Objective To explore the operation code of main diagnosis and treatment methods of lung cancer.Methods The main diagnostic operations in the field of lung cancer mainly include transbronchial needle aspiration biopsy, transbron chial lung biopsy, transbronchoscope pollution protection brush technology and bronchoalveolar lavage, etc.; while the therapeutic operations mainly include high frequency electric cauterization technology, cryoablation, endotracheal stent implantation, radiofrequency ablation of lung cancer and radioactive particle implantation through bronchoscope, etc. In this paper, we analyzed the clinical knowledge of the aforementioned diagnosis and treatment operations, and then sorted out and summarized the corresponding codes based on the classification principles of ICD-9-CM-3.Results Different diagnosis and treatment methods, operation sites and approaches led to different operation classification codes.Conclusion In order to accurately code and fully express medical record information, coders need to fully understand the latest clinical technologies, master the principles of surgical classification, and review the medical records carefully.

 

Objective  To explore the drug treatment compliance and its correlated influencing factors of type 2 diabetes mellitus (T2DM), and provided intervention strategy for it.Methods Totally 220 cases with T2DM were investigated with following-up, and theses patients were divided into the good compliance group and the poor compliance group according to the compliance of treatment. The correlated risk factors were screened by multiple factors non-conditional logistic regression method.ResultsThe poor compliance rate of T2DM patients in this study was 37.3% (82/220). Multiple factors analysis result indicated that the factors of high age (OR=2.537，95%CI：1.118～5.417), bad relationship between doctors and patients (OR=3.387，95%CI：1.777～6.455), lack of knowledge of diabetes (OR=3.428，95%CI：1.897～6.196), adverse drug reaction (OR=3.869，95%CI：2.157～6.938), many kinds of drugs (OR=3.209，95%CI：1.741～5.915) and complications (OR=5.523，95%CI：2.760～11.055).Conclusion There is a bad drug compliance of treatment of T2 DM patients. We should enhance health education, followingup and management of the elderly patients, and medication instruction after leaving hospital, and we should also correctly cope with doctorpatient relationship, in order to raise treatment compliance of patients.

The implementation of the “14th FiveYear Plan” national medical security plan has accelerated the pace of perfecting the multi-tiered medical security system, continuously improved the medical insurance payment mechanism, and announced the implementation of the DRG/DIP payment mode reform three-year action plan in order to improve the quality, reduce the cost and increase the efficiency of the medical insurance funds. Medical insurance payment policy is closely related to performance management of hospitals. Therefore, this paper analyzes the impact of DRG/DIP payment reform on performance management of public hospitals from the perspective of performance management. And from the five aspects of cost accounting and analysis, talent construction, performance evaluation system construction, information construction and culture construction, it explores the performance management path of public hospitals under the background of DRG/DIP payment, and puts forward targeted suggestions for the performance managemet of public hospitals, in order to inject momentum into the high-quality development of public hospitals.

Objective To learn about the pension willingness and factors affecting the choice of home care willingness of the elderly population, and to provide the evidence for improving life status of the elderly population, better providing full cycle services, and promoting healthy aging. Methods Data of the Sixth National Health Service Statistical Survey were used to analyze the nursing situation and pension willingness of the elderly population in different regions, different family conditions, different social demographic characteristics and different disability status, and logistic regression was used to analyze the influencing factors of the elderly population′s choice of homebased care. Results Eleven point eight percent of the elderly needed to be taken care of for their daily life, and daily care was provided 52.5% by spouses and 42.9% by children. And 90.7% of the elderly were inclined to home-based care, urban and rural areas, sex, age, marital status, educational level, family economic status, disability degree, the number of chronic diseases, family size were all influencing factors of elderly pension willingness. The probability of wanting home care was lower in urban areas than in rural areas (OR=0.625,  95%CI: 0.589-0.663); it was higher for men than for wornen(OR=1.122,95%CI:1.062-1.187); and it was higher for 70-79 years old (OR=1.219, 95%CI:1.147-1.296) and 80 years old and above (OR=1.598, 95%CI: 1.433-1.782) than for 60-69 years old; it was lower for unmarried (OR=0.476) and divorce (OR=0.600) than for married; it was lower for the elderly with educational levels of primary school (OR=0.742), junior middle school (OR=0.437), and senior high school or above（OR=0.352）than for those without literacy (P<0.01); it was higher for elderly people not living alone than for those living alone (P<0.01).  Conclusion The elderly with older age, lower educational level, poor family economic status and large family size are more likely to hope for care at home.Combined with the elderly′s pension willingness, in view of the health status, nursing and care needs, it is necessary to establish and improve a service system based on family care, to explore the design of long-term care system, and to protect the health rights and interests of the elderly.

