Analysis of risk factors for recurrence of non-puerperal mastitis and establishment of a prediction model
Objective:To analyze the risk factors of non-puerperal mastitis(NPM)and build a nomogram pre-diction model.Methods:A total of 128 NPM patients from the Surgery Department of Traditional Chinese Medicine of China-Japan Friendship Hospital from September 2021 to March 2023 were retrospectively select-ed and divided into recurrence group and non-recurrence group.Clinical data of patients were collected and compared,univariate analysis and multivariate logistics regression analysis were applied to screen independent risk factors for NPM recurrence,and a nomogram prediction model was built based on the risk factors.Fur-ther,the prediction model was evaluated by drawing the area under ROC curve(AUC),calibration curve and decision analysis curve(DCA)of the predictive model.Results:Among 128 NPM patients,there were 16 cases in the recurrence group and 112 in the non-recurrence group.Univariate and Logistic regression analysis showed that obese/overweight,primary nipple depression,oral contraceptive and surgical history were indepen-dent risk factors for recurrence of NPM(P＜0.05).The nomogram prediction model built based on the above four independent risk factors predicted the recurrence risk of NPM.ROC curve analysis results of the nomo-gram showed that the AUC of the model was 0.818(95％CI:0.710～0.926),and the corresponding specificity at CUT-OFF was 97.32％,sensitivity was 25.00％.Conclusion:Obese/overweight,primary nipple depression,oral contraceptive and surgical history are independent risk factors for NPM recurrence.Early identification of risk factors for recurrence is helpful to avoid the risk of NPM recurrence.The nomogram prediction model es-tablished in this study has good accuracy,calibration degree and clinical practicability for NPM recurrence.
non-puerperal mastitisrecurrencerisk factorsnomogramprediction model
NETL
NSTL
万方数据
