Construction of a risk prediction model for bloodstream infection induced by carbapenem-resistant Klebsiella pneumoniae / 中国药房

OBJECTIVE To construct a risk prediction model for bloodstream infection （BSI） induced by carbapenem-resistant Klebsiella pneumoniae （CRKP）. METHODS Retrospective analysis was conducted for clinical data from 253 patients with BSI induced by K. pneumoniae in the First Hospital of Qinhuangdao from January 2019 to June 2022. Patients admitted from January 2019 to December 2021 were selected as the model group （n＝223）， and patients admitted from January 2022 to June 2022 were selected as the validation group （n＝30）. The model group was divided into the CRKP subgroup （n＝56） and the carbapenem- sensitive K. pneumoniae （CSKP） subgroup （n＝167） based on whether CRKP was detected or not. The univariate and multivariate Logistic analyses were performed on basic information such as gender， age and comorbid underlying diseases in two subgroups of patients； independent risk factors were screened for CRKP-induced BSI， and a risk prediction model was constructed. The established model was verified with patients in the validation group as the target. RESULTS Admissioning to intensive care unit （ICU）， use of immunosuppressants， empirical use of carbapenems and empirical use of antibiotics against Gram-positive coccus were independent risk factors of CRKP-induced BSI （ORs were 3.749， 3.074， 2.909， 9.419， 95%CIs were 1.639-8.572， 1.292- 7.312， 1.180-7.717， 2.877-30.840， P＜0.05）. Based on this， a risk prediction model was established with a P value of 0.365. The AUC of the receiver operating characteristic （ROC） curve of the model was 0.848 [95%CI （0.779， 0.916）， P＜0.001]， and the critical score was 6.5. In the validation group， the overall accuracy of the prediction under the model was 86.67%， and the AUC of ROC curve was 0.926 [95%CI （0.809， 1.000]， P＜0.001]. CONCLUSIONS Admission to ICU， use of immunosuppressants， empirical use of carbapenems and empirical use of antibiotics against Gram-positive coccus are independent risk factors of CRKP- induced BSI. The CRKP-induced BSI risk prediction model based on the above factors has good prediction accuracy.

LASSO regression based risk prediction model establishment for Klebsiella pneumonia infection in patients with severe acute pancreatitis / 中华胰腺病杂志

Objective:To construct a risk prediction model for infection with Klebsiella pneumonia (KP) for patients with severe acute pancreatitis (SAP).Methods:Retrospective analysis was done on the clinical data of 109 SAP patients who were admitted to Shanghai General Hospital, between March 2016 and December 2021. Patients were classified into infection group ( n=25) and non-infection group ( n=84) based on the presence or absence of KP infection, and the clinical characteristics of the two groups were compared. The least absolute shrinkage and selection operator (LASSO) algorithm was used to reduce the dimension of the variables with statistical significance in univariate analysis. A nomogram prediction model was created by incorporating the optimized features from the LASSO regression model into the multivariate logistic regression analysis. Receiver operating characteristic curve (ROC) was drawn and the area under curve (AUC) was calculated; and consistency index (C-index) were used to assess the prediction model's diagnostic ability. Results:A total of 25 strains of KP were isolated from 109 patients with SAP, of which 21(84.0%) had multi-drug resistance. 20 risk factors (SOFA score, APACHEⅡ score, Ranson score, MCTSI score, mechanical ventilation time, fasting time, duration of indwelling of the peritoneal drainage tube, duration of deep vein indwelling, number of invasive procedures, without or with surgical intervention, without or with endoscopic retrograde cholangiopancreatography (ERCP), types of high-level antibiotics used, digestion disorders, abnormalities in blood coagulation, metabolic acidosis, pancreatic necrosis, intra-abdominal hemorrhage, intra-abdominal hypertension, length of ICU stay and total length of hospital stay) were found to be associated with KP infection in SAP patients by univariate analysis. The four variables (APACHEⅡ score, duration of indwelling of the peritoneal drainage tube, types of high-level antibiotics used, and total length of hospital stay) were extracted after reduced by LASSO regression. These four variables were found to be risk factors for KP infection in SAP patients by multiple logistic regression analysis (all P value <0.05). Nomogram prediction model for KP infection in SAP was established based on the four variables above. The verification results of the model showed that the C-index of the model was 0.939, and the AUC was 0.939 (95% CI 0.888-0.991), indicating that the nomogram model had relatively accurate prediction ability. Conclusions:This prediction model establishes integrated the basic clinical data of patients, which could facilitate the risk prediction for KP infection in patients with SAP and thus help to formulate better therapeutic plans for patients.

