<![CDATA[Surgical wound infection is a serious problem in patient care in hospitals. This infection can cause increased morbidity, mortality, increased length of stay as well as costs and patient demands. This can be closely related to the quality and services of hospitals, which will also affect the credibility of health services. Nurses are the people who have the most contact with patients, therefore nurses' behavior is very influential in implementing procedures to prevent surgical wound infections. This study aims to determine the relationship between knowledge and length of work with nurses' behavior in efforts to prevent surgical wound infections at Gatot Soebroto Hospital in 2023. This research method is a quantitative study with a cross-sectional design approach. Using a total sampling technique of 60 people. Data collection was carried out by assessment using knowledge questionnaire instruments and intra-operative behavior checklists. Data were analyzed univariately and bivariately using the Spearman rho test correlation technique. The research results showed that 21 nurses (35%) had less knowledge. Nurses who have worked for less than 6 years are 37%, nurses who have worked for more than 10 years are 40%. Nurses with bad behavior had a higher number of 53.3%. The results of the Spearman rho test showed that (1) there was a relationship between nurses' knowledge of the behavior of preventing surgical wound infections with a p value = 0.000, meaning p > 0.05 (r = 0.616), (2) there was a relationship between the length of service of nurses and the behavior of preventing wound infections. operation with a p value = 0.006 means p > 0.05 (r = 0.349). The conclusion is that there is a significant relationship between knowledge and behavior of nurses and there is a significant relationship between length of service and behavior of nurses in efforts to prevent surgical wound infections. The research suggestion is for hospital management to increase nurses' knowledge in the form of training, workshops and regular supervision,]]>
