Two-stage Method Research Articles

Financial Distress Prediction Using an Artificial Neural Network Integrating a Two-Stage Feature Selection Method

Financial distress prediction (FDP) is critical for companies, banks, and investors, and artificial neural networks (ANN) have been proven to be an efficient method for FDP. However, the “curse of dimensionality” in FDP not only increases the computational complexity, but also reduces the prediction accuracy. To solve this problem, this paper takes an ANN model as the basic classifier and presents a new two-stage feature selection method integrated with multiple filters and a wrapper method. The financial data of Chinese listed companies are applied for comparative analysis to verify the effectiveness of the constructed method. The results demonstrate that the proposed method achieves a smaller feature subset and better predictive effect than other methods, thus solving the “curse of dimensionality” more effectively and improving the accuracy. In addition, SHapley Additive exPlanations (SHAP) and Partial Dependence Plots (PDP) are employed to investigate the relative importance of selected features. Their results increase the credibility of the proposed model, giving users more confidence in using this “black box” model.

Examining Lifestyle Risk Factor Clusters and Mental Health in Adults

Abstract Background Lifestyle risk factors often cluster, impacting both physical and mental health. Studies reveal bidirectional relationships between lifestyle factors and health. The aim of this study is to examine the relationship between mental health problems, self-reported depression, and lifestyle risk factors in a sample of the adult population in Turkey. Methods 2022 Turkey Health Survey (THS) data were used. The sample group was made using cross-sectional data of approximately 22742 individuals in the adult population over 15 years of age. Two-stage clustering method was used to determine vital risk factor clusters. Regression analysis was used to determine the relationship between vital risk factor clusters and self-reported depression and mental health problems. Results Lifestyle risk factors were grouped into 4 clusters. It was observed that unhealthy diet, inadequate physical activity, and smoking, which are modifiable lifestyle risk factors, were together in the clusters. It was also observed that self-reported depression and mental problems were more associated with clusters with more than one lifestyle risk factor. In addition, women, married individuals, and individuals over 65 years of age were more associated with mental problems. Conclusions As a result of the findings, there was an association between people with mental problems and lifestyle risk factors such as smoking, unhealthy diet, and insufficient physical activity. These findings may help to plan and implement more effective health policies and individual interventions targeting groups with multiple lifestyle risk factors in the future. Key messages • The target group was determined to improve community mental health and to make necessary interventions. • Since inadequate physical activity, which is one of the vital risk factors, and unhealthy nutrition and smoking tend to cluster together, it was determined that these problems should be prioritised

Primary clinics achieving regular follow-up of patients: Teamwork aspects among healthcare staff

Abstract Backgrounds Studies have shown that more temporally regular primary care visits, or visits more evenly spaced in time, are associated with improved patient outcomes. However, no study has examined what clinic staff can do to encourage greater temporal regularity (TR). We aimed to understand factors related to healthcare staff dynamics that contribute to higher TR for adults with chronic conditions. Methods The study is part of a two-stage mixed methods project. In the first stage, we ranked clinics according to their TR level. In this qualitative stage, we interviewed staff at 6 clinics with extremely high TR (regular follow-up with primary care physician, or PCP), and 6 clinics with extremely low TR. Interviews were conducted with 15 PCPs, 12 nurses, 15 administrative staff, and 4 pharmacists between July 2022- September 2023. Interviews were audiotaped, transcribed, and analyzed using Atlas qualitative analysis software. Findings Themes emerged regarding best ways to promote regular follow-up of patients with chronic conditions. Strategies included having a defined system to reach out to no-show patients, having routine staff meetings to discuss no-shows, adapting workflow to meet the needs of challenging patients, dealing with bureaucracy on patients’ behalf, informal channels of communication with patients, and consulting social workers more often. Relational aspects among staff members at high TR clinics emphasized flat as opposed to hierarchical relationships, as well as high levels of teamwork. Conclusions Teamwork between staff members in primary care settings can contribute to more proactive care delivery, with greater potential to prevent long-term complications. The findings suggest that a high-functioning multidisciplinary care team that focuses on creating the right sorts of interactions and teamwork among staff members, and uses proactive strategies to help engage patients, can contribute to more temporally regular care and thus improved outcomes. Key messages • Stronger teamwork and flat relationships between staff members in primary care settings can result in proactive medicine and developing strategies that improve patients’ regular follow-ups. • Programs focusing on how medical staff utilize their dynamics/relations to improve health outcomes and how doctors’ role align with the staff, can facilitate regular follow-ups of challenging patients.