Epidemiological characteristics and risk prediction of pulmonary infection in elderly patients with chronic obstructive pulmonary disease / 公共卫生与预防医学

Objective To explore the epidemiological characteristics of pulmonary infection in elderly patients with chronic obstructive pulmonary disease (COPD), and to construct a risk prediction model. Methods Among of 125 elderly patients with COPD from May 2020 to June 2022 were selected as the research subjects. The epidemiological characteristics of infected patients were counted, and the risk factors of pulmonary infection in patients were analyzed and a prediction model was constructed. Results A total of the 125 elderly patients with COPD, there were 46 cases of pulmonary infection, with the infection rate of 36.80%. The detection rate of Gram-negative bacteria was higher than that of Gram-positive bacteria or fungi (64.44% vs 33.33% or 2.22%, P2=0.812 and P=0.295. ROC curve analysis revealed that the AUC value of the prediction model on predicting the pulmonary infection in elderly patients with COPD was 0.802. Conclusion The pathogenic bacteria of elderly patients with COPD complicated with pulmonary infection are mainly Gram-negative bacteria. The prediction model constructed according to the risk factors of pulmonary infection in patients has predictive value on pulmonary infection in patients.

Development and validation of risk prediction model for new-onset cardiovascular diseases among breast cancer patients: Based on regional medical data of Inner Mongolia / 北京大学学报(医学版)

OBJECTIVE@#To develop and validate a three-year risk prediction model for new-onset cardiovascular diseases (CVD) among female patients with breast cancer.@*METHODS@#Based on the data from Inner Mongolia Regional Healthcare Information Platform, female breast cancer patients over 18 years old who had received anti-tumor treatments were included. The candidate predictors were selected by Lasso regression after being included according to the results of the multivariate Fine & Gray model. Cox proportional hazard model, Logistic regression model, Fine & Gray model, random forest model, and XGBoost model were trained on the training set, and the model performance was evaluated on the testing set. The discrimination was evaluated by the area under the curve (AUC) of the receiver operator characteristic curve (ROC), and the calibration was evaluated by the calibration curve.@*RESULTS@#A total of 19 325 breast cancer patients were identified, with an average age of (52.76±10.44) years. The median follow-up was 1.18 [interquartile range (IQR): 2.71] years. In the study, 7 856 patients (40.65%) developed CVD within 3 years after the diagnosis of breast cancer. The final selected variables included age at diagnosis of breast cancer, gross domestic product (GDP) of residence, tumor stage, history of hypertension, ischemic heart disease, and cerebrovascular disease, type of surgery, type of chemotherapy and radiotherapy. In terms of model discrimination, when not considering survival time, the AUC of the XGBoost model was significantly higher than that of the random forest model [0.660 (95%CI: 0.644-0.675) vs. 0.608 (95%CI: 0.591-0.624), P < 0.001] and Logistic regression model [0.609 (95%CI: 0.593-0.625), P < 0.001]. The Logistic regression model and the XGBoost model showed better calibration. When considering survival time, Cox proportional hazard model and Fine & Gray model showed no significant difference for AUC [0.600 (95%CI: 0.584-0.616) vs. 0.615 (95%CI: 0.599-0.631), P=0.188], but Fine & Gray model showed better calibration.@*CONCLUSION@#It is feasible to develop a risk prediction model for new-onset CVD of breast cancer based on regional medical data in China. When not considering survival time, the XGBoost model and the Logistic regression model both showed better performance; Fine & Gray model showed better performance in consideration of survival time.

Development and validation of ischemic heart disease and stroke prognostic models using large-scale real-world data from Japan

BACKGROUND@#Previous cardiovascular risk prediction models in Japan have utilized prospective cohort studies with concise data. As the health information including health check-up records and administrative claims becomes digitalized and publicly available, application of large datasets based on such real-world data can achieve prediction accuracy and support social implementation of cardiovascular disease risk prediction models in preventive and clinical practice. In this study, classical regression and machine learning methods were explored to develop ischemic heart disease (IHD) and stroke prognostic models using real-world data.@*METHODS@#IQVIA Japan Claims Database was searched to include 691,160 individuals (predominantly corporate employees and their families working in secondary and tertiary industries) with at least one annual health check-up record during the identification period (April 2013-December 2018). The primary outcome of the study was the first recorded IHD or stroke event. Predictors were annual health check-up records at the index year-month, comprising demographic characteristics, laboratory tests, and questionnaire features. Four prediction models (Cox, Elnet-Cox, XGBoost, and Ensemble) were assessed in the present study to develop a cardiovascular disease risk prediction model for Japan.@*RESULTS@#The analysis cohort consisted of 572,971 invididuals. All prediction models showed similarly good performance. The Harrell's C-index was close to 0.9 for all IHD models, and above 0.7 for stroke models. In IHD models, age, sex, high-density lipoprotein, low-density lipoprotein, cholesterol, and systolic blood pressure had higher importance, while in stroke models systolic blood pressure and age had higher importance.@*CONCLUSION@#Our study analyzed classical regression and machine learning algorithms to develop cardiovascular disease risk prediction models for IHD and stroke in Japan that can be applied to practical use in a large population with predictive accuracy.

