LYU Yangyang,HUANG Haiqun,LIANG Chengying, et al. Risk factors for emergency room patients with catheter-associated urinary tract infections and the establishment of prediction model[J]. JOURNAL OF WEZHOU MEDICAL UNIVERSITY, 2022, 52(2): 126-131，138.
Abstract：Objective: To explore the risk factors for patients with catheter-associated urinary tract infection (CAUTI) in the emergency room and to establish a risk prediction model. Methods: A total of 1 780 patients with indwelling catheter in the emergency room of our unit from June 2018 to December 2020 were retrospectively analyzed by single factor and multivariate Logistic regression analysis to find out the independent risk factors for CAUTI. Based on the results, a risk prediction model was established, the fitting degree of the model tested by Hosmer-Lemeshow and predictive value of the regression model evaluated by the receiver operating characteristic (ROC) curve. Results: The incidence of CAUTI was 7.98% during the period of this study. Compared with reference, Age≥60 years old, BMI≥30 kg/m2, indwelling time of catheter≥3 d and HbA1c were independent risk factors for CAUTI in emergency room patients, while serum Albumin level was protective factor for CAUTI (OR=1.637, 3.052, 3.135, 1.563, 0.896, respectively, all P<0.05). According to the results, the risk prediction model was constructed and the fitting degree was acceptable (χ2=13.383, P=0.099). ROC analysis showed that the area under curve (AUC) of the regression model was 0.870, with the corresponding sensitivity being 83.10% and specificity 78.45%. In addition, ROC analysis was made for HbA1c,serum albumin and the two combined detection, with the corresponding AUC being 0.759, 0.618 and 0.842, respectively, and sensitivity being 52.82%, 35.92%, 67.61% and specificity 90.54%, 90.72%, 90.90% accordingly. Conclusion: CAUTI in emergency room remains a major source of nosocomial infection. The regression model constructed in this study has good predictive effect, and the combined detection for HbA1c and serum albumin could be of better predictive value than any single index. In clinical practice, it is necessary to take comprehensive nursing preventive measures according to the above-mentioned factors.
