CHAID Research Articles

Navigating Digital Inequality: Examining Factors Affecting Rural Customers’ Internet Banking Adoption in Post-COVID Bangladesh

As the world continues to navigate the new normal brought about by the COVID-19 pandemic, one issue that has come to the forefront is digital inequality. In Bangladesh, where a significant portion of the population resides in rural areas, the adoption of internet banking has been hindered by various factors. However, understanding these factors is crucial, especially now that digital transactions have become more important. This study aims to understand the factors influencing the adoption of internet banking services among rural customers in Bangladesh. To acquire data, a questionnaire was administered to 443 rural bank customers in the district of Barisal. The Exploratory Factor Analysis (EFA) revealed three primary factors: trust compatibility, service benefit, and access to consumer education. In addition, the research sought to determine if the identified factors, particularly access to consumer education, varied according to the occupation and income level of rural consumers. Using exhaustive Chi-squared Automatic Interaction Detection (CHAID) analysis, the findings revealed that access to consumer education differs significantly by occupation level, with business and service holders being more likely than farmers to have access to consumer education. This research contributes to the literature by providing insights into the adoption of internet banking by rural customers and informing policymakers about the special needs of this demographic.

Preschool teachers display a flexible pattern of pedagogical actions in promoting healthy habits in children.

The school represents the optimal setting for promoting the physical, emotional, and social health of children, especially during the first years of life. Understanding the pedagogical actions of teachers to address health education is an important first step in promoting healthy behaviors in children. We inhere analyzed the pedagogical action patterns in the preschool teaching of healthy habits from a holistic health perspective. We used photography as a strategy for data collection and applied a Chi-square automatic interaction detection (CHAID) classification tree, a data mining procedure, to generate a pattern model. We found that the school space and the learning playfulness strategies for the development of executive functions, classified according to the exercise, symbolic, assembly, rules (ESAR) model, were the main factors that influence the pedagogical actions fostering healthy habits. By contrast, the school and the pedagogical resources of the classroom are factors with a much smaller impact on working with healthy habits. This pedagogical action pattern is flexible, since teachers conduct a multiplicity of pedagogical actions through different strategies, in different school spaces, at any time. In conclusion, our results unmask the interdependent relationships between the different factors that determine the teacher's actions at the preschool. It also contributes to the understanding of the teacher's practices in fostering healthy habits in a healthy learning environment.

Investigating machine learning models in predicting lake water quality parameters as a 3-year moving average

Lake water quality plays a vital role in the lake ecosystem, including biotic (for living creatures, such as plants, animals, and micro-organisms) and abiotic interactions. In this research, various types of machine learning (ML) methodologies, such as classification and regression tree (CART), chi-squared automatic interaction detector (CHAID), C5 tree, quick, unbiased, and efficient statistical tree (QUEST), along with multilayer perceptron (MLP) neural network, and radial basis function (RBF) neural network, are employed to predict the concentration of water quality parameters (P, EC, TDS, pH, DO, NH3, SO4, and θ). Lake Erie is situated at the international border of the USA and Canada. The C5 tree and QUEST tree are used to classify data and predict the number of groups, while the other methods are used to predict the concentration of water quality parameters in the form of a 3-year moving average. The greater matching between the observed and predicted data of dissolved oxygen (NSE = 0.978, bias = 0.126) shows that the CART decision tree has higher accuracy in correctly detecting the concentration of this parameter. The C5 tree could identify 33 groups correctly out of 36 total groups, which shows better accuracy for the C5 tree in classifying the data for this parameter.