Objective To develop a simple random sampling method suitable for researchers at all levels.Methods By using the inherent functions of rank (), rand (), vlookup () and if () of Excel and WPS softwares, the simple random sampling method is designed and the reusable sampling tool is customized.Results The simple random sampling method based on the function design of Excel or WPS can realize the simple random sampling of millions of research population in a few seconds. At the same time, it has the characteristics of simple design and use, short development time and reusability.Conclusion Compared with the sampling tool of Excel, which is only applicable to the data of digital population, the simple random sampling method based on the function design of Excel or WPS can not only realize the sampling of all character types, but also does not need to install add-on items, which is more suitable for health personnel at all levels.

Data analysis and data visualization has been widely used in various industries. This article introduces the main content of IBCS, Material Design data  visualization design specifications and AntV data visualization specifications, and discusses the key elements of data visualization specifications.

Objective To introduce the basic principle of 10-fold cross-validation, and use R to show the application of 10-fold cross-validation in the evaluation of the generalization ability of prediction models in an example. Methods The logistic regression model was used with 10-fold cross-validation as the evaluation method to predict the treatment outcome of drug-resistant tuberculosis patients. At the same time, the 5-fold, 10-fold, and leave-one-out cross-validation were used to compare the parameters of evaluation. Results Different divisions of training and test sets have influence on the parameters of evaluation. Compared with other cross-validation methods, the parameters of evaluation from 10-fold cross-validation have a higher stability and efficiency than other numbers of fold. Conclusion In the evaluation of the application effects of machine learning models, the comprehensive performance of 10-fold cross-validation is prominent and the generalization ability of different models can be objectively measured.

Since 2019, tertiary public hospitals across the country have carried out performance appraisal. Under the new situation of medical reform, how to improve the quality management level is a challenge to the hospital. This paper makes a comprehensive analysis on the relationship between performance appraisal and quality management of tertiary public hospitals and its impact on quality management. And it discusses the strategies to improve the performance assessment of tertiary public hospitals through continuous improvement and prospective management by taking the performance assessment policy of tertiary public hospitals as the work guide, perfecting the quality management system, establishing the quality management database, realizing information management, playing the role of quality control, and improving the hospital performance assessment results through continuous improvement and prospective management.

Objective To determine the annual priority improvement indexes of hospitals' quality and safety with delphi method and analytic hierarchy process.Methods Delphi method is used to establish the annual index system for optimization, and AHP is used to establish evaluation index system and check consistency of the judgment matrix.Results Nine hospitals′ annual improvement priority indexes are selectedand consistency check results are good.Conclusion Using Delphi method and AHP to determine the hospitals′ improved indicators is quite scientific and feasible. These two methods are effective tools for hospital management.

Objective To provide a fast and rich visualized-template analysis method for the hospital statisticians. Methods The hospital performance data from January 1, 2017 to June 30, 2018 were extracted. After data import, data type conversion, data relationship, and data modeling method, using the builtin visualization template, combined with the actual demand and display effect, the study constructed a series of delicate visualized reports from the overall technical difficulty of medical services, medical efficiency, medical safety, and key specialist capabilities. Results The study produced several reports such as an overview of clinical department data indicators, RW distribution of clinical department, clinical department surgery, and detailed query of clinical department medical records. These reports have the characteristics of high degree of visualization, rich chart, convenient and quick customization. Conclusion Power BI has the characteristics of not relying on IT department, low threshold and a high degree of customization, rich visual template and easy function to understand, and can be used as the preferred tool for hospital statisticians to analyze hospital performance data.

Objective To analyze the defective problems existing in the data filling in the front page of medical records of the high rate cases whose cost were more than twice the average cost based on DRGs payment in a hospital, and to put forward improvement measures.Methods The hospital received the high rate cases which were grouped by DRGs and settled expenses between January, 2018 and June, 2018 from the Social Security Bureau's feedback. Combined with DRGs payment mode, under the guidance of the diagnosis operation selection standard announced by the Health Development Planning Commission and the Principles of International Classification of Diseases, selfcheck and retrospective analysis were conducted, and the defects were classified and counted with 6 cases of high rate paying.Results Totally 111 of 526 high rate cases had defect problems related to clinicians and coders, with defective rate 21.11%, and among them, 47, 20, 12 and 29 cases were clinicians' main diagnostic error selection, main operation error selection , main diagnosis and operation error selection, and missing of diagnosis or operation, respectively. The percentage of errors was 42.34%, 18.02%, 10.81% and 26.13% respectively. The coders' main diagnosis coding error and main surgical coding error were 2 and 1 cases respectively, with error rate 1.8% and 0.9% respectively.Conclusion The high rate paying cases caused by the defects of the filling diagnosis and operation and the coding defects on the front page of medical records were closely related to the quality of filling by clinicians and the coding level of coders. Effective measures are to train clinicians to correctly select diagnosis and operation, improve coders' professional skills, and optimize coders' work, so as to reduce high rate cases caused by defective problems.