Effects of preoperative lipid metabolism indexes on the prognosis of patients with non-muscular invasive bladder cancer / 现代泌尿外科杂志

【Objective】 To investigate the effects of preoperative lipid metabolism level on the postoperative prognosis of non-muscular invasive bladder cancer (NMIBC). 【Methods】 Clinical data of NMIBC patients who underwent surgical treatment in our hospital during Mar.2014 and May 2021 were retrospectively analyzed. Based on receiver operating characteristic (ROC) curve, the optimal cutoff values of all lipid metabolism indicators were determined and patients were classified accordingly. The independent risk factors for postoperative recurrence were identified with Cox regression model. The survival was analyzed with Kaplan-Meier, and recurrence-free survival (RFS) was compared using log-rank tests. A recurrence risk prediction model was established based on the high-density lipoprotein (HDL) and other clinic pathological factors and the accuracy of prediction was evaluated with the area under the ROC curve (AUC). 【Results】 Cox multivariate analysis showed HDL, tumor number, tumor size and histological grade were independent risk factors for recurrence (P<0.05). Kaplan-Meier analysis showed that RFS was significantly longer in the high-HDL group than in the low-HDL group (P<0.001). Incorporating HDL, tumor number, tumor size, histological grade, and tumor stage into the recurrence risk model, the AUC was 0.706, and internal cross validation showed the AUC was 0.711. 【Conclusion】 Preoperative HDL is an independent risk factor affecting the RFS of patients with NMIBC, and combining it with clinic pathological factors will improve the prediction of tumor recurrence.

Establishment and evaluation of the polygenic disease risk prediction model / 上海预防医学

To establish a disease risk prediction model based on genetic susceptibility genes and environmental risk factors， which can target high-risk population as early as possible， and intervene in the environmental risk factors in this population. Moreover， accurate screening of genetically susceptible populations can enhance the efficiency of health system. In recent years， with the maturation and cost reduction of high-throughput gene testing， gene testing has been widely used in individual clinical decision-making and will play a more important role in medical and health decision-making. The correlation between genetic testing and disease risk prediction is increasing， making it a prominent research topic in this field. This review summarizes the approaches for establishing and evaluating risk prediction models and discusses potential future challenges and opportunities.

Risk factors of mortality in Klebsiella pneumoniae bloodstream infection and construction of a prediction model for prognosis of patients / 中华临床感染病杂志

Objective:To explore the risk factors of mortality in patients with Klebsiella pneumoniae bloodstream infection, and to construct a predictive model. Methods:The clinical data of 234 patients with Klebsiella pneumoniae bloodstream infection admitted in the First Hospital of Qinhuangdao from January 2020 to December 2022 were retrospectively analyzed, including 202 cases admitted during January 2020 to June 2022 (model set), and 32 cases admitted during July to December 2022 (validation set). There were 64 cases died (fatal group) and 138 cases survived (survival group) within 28 d after admission in model set. Multivariate Logistic regression was used to analyze the risk factors of death in patients with Klebsiella pneumoniae bloodstream infection and a mortality prediction model was constructed. The constructed model was applied in validation set, and the consistency between predicted mortality and real mortality was analyzed. Results:Multivariate Logistic regression analysis showed that male sex ( OR=2.598, 95% CI 1.179-5.725, P=0.018), age≥65 years ( OR=4.420, 95% CI 2.029-9.627, P<0.001), admitted to intensive care unit (ICU) ( OR=10.299, 95% CI 4.752-22.321, P<0.001), and the empirical use of quinolones antibiotics ( OR=4.288, 95% CI 1.127-16.317, P=0.033) were independent risk factors for 28-day mortality in Klebsiella pneumoniae bloodstream infection patients. The regression equation for predicting the risk of death was -3.469+ male × 0.955+ age ≥ 65 years × 1.486+ admitted to ICU × 2.332+ empirical use of quinolone antibiotics × 1.456. The area under the ROC curve (AUC) for predicting death in the model set was 0.831, with sensitivity and specificity of 71.9% and 80.4%, respectively. The AUC for predicting death in the validation set was 0.881, with sensitivity and specificity of 91.7% and 75.0%, respectively. Conclusion:The constructed mortality prediction model in the study has good application value for the prognosis of patients with Klebsiella pneumoniae bloodstream infection.