Reasons for medication administration errors, barriers to reporting them and the number of reported medication administration errors from the perspective of nurses: A cross-sectional survey

The aim of the study was to identify the reasons for medication administration errors, describe the barriers in their reporting and estimate the number of reported medication administration errors. BackgroundProviding quality and safe healthcare is a key priority for all health systems. Medication administration error belongs to the more common mistakes committed in nursing practice. Prevention of medication administration errors must therefore be an integral part of nursing education. DesignA descriptive and cross-sectional design was used for this study. MethodsSociological representative research was carried out using the standardized Medication Administration Error Survey. The research study involved 1205 nurses working in hospitals in the Czech Republic. Field surveys were carried out in September and October 2021. Descriptive statistics, Pearson’s and Chi-square automatic interaction detection were used to analyze the data. The STROBE guideline was used. ResultsAmong the most frequent causes of medication administration errors belong name (4.1 ± 1.4) and packaging similarity between different drugs (3.7 ± 1.4), the substitution of brand drugs by cheaper generics (3.6 ± 1.5), frequent interruptions during the preparation and administration of drugs (3.6 ± 1.5) and illegible medical records (3.5 ± 1.5). Not all medication administration errors are reported by nurses. The reasons for non-reporting of such errors include fear of being blamed for a decline in patient health (3.5 ± 1.5), fear of negative feelings from patients or family towards the nurse or legal liability (3.5 ± 1.6) and repressive responses by hospital management (3.3 ± 1.5). Most nurses (two-thirds) stated that less than 20 % of medication administration errors were reported. Older nurses reported statistically significantly fewer medication administration errors concerning non-intravenous drugs than younger nurses (p < 0.001). At the same time, nurses with more clinical experience (≥ 21 years) give significantly lower estimates of medication administration errors than nurses with less clinical practice (p < 0.001). ConclusionPatient safety training should take place at all levels of nursing education. The standardized Medication Administration Error survey is useful for clinical practice managers. It allows for the identification of medication administration error causes and offers preventive and corrective measures that can be implemented. Measures to reduce medication administration errors include developing a non-punitive adverse event reporting system, introducing electronic prescriptions of medicines, involving clinical pharmacists in the pharmacotherapy process and providing nurses with regular comprehensive training.

Application of decision tree algorithms to predict central lymph node metastasis in well-differentiated papillary thyroid carcinoma based on multimodal ultrasound parameters: a retrospective study.

Prophylactic central neck dissection (pCND) in patients with well-differentiated primary papillary thyroid carcinoma (PTC) has become controversial. Several attempts have been made to predict central compartment lymph node metastasis (CLNM) based on clinical and conventional ultrasonic parameters. This study aimed to develop a decision tree (DT) model for predicting the risk of CLNM in patients with PTC based on clinical and preoperative multimodal ultrasound (US) characteristics. A total of 148 PTC nodules confirmed by surgical pathology at Beijing Tiantan Hospital were retrospectively analyzed. All nodules underwent multimodal US examinations preoperatively from January 2020 to September 2021. Correlation analysis of CLNM with clinical characteristics as well as multimodal US parameters of PTC lesions based on gray-scale US, color Doppler flow imaging (CDFI), superb microvascular imaging (SMI), contrast-enhanced ultrasound (CEUS), and shear wave elastography (SWE) technology was carried out. Finally, the chi-squared automatic interaction detector (CHAID) with a 10-fold cross-validation was used to establish DTs for CLNM prediction. The area under the curve was calculated to compare the predictive performance. Univariate analysis indicated that CLNM was positively correlated with thyroglobulin level, maximum size, taller-than-wide, the number of microcalcifications greater than or equal to 5, contact capsule, abnormal cervical lymph node on conventional US, noncentripetal perfusion, delayed clearance, the average shear wave velocity (SWV mean), and the SWV ratio (P<0.05). The multimodal US DT based on taller-than-wide, contact capsule, abnormal cervical lymph node on conventional US, and centripetal enhancement as independent variables showed good discrimination: the sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were 80.0%, 76.7%, 78.4%, and 0.837 [95% confidence interval (CI): 0.771-0.902]. There was a significant difference between the multimodal and conventional US DTs (P=0.009). Our results indicated that the DT based on the preoperative multimodal US characteristics of PTCs has a reasonable predictive ability for CLNM and can be conveniently used for clinical decision-making of individualized treatment in patients with well-differentiated PTC.

