In the rapid development of information construction of medical system, more and more work depends on information. Data mining can extract useful information hidden behind a large amount of information, so it is used more and more widely in hospital information construction. Decision tree classification is an important method in data mining. Decision tree mining technology is applied to medical and nursing and health data to analyze and mine the factors affecting medical and nursing staff's medical nursing and health. The medical nursing and health of medical staff is an important factor affecting their lives, work and safety. In recent years, they have received widespread attention at home and abroad. Therefore, it is urgent to evaluate the medical staff and provide health and psychological interventions to support this group. This paper studies the application of decision tree algorithm in the medical and nursing psychological personality health evaluation system, and understands the relevant theoretical knowledge of psychological evaluation on the basis of literature data, and then designs the medical nursing and health evaluation system based on the decision tree algorithm. The decision tree algorithm cited in the system is tested, and the test results show that after the improved algorithm in this paper is added with a balance coefficient of 0<λ≤1, the distance distance between the attribute and the root node has been shortened to varying degrees.