Influencing factors and risk prediction model for cervical cancer recurrence / 中南大学学报(医学版)

OBJECTIVES@#Cervical cancer is the most common malignant tumor in the female reproductive system worldwide. The recurrence rate for the treated cervical cancer patients is high, which seriously threatens women's lives and health. At present, the risk prediction study of cervical cancer has not been reported. Based on the influencing factors of cervical cancer recurrence, we aim to establish a risk prediction model of cervical cancer recurrence to provide a scientific basis for the prevention and treatment of cervical cancer recurrence.@*METHODS@#A total of 4 358 cervical cancer patients admitted to the Hunan Cancer Hospital from January 1992 to December 2005 were selected as research subjects, and the recurrence of cervical cancer patients after treatment was followed up. Univariate analysis was used to analyze the possible influencing factors. Variables that were significant in univariate analysis or those that were not significant in univariate analysis but may be considered significant were included in multivariate Cox regression analysis to establish a cervical cancer recurrence risk prediction model. Line graphs was used to show the model and it was evaluated by using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis.@*RESULTS@#Univariate analysis showed that the recurrence rates of cervical cancer patients with different age, age of menarche, parity, miscarriage, clinical stage, and treatment method were significantly different (all P<0.05). Multivariate Cox regression analysis showed that RR=-0.489×(age≥55 years old)+0.481×(age at menarche >15 years old)+0.459×(number of miscarriages≥3)+0.416×(clinical stage II)+0.613×(clinical stage III/IV)+0.366×(the treatment method was surgery + chemotherapy) + 0.015×(the treatment method was chemotherapy alone). The area under the ROC curve (AUC) of the Cox risk prediction model for cervical cancer recurrence constructed was 0.736 (95% CI 0.684 to 0.789), the best prediction threshold was 0.857, the sensitivity was 0.576, and the specificity was 0.810. The accuracy of the Cox risk model constructed by this model was good. From the clinical decision curve, the net benefit value was high and the validity was good.@*CONCLUSIONS@#Patient age, age at menarche, miscarriages, clinical stages, and treatment methods are independent factors affecting cervical cancer recurrence. The Cox proportional hazards prediction model for cervical cancer recurrence constructed in this study can be better used for predicting the risk of cervical cancer recurrence.

Risk factors analysis and construction of risk prediction model for unplanned readmission in patients with acute myocardial infarction / 中国实用护理杂志

Objective:To explore the risk factors of unplanned readmission in patients with acute myocardial infarction, and to construct a risk prediction model.Methods:This study used cross-sectional survey method. A total of 270 acute myocardial infarction patients admitted from Tianjin Union Medical Cencer from March 2020 to March 2021 were evaluated in a cardiology department. We used the electronic medical record system to collect the patients′ data. Patients were divided into two groups according to the occurrence of readmission within 1 year or not. Logistic regression analysis was performed to identify risk factors and formulated prediction model.Results:Totally 81 patients (30%) were readmitted. Binary Logistic regression model showed that the independent influencing factors of unplanned readmission in acute myocardial infarction patients included smoking ( X1), hypertension ( X2), marital status ( X3), hospitalization days ( X4), percutaneous coronary intervention ( X5), and heart failure ( X6). Area under ROC curve was 0.840, the maximum value of the Youden index was 0.560, and the sensitivity was 85.2%, the specificity was 70.8%, and the cutoff value was 0.377. Prediction model expression of unplanned readmission risk in patients with acute myocardial infarction was Logit(p/1-p)=-4.012+1.172 X1+1.104 X2+0.992 X3+0.118 X4+1.191 X5+1.093 X6. Conclusions:The risk prediction model of unplanned readmission in patients with acute myocardial infarction established in this article was with a good predictive effect, and it could be used in early identification of those patients with high-risk in unplanned readmission. At the same time, combined with the risk factors of depression, targeted intervention measures can be formulated.

Analysis of risk factors and construction of risk prediction model of cognitive dysfunction in patients with atrial fibrillation / 中国实用护理杂志

Objective:To identify the risk factors of cognitive dysfunction in patients with atrial fibrillation and to establish a risk prediction model.Methods:The convenience sampling method was used to evaluate 260 patients with atrial fibrillation who were hospitalized in the Department of Cardiology of the Affiliated Hospital of Jining Medical College from January to December 2020. The cognitive function of the patients was evaluated with the Montreal Cognitive Function Assessment Scale (MoCA). Univariate analysis was used to screen the independent variables that had influence on the occurrence of cognitive dysfunction, and the statistically significant variables were included in the multivariate Logistic regression model. According to the regression coefficients of statistically significant variables, a line map was drawn to construct the risk prediction model of cognitive dysfunction in patients with atrial fibrillation.Results:There were 209 cases with cognitive impairment and 51 cases without cognitive impairment. Univariate analysis showed that sex, age, smoking history, drinking history, education level, free thyroxine, hemoglobin, D-dimer and BMI ( χ2 values were 4.08-18.83, t values were -6.04-2.94, Z=-2.76) were significantly different between the patients with or without cognitive dysfunction. The results of multivariate Logistic regression analysis showed that age ( OR values were 1.13), education level ( OR=0.01-0.05), quit smoking history ( OR=0.36), drinking history ( OR=0.35) and free thyroxine( OR=1.14) had significantly statistical significance ( P<0.05). The area under ROC curve (AUC) = 0.878 and AUC>0.8, this model had good clinical prediction ability. Conclusions:The construction of cognitive dysfunction risk prediction model for patients with atrial fibrillation can prevent or intervene high risk factors in advance, facilitate clinical use, and provide data support for the improvement of cognitive function in patients with atrial fibrillation.