Variable selection of partially functional linear spatial autoregressive model with a diverging number of parameters

In this article, we consider variable selection of partially functional linear spatial autoregressive model with a diverging number of parameters, in which the explanatory variables include infinite dimensional predictor procedure, treated as functional data, and multiple real valued scalar variate. By combining series approximation method, two-stage least squares method and a class of non convex penalty function, we propose a variable selection method to simultaneously select significant explanatory variables in the parametric component and estimate the corresponding parameter related to spatial lag of the response variable. Under appropriate conditions, we derive the rate of convergence of the series estimator of the functional and parametric component, and show that the proposed variable selection method processes the oracle property. That is, it can estimate the zero components as exact zero with high probability, and estimate the non zero components as efficiently as if the true model was known beforehand. Simulation result show that our proposed variable selection method has better finite sample property.Notably, in the case where the correlation among the explanatory variables in the parametric component is low, the proposed variable selection method performs well. An application of the proposed variable selection method serves as a practical illustration.

Two-stage pseudo maximum likelihood estimation of semiparametric copula-based regression models for semi-competing risks data.

We propose a two-stage estimation procedure for a copula-based model with semi-competing risks data, where the non-terminal event is subject to dependent censoring by the terminal event, and both events are subject to independent censoring. With a copula-based model, the marginal survival functions of individual event times are specified by semiparametric transformation models, and the dependence between the bivariate event times is specified by a parametric copula function. For the estimation procedure, in the first stage, the parameters associated with the marginal of the terminal event are estimated using only the corresponding observed outcomes, and in the second stage, the marginal parameters for the non-terminal event time and the copula parameter are estimated together via maximizing a pseudo-likelihood function based on the joint distribution of the bivariate event times. We derived the asymptotic properties of the proposed estimator and provided an analytic variance estimator for inference. Through simulation studies, we showed that our approach leads to consistent estimates with less computational cost and more robustness than the one-stage procedure developed in Chen YH (Lifetime Data Anal 18:36-57, 2012), where all parameters were estimated simultaneously. In addition, our approach demonstrates more desirable finite-sample performances over another existing two-stage estimation method proposed in Zhu H et al., (Commu Statistics-Theory Methods 51(22):7830-7845, 2021) . An R package PMLE4SCR is developed to implement our proposed method.

A Two-Stage Damage Localization Method for Structural Sealants Based on Boundary Modal Curvature

To facilitate the maintenance of hidden frame glass curtain walls, it is necessary to identify the damage location of the structural sealant accurately. However, detecting concealed sealant damage directly is challenging. This paper proposes a two-stage damage localization method to pinpoint the center of damaged structural sealants. In the first stage, an assurance criterion of boundary modal curvature (ACBMC) is developed to identify damaged edges. In the second stage, the exact damage location is determined by comparing the differences in boundary modal curvature between the undamaged and damaged edges. Two-panel units are analyzed by numerical and model experiments. The results show that the proposed method can effectively identify the damage location of the structural sealant when the damage severity is higher than 4%.

Whole slide cervical cancer classification via graph attention networks and contrastive learning

Cervical cancer is one of the most common cancers among women, which seriously threatens women’s health. Early screening can reduce the incidence rate and mortality. Thinprep cytologic test (TCT) is one of the important means of cytological screening, which has high sensitivity and specificity, and has been widely used in the early screening of cervical cancer. The automatic diagnosis of whole slide images (WSIs) by computers can effectively improve the efficiency and accuracy of doctors’ diagnoses. However, current methods ignore the intrinsic relationships between cervical cells in WSIs and neglect contextual information from the surrounding suspicious areas, and therefore limit their robustness and generalizability. In this paper, we propose a novel two-stage method to implement the automatic diagnosis of WSIs, which constructs Graph Attention Networks (GAT) based on local and global fields respectively to capture their contextual information in a hierarchical manner. In the first stage, we extract representative patches from each WSI through suspicious cell detection, and then employ a Local GAT to classify cervical cells by capturing correlations between suspicious cells in image tiles. This classification process provides the confidence and feature vectors for each suspicious cell. In the second stage, we perform WSI classification using a Global GAT model. We construct graphs corresponding to top-Kg and bottom-Kg cells for each WSI based on results from Local GAT, and introduce a supervised contrastive learning strategy to enhance the discriminative power of the extracted features. Experimental results demonstrate that our proposed method outperforms conventional approaches and effectively showcases the benefits of supervised contrastive learning. Our source code and example data are available at https://github.com/feimanman/Whole-Slide-Cervical-Cancer-Classification.