Objective To explore the relationship between sleep quality and well-being in the elderly and the intermediary effect of state-anxiety between them. Methods One hundred and fifty-five elderly patients with insomnia in the department of psychosomatic medicine of the First Affiliated Hospital of Nanchang University from September 2018 to May 2019 were investigated with Pittsburgh sleeping quality index (PSQI), state-trait anxiety questionnaire (STAI) and memorial university of Newfoundland scale of happiness (MUNSH). Results The STAI score of female patients was higher than that of male patients, and the MUNSH score of patients living alone was lower than that of non-living alone patients (P<0.05). Sleep quality index score was negatively correlated with well-being (r=-0.557, P<0.05), state-trait anxiety was negatively correlated with well-being (r=-0.634, P<0.05), and state-trait anxiety was positively correlated with sleep quality index score (r=0.510, P<0.05). Structural equation model showed that sleep quality had a negative effect on happiness (β=-0.28, P<0.05) and a positive effect on state-trait anxiety (β=0.70, P<0.05), and state-trait anxiety had a negative effect on well-being (β=-0.64, P<0.05). Conclusion State-trait anxiety plays an intermediary role in elderly patients with insomnia. Sleep quality not only directly affects happiness, but also affects the happiness of elderly patients with insomnia through state-trait anxiety.

Objective To explore the application status and hot research trend of DRG in hospital management in China and provide reference and thinking for the refined management of my country.Methods The relevant literatures of DRG in hospital management in China collected by CNKI and Wanfang database were taken as the research objects, and CiteSpace software was used to analyze the authors, institutions and key words in the literatures.Results Based on the analysis of 724 articles after weight removal, it was found that the number of articles has increased rapidly in recent years, and the research team of core authors was formed, but the research institutions were scattered and inter-agency cooperation needed to be strengthened. Sixteen clusters have been formed. Performance evaluation, cost management and DRG payment have become hot issues in hospital-management  in recent years.Conclusion The role of DRG in hospital management is a process of continuous improvement and dynamic improvement. It is of great significance for hospital managers to strengthen their active exploration in DRG for the fine management of hospitals under the current background.

Objective To implement the procedure of distributed lag non-linear model (DLNM) and multivariate meta-analysis in R software.Methods A specific step of two-stage analysis of DLNM and multivariate meta-analysis was carried out in R software through a time series data.Results The two-stage analysis of DLNM and multivariate meta-analysis was successfully implemented in R software.Conclusion R software can be used to construct DLNM and conduct multivariate meta-analysis flexibly and conveniently, and the method can be widely used in environmental epidemiology.

Objective There are more variant subtypes of thyroid papillary carcinoma, which cause difficult coding of thyroid papillary carcinoma and are prone to miscoding. According to the histological classification criteria of thyroid tumors and the ICD-O-3 principle, the coding characteristics of thyroid papillary carcinoma are analyzed to improve the coding accuracy rate. Methods  The hospital medical record information inquiry system was used to retrieve 417 cases of thyroid papillary cancer diagnosed from January 2016 to December 2018 in a hospital. The ICD-O-3 classification rule was used to analyze the morphological coding situation of thyroid papillary carcinoma in a hospital for nearly 3 years. The existing problems were summarized and analyzed. Results There were 159 coding errors and the error rate was 38.13%. Among them, 2 cases of NOS were misclassified in M8000/3 malignant tumor, accounting for 0.48%; 149 cases were misclassified in M8050/3 papillary carcinoma, accounting for 35.73%; the disease classifier was not familiar with the pathological variant subtype of papillary thyroid carcinoma. Pathological diagnosis had given specific subtypes, but was misclassified in M8260/3 papillary adenocarcinoma in 8 cases, accounting for 1.92%; missing morphological morphology was 1 case, accounting for 0.24%.Conclusion Disease classifiers are unfamiliar with the pathological features and variant subtypes of thyroid papillary carcinoma, lack the homepage quality control link, and the coding library is relatively lagging behind clinical development. These reasons are likely to cause coding errors, so disease classification personnel must carefully read the medical records and pathology reports, actively communicate with clinicians and patho logists, update relevant knowledge structures, and continuously improve the accuracy of thyroid papillary cancer coding.