Cognitive dysfunction status and risk prediction of patients with coronary heart disease / 中国实用护理杂志

Objective:To understand the current situation of cognitive dysfunction in patients with coronary heart disease, and explore the risk prediction model of the onset of cognitive dysfunction in patients with coronary heart disease.Methods:A total of 448 patients with coronary heart disease admitted to the North China University of Science and Technology Affiliated Hospital from January 2019 to June 2020 were prospectively selected as study subjects. Patients with coronary heart disease were divided into the cognitive dysfunction group ( n=185) and the normal cognitive function group ( n=263) according to whether they were accompanied by cognitive dysfunction. Demographic characteristics, cognitive function, disease history, blood pressure, blood glucose, blood lipid and vascular lesions were compared between the two groups. Montreal Cognitive Assessment (MoCA) was used to evaluate cognitive function. Logistic regression was used to analyze the risk factors of cognitive dysfunction in coronary heart disease patients, and the prediction model of the above risk factors was constructed. The value of the prediction model was evaluated by C-index and cilibration curve. Results:The language, abstraction, visual space and execution, delayed memory and total scores of the cognitive dysfunction group were 1.81 ± 0.59, 1.12 ± 0.33, 3.01 ± 0.90, 2.61 ± 0.79 and 22.32 ± 1.70, respectively, which were lower than those of the normal cognitive function group (2.68 ± 0.47, 1.82 ± 0.38, 4.54 ± 0.50, 4.77 ± 0.42, 27.67 ± 0.76), and the differences were statistically significant ( t values were 17.39-40.00, all P<0.05). The age, fasting blood glucose, systolic blood pressure, proportion of alcohol drinking, proportion of diabetes mellitus in the cognitive dysfunction group were (62.86 ± 5.21) years, (6.19 ± 0.89) mmol/L, (144.00 ± 17.16) mmHg (1 mmHg=0.133 kPa), 36.76% (68/185), 16.22% (30/185), respectively, which were higher than (58.77 ± 5.63) years, (5.46 ± 0.95) mmol/L, (133.74 ± 15.90) mmHg, 27.38% (72/263), 6.84% (18/263) in the normal cognitive function group, the differences were statistically significant ( t=7.81, 8.25, 6.42, χ2=4.45, 9.97, all P<0.05). The rates of single vessel, double vessel and three vessel lesions in the cognitive dysfunction group were 49.73% (92/185), 27.03% (50/185) and 23.24% (43/185), respectively, and those in normal cognitive function group were 46.39% (122/263), 39.92% (105/263) and 13.69% (36/263), respectively ( χ2=11.10, P<0.05) . Logistic regression analysis showed that age, fasting blood glucose, systolic blood pressure and number of vascular lesions were independent risk factors for coronary heart disease patients with cognitive impairment ( OR values were 1.038-2.216, all P<0.05). The correction curve of the prediction model composed of age, fasting blood glucose, systolic blood pressure and number of vascular lesions was in good agreement with the ideal curve, and the C-index of the model was 0.807 for the diagnosis of cognitive dysfunction in patients with coronary heart disease. Conclusions:The cognitive dysfunction of patients with coronary heart disease is mainly manifested in language, abstraction, visual space and execution and delayed memory. The prediction model composed of age, fasting blood glucose, systolic blood pressure and number of vascular lesions has a certain degree of discrimination and accuracy for patients with coronary heart disease complicated by cognitive dysfunction, and can be used for the screening of coronary heart disease complicated by cognitive dysfunction.