Enhancing Mobile Central Bank Digital Currency Payment Participation Through Creative Communication on Social Media

With the development of mobile technology, mobile devices have dramatically changed the way people entertain and pay. Social media platforms are booming, and social media is gradually becoming the main form of entertainment for people. Enterprises and organisations with promotional needs are paying more attention to the high efficiency and high input–output ratio of social media communication. Social media advertising has shown more vitality than traditional advertising. Meanwhile, mobile devices have also changed the way people pay, and mobile payment is gradually becoming popular. China has taken the lead in launching mobile Central Bank Digital Currency (CBDC) payment, a new payment technology, and social media communication also plays an important role in promoting CBDC payment. The purpose of this article is to explore the impact of social media creative communication on mobile CBDC payment participation. A research model suitable for this study was created by integrating privacy concern and elaboration likelihood model theory. A valid response from 300 respondents was obtained through an online self-administered questionnaire. Hypotheses were tested through a two-stage research method of partial least squares structural equation modelling with the method of artificial neural network. The results of the study showed that social media advertisements’ creativity and social media information quality significantly and positively influence users’ social media engagement. Users’ social media engagement significantly and positively influences users’ mobile CBDC payment participation. The study provides valuable practical guidance for managers of mobile CBDC payments to promote mobile CBDC payments through social media.

A multitask framework with self-attention mechanism for simultaneous detection of axial pressure, shear deformation, and rupture damage in rubber bearings

Base-isolated structures can effectively control structural response, of which rubber bearings are key components. To ensure the proper functioning of rubber bearings, it is important to accurately detect their state. In previous studies, a smart rubber bearing has been proposed to monitor the axial pressure, shear deformation, and rupture damage of rubber bearings. However, owing to the small dataset and limited ability of the prediction model, two problems exist in previous research: (1) the axial pressure, shear deformation, and rupture damage cannot be detected simultaneously and (2) the detection accuracy is not sufficiently high for practical applications. This study solves these problems in three aspects. First, it addresses the issue of limited data by researching and comparing three data augmentation methods. Second, a multitask framework with a self-attention mechanism is established to simultaneously detect axial compression, shear deformation, and rupture damage. Finally, we introduce a two-stage optimization method based on Bayesian optimization and a grid search for the network structure and optimization. The root-mean-square errors for predicting the axial pressure and shear deformation in the test set are 0.19 MPa and 2.04%, which reduced the error by up to 56.8% and 73.0%, respectively, compared to conventional deep neural network models. In addition, the accuracy of rupture damage detection reached 100%.

Efficacy of new two-stage methods of surgical and laser treatment for primary open-angle glaucoma combined with cataract

