Objective To develop a nomogram to predict surgical site infection (SSI) after spinal surgery, so as to help medical staff to plan preventive strategies to reduce the incidence of SSI. Methods We retrospectively selected 2 348 patients undergoing spinal surgery at Wenzhou People's Hospital from May 2012 to May 2019. Univariate and multivariate logistic regression models were used to determine independent predictors for SSI after spinal surgery and a nomogram was constructed to predict SSI after spinal surgery based on each predictive factor. The area under the receiver operating characteristic curve (AUC) was used to assess the discriminatory ability of the nomogram and the consistency of the nomogram was tested using calibration plot and Hosmer-Lemeshow  goodness of fit test (H-L test). Results In univariate and multivariate analysis, current smoking (OR=2.242, 95%CI:1.125-4.469), diabetes (OR=0.932, 95%CI:1.260-5.113), operation time≥180 min (R=7.256, 95%CI:3.537-14.886), ASA ≥ grade III (OR=2.963, 95% CI:1.454-6.037) and blood transfusion (autologous blood OR=3.685, 95%CI:1.203-11.284; allogeneic blood OR=6.443, 95%CI:2.765-15.017) were identified as significant independent predictors. The nomogram was developed using these independent predictors. The AUC of the multivariate model for discrimination was 0.818 (95%CI:0.802-0.863). The calibration plot and H-L test (χ2=12.337, P=0.195) showed good consistency between predicted and observed outcomes. Conclusion The nomogram based on smoking, diabetes, operation time, ASA classification and blood transfusion has potential as a clinically useful predictive tool of SSI after spinal surgery.

 

Objective To understand the distribution characteristics of inpatient deaths in a hospital, find out the cause of the changing trend of inpatient death rate, and provide reference for hospital management and area public health service. Methods The data on first pages of medical records were collected in a tertiary hospital from 2010 to 2019, and death causes were classified according to the classification criteria of ICD-10. The cross-table analysis, chi-square test and correlation analysis were used to analyze the basic information about patients, the rank of death causes, the changing of inpatient mortality. Results Cancer and cardiovascular disease were the top two causes of death for both men and women. The nosocomial mortality had been falling in the last decade. This falling had an intense relationship with decrease in quantities of cardiovascular disease and tumor. The deaths from respiratory diseases in men were on the rise. Conclusion Hospitals should pay more attention to the prevention, diagnosis, treatment about tumor, cardiovascular disease and respiratory diseases to reduce the incidence and mortality of these diseases and improve the qualify of life.

In this article, we establish the multidimensional assessment rules covering ethics, ability, frequency, performance and other dimensions, by taking quality of care as the core, supplemented by taking quantity and benefit into consideration. The assessment results will be converted into percentages in proportion and linked to the performance bonus. After implementing the performance bonus in ward nursing, the integrated quality of nurses has been comprehensively improved, thus guaranteeing the nursing safety and promoting the sustainable development of the hospital.

Objective To evaluate the implementation of the open medical project in Shaoxing tertiary general hospitals by using entropy weight method. Methods Ten representative indicators were selected as the evaluation index system, and the entropy weight method was used to determine the weight of each evaluation index. Results According to the entropy weight method, the ranking of 7 tertiary general hospitals were D, C, F, G, B, E and A. Conclusion The evaluation results obtained by using the entropy weight method were consistent with the current status of the implementation of open medical project in various hospitals. The entropy weight method can objectively reflect the implementation of open medical project.

Objective To understand the cognition of doctors and patients on graded healthcare, and to explore the factors affecting the patient's willingness to seek medical treatment and the awareness of the doctors and patients on the classification and treatment. Methods The accidental sampling method was adopted. One hundred and eighty-three patients and 92 medical staff of a tertiary hospital in Binzhou were investigated with questionnaires and unstructured interviews. According to logistic regression, the factors affecting the cognition of both doctors and patients on graded healthcare were analyzed. Results Among the surveyed subjects, 38.2% of the patients said they understood the hierarchical diagnosis and treatment system, and 44.1% of the patients chose the primary hospital for the first consultation when they were sick.The higher the educational background, the higher the awareness of graded diagnosis and treatment (P<0.001).And patients who understand the hierarchical diagnosis and treatment system are more inclined to choose primary hospitals for the first diagnosis.70.6% of medical staff said they understood the hierarchical diagnosis and treatment system, but medical staff in different positions have different perceptions of hierarchical diagnosis and treatment, and clinicians have higher awareness of hierarchical diagnosis and treatment (P<0.001). Conclusion The awareness of both doctors and patients on hierarchical diagnosis and treatment needs to be improved, and the willingness of patients to choose primary care at the grassroots level is also low.Both doctors and patients should strengthen the promotion of graded diagnosis and treatment, increase the awareness of both doctors and patients, and increase investment in primary medical institutions, actively improve the primary diagnosis and treatment environment, comprehensively improve primary service capabilities, and attract patients to choose primary diagnosis.