Construction and verification of the risk prediction model for acute exacerbation within 6 months in patients with chronic obstructive pulmonary disease: a secondary analysis based on previous research data / 中华危重病急救医学

Objective:To construct the risk prediction model of acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and verify its effectiveness based on deep learning and back propagation algorithm neural network (BP neural network).Methods:Based on the relevant data of 1 326 patients with chronic obstructive pulmonary disease (COPD) in the team's previous clinical study, the acute exacerbation, and its risk factors during the stable period and 6 months of follow-up were recorded and analyzed. Combined with previous clinical research data and expert questionnaire results, the independent risk factors of AECOPD after screening and optimization by multivariate Logistic regression including gender, body mass index (BMI) classification, number of acute exacerbation, duration of acute exacerbation and forced expiratory volume in one second (FEV1) were used to build the BP neural network by Python 3.6 programming language and Tensorflow 1.12 deep learning framework. The patients were randomly selected according to the ratio of 4∶1 to generate the training group and the test group, of which, the training group had 1 061 sample data while the test group had 265 pieces of sample data. The training group was used to establish the prediction model of neural network, and the test group was used for back-substitution test. When using the training group data to construct the neural network model, the training group was randomly divided into training set and verification set according to the ratio of 4∶1. There were 849 training samples in the training set and 212 verification samples in the verification set. The optimal model was screened by adjusting the parameters of the neural network and combining the area under the receiver operator characteristic curve (AUC), and the sample data of the test group was substituted into the model for verification.Results:The independent risk factors including gender, BMI classification, number of acute exacerbation, duration of acute exacerbation and FEV1 were collected from the team's previous clinical research, and the AECOPD risk prediction model was constructed based on deep learning and BP neural network. After 10 000 training sessions, the accuracy of the AECOPD risk prediction model in the validation set of the training group was 83.09%. When the number of training times reached 8 000, the accuracy basically tended to be stable and the prediction ability reached the upper limit. The AECOPD risk prediction model trained for 10 000 times was used to predict the risk of the validation set data, and the receiver operator characteristic curve (ROC curve) analysis showed that the AUC was 0.803. When using this model to predict the risk of the data of the test group, the accuracy rate was 81.69%.Conclusion:The risk prediction model based on deep learning and BP neural network has a medium level of prediction efficiency for acute exacerbation within 6 months in COPD patients, which can evaluate the risk of AECOPD and assist the clinic in making accurate treatment decisions.

Research progress on risk prediction models for the elderly frailty / 中国实用护理杂志

The research progress of the frailty risk prediction model for the elderly were reviewed in order to analyze its evaluation content, prediction effect and clinical application, so as to provide reference for the risk prediction model construction and application, reduce the risk of the elderly frailty.

Epidemiological characteristics and risk prediction model of pulmonary infection in elderly patients in a hospital in Hebei Province / 公共卫生与预防医学

Objective To analyze the epidemiological characteristics and influencing factors of pulmonary infection in the elderly, and to construct a risk prediction model. Methods Stratified cluster sampling was used to randomly select 683 elderly patients in Zhangjiakou First Hospital as the investigation subjects. Sputum specimens were collected and sent for bacterial isolation, culture, identification, and drug sensitivity test. According to whether the patients had pulmonary infection, they were divided into pulmonary infection group (n=315) and non-pulmonary infection group (n=368). The clinical data of the two groups such as age, sex, COPD, and ICU admission were analyzed. Univariate analysis and logistic regression analysis were used to analyze the influencing factors of pulmonary infection in elderly patients, and a risk prediction model was established. Results A total of 331 strains of pathogenic bacteria were detected in 315 patients with pulmonary infection, and there were 207 strains (62.54%) of gram-negative bacteria detected, mainly including 95 strains (28.70%) of Acinetobacter baumannii and 71 strains (21.45%) of Klebsiella pneumoniae. There were 169 strains (26.28%) of gram-positive bacteria detected, mainly 68 strains (20.54%) of Staphylococcus aureus. In addition, there were 25 strains of fungi (7.55%). There were no significant differences in gender, smoking history, history of COPD, asthma, and stroke between the two groups (P>0.05). The proportion of patients aged≥70, mechanical ventilation, admission to ICU and recent respiratory tract infection in the experimental group was significantly higher than that in the control group (P<0.05). Multivariate logistic regression analysis showed that age, smoking history, mechanical ventilation, and ICU admission were independent risk factors for pulmonary infection in elderly patients (P<0.05). According to the above four independent influencing factors and corresponding regression coefficient of each factor, the prediction model of pulmonary infection in elderly patients was constructed, Z=-5.948+1.198× (age) +1.281×(smoking history) +2.029×(mechanical ventilation) +1.211×(ICU admission). Conclusion Lung infection in elderly patients in our hospital is dominated by gram-negative bacilli. Antibiotics should be rationally selected according to drug sensitivity results. Age≥70 years old and COPD can increase the risk of pulmonary infection in elderly patients, and the prediction model constructed can effectively predict the occurrence of pulmonary infection in elderly patients.