Background. Lowering the intraocular pressure (IOP) is the only proven method to slow the progression of glaucomatous optic neuropathy. Successful conservative treatment or surgery do not guarantee the preservation of visual fields throughout life. Despite the current advances in conservative treatment of primary open-angle glaucoma, about 5.3 % of patients still require surgical or combined treatment. There is no consensus yet on approaches to the treatment of patients with glaucoma and cataract. The objective was to evaluate the effectiveness of new two-stage methods for surgical and laser treatment of patients with primary open-angle glaucoma associated with cataract. Materials and methods. Seventy patients (70 eyes) with primary open-angle glaucoma stages I–IV complicated by cataract were included in this study. There were two observation groups. All patients received a two-stage treatment — micropulse transscleral cyclophotocoagulation (MP-TSCPC) and phacoemulsification with a monofocal intraocular lens (IOL) implantation. The first group (n = 34) included participants who underwent phacoemulsification with IOL implantation (stage 1) one month prior to MP-TSCPC (stage 2). The second group (n = 36) consisted of patients who underwent MP-TSCPC (stage 1) and phacoemulsification with IOL implantation a month after (stage 2). These groups did not statistically differ in terms of age, gender, baseline IOP, and stage of primary open-angle glaucoma. The primary outcome measure was the normalization of IOP with or without the use of antiglaucoma medications and without a need for further laser and incisional procedures in 1, 3 and 6 months. The secondary outcome was stabilization or expansion of visual fields when evalua­ting 1, 3, and 6 months after stage 2. Results. The average IOP level with antihypertensive medications before the treatment in group 1 was 27.5 ± 1.3 mmHg, in group 2 — 27.0 ± 1.1 mmHg. One month after the treatment, patients of group 1 had normal IOP in 94.2 % of eyes and in 5.8 % of eyes with local antihypertensive medications; in group 2, IOP was normal in 100 % of eyes without antiglaucoma medications (p < 0.05). In 3 and 6 months, better hypotensive results were in group 2, but outcomes were not statistically significant. The visual fields were preserved at the preoperative level or expan­ded in patients from groups 1 and 2 one month after the treatment in 91.2 and 97.2 % of the eyes, respectively, 3 months after — in 91.2 and 97.2 %, 6 months after — in 88.2 and 94.4 % of the eyes, respectively. In group 2, better functional results were obtained at all time points, but it was not statistically significant. Conclusions. The studies have established a high efficacy and safety of new two-stage methods for surgical and laser treatment of patients with primary open-angle glaucoma and cataract with a follow-up period of 6 months. There was a tendency to greater efficacy of the method, where MP-TSCPC as stage 1 was followed by phacoemulsification with IOL implantation as stage 2. The nature and frequency of surgical and postoperative complications in both groups did not differ.

Assessment Beliefs, Practices and Competence Vis-À-Vis Demographic Characteristics of Senior High School Social Studies Teachers in Ghana

This study examined the impact of demographic factors on the assessment beliefs, practices and competence of Social Studies teachers in selected Senior High / Technical Schools in the Volta and Oti Regions of Ghana. The study was guided by the Human Capital Theory (HCT). It utilized a quantitative research approach and a cross-sectional research design. Employing a two-stage simple random sampling method, 196 Social Studies teachers were randomly selected from 80 Senior High Schools (SHS) and Senior High Technical Schools (SHTS) in the Regions. Data was collected through a questionnaire, covering the participants’ demographic information and assessment beliefs, practices and competence. Statistical data treatment was done through frequency distributions. It was displayed using a contingency table and Cramer's V analysis. Findings indicated no significant correlation between the teachers’ gender, educational background, teaching experience, assessment beliefs, practices and competence. The study concludes that the demographic did not influence the teachers’ assessment beliefs. Based on the conclusions, it is recommended that professional development should focus on individualized training that targets specific needs of teachers in assessment literacy rather than generalizing based on demographic factors. This could help foster more robust and consistent assessment beliefs across the teaching population. In addition, teacher education programs should prioritize skills-based training on contemporary assessment methodologies and address competency gaps regardless of demographic factors.

A Method for Predicting the Impact of Labor Education at Institutions of Higher Education Based on a Two-Stage Clustering Algorithm

In order to develop more accurate teaching plans and methods based on students’ characteristics and needs, and improve the effectiveness of labor education, a two-stage clustering algorithm based method for predicting the effectiveness of labor education in universities is studied. A two-stage clustering algorithm-based method for predicting the effectiveness of labor education in universities has been proposed. Through in-depth analysis of labor education data in universities, principal component analysis (PCA) was used to select 16 most representative variables from the original features. Subsequently, self-organizing map (SOM) neural network was used for preliminary clustering, and the optimal number of clusters was determined to be 8. Furthermore, the K-means clustering algorithm performs fine clustering based on the initial centers provided by SOM, significantly improving the stability and accuracy of clustering. At the same time, processing the outliers in the clustering results has reduced their impact on the clustering effect. Finally, the processed data was input into an improved XGBoost prediction model, which achieved a true positive rate (TPR) of over 85% on the ROC curve when predicting the effectiveness of labor education in universities. While maintaining a low false positive rate (FPR), the model outperformed other comparison methods significantly. This achievement not only validates the effectiveness of the proposed method in feature selection, clustering optimization, and prediction model construction, but also confirms its ability to accurately predict the labor education effectiveness of students in different grades, providing strong support for higher education institutions to optimize teaching plans and improve educational outcomes.