目的 探讨多组等级资料的秩和检验及组间两两比较的SAS实现。方法 结合实例介绍多组等级资料Kruskal-Wallis H检验和两两比较Nemenyi检验的计算方法和步骤,通过编制SAS程序一次性完成Kruskal-Wallis H检验和Nemenyi检验。结果 运行SAS程序得到可靠结果。结论编制SAS程序可以有效实现多组等级资料的秩和检验及组间两两比较。

Schwannoma is a benign tumor of peripheral and central nerves, which belongs to the nerve tumors. Due to the lack of clinical medical knowledge, the lack of familiarity with the disease types and surgical methods of schwannoma, the over-reliance on computer coding, the lack of code review and quality control in the coding system, and other reasons, schwannoma is often miscoded as benign tumor at the site of disease. This paper introduces the clinical features, diseases and surgical codes of schwannoma, and analyzes the reasons of coding errors through a retrospective study of 3 typical cases of schwannoma, and summarizes the coding ideas of schwannoma. Medical institutions should attach great importance to coding, constantly improve the professional ability and responsibility of the coders, and improve the accuracy of disease and surgical coding through the establishment of specialist coders and the establishment of the code review mechanism on the front page of medical records.

目的 研究影响医院业务收入的因素,为医院管理决策提供依据。 方法用因素分析法对医院门诊和住院业务收入作相关因素分析。 结果业务收入受就诊人次数、出院病人数及相应次均费用和住院天数的影响。 结论医院应着力提高医疗技术、提供优质服务、扩大服务范围、减轻患者负担,实现“优质、低耗、高效”的开源节流型、质量效益型的现代化医院。

Many diseases in ophthalmology are named after people's names. These diseases are usually named after foreign ophthalmologists who reported the disease for the first time. Clinicians often use English names directly as the diagnosis of these diseases, such as Coats′ disease, Fuchs syndrome, Leber's hereditary optic neuropathy, Duane syndrome, etc. In order to search for the correct codes of these diseases, this paper explores the coding search methods of these ophthalmic diseases, and finds that their codes are H35.0, H20.8, H47.2 and H50.8, respectively. It is concluded that there are generally two methods of selecting leading words for disease coding named by human names: name and clinical manifestations. Choosing persons' names as leading words requires searching for the authoritative Chinese transliteration names of these foreign ophthalmologists in the literature, and using authoritative Chinese transliteration names as leading words to search for codes; choosing clinical manifestations as leading words requires a full understanding of the etiology, development, diagnosis, treatment and prognosis of these diseases in the medical literature, to grasp the clinical manifestations of these diseases.

This article analyzes and elaborates on the connotation of elderly health, the narrow and broad content of elderly health services, as well as safeguard measures. Meanwhile, it combines with long-term practical experience and practice to establish and improve the elderly health service system with a large amount of data and examples. The framework of this system is to establish a comprehensive and continuous elderly health service system covering urban and rural areas, which is based on the "chain-type" medical services and supported by "all-round" health services.

Objective To understand the current status and hotspots of DRG research in China in order to provide reference for the focus and direction of subsequent research.Methods The literatures related to DRG were retrieved from China National Knowledge Network (CNKI). Descriptive analysis was made on the number of DRG literatures and the years of publication. Social network analysis was used to analyze the core authors, cooperative groups, high-frequency keywords and their relationships in DRG literatures.Results There were 2 602 journal articles with the theme of ‘DRG’ or ‘diagnosis related groups’. After 2015, the number of DRG literatures in China showed a multiple increase, with the annual literatures in 2021 and 2022 exceeding 500. There were 1 072 articles from core journals (41.20%) and 1 530 articles from non-core journals (58.80%). There were 188 core authors in the field of DRG research, mainly forming 7 cooperative groups led by Hu Mu and Guo Moning, with a total of 1 425 publications (54.77%). The network cograph and centrality analysis of high-frequency keywords showed that payment method, DRG prepaid, first page of medical records , hospital management, hospitalization cost and performance evaluation were the high-frequency keywords in the field of DRG research, and their point degree centrality and proximity centrality ranked first. However, the first page of medical records and CMI were more mediation-centered.Conclusion The number of literatures in the field of DRG research increases rapidly in recent years, but the quality of literatures needs to be evaluated and improved. The core authors are relatively concentrated and have strong overall influence and academic leading role. The research hotspot of DRG focuses on DRG payment and DRG performance evaluation and hospital management.

