A study on predicting the length of hospital stay for Chinese patients with ischemic stroke based on the XGBoost algorithm.

BACKGROUND: The incidence of stroke is a challenge in China, as stroke imposes a heavy burden on families, national health services, social services, and the economy. The length of hospital stay (LOS) is an essential indicator of utilization of medical services and is usually used to assess the efficiency of hospital management and patient quality of care. This study established a prediction model based on a machine learning algorithm to predict ischemic stroke patients' LOS. METHODS: A total of 18,195 ischemic stroke patients' electronic medical records and 28 attributes were extracted from electronic medical records in a large comprehensive hospital in China. The prediction of LOS was regarded as a multi classification problem, and LOS was divided into three categories: 1-7 days, 8-14 days and more than 14 days. After preprocessing the data and feature selection, the XGBoost algorithm was used to build a machine learning model. Ten fold cross-validation was used for model validation. The accuracy (ACC), recall rate (RE) and F1 measure were used to evaluate the performance of the prediction model of LOS of ischemic stroke patients. Finally, the XGBoost algorithm was used to identify and remove irrelevant features by ranking all attributes based on feature importance. RESULTS: Compared with the naive Bayesian algorithm, logistic region algorithm, decision tree classifier algorithm and ADaBoost classifier algorithm, the XGBoot algorithm has higher ACC, RE and F1 measure. The average ACC, RE and F1 measure were 0.89, 0.89 and 0.89 under the 10-fold cross-validation. According to the analysis of the importance of features, the LOS of ischemic stroke patients was affected by demographic characteristics, past medical history, admission examination features, and operation characteristics. Finally, the features in terms of hemiplegia aphasia, MRS, NIHSS, TIA, Operation or not, coma index etc. were found to be the top features in importance in predicting the LOS of ischemic stroke patients. CONCLUSIONS: The XGBoost algorithm was an appropriate machine learning method for predicting the LOS of patients with ischemic stroke. Based on the prediction model, an intelligent medical management prediction system could be developed to predict the LOS based on ischemic stroke patients' electronic medical records.

Culturally adapted and lay-delivered cognitive behaviour therapy for older adults with depressive symptoms in rural China: a pilot trial.

BACKGROUND: Late-life depression issues in developing countries are challenging because of understaffing in mental health. Cognitive behavioural therapy (CBT) is effective for treating depression. AIM: This pilot trial examined the adherence and effectiveness of an eight-session adapted CBT delivered by trained lay health workers for older adults with depressive symptoms living in rural areas of China, compared with the usual care. METHOD: Fifty with screen-positive depression were randomly assigned to the CBT arm or the care as usual (CAU) arm. The primary outcomes were the session completion of older adults and changes in depressive symptoms, assessed using the Geriatric Depression Scale (GDS). RESULTS: The majority (19/24) of participants in the CBT arm completed all sessions. Mixed-effect linear regression showed that the CBT reduced more GDS scores over time compared with CAU. CONCLUSION: Lay-delivered culturally adapted CBT is potentially effective for screen-positive late-life depression.

Long-Term Health Consequences Among Wenchuan Earthquake Adult Survivors: Implications of a Framework for Postearthquake Reconstruction.

The objective of this study was to examine the longitudinal health consequences of the Wenchuan earthquake. Based on descriptive analyses of national-level data and multivariate analyses on a six-wave repeated cross-sectional survey, the findings suggested that after 8 years health risks remained high among earthquake-affected survivors; however, a process of recovery existed. To conceptualize these findings, in this study, we proposed a three-stage recovery model in which the postdisaster health status was divided into three stages: acute, stagnant, and adaptive. At each stage, the health risk varied, and over time, associations between health outcomes and protective factors varied. The three-stage recovery model identified the trends of long-term health consequences among adult earthquake survivors and provided guidance for postdisaster reconstruction in China on the basis of protective factors analyses.

The effect of smoke-free legislation on the mortality rate of acute myocardial infarction: a meta-analysis.