A hierarchical deep reinforcement learning method for solving urban route planning problems under large-scale customers and real-time traffic conditions

As urbanization and economic growth advance, large-scale customers and real-time traffic conditions have become crucial factors in urban route planning. Deep reinforcement learning is considered the most effective method for solving urban route planning problems involving large-scale customers and real-time traffic conditions. Due to memory usage limitations, existing deep reinforcement learning methods cannot identify candidate customers or determine optimal travel routes in large-scale and real-time environments. To tackle these problems, this study introduces a hierarchical deep reinforcement learning method utilizing an improved transformer model (HDRLITF) based on the divide-and-conquer concept. Graph attention networks and gate mechanisms are integrated into the transformer model to capture dynamic features and improve the model’s performance. A two-stage training method, based on the actor-critic algorithm, is proposed to determine the optimal policy function. To evaluate the HDRLITF method, experiments were conducted using datasets from the cities of Shenzhen and Chengdu in China. The experimental results suggest that the HDRLITF method can effectively interact with real-time traffic environments and obtain high-quality solutions compared to other deep reinforcement learning methods. The robustness and reliability of HDRLITF were further validated across multiple traffic scenarios and indicators.

Assessment of the Effect of Community-Based Health Insurance Scheme on Health-Related Outcomes in Ethiopia: A Systematic Review.

We aimed to review the effect of community-based health insurance on health-related outcomes in Ethiopia. A systematic review was undertaken utilizing a major relevant published literature review from September 2017 to June 15, 2023. PubMed, Scopus, Web of Science, Science Direct, Google Scholar, Semantic Scholar, EMBASE, ProQuest, Hinari, and the Cochrane Library were used to search for relevant literature. Moreover, the Prisma flow model was used to select eligible findings. Overall, 72% of the articles employed cross-sectional comparative study designs and procedures, and 36% of them employed samples ranging in size from 501 to 1000 participants. Furthermore, 76% were studied using descriptive statistics and logistic regression, whereas fewer utilized a random model, a probity model, or a correlation model. Similarly, 32% of the research used two-stage stratified sampling methods, and around 40% of the data revealed that the scheme increased healthcare utilization services. About 72 % of the reviewed study results showed that the scheme reduced catastrophic health expenditure and increases utilization of healthcare services. And the 20% reviewed studies stated that the CBHI boosts household satisfaction level. Moreover 12% of the reviewed studies stated that, CBHI increased QoL (quality of life). Most of the studies provide evidence of the positive effect of CBHI in Ethiopia. Mainly, its membership improved the utilization of health services and decreased the incidence of catastrophic health expenditures. Thus, all actors should cooperate to strengthen it to solve the effective attribute of the deprived value of health care and continuity of care delivery system related to the country's new policy.

Specific depressive symptoms, body mass index and diabetes in middle-aged and older Chinese adults: Analysis of data from the China Health and Retirement Longitudinal Study (CHARLS)

BackgroundThis study examined the association between specific depressive symptoms and incident diabetes, and whether overweight or obesity mediates this relationship among middle-aged and older adults in China. MethodsIn a nationally representative prospective cohort study of 11,893 middle-aged and older Chinese adults without baseline diabetes, we used Cox models to assess the association between depressive symptoms and diabetes. The quantile g-computation (qgcomp) model evaluated the contribution of 10 specific depressive symptoms to diabetes risk, and a two-stage regression method explored the mediation effect of overweight or obesity. ResultsOver a median follow-up of 7.1 years, 1,314 cases of diabetes were identified. Elevated depressive symptoms were associated with increased diabetes risk (HR 1.23; 95 % CI 1.09–1.38). Eight out of 10 depressive symptoms were significantly associated to diabetes, with loneliness (weight = 18 %; HR 1.23; 95 % CI 1.10–1.39), restless sleep (weight = 17 %; HR 1.16; 95 % CI 1.04–1.29), and bother (weight = 15 %; HR 1.19; 95 % CI 1.07–1.33) being the primary contributors. Mediation analysis showed that overweight and obesity reduced the depression-diabetes risk association by 8.21 % and 12.61 %, respectively. LimitationsDiagnosis of diabetes was self-reported. ConclusionsEight out of ten specific depressive symptoms were associated to diabetes, overweight and obesity may partially mitigate the effect of depressive symptoms on diabetes among middle-aged and older adults in China. Clinical implicationsOur results highlight the importance of tailoring diabetes prevention and management strategies according to specific depressive symptoms among middle-aged and older adults in China.