Establishment and effect evaluation of risk prediction model for lung infection after kidney transplantation / 器官移植

Objective To establish and evaluate the predictive value of the risk prediction model for lung infection within postoperative 1 year in kidney transplant recipients. Methods Clinical data of 197 kidney transplant recipients were retrospectively analyzed. All recipients were divided into the infection group (n=42) and non-infection group (n=155) according to the incidence of lung infection within postoperative 1 year. The incidence and risk factors of lung infection after kidney transplantation were analyzed. Risk prediction model was established by multiple logistic regression analysis. Forty-five kidney transplant recipients who met the inclusion criteria, including 8 cases in the infection group and 37 cases in the non-infection group, were selected to verify the predictive effect of the established model. Results The incidence of lung infection within 1 year after kidney transplantation was 21.3% (n=42), including 38 cases (90%) of pneumonia severity index (PSI) class Ⅰ, 1 case (2%) of PSI class Ⅲ and 3 cases (8%) of PSI class Ⅴ. Lung infection occurred within 1 month after operation in 13 cases, within postoperative 2-6 months in 22 cases and after postoperative 6 months in 7 cases. Nineteen recipients were diagnosed with bacterial infection, 7 cases of fungal infection, 10 cases of viral infection and 6 cases of mixed infection. Smoking history, diabetes mellitus history, pulmonary disease history and albumin level of < 35 g/L were the independent risk factors for lung infection after kidney transplantation (all P < 0.05). The equation of risk prediction model for postoperative lung infection in kidney transplant recipients was logit (lung infection within postoperative 1 year in kidney transplant recipients)=-1.891+1.063×smoking history (yes=1, no=0)+1.398×diabetes mellitus history (yes=1, no=0)+1.732×pulmonary disease history (yes=1, no=0)+1.269×albumin level (< 35 g/L=1, ≥35 g/L=0). The area under the curve (AUC) of receiver operating characteristic (ROC) was 0.788, the sensitivity was 0.786, the specificity was 0.645, and the Youden index was 0.431, respectively. Hosmer-Lemeshow goodness-of-fit test demonstrated that the predicted value of this model yielded relatively high consistency with the observed value. The AUC in the verification group was 0.834. Hosmer-Lemeshow goodness-of-fit test validated high degree of calibration of this model. Conclusions The risk prediction model, consisting of smoking history, diabetes mellitus history, pulmonary disease history and albumin level as predictors, may effectively predict the incidence of lung infection within postoperative 1 year in kidney transplant recipients.

Construction and application of a risk prediction model for moderate to severe cancer-related fatigue in patients undergoing chemotherapy of prostate cancer / 中国实用护理杂志

Objective:To investigate the risk factors of moderate to severe cancer-related fatigue (CRF) in patients undergoing chemotherapy of prostate cancer, and to construct a nomogram model to predict the occurrence of CRF.Methods:Using the case data questionnaire, Brief Fatigue Inventory, Social Support Rating Scale and International Prostate Symptom Scores, 724 patients of prostate cancer treated by chemotherapy in Shanghai Tenth People′s Hospital from August 2016 to June 2021 were selected and were treated with 1∶1 ratio, and the indexes of the moderate and severe CRF group (216 cases) and the non-moderate and severe CRF group (216 cases) were compared. According to the ratio of 7∶3, the envelope method was used to divide into training set and validation set. The independent risk factors of moderate and severe CRF were explored by univariate analysis and multivariate Logistic regression analysis, and the risk prediction model was established and the nomogram model was constructed. The C-index and area under ROC curve were used to verify the prediction effect of the model.Results:Multivariate Logistic regression analysis showed that BMI ranged from 24.0 to 27.9 kg/m 2 ( OR=1.733), BMI≥28.0 kg/m 2 ( OR=3.126), neutropenia occurred during chemotherapy ( OR=1.747), chemotherapy course >6 months ( OR=1.893), moderate social support level ( OR=1.244), low social support level ( OR=2.434), mild urinary tract symptoms ( OR=1.264), moderate urinary tract symptoms ( OR=3.371) and severe urinary tract symptoms ( OR=5.297) were independent risk factors for moderate and severe CRF. The nomogram model constructed according to the above risk factors was internally verified by the training set and the validation set, and its C-index was 0.854 and 0.741 respectively. The area under ROC curve training set was 0.823, and the validation set was 0.733. Conclusions:The nomogram model can effectively predict the occurrence of moderate to severe CRF in patients with prostate cancer undergoing chemotherapy.