Decision tree analysis as predictor tool for in-hospital mortality in critical SARS-CoV-2 infected patients

Abstract Identification of predictive biomarkers for the evolution of critically ill COVID-19 patients would represent a milestone in the management of patients and in human and financial resources prioritization and allocation. This retrospective analysis performed for 396 critically ill COVID-19 patients admitted to the intensive care unit aims to find the best predictors for fatal outcomes in this category of patients. The inflammatory and metabolic parameters were analyzed and Machine Learning methods were performed with the following results: (1) decision tree with Chi-Square Automatic Interaction Detector (CHAID) algorithm, based on the cut-off values using ROC Curve analysis, indicated NLR, IL-6, comorbidities, and AST as the main in-hospital mortality predictors; (2) decision tree with Classification and Regression Tree (CRT) algorithm confirmed NLR alongside CRP, ferritin, IL-6, and SII (Systemic Inflammatory Index) as mortality predictors; (3) neural networks with Multilayer Perceptron (MLP) found NLR, age, and CRP to be the best mortality predictors. Structural Equation Modeling (SEM) analysis was complementarily applied to statistically validate the resulting predictors and to emphasize the inferred causal relationship among factors. Our findings highlight that for a deeper understanding of the results, the combination of Machine Learning and statistical methods ensures identifying the most accurate predictors of in-hospital mortality to determine classification rules for future events.

Predicting early versus late recovery from sport-related concussion using decision tree analysis.

Accurately predicting early (≤ 14 days) versus typical (15-27 days) or delayed (≥ 28 days) recovery from sport-related concussion (SRC) may allow for improved resource utilization and precision in planning and carrying out rehabilitation. In this study, the authors sought to develop an algorithm that enables accurate differentiation of recovery periods and duration after SRC. The authors hypothesized that data regarding initial symptom burden as quantified by a Post-Concussion Symptom Scale (PCSS) score, time to presentation, and number of prior concussions would be the most useful for analyzing predictive factors for concussion recovery duration. A retrospective case-control study was conducted to assess the primary outcome of days to clinical recovery following SRC in pediatric patients. Data from patients 12-18 years old presenting within 28 days of injury to an SRC clinic between November 11, 2017, and October 10, 2020, were analyzed. Patients with positive evidence of injury on head imaging or incomplete records were excluded. The primary outcome was duration of clinical recovery, grouped as early (≤ 14 days), typical (15-27 days), or delayed (≥ 28 days). Recovery was defined as follows: 1) symptom resolution or return to baseline, or 2) initiation of graduated return to play. CHAID (chi-square automatic interaction detection) analysis was used to optimize a decision tree based on 16 input factors, including age, sex, initial PCSS score, time to clinic presentation, number of prior concussions, and presence of defined symptom clusters. The cohort was randomized into training (70%) and test (30%) samples for algorithm validation. A total of 493 patients met the inclusion criteria (mean age 15.7 ± 1.5 years, 68.2% male, 70.0% White). The median time to presentation was 5 days (IQR 2-10 days). Most patients (52.3%) recovered within 14 days of injury, 21.5% recovered within 15-27 days, and 26.2% had a recovery period of 28 days or longer. The variables most predictive of recovery were initial PCSS score (cutoffs ≤ 6, 7-28, or ≥ 29), time to presentation (≤ 7 vs > 7 days), or prior concussions (0 vs ≥ 1). The model accurately discriminated between early versus typical or delayed recovery duration groupings (area under the curve 0.80, Youden index 0.44), and correctly classified > 90% of patients who recovered early. This novel three-factor predictive tool enabled accurate discrimination of early versus typical or delayed SRC recovery to better allocate resources, counsel patients, and make timely referrals.

Machine Learning-Based Prediction of Digoxin Toxicity in Heart Failure: A Multicenter Retrospective Study.