Environmental, social and governance (ESG) performance and abnormal positive tone

Purpose The purpose of this study is to explore the relationship between environmental, social and governance (ESG) performance and tone management in the annual report. This is based on the notion that managers, driven by personal interests, may use their ESG accomplishments by using an abnormal positive tone to enhance their reputation or career prospects. Design/methodology/approach Using panel data from Chinese listed companies from 2010 to 2022, this study first investigates the relationship between ESG performance and abnormal tone management. The study then uncovers this relationship is mediated through the mechanisms of equity-based incentive and analyst coverage. The conclusions of this paper hold even after a series of robustness tests, such as propensity score matching, Heckman two-stage method and two-stage least squares with instrumental variables. Findings This study finds a positive correlation between ESG performance and the presence of abnormal positive tone in annual reports. Furthermore, the mechanistic analysis reveals that managers in companies with strong ESG performance are motivated to use an overly positive tone, largely due to their vested interests in equity-based compensation. Moreover, in an effort to alleviate the pressure stemming from heightened financial analyst coverage and enhance the impression conveyed through analysts' reports, managers with superior ESG performance also tend to inflate the tone within their annual reports. Practical implications This study provides significant insights into the ongoing dialogue surrounding ESG-related equity incentives, which incentivize managerial manipulation of stock prices through the use of abnormal positive tone. The findings call upon investors to exercise greater vigilance in examining narrative information in annual reports, as abnormally positive tones may not always faithfully represent performance but rather reflect managerial self-interest. Social implications There is an emphasis on the importance of robust oversight mechanisms within corporate governance bodies to curb the manipulation of tone for managers’ personal gain. Originality/value This study enhances the theoretical foundation of ESG studies, offering a holistic perspective on the intricate interplay among ESG performance, managerial behavior and financial markets, with potential implications for researchers, investors and regulators.

Using Survey Data to Estimate Intergenerational Mobility in Income and Education in Portugal

AbstractPrevious studies about intergenerational mobility for the Portuguese economy find that education and income persistence is very high in comparison with other developed economies. We construct relative, absolute, global and local measures of mobility for Portugal, comparing them with existing evidence for this and other countries. These are the intergenerational income elasticity (computed using the two-sample two-stage least squares method), income correlation, rank-rank slope, bottom to top income level probability, the share of individuals earning more than their fathers and also the intergenerational education correlation, the low to high education level probability, and the share of individuals with a higher education level than their fathers. We consider the 1968–1988 cohorts and the 1995 and 2019 waves of the European Community Household Panel and the European Union Statistics on Income and Living Conditions, respectively. Overall, based on the point estimates, women seem to present more mobility in income. Upward income mobility is verified at the bottom while persistence exists at the top. Women present a greater absolute educational mobility. More than 80% of individuals have a higher education than their fathers and full upward education mobility exists for children of low-educated fathers. Mobility in education is higher for the offspring of medium–high-income fathers. Individuals with a high education level, in the medium–high income level or with occupations requiring a higher education level show higher mobility.

Two-Stage Multiple-Vector Model Predictive Control for Multiple-Phase Electric-Drive-Reconstructed Power Management for Solar-Powered Vehicles

Electric-drive-reconstructed onboard chargers (EDROCs), also known as electric-drive-reconstructed power management systems, are a promising alternative to conventional onboard chargers due to their characteristics of low cost and high power density. The model predictive control offers a fast dynamic response, simple implementation, and the ability to control multiple targets simultaneously. In this paper, a two-stage multi-vector model predictive current control (MPCC) of a six-phase EDROC for solar-powered electric vehicles (EVs) is proposed. Firstly, the topology for the EDROC incorporating a six-phase symmetrical permanent magnet synchronous machine (PMSM) is introduced, and the operation principles of the DC charge mode, the drive mode, and, especially, the in-motion charge mode are analyzed in detail. After that, a two-stage multi-vector MPCC method is proposed by using the multi-vector MPC technique and designing a two-stage MPC structure to eliminate the regulation of the weighting factor of the MPC. Finally, the effectiveness of the proposed method is verified on a self-designed 2 kW EDROC platform.