Objective To improve the ability of traditional Chinese medicine(TCM) medical service by retrospectively describing and analyzing the clinical diagnosis and treatment characteristics of discharged patients treated mainly by  TCM in a hospital.Methods Discharged patients were divided into the TCM dominant group and the non-TCM dominant group for analysis and comparison, according to the requirement of national performance assessment indicators.Results The differences in population distribution and cost between the two groups were statistically significant (P＜0.05); the number of discharged patients and the proportion of TCM dominant patients were higher for diseases such as posterior circulation ischemia, malignant tumor support treatment and lumbar intervertebral disc herniation; the proportion of DRG group diseases was high, such as alzheimer′s disease or other chronic brain function lesions, coronary atherosclerosis without complications and concomitant diseases.Conclusion Discharged patients treated mainly by TCM had a significant cost advantage, but the average hospitalization days were longer; the diseases with high proportion and high number of TCM dominant discharged patients can be candidate diseases of TCM superiority. It is necessary to actively explore the DRG management scheme in accordance with the characteristics of TCM therapy.

Objective To evaluate the bed efficiency index after correction of each month and each department of a tertiary hospital in 2018, to analyze the reasons for its high efficiency or low efficiency operation, and analyze the influencing factors of the adjusted bed efficiency index, so as to provide reference for the rational allocation of beds in the hospital. Methods The "normalization method" was adopted to calculate the bed efficiency index of each month and each department of a tertiary hospital in 2018, and the CD rate was used for correction.Multiple linear regression method was used to analyze the influencing factors of bed efficiency index after correction. Results For the beds efficiency index in every month, the highest was in July and the lowest was in February.Indexes in every month except that in February were all higher than 0.9.In the adjusted bed efficiency index of all departments, while some department beds were running with high efficiency, there were also some departments with vacant beds. The efficiency index of ICU was the highest (4.06). The linear regression analysis found that the higher the initial number of patients, the number of patients admitted, and the number of patients undergoing two or three types of incision surgery, the higher the emergency admission rate, the higher the bed efficiency index after correction, and the higher the bed efficiency. The more cases of grade 3 and 4 operations, the lower the bed efficiency index. Conclusion Reasonable allocation of hospital beds could be carried out efficiently through administrative intervention and the improvement of medical service mode, and through making use of affecting factors of bed efficiency index.

Objective Based on C# language and R language, a software is developed, which is convenient for clinicians to use. It is used to realize the automation of variance analysis of repeated measurement design data. Methods First, the statistical analysis data are imported into the software by C# language, then the R language command script is invoked to complete the ANOVA and the result output of the repeated measurement design data. Finally, the example in the 5th edition of medical statistics published in 2020 is used, combined with the example results and SPSS operation results, to verify the effect of the automation of the software. Results The results of statistical analysis included basic statistical description, repeated measurement ANOVA and result chart of pairwise comparison within the group. Based on the combination of C# language and R language, the running results of the statistical software are completely consistent with the results of manual calculation and analysis of teaching materials and SPSS operation. The operation is simple and the results are intuitive, which greatly surpasses the repeated operation of manual analysis and SPSS menu selection in efficiency; at the same time, the low-level errors in manual analysis and SPSS operation steps are avoided to the greatest extent, and the results are presented automatically, which reduces the errors in judgment and selection of results, and ensures the accuracy and standardization of results. Conclusion The software can automatically realize the single factor, two factor and two / multi-level ANOVA of repeated measurement design data, which is worthy of clinical application.

Objective It is hoped that the risk of 10-year cardiovascular diseases can be calculated quickly by compiling SAS macro program, so as to identify high-risk groups of cardiovascular diseases.Methods The cardiovascular risk was calculated by runningthe macro program on Framingham, UKPDS and ASCVD risk scores using the data of the China Health and Retirement Longitudinal Study.Results Running the SAS macro program to analyze the partial data of the China Health and Retirement Longitudinal Studyin2011,and the Framingham, UKPDS and ASCVD risk predictions of the sample population were 9.5%(95%CI:8.4%～10.6%),7.1%(95%CI:6.3%～7.8%) and 16.4% (95%CI:15.1%～17.7%).Conclusion The SAS macro program we compiled can effectively calculate the prediction of cardiovascular disease risk by Framingham, UKPDS and ASCVD risk scores in large sample population.