Digoxin toxicity (plasma digoxin concentration ≥0.9 ng/mL) is associated with worsening heart failure (HF). Decision tree (DT) analysis, a machine learning method, has a flowchart-like model where users can easily predict the risk of adverse drug reactions. The present study aimed to construct a flowchart using DT analysis that can be used by medical staff to predict digoxin toxicity. We conducted a multicenter retrospective study involving 333 adult patients with HF who received oral digoxin treatment. In this study, we employed a chi-squared automatic interaction detection algorithm to construct DT models. The dependent variable was set as the plasma digoxin concentration (≥ 0.9 ng/mL) in the trough during the steady state, and factors with p < 0.2 in the univariate analysis were set as the explanatory variables. Multivariate logistic regression analysis was conducted to validate the DT model. The accuracy and misclassification rates of the model were evaluated. In the DT analysis, patients with creatinine clearance <32 mL/min, daily digoxin dose ≥1.6 µg/kg, and left ventricular ejection fraction ≥50% showed a high incidence of digoxin toxicity (91.8%; 45/49). Multivariate logistic regression analysis revealed that creatinine clearance <32 mL/min and daily digoxin dose ≥1.6 µg/kg were independent risk factors. The accuracy and misclassification rates of the DT model were 88.2 and 46.2 ± 2.7%, respectively. Although the flowchart created in this study needs further validation, it is straightforward and potentially useful for medical staff in determining the initial dose of digoxin in patients with HF.

Study on Classification Method of Soil Liquefaction Potential Based on Decision Tree

A new method for evaluating the liquefaction probability of saturated soils is presented in this paper. The existing methods are mainly expressed in equation form, and the evaluation process is always complex. Therefore, to address these shortcomings, a new method based on a tabular form is constructed using a hybrid approach combining the logistic equation and decision trees (classification and regression tree and chi-squared automatic interaction detector) based on the shear wave velocity test databases. The effectiveness of the new method was verified by comparing the predicted liquefaction results with actual liquefaction results. The new method improved the accuracy of the liquefaction evaluation method based on shear wave velocity; the classification was correct for more than 75% of the samples. In addition, the importance of the liquefaction factors was considered during the method construction, which ultimately simplified the evaluation process while ensuring accuracy.

Predicting the Behavioral Health Needs of Asian Americans in Public Mental Health Treatment: A Classification Tree Approach.

Given the fact that experiencing pandemic-related hardship and racial discrimination worsen Asian Americans' mental health, this study aimed to identify unique characteristics of behavioral health needs among Asian Americans (N = 544) compared to White Americans (N = 78,704) and Black Americans (N = 11,252) who received publicly funded behavioral health services in Indiana before and during the COVID-19 pandemic. We used 2019-2020 Adults Needs and Strengths Assessment (ANSA) data for adults eligible for Medicaid or funding from the state behavioral health agency. Chi-squared automatic interaction detection (CHAID) was used to detect race-specific differences among demographic variables, the pandemic status, and ANSA items. Results indicated that, regardless of age, gender, or pandemic status, Asian Americans who received behavioral health services, struggled more with cultural-related factors compared to White and Black individuals. Within this context, intersections among behavioral/emotional needs (psychosis), life functioning needs (involvement in recovery, residential stability, decision making, medical/physical health), and strengths (job history, interpersonal, and spiritual) further differentiated the mental health functioning of Asian from White and Black Americans. Classification tree algorithms offer a promising approach to detecting complex behavioral health challenges and strengths of populations based on race, ethnicity, or other characteristics.

Risk evaluation of carbapenem-induced liver injury based on machine learning analysis