BACKGROUND: Several studies have demonstrated that smoke-free legislation is associated with a reduced risk of mortality from acute myocardial infarction (AMI). This study aimed to examine and quantify the potential effect of smoke-free legislation on AMI mortality rate in different countries. METHODS: Studies were identified using a systematic search of the scientific literature from electronic databases, including PubMed, Web of Science, ScienceDirect, Embase, Google Scholar, and China National Knowledge Infrastructure (CNKI), from their inception through September 30, 2017. A random effects model was employed to estimate the overall effects of smoke-free legislation on the AMI mortality rate. Subgroup analysis was performed to explore the possible causes of heterogeneity in risk estimates based on sex and age. The results of meta-analysis after excluding the studies with a high risk of bias were reported in this study. RESULTS: A total of 10 eligible studies with 16 estimates of effect size were included in this meta-analysis. Significant heterogeneity in the risk estimates was identified (overall I2 = 94.6%, p < 0.001). Therefore, a random effects model was utilized to estimate the overall effect of smoke-free legislation. There was an 8% decline in AMI mortality after introducing smoke-free legislation (RR = 0.92, 95% confidence interval (CI): 0.90-0.94). The results of subgroup analyses showed that smoke-free legislation was significantly associated with lower rates of mortality for the following 5 diagnostic subgroups: smoke-free in workplaces, restaurants and bars (RR = 0.92, 95% CI: 0.90-0.95), smaller sample size (RR = 0.92, 95% CI: 0.89-0.95), study location in Europe (RR = 0.90, 95% CI: 0.85-0.94), regional study area (RR = 0.92, 95% CI: 0.89-0.94), and no previous local smoke-free legislation (RR = 0.91, 95% CI: 0.90-0.93). However, there was not much difference in AMI mortality rates after the legislation between the longer (RR = 0.92, 95% CI: 0.86-0.98) and shorter follow-up duration subgroups (RR = 0.92, 95% CI: 0.89-0.94). CONCLUSION: Smoke-free legislation could significantly reduce the AMI mortality rate by 8%. The reduction in the AMI mortality rate was more significant in studies with more comprehensive laws, without prior smoke-free bans, with a smaller sample size, at the regional level, and with a location in Europe.

The impact of contract service policy and doctor communication skills on rural patient-doctor trust relationship in the village clinics of three counties.

BACKGROUND: Trust is regarded as the cornerstone of the doctor-patient relationship in the world of medicine; it determines the decisions patients make when choosing doctors and influences patients' compliance with recommended treatments. In China, patient-doctor trust acts as a thermometer measuring harmony in the doctor-patient relationship. The objective of this study is to explore the relationship between the contract service and patient-doctor trust-building in 25 village clinics of rural China. METHOD: The research was carried out in village clinics in rural China. A simple random sampling method was used to choose clinics and subjects. Based on feasibility and financial support, we chose three counties as our study settings: Dafeng District, Jiangsu Province; Yinan County, Shandong Province; and Wufeng Tujia Autonomous County, Hubei Province. Twenty-five village clinics and 574 subjects were selected in the three areas from the contract service and patient list. Descriptive statistics, t-tests, MANOVA, SEM, and multiple regression statistical analysis were employed to analyze the data. RESULT: Statistical analysis showed that contract service directly and indirectly influenced patient-doctor trust-building in village clinics. The patient perception of doctor communication skills was a mediator in the relationship between contract service policy and patient-doctor trust-building. CONCLUSIONS: Building patient-doctor trust is important in developing and enhancing rural health. The policy of contract service plays a significant role in building relationships. Well-developed communication skills of doctors contribute to the implementation of the contract service policy and to establishing patient-doctor trust.

Performance of Between-study Heterogeneity Measures in the Cochrane Library.

The growth in comparative effectiveness research and evidence-based medicine has increased attention to systematic reviews and meta-analyses. Meta-analysis synthesizes and contrasts evidence from multiple independent studies to improve statistical efficiency and reduce bias. Assessing heterogeneity is critical for performing a meta-analysis and interpreting results. As a widely used heterogeneity measure, the I statistic quantifies the proportion of total variation across studies that is caused by real differences in effect size. The presence of outlying studies can seriously exaggerate the I statistic. Two alternative heterogeneity measures, the (Equation is included in full-text article.)and (Equation is included in full-text article.)have been recently proposed to reduce the impact of outlying studies. To evaluate these measures' performance empirically, we applied them to 20,599 meta-analyses in the Cochrane Library. We found that the (Equation is included in full-text article.)and (Equation is included in full-text article.)have strong agreement with the I, while they are more robust than the I when outlying studies appear.

Empirical Comparison of Publication Bias Tests in Meta-Analysis.