Objective To investigate the prevalence and risk factors of prenatal depression among pregnant women in different stages of pregnancy.MethodsA total of 1,280 pregnant women who came to our obstetrics outpatient clinic during June 2016 to August 2017 was enrolled in this study. The general data of pregnant women were collected and the prenatal depression was analyzed by the Edinburgh Depression Scale (EPDS). The prevalence of depression in early pregnancy, second trimester and third trimester was analyzed, and the risk factors of prenatal depression in pregnant women were explored. Results Among 1 280 pregnant women, there were 417 cases (32.58%) with prenatal depression. There was no significant difference in the prevalence of prenatal depression between first trimester, second trimester and third trimester (P>0.05). Multivariate logistic regression analysis showed no work (OR=3.842,95% CI: 2.097～7.040), disharmonious spouse relationship (OR=4.242, 95% CI 1.561～11.526), abortion history(OR=3.629，95%CI：1.094～12.043). prenatal depression history (OR=6.573, 95% CI: 2.819～15.329), and gender role expectation (OR=2.171, 95% CI:1.080～4.361) were independent risk factors for prenatal depression in pregnant women (all P<0.05).Conclusion Pregnancy women have similar possibility of prenatal depression in different pregnancy stages. No work, disharmonious spouse relationship, abortion history, history of maternity depression, and expectation of fetus gender are all risk factors of prenatal depression in pregnant women, which should be taken precaution.

Objective To develop a SAS macro program to sort data about time stratified case-crossover design.Methods An example of data sorting about child's pneumonia incidence, temperature, PM10 and SO2 was used to design and verify the SAS macro program named time stratified (IN_DSN=, OUT_DSN=, DATE_VAR=, LAG_EFFECT=, LAG_VARS=, COUNTS_VAR=).Results Parameters including input data set, output data set, date variable, lag effect, lag effect related variables and frequency variable can be specified respectively in the SAS macro program, and then SAS will sort the data automatically and output the time stratified casecrossover design data set.Conclusion The application of SAS macro program time stratified is simple, convenient, flexible and efficient. It can avoid errors in data sorting process of time stratified casecrossover design and has certain practicality.

Objective To analyze the level and distribution of the direct economic burden caused by smoking in China, and to provide the basis for making relevant policies. Methods Using epidemiological methods, data from the Sixth National Health Service Survey 2018 were used to analyze the direct economic burden attributable to smoking, as well as the direct economic burden of diseases in different age groups and diseases. Results The direct economic burden of smoking was 82.63 billion yuan in 2018, up 93.0% from 42.80 billion yuan in 2008, with an average annual growth rate of 6.8%. The economic burden of cardiovascular and cerebrovascular diseases was 41.34 billion yuan, accounting for 50.1% of the total economic burden; the economic burden of malignant tumors was 25.86 billion yuan, accounting for 31.3% of the total economic burden. Conclusion The direct economic burden caused by smoking is rising rapidly, so it is necessary to further strengthen policies related to tobacco control and disease prevention, in order to reduce the burden of diseases, and improve the health status of the whole population.

Objective To evaluate the performance of hospital clinic departments scientifically and accurately.Methods A reasonable indicator system was established, twelve indicators were selected which were scientific, ingenious, practical, independent, and operational. Efficacy coefficient method which was weighted through entropy method was applied to performance evaluation. Results The total efficiency coefficients of all clinic departments in the hospital were sorted high to low as the 3^rd Medicine Department, Otolaryngology Department, Dermatology Department, the 2^nd Medicine Department, the 3^rd Surgery Department, the 1^st Medicine Department, the 2^nd Surgery Department, the 1^st Surgery Department, TCM Department and Ophthalmology Department.Conclusion The efficacy coefficient method, which was weighted through entropy method, was used to evaluate the performance of hospital clinic departments. It makes the evaluation more scientific and accurate, and thereby hospital clinic departments’ actual level can be reflected objectively. The efficacy coefficient method, which is based on the principle of multi-objective programming, shows obvious advantages.
   

目的 探讨组段频数资料用SPSS计算百分位数的方法。 方法选用文献数据和百分位数计算方法。对于组段频数资料,为了描述方便,按照计算数据和计算方法的不同,将百分位数的计算方法细分为组中值法、非组中值法和公式法3种。通过作者编制的SPSS百分位数运算程序,得到运算结果。 结果指出了对于相同的组段频数资料用这3种方法得到的百分位数结果存在明显差别,其中组中值法与公式法计算结果比较接近,非组中值法与公式法计算结果差别比较明显。分析了造成这种明显差别的原因是由于SPSS 系统将组中值当作测量值而不是当作组中值进行计算。为了消除这种差别,作者编写了针对组段频数资料用SPSS计算百分位数的通用程序,提出了解决此类问题的正确方法。实证分析表明,该SPSS百分位数运算程序结果正确。 结论对于组段频数资料运用该SPSS百分位数运算程序计算结果准确,避免了用SPSS计算百分位数结果不一致的现象,这对于更好地运用SPSS统计分析软件促进科研工作和管理工作的顺利开展具有重要意义。