IntroductionInformation regarding carbapenem-induced liver injury is limited, and the rate of liver injury caused by meropenem (MEPM) and doripenem (DRPM) remains unknown. Decision tree (DT) analysis, a machine learning method, has a flowchart-like model where users can easily predict the risk of liver injury. Thus, we aimed to compare the rate of liver injury between MEPM and DRPM and construct a flowchart that can be used to predict carbapenem-induced liver injury. MethodsWe investigated patients treated with MEPM (n = 310) or DRPM (n = 320) and confirmed liver injury as the primary outcome. We used a chi-square automatic interaction detection algorithm to construct DT models. The dependent variable was set as liver injury from a carbapenem (MEPM or DRPM), and factors including alanine aminotransferase (ALT), albumin-bilirubin (ALBI) score, and concomitant use of acetaminophen were used as explanatory variables. ResultsThe rates of liver injury were 22.9% (71/310) and 17.5% (56/320) in the MEPM and DRPM groups, respectively; no significant differences in the rate were observed (95% confidence interval: 0.710–1.017). Although the DT model of MEPM could not be constructed, DT analysis showed that the incidence of introducing DRPM in patients with ALT >22 IU/L and ALBI scores > −1.87 might be high-risk. ConclusionsThe risk of developing liver injury did not differ significantly between the MEPM and DRPM groups. Since ALT and ALBI score are evaluated in clinical settings, this DT model is convenient and potentially useful for medical staff in assessing liver injury before DRPM administration.

Integrating the valence theory and the norm activation theory to understand consumers’ e-waste recycling intention

Electrical and electronic waste (e-waste) is a growing challenge, matching the widespread boom in the use of information and communication technology. Opposite to an alarming increasing amount of e-waste, a low rate of consumer engagement in ensuring the proper disposal of such materials intensifies the pressure on the existing e-waste crisis. To deal with this thorny problem, it is of great interest to grasp consumers’ disposal and recycling behavioral intentions. Therefore, this study attempts to understand complementary perspectives around consumers’ e-waste recycling intention based on the integration of the valence theory and the norm activation theory. Four data mining models using classification and prediction-based algorithms, namely Chi-squared automatic interaction detector (CHAID), Neural network, Discriminant analysis, and Quick, unbiased, efficient statistical tree (QUEST), were employed to analyze a set of the 398 data collected in Vietnam. The results revealed that the social support value is by far the most critical predictor, followed by the utilitarian value, task difficulty, and monetary risk. It is also noteworthy that the awareness of consequences, education background, the ascription of responsibility, and age were also ranked as critical affecting factors. The lowest influential predictors found in this study were income and gender. In addition, a comparison was made in terms of the classification performance of the four utilized data mining techniques. Based on several evaluation measurements (confusion matrix, accuracy, precision, recall, specificity, F-measure, ROC curve, and AUC), the aggregated results suggested that CHAID and Neural network performed the best. The findings of this research are expected to assist policymakers and future researchers in updating all information surrounding consumer behavioral intention-related topics focusing on e-waste. Furthermore, the adoption of data mining algorithms for prediction is another insight of this study, which may shed the light on data mining applications in such environmental studies in the future.

Developing supervised machine learning algorithms to evaluate the therapeutic effect and laboratory-related adverse events of cyclosporine and tacrolimus in renal transplants.

Single nucleotide polymorphisms influence the effects of tacrolimus and cyclosporine in renal transplants. We set out to use machine learning algorithms (MLAs) to identify variables that predict the therapeutic effects and adverse events following tacrolimus and cyclosporine administration in renal transplant patients. We sampled 120 adult renal transplant patients (on cyclosporine or tacrolimus). Generalized linear model (GLM), support vector machine (SVM), artificial neural network (ANN), Chi-square automatic interaction detection, classification and regression tree, and K-nearest neighbors were the chosen MLAs. The mean absolute error (MAE), relative mean square error (RMSE), and regression coefficient (β) with a 95% confidence interval (CI) were used as the model parameters. For a stable dose of tacrolimus, the MAEs (RMSEs) of GLM, SVM, and ANN were 1.3 (1.5), 1.3 (1.8), and 1.7 (2.3) mg/day, respectively. GLM revealed that the POR*28 genotype and age significantly predicted the stable dose of tacrolimus as follows: POR*28 (β -1.8; 95% CI -3, -0.5; p = 0.006), and age (β -0.04; 95% CI -0.1, -0.006; p = 0.02). For a stable dose of cyclosporine, MAEs (RMSEs) of 93.2 (103.4), 79.1 (115.2), and 73.7 (91.7) mg/day were observed with GLM, SVM, and ANN, respectively. GLM revealed the following predictors of a stable dose of cyclosporine: CYP3A5*3 (β -80.8; 95% CI -130.3, -31.2; p = 0.001), and age (β -3.4; 95% CI -5.9, -0.9; p = 0.007). We observed that various MLAs could identify significant predictors that were useful to optimize tacrolimus and cyclosporine dosing regimens; yet, the findings must be externally validated.