Establishment and analysis of risk prediction model and nomogram of pediatric outpatient nursing medication errors / 中国实用护理杂志

Objective:To analyze the influencing factors of nursing medication errors in pediatric outpatient to establish risk prediction model and nomogram for helping pediatric outpatient nurses prevent nursing medication errors.Methods:Totally 172 cases of pediatric outpatient nursing children with nursing medication errors were selected from January 2013 to December 2020 in the Children′s Hospital Affiliated to Zhejiang University School of Medicine as the medication error group and 344 cases of children without medication errors as the non-medication error group, and the indicators of the two groups were compared and analyzed.Results:According to Logistic regression analysis, children′s age, children′s enrollment, administration of narcotic analgesics, administration of anti-fever drugs, micro pump intravenous route of administration, subcutaneous route of administration, nursing age of administration nurse, title of administration nurse were all significant influencing factors of pediatric outpatient nursing administration errors ( P<0.05). To construct the risk prediction model of pediatric outpatient nursing administration error and its AUC area of the model was 0.89 (95% CI 0.84-0.94, P<0.01). The total score of the nomogram was 35-259, and the risk rate was 0.001-0.999. The higher the total score was, the higher the risk of medication errors was. Conclusions:The establishment of pediatric outpatient nursing medication error prediction model and nomogram can provide reference model and tools for the prevention of pediatric outpatient nursing medication error, which has certain clinical value for ensuring the safety of pediatric outpatient medication.

Risk-Prediction Model for Transfusion of Erythrocyte Concentrate During Extracorporeal Circulation in Coronary Surgery

Abstract Introduction: Our objective was to identify preoperative risk factors and to develop and validate a risk-prediction model for the need for blood (erythrocyte concentrate [EC]) transfusion during extracorporeal circulation (ECC) in patients undergoing coronary artery bypass grafting (CABG). Methods: This is a retrospective observational study including 530 consecutive patients who underwent isolated on-pump CABG at our Centre over a full two-year period. The risk model was developed and validated by logistic regression and bootstrap analysis. Discrimination and calibration were assessed using the area under the receiver operating characteristic curve (AUC) and the Hosmer-Lemeshow (H-L) test, respectively. Results: EC transfusion during ECC was required in 91 patients (17.2%). Of these, the majority were transfused with one (54.9%) or two (41.8%) EC units. The final model covariates (reported as odds ratios; 95% confidence interval) were age (1.07; 1.02-1.13), glomerular filtration rate (0.98; 0.96-1.00), body surface area (0.95; 0.92-0.98), peripheral vascular disease (3.03; 1.01-9.05), cerebrovascular disease (4.58; 1.29-16.18), and hematocrit (0.55; 0.48-0.63). The risk model developed has an excellent discriminatory power (AUC: 0,963). The results of the H-L test showed that the model predicts accurately both on average and across the ranges of deciles of risk. Conclusions: A risk-prediction model for EC transfusion during ECC was developed, which performed adequately in terms of discrimination, calibration, and stability over a wide spectrum of risk. It can be used as an instrument to provide accurate information about the need for EC transfusion during ECC, and as a valuable adjunct for local improvement of clinical practice. OR=odds ratio Key Question: What is the risk of the need for use of erythrocyte concentrate (EC) during cardiopulmonary bypass? Key Findings: Risk factors with the greatest prediction for EC transfusion. Take-Home Message: The implementation of this model would be an important step in optimizing and improving the quality of surgery.

Construction and validation of a risk prediction model for critically ill patients with stress injury / 中国实用护理杂志

Objective:To construct a pressure injury risk prediction model for critical patients and verify its prediction effect.Methods:A cohort study was conducted to collect relevant data of critical patients hospitalized in the Intensive Care Unit from February 2019 to September 2019. The occurrence of pressure injuries was used as a dependent variable to conduct a single factor and multiple factor analyses of relevant data and establish predictive models. The risk stratification and predictive effect tests were also performed.Results:There were 329 critical patients and 48 cases of pressure injuries. The single factor analysis of 11 factors showed that blood lactate, body temperature, ICU hospitalization days, Braden score, consciousness state, age and booster drug treatment were the suspicious factors of stress injury, and the difference was statistically significant ( Z value was 2.575-3.694, χ 2 values were 6.800, 30.510, 6.344, P<0.05 or 0.01); The results of the binary Logistic regression analysis showed that the independent influencing factors for the occurrence of pressure injuries included the patient′s body temperature within 24 hours after entering the ICU, the Braden score, state of consciousness, age and ICU hospitalization duration ( P<0.05 or 0.01). A prediction model was established. The likelihood ratio chi-square proved that the model was statistically significant and fitted well. The sensitivity was 66.7% and the specificity was 72.2%. The risk stratification of the model was performed. The difference between the high-risk group and the low-risk group was statistically significant ( t value was -33.371, P<0.01); the validation set was used to test the prediction effect, and the area under the ROC curve was 0.758. Conclusions:The constructed prediction model is a scientific combination of objective indicators of the clinical characteristics of critical patients, which is statistically significant; the model can predict critical patients’ risks of pressure injuries; it also has a good degree of discrimination, which can provide a theoretical basis for the risk management of critical patients with great clinical application value.