Abstract: OBJECTIVE To investigate the distribution of pathogens causing nosocomial infection in patients with non-Hodgkin’s lymphoma（NHL） and the establishment of risk nomogram model. METHODS A survey of 313 NHL inpatients admitted to the Nantong University Affiliated Hospital from Dec 2017 to Dec 2019 was conducted to analyze the impact of the patients’ clinical data on nosocomial infections, analyze the distribution characteristics of pathogenic bacteria, and screen out risk factors to establish a nomogram prediction model. RESULTS Nosocomial infections of NHL patients were mainly respiratory tract infections. A total of 23 stains of Gram-negative bacteria, 14 stains of Gram-positive bacteria and 11 stains of fungi were detected. Age ≥60 years, diabetes, underlying lung diseases, hospitalization time ≥30 days, invasive procedures, and white blood cell count ＜2.0×10⁹/L were independent risk factors for nosocomial infection in NHL patients. The verification result of the nomogram model showed that the predicted value was basically equivalent to the measured value, and the consistency index（C-index） was 0.751（95%CI: 0.714～0.788）. CONCLUSION Respiratory tract are the most common nosocomial infection sites and gram-negative bacteria are the main bacterial species in NHL patients. The nomogram model based on nosocomial infection risk factors for NHL has good predictive ability and accuracy.