Predictive power of the DSM-5 criteria for internet use disorder: A CHAID decision-tree analysis.

Although the majority of internet users enjoy the internet as a recreational activity, some individuals report problematic internet use behaviors causing negative psychosocial consequences. Therefore, it is important to have precise and valid diagnostic criteria to ensure suitable treatment for those affected and avoid over-pathologization. The aim of the present study was to determine which of the nine DSM-5 criteria of internet gaming disorder (IGD) are crucial in distinguish pathological from non-pathological internet use based on the questionnaire-based response behavior of the participants by applying the Chi-squared automatic interaction detection (CHAID) decision tree analysis. Under consideration of the nine DSM-5 criteria for IGD and according to the short-form scale to assess Internet Gaming Disorder (IGDS-SF9) the DSM-5 criteria were formulated as questions and applied to the broader concept of Internet Use Disorder (IUD). The nine questions were answered on a 5-point Likert scale from "never" to "very often." In accordance with the IGDS-SF9 participants were assigned to IUD-5plus if at least 5 of the 9 criteria were answered with "very often." The study was conducted in Germany (N = 37,008; mean age: 32 years, SD = 13.18, 73.8% male). Although "loss of control," "continued overuse" and "mood regulation" were the most endorsed criteria, the analysis indicated that the criterion "jeopardizing" was found as the best predictor for IUD-5plus, followed by "loss of interest" and "continued overuse." Overall 64.9% of all participants who were in the IUD-5plus, could been identified by the fulfillment of the three criteria mentioned above. The results found support for adjustment of the DSM-5 criteria of IGD in accordance to ICD-11. If the predictive power of the three criteria can be replicated in future representative studies, such a decision tree can be used as guidance for diagnostics to capture the particularly relevant criteria.

Assessment of Factors Contributing to Consumer Attitudes and Preferences toward Chantenay Carrots using the CHAID Model

Increasing the consumption of fresh vegetables to the recommended dietary guidelines by the World Health Organization (WHO) is the top priority for many national policy makers and horticulture businesses. One growth strategy which has been shown to promote vegetable intake is introducing different vegetable varieties into the market. This research analyses factors affecting consumer attitudes and willingness to purchase a new variety of carrots imported to Australia from the UK—the Chantenay carrot. To better inform marketing tactics for this new variety and to provide insights for predictions for uptake, a general carrot usage and attitude survey was conducted in four Australian states (N = 300). Results indicated that: sustainable packaging options; the size of carrots; cooking tips; taste; and place of origin were attitudinal triggering factors for purchasing fresh carrots in this sample. These factors were compared to the characteristics of the Chantenay variety using a CHi-squared Automatic Interaction Detection (CHAID) model to gauge the likelihood of acceptance and uptake of the variety within the larger Australian population. This study benefits marketing practitioners and horticultural businesses to develop a new vegetable variety that meets consumer preferences successfully.