BACKGROUND: Decision makers rely on meta-analytic estimates to trade off benefits and harms. Publication bias impairs the validity and generalizability of such estimates. The performance of various statistical tests for publication bias has been largely compared using simulation studies and has not been systematically evaluated in empirical data. METHODS: This study compares seven commonly used publication bias tests (i.e., Begg's rank test, trim-and-fill, Egger's, Tang's, Macaskill's, Deeks', and Peters' regression tests) based on 28,655 meta-analyses available in the Cochrane Library. RESULTS: Egger's regression test detected publication bias more frequently than other tests (15.7% in meta-analyses of binary outcomes and 13.5% in meta-analyses of non-binary outcomes). The proportion of statistically significant publication bias tests was greater for larger meta-analyses, especially for Begg's rank test and the trim-and-fill method. The agreement among Tang's, Macaskill's, Deeks', and Peters' regression tests for binary outcomes was moderately strong (most κ's were around 0.6). Tang's and Deeks' tests had fairly similar performance (κ > 0.9). The agreement among Begg's rank test, the trim-and-fill method, and Egger's regression test was weak or moderate (κ < 0.5). CONCLUSIONS: Given the relatively low agreement between many publication bias tests, meta-analysts should not rely on a single test and may apply multiple tests with various assumptions. Non-statistical approaches to evaluating publication bias (e.g., searching clinical trials registries, records of drug approving agencies, and scientific conference proceedings) remain essential.

Risk factors of the stigma towards psychiatric patients among primary healthcare workers in China: a county study.

BACKGROUND: Attitude towards psychiatric patients among healthcare workers has an impact on quality of medical care and rehabilitation of patients. In China, primary healthcare workers play an important role in mental health care, but little is known about the attitude of them towards psychiatric patients. This study aims to examine the risk factors associated with stigma among primary healthcare workers in West China. METHODS: This cross-sectional study randomly recruited 395 primary healthcare workers in Mianzhu County, China. Data were collected via self-reported questionnaires. Descriptive analyses, bivariate analyses, and hierarchical linear regressions were performed by SPSS 17.0 to test the factors that accounted for the variation of stigma towards psychiatric patients. RESULTS: Several risk factors were confirmed, including the satisfaction of income, work experience in psychiatric/ psychological departments, rehabilitation of patients, contact quality, and the attitude of mass media. However, demographic factors, the rest of work-related factors, and contact frequency might not be related with primary healthcare workers' attitude towards psychiatric patients. CONCLUSIONS: The findings suggested that the quality of contact between primary healthcare workers and psychiatric patients be enhanced in order to decrease the stigma of healthcare staff. On-the-job training and institutional medical education needs a further exploration and development.

How China's new health reform influences village doctors' income structure: evidence from a qualitative study in six counties in China.

BACKGROUND: In 2009, health-care reform was launched to achieve universal health coverage in China. A good understanding of how China's health reforms are influencing village doctors' income structure will assist authorities to adjust related polices and ensure that village doctors employment conditions enable them to remain motivated and productive. This study aimed to investigate the village doctors' income structure and analyse how these health policies influenced it. METHODS: Based on a review of the previous literature and qualitative study, village doctors' income structure was depicted and analysed. A qualitative study was conducted in six counties of six provinces in China from August 2013 to January 2014. Forty-nine village doctors participated in in-depth interviews designed to document their income structure and its influencing factors. The themes and subthemes of key factors influencing village doctors' income structure were analysed and determined by a thematic analysis approach and group discussion. RESULTS: Several policies launched during China's 2009 health-care reform had major impact on village doctors. The National Essential Medicines System cancelled drug mark-ups, removing their primary source of income. The government implemented a series of measures to compensate, including paying them to implement public health activities and provide services covered by social health insurance, but these have also changed the village doctors' role. Moreover, integrated management of village doctors' activities by township-level staff has reduced their independence, and different counties' economic status and health reform processes have also led to inconsistencies in village doctors' payment. These changes have dramatically reduced village doctors' income and employment satisfaction. CONCLUSIONS: The health-care reform policies have had lasting impacts on village doctors' income structure since the policies' implementation in 2009. The village doctors have to rely on the salaries and subsidies from the government after the drug mark-up was cancelled. China's national health reforms are attempting to draw village doctors into the national health workforce, but the policies have impacted their income and independence. Further research into these concerns and monitoring of measures to adequately compensate village doctors should be undertaken. Reasonable compensation strategies should be established, and sufficient subsidies should be allocated in a timely manner.

The symptoms of posttraumatic stress disorder and depression among adult earthquake survivors in China.

The objective of the study was to examine the relationships between mental health conditions (posttraumatic stress disorder [PTSD] only, depression only, and PTSD and depression) and related factors. A cross-sectional survey was conducted among 1362 adults from two severely affected townships at 6 months after the earthquake. The results of the analyses showed that the prevalence of depression and PTSD were 31.4% and 22.1%, respectively, 6 months after the earthquake. When PTSD and depression were treated as two separate dependent variables, PTSD and depression share almost similar sets of predictive factors. After its four categories (none, PTSD only, depression only, and PTSD and depression) were used as categorical dependent variables, there are different predictive factors. The findings suggest that there are two different groups of individuals, those who develop depression only in response to earthquake exposure and those who develop both depression and PTSD.