Objective To explore positive screening rate of syphilis in the blood donors in Wenzhou, analyze its epidemiological characteristics and provide scientific basis for its prevention and control. Methods ELISA was used to detect the syphilis antibodies in the blood samples from voluntary donors in Wenzhou from 2016 to 2018. The detection and epidemiological characteristics of syphilis in the blood donors were analyzed. Results Totally 286 322 valid blood donation samples were received in Wenzhou from 2016 to 2018, of which 1 011 were positive in the primary screening of syphilis and the positive rate was 0.35%. There was lack of significance between the male (0.35%) and the female (0.36%) regarding the prevalence of syphilis (P>0.05). The positive rates of different age groups were significantly different. Results from pairwise comparison suggested that the positive rate of syphilis in the 31-45 years old group (0.40%) was significantly higher than that in the 18-30 years old group (0.33%) and the 46-60 years old group (0.31%). There were significant differences among different occupation groups and pairwise comparisons suggested: others (0.62%)>businessmen/the self-employed (0.38%)>teachers/cadres (0.16%)>students (0.09%), (P<0.001). There were significant differences among different education groups and pairwise comparisons suggested: junior high school and below (0.53%)>senior high School / technical secondary school (0.37%)>junior college and above (0.24%), (P<0.001). There was lack of significance among different blood types, different blood collection sites regarding the prevalence of syphilis (P>0.05). Conclusion According to the characteristics of syphilis in Wenzhou, the key prevention and control measures should be taken for the female, low age group, high age group, other occupations, businessmen/selfemployed, and low education population, so as to further reduce the incidence of syphilis among the population.

Objective  To understand the current status of social support for medical staff in the epidemic of the corona virus disease 2019 (COVID-19) and explore its influencing factors, in order to improve the social support of medical staff and provide a scientific basis for improving their physical and mental health. Methods General information questionnaires and social support rating scale (SSRS) were used to conduct an online questionnaire survey on 707 medical staff in 6 secondary and higher hospitals in Taizhou to understand the current social support status of medical staff in the epidemic situation. Univariate analysis and binary logistic regression analysis were used to analyze the influencing factors. Results The recovery rate of the effective questionnaires was 98.16%, and the total score of the social support scale of 694 medical staff was (35.20±7.33), which was not statistically different from the domestic norm score (34.56±3.73) (P=0.335). Medical staff with old age, high professional title and high family income per capita and married had higher total scores of social support, subjective support dimensions, and objective support dimensions than others (P<0.05); the total score of social support, the scores of subjective support dimension and objective support dimension, and support utilization dimension score of medical staff without fear of new coronary pneumonia, without sense of conscious uselessness, and with high sleep quality were higher than those in other groups (P<0.05). The analysis results of the influencing factors of low-level social support of medical staff showed that married medical workers (OR=0.25, 95%CI: 0.10-0.62) and those with high sleep quality (OR=0.34, 95%CI: 0.11-1.00) were less likely to obtain low level of social support, which was a protective factor. Conscious uselessness (OR=2.59, 95% CI: 0.93-7.19) was a risk factor. Conclusion Under the epidemic situation, medical staff had received due social support, and there were many influencing factors. Emphasis should be placed on the social support of medical staff who are unmarried/ divorced, poor sleep quality or have a strong sense of uselessness to better meet the needs of their physical and mental health.

Objective To investigate nurses′ intent to stay during COVID-19 and explore its influencing factors. Methods With the help of the wenjuanxing online survey platform, the general information questionnaire, nurses′ job stressor scale and nurses′ intent to stay questionnaire were used for the survey. Results Totally 519 valid questionnaires were received in this study. The total score of nurses′ intent to stay was (22.12±4.36), which was higher than the results of domestic research in recent 5 years (P<0.05). The independent influencing factors of nurses′ intent to stay were professional choice, working years, whether the department had colleagues to help Hubei, whether to let their children study medicine, problems in nursing profession and work, problems in patient care, and problems in management and interpersonal relationship (P<0.05). Conclusion The perceived intent to stay is above the middle level, which is higher than many domestic researches in recent 5 years. However, nursing managers still need to pay attention to the reasonable needs of nurses, optimize manpower allocation, and further improve the nurses′ intent to stay.

Objective  Ensemble learning is a novel approach to improving learning accuracy in machine learning field recently. This paper aims to introduce the application of ensemble learning method based on super learner algorithm in the prediction modeling of longitudinal censored data and its implementation of R language. Methods This paper introduced the principle in modeling longitudinal censored data based on super learner algorithm and its implementation method with R-programming language. In addition, tumor survival data from TCGA database were used for real data analysis to illustrate its performance in practice. Results The estimation methods for model parameters and definition of ensemble learning parameters based on super learner algorithm are more flexible. In actual data analysis, super learner algorithm can make full use of the obtained data to establish the prediction model. The prediction accuracy of the model is 0.8737 (95% CI: 0.7897-0.9330) and the C-index is 0.883, so the prediction performance is good. Conclusion The ensemble learning approach with super learner algorithm provides a new choice for the prediction analysis based on longitudinal censored data.

Objective To investigate if the class imbalance poses additional challenges when dealing with class prediction of gene expression data, to evaluate the performance of seven types of classifiers on class-imbalanced data on a publicly available data set, in order to provide a theoretical basis for subsequent research. Methods We obtained different levels of class imbalance by repeatedly randomly selecting subsets of the samples from the public data set: training sets (N_n=10 vs N_p=10,15,20,30,35, and N_n=15 vs N_p=5,10,15,25,30); the test sets were balanced (N_n=20 vs N_p=20). We trained the classifiers on them and compare the classification effect of single sampling classification and 100 sampling average. Results As the number of positive samples in the data set increases, the overall sensitivity of the classification algorithm shows an upward trend, while the specificity shows a downward trend. In the colon cancer data set, AdaBoost, NB, and RF algorithms performed well, while support vector machines performed poorly and were unstable. In the leukemia data set, the overall performance of the NB algorithm is the best and stable, and the classification effect of AdaBoost, C4.5 and RF algorithms is better but fluctuates greatly.Conclusion Class imbalance, data characteristics and algorithm selection affect classification for high-dimensional data. And the results of a single analysis are contingent, with poor reproducibility. Researchers using class-imbalanced data should be careful in selecting algorithm and they should always use an appropriate method for dealing with the class imbalance problem.