Towards the ideal vascular implant: Use of machine learning and statistical approaches to optimise manufacturing parameters

Introduction: Electrospinning is a manufacturing technique that creates a net of nano and microfibres able to mimic the natural extracellular matrix (ECM) of biological tissue. Electrospun scaffolds' morphology and mechanical behaviour can be tailored by modifying the environmental, solution and process parameters. This study aims to produce biomimetic vascular implants optimising the manufacturing set up through two machine learning techniques and statistical approaches.Methods: Polyvinyl alcohol (PVA) based scaffolds were produced by modifying the concentration of the polymer, flow rate, voltage, type of collector, diameter of the needle, distance between needle and collector and revolutions of the mandrel. The scaffolds were morphologically and mechanically characterised using scanning electron microscopy and mechanical testing respectively to inform the morphological model (simultaneously predicting diameter of the fibres and inter-fibre separation) and mechanical model (predicting strain at rupture and ultimate tensile strength).Results: Prediction and traditional techniques led to an optimum set up of: 12% PVA, 1 ml/h flow rate, 20 kV, 8 cm between the needle, 18 G gauge needle, rotational mandrel of 15 cm and 2000 rpm. Optimised PVA scaffolds replicated the mechanical properties and morphology of the vascular tissue with an ultimate tensile strength of 6.17 ± 0.18 MPa, a strain at break of 97.39 ± 5.06, fibre diameters of 126 ± 6.11 nm and inter-fibre separation of 1488 ± 91.99 nm.Discussion: This work revealed for the first time that machine learning Chi-squared Automatic Interaction Detection (CHAID) models are a novel and visual route to elect the optimum manufacturing set up to develop biomimetic vascular implants. Novel two-output Artificial Neural Networks (ANN) and multivariate analysis of variance and covariance (MANOVA, MANCOVA) models presented comparable prediction results (R2=0.91); however, two-output ANN predicted models demonstrated to be the most powerful tool for non-parametric conditions, showing cross-validation mean squared errors (MSE) of 0.0001943. Multi Linear Regression models (MLR) exhibited the lowest accuracy in their predictions (R2=0.6). Machine learning, statistical approaches and traditional characterisation methods were studied to successfully achieve vascular substitutes with analogous biomechanical behaviour and physical structure to the native vascular tissue.

A decision tree analysis to predict clinical outcome of minimally invasive lumbar decompression surgery for lumbar spinal stenosis with and without coexisting spondylolisthesis and scoliosis

Implementing machine learning techniques, such as decision trees, known as prediction models that use logical construction diagrams, are rarely used to predict clinical outcomes. To develop a clinical prediction rule to predict clinical outcomes in patients who undergo minimally invasive lumbar decompression surgery for lumbar spinal stenosis with and without coexisting spondylolisthesis and scoliosis using a decision tree model. A retrospective analysis of prospectively collected data. This study included 331 patients who underwent minimally invasive surgery for lumbar spinal stenosis and were followed up for ≥2 years at 1 institution. Self-report measures: The Japanese Orthopedic Association (JOA) scores and low back pain (LBP)/leg pain/leg numbness visual analog scale (VAS) scores. Physiologic measures: Standing sagittal spinopelvic alignment, computed tomography, and magnetic resonance imaging results. Low achievement in clinical outcomes were defined as the postoperative JOA score at the 2-year follow-up <25 points. Univariate and multiple logistic regression analysis and chi-square automatic interaction detection (CHAID) were used for analysis. The CHAID model for JOA score <25 points showed spontaneous numbness/pain as the first decision node. For the presence of spontaneous numbness/pain, sagittal vertical axis ≥70 mm was selected as the second decision node. Then lateral wedging, ≥6° and pelvic incidence minus lumbar lordosis (PI-LL) ≥30° followed as the third decision node. For the absence of spontaneous numbness/pain, sex and lateral olisthesis, ≥3mm and American Society of Anesthesiologists physical status classification system score were selected as the second and third decision nodes. The sensitivity, specificity, and the positive predictive value of this CHAID model was 65.1, 69.8, and 64.7% respectively. The CHAID model incorporating basic information and functional and radiologic factors is useful for predicting surgical outcomes.

Inequities in home care use among older Canadian adults: Are they corrected by public funding?

