO₃^--N). As EC concentration increased, the positive response relationship between d and NH₃-N, and the negative response relationship between d and NO₃^- were enhanced. Therefore, the coupling effect of multi-environmental factors on aquatic organisms should not be ignored, and more attention should be paid on the contribution rate of various environmental factors in the biodiversity protection of macroinvertebrates.