Objective To analyze the characteristics, influencing factors and potential risks of adverse drug reactions (ADR) in a tertiary psychiatric hospital, so as to provide reference for clinical prevention and disposal of adverse drug reactions of psychotropic drugs.Methods A retrospective analysis was conducted on 265 ADR case reports due to the use of psychotropic drugs reported by a psychiatric hospital from October 2019 to October 2021.Results Among the 265 reported cases with ADR, 88 were males and 177 were females, with a higher incidence among females. The average age of the patients was (42.23±17.54) years old, with the majority being young and middle-aged. Forty drugs were involved in ADRs reports, of which the number of adverse reaction reports of antipsychotics was the largest, accounting for 66.42%, followed by antidepressants, accounting for 20.38%. The top three drugs in the cumulative number of reported ADR cases were Olanzapine, Quetiapine and Risperidone respectively. The system most affected by ADRs was the digestive system, with 105 cases reported, accounting for 37.23%, followed by the nervous system, with 88 cases reported, accounting for 31.21%.Conclusion The forms of adverse reactions caused by psychiatric drugs are various, and it is difficult to identify them. Clinicians should strengthen the monitoring of the use of psychiatric drugs and use the psychiatric drugs reasonably to ensure the medication safety.

Objective The treatment and overall management of breast cancer is a longterm process, and the use of a computer database system to manage patients can help improve work efficiency. Based on the breast cancer database from the cloud platform, this article discussed its design and application experience.  Methods The system adopted browser/server architecture, the server adopted one cloud server from Alibaba cloud and one cloud server from Tencent cloud, the database software adopted MySQL, and the server-side scripting language adopted PHP. The coders established a variety of tables in the database to store relevant data, and wrote dynamic webpage code to achieve functions like data entry, modification, and database management. Results A breast cancer database management system based on cloud platform was successfully established (computer software copyright registration No.: 2021SR0484007). The main database was run by the Alibaba cloud server and the backup database was run by the Tencent cloud server. The two databases were synchronized in real-time via the Internet. Thirteen patient-related data tables, 1 user permission data table, 6 database log tables, and 4 scientific research data tables were established. And 118 software code files were created, with a total of more than 20000 lines of code. By February 24, 2021, data of 2 763 breast cancer patients were stored. The database runs stably and meets the design requirements. Conclusion The breast cancer database management system based on cloud platform meets the design requirements, and the system runs stably with high security.

Objective To explore the risk factors of diabetic nephropathy complicated with proteinuria, and provide evidence for relevant departments to develop preventive measures.Methods Totally 92 patients with diabetes mellitus admitted to a hospital from July 2016 to December 2018 were enrolled. Selfmade questionnaires were used to collect patientrelated data and laboratory tests and multivariate logistic regression was used to screen for risk factors for diabetic nephropathy complicated with proteinuria.Results Among 92 patients with diabetes, 28 cases of diabetic nephropathy with proteinuria, the incidence rate was 30.43%. Multivariate logistic stepwise regression analysis showed that the course of disease (OR=2.195), BMI (OR=1.634), LDL-C (OR=1.829), TG (OR=1.259), HbAlc (OR=1.642), UA (OR= 1.579) and hypertension (OR=1.381) were risk factors for diabetic nephropathy with proteinuria (P<0.05).Conclusion Active health education, shortening the course of disease, reduction of body weight, and control of blood pressure and blood lipids can reduce the incidence of diabetic nephropathy with proteinuria.

Objective To establish a comprehensive performance appraisal management system for large-scale medical equipments which value more than 5 million yuan in our hospital.Methods Based on the data of 2016 and 2017 of large-scale medical equipment in our hospital, the performance appraisal system was constructed from three dimensions of economic benefit, social benefit and technical state. The mathematical model of comprehensive performance appraisal of large-scale medical equipment was established by using weighted coefficient method and fuzzy mathematics appraisal method, which turns the index which is difficult to be quantified and evaluated into quantifiable one.Results In this study, 59.1% of the 15 large-scale medical equipment had good comprehensive performance, 18.6% had general comprehensive performance, and 22.3% had poor comprehensive performance.Conclusion Implementing the comprehensive performance appraisal and management system of large-scale medical equipment can help to change the management mode from extensive to refined, reduce the waste of health resources and improve the management level of large-scale medical equipment.
   