Although care use should parallel needs, enabling and predisposing circumstances including the socio-demographic inequities of socioeconomic status (SES), gender, or isolation often intervene to diminish care. We examine whether availability of state-funded medical and support services at home can rebalance these individual and social inequities, and do this by identifying if and how intersecting social identities predict homecare use among older Canadian adults. Using the Canadian Longitudinal Study on Aging (CLSA) of 30,097 community-dwelling adults aged 45 to 85, we performed recursive partitioning regression tree analysis using Chi-Squared automatic interaction detection (CHAID). Combinations of individual and social characteristics including sociodemographic, family-related, physical and psychological measures and contextual indicators of material and social deprivation were explored as possible predictors of formal and informal care use. Diminished function i.e. increased need, indicated by Activities of Daily Living, was most strongly aligned with formal care use while age, living arrangement, having no partner, depression, self-rated health and chronic medical conditions playing a lesser role in the pathway to use. Notably, sex/gender, were not determinants. Characteristics aligned with informal care were first-need, then country of birth and years since immigration. Both 'trees' showed high validity with low risk of misclassification (4.6% and 10.8% for formal and informal care, respectively). Although often considered marginalised, women, immigrants, or those of lower SES utilised formal care equitably. Formal care was also differentially available to those without the financial or human resources to receive informal care. Need, primarily medical but also arising from living arrangement, rather than SES or gender predicted formal care, indicating that universal government-funded services may rebalance social and individual inequities in formal care use.

Spatial assessment of landslide susceptibility mapping generated by fuzzy-AHP and decision tree approaches

The current study aimed to elaborate a landslide susceptibility map (LSM) in the region that has high landslide risk and is located within the borders of the Atakum district of Samsun province, Turkey. For this aim, topographic, geological, land use, and soil indicators were considered in terms of landslide-conditioning and landslide-triggering parameters and they were weighted through the Fuzzy-Analytic Hierarchy Process (Fuzzy-AHP) approach. Then landslide susceptibility maps were generated at 4 different class levels (Very low-H1, Low-H2, Moderate-H3, High-H4) by using a weighted linear combination technique integrated with the Geographic Information System (GIS). In addition, it was investigated the predictability of susceptibility maps by using a decision tree algorithm named CHAID (Chi-Square Automatic Interaction Detection). According to the results, the 'very low' and 'low' susceptibility class, corresponding to 29.8 % of the total area in the susceptibility map, was estimated with 100 % accuracy through the decision tree algorithm. 70.2 % of the total area, specified in ‘medium’ (H3-68.6 %) and ‘high’ (H4-1.6 %) susceptibility classes in the map created Fuzzy-AHP, was found to be in the ‘medium’ susceptibility class, as a result of the estimation made with the decision tree. Although the H1, H2, and H3 classes were successfully estimated (p < 0.05) by using the weights obtained through the Fuzzy-AHP approach with the help of the decision tree algorithm, the estimation accuracy of the H4 class was 'low' (AUC [Area Under Curve]: 0.773; p > 0.05), at the end of the research.

Identity profile of young people experiencing a sense of risk on the internet: A data mining application of decision tree with CHAID algorithm

Social media, interactive games, and digital platforms are just some of the ways in which the internet has become part of our lives, and more specifically so among young people. Among other things, this technology is the reason young people are spending increasingly more time online than offline, prompting numerous risks that they are sometimes unaware of. The following quantitative research is based on this premise, and it pursues two objectives, namely, to discover whether or not young people experience a sense of risk linked to their use of the internet and identify the kind of young people that say they do or do not perceive any such risk. A total of 1991 young people answered the Technology Use Studies questionnaire. The classification tree technique based on the Chi-square Automatic Interaction Detector (CHAID) was applied to identify the predictive variables associated with the sense of risk in young people. Two highly significant results have been forthcoming: firstly, we present the factors with the biggest impact on a young person's internet-related sense of risk, involving isolation when they cannot go online, followed by variables such as gender, the use of TikTok, and videogames, the consumption of pornography, and distraction from study time. Secondly, our results indicate that the young people that feel isolated when they cannot go online are precisely the ones that do not experience a sense of online risk. It is therefore of vital importance for education to generate and provide young people with strategies that allow them to establish a healthy and responsible relationship with technologies such as the internet.