Association Between Quality of Life and Drug Adherence Among Patients With Diabetes in India.

Diabetes represents a significant global health concern. Effective diabetes management necessitates substantial behavioral and psychological modifications, including regular glucose monitoring, dietary adjustments, and medication adherence. However, treatment adherence remains suboptimal, particularly in developing countries, leading to poor glycaemic control, increased healthcare costs, and diminished quality of life (QoL). Common factors influencing medication nonadherence include lack of medication knowledge, age, education of patients, polypharmacy, duration of disease, cost of therapy, complexity of dosing regime, and the presence of comorbidities. So, the study aims to assess the adherence status to antidiabetic medication and its association with QoL among diabetes patients in the Anand district of Gujarat state. A study was carried out in Anand district, Gujarat, from 2023 to 2024, focusing on community-based cross-sectional research. In the eight blocks of Anand district, a two-stage cluster sampling method was utilized, with one village or town randomly chosen from each taluka, and 25 diabetes patients interviewed per area. Data collection involved visiting 200 participants' houses, starting with a randomly selected house and continuing until the target was met. The interviews, conducted in the local language, lasted around 45 minutes and utilized a pre-validated and pretested questionnaire. The questionnaire covered demographic details, disease characteristics, and medication adherence using the semi-structured scale from Adherence to Refills and Medications Scale (ARMS-D), as well as the assessment of QoL using the WHOQoL-BREF questionnaire. Descriptive and inferential statistics were used for data analysis. Sociodemographic attributes were presented in terms of percentages (%) and frequencies (N). The means and standard deviation (SD) of health-related QoL and adherence to anti-diabetic medication were reported. Multiple linear regression analysis was used to assess the impact of adherence to antidiabetic medication within each QoL area after adjusting for a few sociodemographic variables. P values below 0.05 were considered statistically significant. The study included predominantly older participants (58.1 years) with 53.5% being male. Most participants (65.5%) had diabetes for more than 5 years, 60.0% had comorbidities, and 29.0% experienced complications. Nonadherence to treatment was observed in 37.0% of patients. The study found that factors like age, education, occupation, family history of diabetes, smoking, alcohol consumption, exercise habits, and medication adherence significantly influenced the QoL in diabetic patients, with medication adherence showing the strongest positive association with all QoL domains. This study shows a high percentage of nonadherence to antidiabetic medications. Poor adherence had significantly lower mean scores across all domains of QoL. Community-based educational programs targeting older and less-educated individuals should promote regular physical activity. Developing strategies to improve medication adherence and ensuring affordable access to essential medications, along with prioritizing enhancements in the QoL through psychosocial support and lifestyle counseling are imperative for diabetic patients.

Feature Contributions and Predictive Accuracy in Modeling Adolescent Daytime Sleepiness Using Machine Learning: The MeLiSA Study.

Background: Excessive daytime sleepiness (EDS) among adolescents poses significant risks to academic performance, mental health, and overall well-being. This study examines the prevalence and risk factors of EDS in adolescents in Bangladesh and utilizes machine learning approaches to predict the risk of EDS. Methods: A cross-sectional study was conducted among 1496 adolescents using a structured questionnaire. Data were collected through a two-stage stratified cluster sampling method. Chi-square tests and logistic regression analyses were performed using SPSS. Machine learning models, including Categorical Boosting (CatBoost), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), and Gradient Boosting Machine (GBM), were employed to identify and predict EDS risk factors using Python and Google Colab. Results: The prevalence of EDS in the cohort was 11.6%. SHAP values from the CatBoost model identified self-rated health status, gender, and depression as the most significant predictors of EDS. Among the models, GBM achieved the highest accuracy (90.15%) and precision (88.81%), while CatBoost had comparable accuracy (89.48%) and the lowest log loss (0.25). ROC-AUC analysis showed that CatBoost and GBM performed robustly in distinguishing between EDS and non-EDS cases, with AUC scores of 0.86. Both models demonstrated the superior predictive performance for EDS compared to others. Conclusions: The study emphasizes the role of health and demographic factors in predicting EDS among adolescents in Bangladesh. Machine learning techniques offer valuable insights into the relative contribution of these factors, and can guide targeted interventions. Future research should include longitudinal and interventional studies in diverse settings to improve generalizability and develop effective strategies for managing EDS among adolescents.