Objective To investigate the psychological health status of parents of autistic children, and to explore the influence factors of psychological health of parents of autistic children and the relationship among different factors. Methods Questionnaire survey on the parents of autistic children treated at Zao Zhuang mental health center was conducted, using the psychological test scales (Simplified Mental State Rating Scale Kessler 10 Chinese Version, Simple Coping Style Questionnaire, and Social Support Scale by Xiao Shuiyuan) with high reliability and validity. Results ①The only child, the way of payment, the child′s age, and the illness time had a statistical significance in the mental health of parents of children with autism. ②The time of illness and family month income had a definite effect on social support of parents of autistic children. ③There was a significant negative correlation between social support and mental health of the parents of autistic children, and there was a significant positive correlation between coping inclination and mental health. ④Support had mediating effects on coping style and mental health. Conclusion The mental health of parents of children with autism is affected by many factors, and social support had mediating effect on mental health and coping style.

Objective To investigate the residents′ cognition, attitude, medical seeking behavior and its influencing factors through questionnaire survey, and provide reference for effectively promoting and improving hierarchical medical system.Methods A selfdesigned questionnaire was used to conduct an onsite questionnaire survey on some urban and rural residents in a city. Descriptive statistics methods were used to analyze the residents′ cognition, attitude and medical behavior choice, and multiple correspondence analysis was used to analyze the influencing factors of residents′ willingness to transfer to lower level hospitals.Results (1) The awareness rate of hierarchical medical system was 36.06%, and the awareness rates of twoway referral (9.83%) and medical association (18.03%) were relatively low. And 64.75% of the respondents were very satisfied or relatively satisfied with the hierarchical medical system. The main reasons for dissatisfaction were that the medical conditions of primary medical institutions were poor, the hierarchical medical system in different regions was lack of unified norms, and the medical security system was not perfect. (2) The top three medical institutions preferred by residents were: County/city/district hospital (31.97%), health room/station (22.13%), and private clinic (21.31%). The main reasons affecting their choice of medical institutions were: severity of diseases, advanced level of medical equipment and medical insurance reimbursement. (3)Multiple correspondence analysis showed that residents living in rural areas with family income less than 20　000 yuan would refer to lower level hospitals for treatment depending on the situation. Male, inservice, with junior high school education or below, income of 20 000 to less than 40 000 yuan or 60 000 to less than 80 000 yuan, who did not agree with the implementation of hierarchical medical system in China, were more reluctant to transfer to lower level hospitals for the treatment. Conclusions Residents′ awareness of the hierarchical medical system was low. The main reasons for not choosing grassroot medical treatment or referral were that the Internet mechanism of hierarchical medical system was imperfect, the quality of primary medical services was not trusted, and the lack of grassroot talents. It was suggested that we should strengthen the Internet construction in the process of hierarchical medical system, improve the level of primary medical service and medical environment, and promote the quality of medical resources to the primary medical institutions.

Objective To construct a machine learning model to predict the risk of deep venous thrombosis (DVT) in patients undergoing total hip arthroplasty (THA), and to identify key risk factors influencing DVT using shapley additive explanations (SHAP) method.Methods We retrospectively analyzed data from 416 patients who underwent THA in Wenzhou People′s Hospital from January 1, 2017 to July 31, 2022, and randomly divided them into a training set and a test set in a 4∶〖KG-*2/3〗1 ratio. Recursive feature elimination and five-fold cross-validation were used to select the best features. Six machine learning algorithms were utilized to develop predictive models, and various performance metrics were employed to evaluate them. The SHAP method was used to analyze the interpretability of the optimal model.Results Four hundred and sixteen patients were included in the final study, including 333 in the training set and 83 in the test set. The XGBoost model was the most accurate on the test dataset, achieving a sensitivity of 0.817, specificity of 0.783, F1 score of 0.860, ROC-AUC of 0.800, and a Brier score of 0.106. SHAP summary plots showed that age, cholesterol, postoperative bed time, fibrinogen, and preoperative plasma D-dimer levels were the top five determinants for post-THA DVT. SHAP values feature dependence plots revealed complex non-linear effects of these factors on DVT risk, with age, bed rest, and fibrinogen showing an inverted U-shaped relationship, and cholesterol displaying a positive correlation. Individual SHAP values offered insights into each predictor′s role in DVT risk.Conclusion This study developed an efficient and interpretable machine learning model to predict DVT risk in THA patients, which is helpful for clinical health professionals in identifying high-risk patients and providing personalized intervention.