Psychological Assessment in Anorexia Nervosa: A Single Case Study on Predictive Factors of Dropout.

Anorexia nervosa (AN) has the highest mortality rate among psychiatric disorders. Adult AN patients have a chronic history of treatment dropout due to denial of their psychological and physical disease states, which may be connected to defense mechanisms. We developed an assessment protocol to evaluate the psychological functioning of patients undergoing a psychodynamic approach for eating disorders (PAED), aimed at identifying the psychological factors associated with intervention success or dropout. We analyzed the case of an adult patient who quit treatment at the start and discussed her psychological functioning profile. We present the case of a 45-year-old woman with enduring AN, who entered the PAED program at an Italian hospital. In adult AN patients, denial and acting out may have significant impacts on clinic compliance. This hampers establishing a relationship with the clinic and the success of the psychological work aimed at promoting mental awareness and insights into the disorder. This highlights the need to consider which aspects of the initial psychological assessment are predictive of dropout in AN patients.

Predictors and Patterns of Dropout From Psychiatric Treatment Among Trauma-Affected Refugees: A Large Data Pool Analysis.

Dropout from mental health treatment is a substantial hindrance to relevant and effective treatment. Despite the high prevalence of PTSD among refugees, research into their treatment dropout has received limited attention. This study aimed to identify patterns and predictors of treatment dropout versus completion through different treatment stages. The sample included 940 patients with a refugee background undergoing outpatient treatment for PTSD in Denmark. All patients were offered 10 medical doctor sessions and 16-20 psychotherapy sessions. Dropout was analysed in three stages: (1) during the first six MD sessions, (2) during the first eight psychotherapy sessions upon completion of Stage 1, and (3) during psychotherapy sessions 9 to 16. A stepwise multiple regression analysis was conducted for each stage to identify predictors of stage-specific dropout. Counter to expectations, both early dropout and full completion were associated with better symptom outcomes, relative to late-treatment dropout. Key predictors varied by stage, with younger age predicting early dropout, whereas chronic pain and poor Danish proficiency predicted late dropout. Female gender and a clearly articulated motivation for active participation were predictors for full treatment completion. Practical advice is suggested to accommodate at-risk patients and to re-evaluate patient engagement after familiarisation with treatment.

Dropout From Psychological Interventions for Pathological Health Anxiety: A Three-Level Meta-Analysis of Randomized Controlled Trials.

This study aimed to provide the first comprehensive evidence on the prevalence and predictors of dropout in psychological interventions for pathological health anxiety. A database search in Web of Science, EMBASE, PubMed, Scopus, PsycINFO and the Cochrane Central Register of Controlled Trials identified 28 eligible randomized controlled trials (40 intervention conditions; 1783 participants in the intervention condition), published up to 18 June 2024. Three-level meta-analytic results showed a weighted average dropout rate of 9.67% (95% confidence interval [CI] [6.49%, 14.17%]), with dropout equally likely from treatment and control conditions (odds ratio = 1.07, 95% CI [0.80, 1.44]). Moderator analyses indicated no statistically significant effects of study, participant, treatment or therapist characteristics, except for the country of study. These findings suggest that the average dropout rate is relatively low compared with those reported for other mental health conditions and highlight the importance of considering cultural and societal factors when evaluating treatment adherence. Future research should continue to explore the complex and multifaceted factors influencing dropout to improve the design and implementation of psychological interventions for pathological health anxiety.

SAE-Impute: imputation for single-cell data via subspace regression and auto-encoders.

BACKGROUND: Single-cell RNA sequencing (scRNA-seq) technology has emerged as a crucial tool for studying cellular heterogeneity. However, dropouts are inherent to the sequencing process, known as dropout events, posing challenges in downstream analysis and interpretation. Imputing dropout data becomes a critical concern in scRNA-seq data analysis. Present imputation methods predominantly rely on statistical or machine learning approaches, often overlooking inter-sample correlations. RESULTS: To address this limitation, We introduced SAE-Impute, a new computational method for imputing single-cell data by combining subspace regression and auto-encoders for enhancing the accuracy and reliability of the imputation process. Specifically, SAE-Impute assesses sample correlations via subspace regression, predicts potential dropout values, and then leverages these predictions within an autoencoder framework for interpolation. To validate the performance of SAE-Impute, we systematically conducted experiments on both simulated and real scRNA-seq datasets. These results highlight that SAE-Impute effectively reduces false negative signals in single-cell data and enhances the retrieval of dropout values, gene-gene and cell-cell correlations. Finally, We also conducted several downstream analyses on the imputed single-cell RNA sequencing (scRNA-seq) data, including the identification of differential gene expression, cell clustering and visualization, and cell trajectory construction. CONCLUSIONS: These results once again demonstrate that SAE-Impute is able to effectively reduce the droupouts in single-cell dataset, thereby improving the functional interpretability of the data.

Uncertainty quantification in multi-class image classification using chest X-ray images of COVID-19 and pneumonia.

This paper investigates uncertainty quantification (UQ) techniques in multi-class classification of chest X-ray images (COVID-19, Pneumonia, and Normal). We evaluate Bayesian Neural Networks (BNN) and the Deep Neural Network with UQ (DNN with UQ) techniques, including Monte Carlo dropout, Ensemble Bayesian Neural Network (EBNN), Ensemble Monte Carlo (EMC) dropout, across different evaluation metrics. Our analysis reveals that DNN with UQ, especially EBNN and EMC dropout, consistently outperform BNNs. For example, in Class 0 vs. All, EBNN achieved a UAcc of 92.6%, UAUC-ROC of 95.0%, and a Brier Score of 0.157, significantly surpassing BNN's performance. Similarly, EMC Dropout excelled in Class 1 vs. All with a UAcc of 83.5%, UAUC-ROC of 95.8%, and a Brier Score of 0.165. These advanced models demonstrated higher accuracy, better discriaminative capability, and more accurate probabilistic predictions. Our findings highlight the efficacy of DNN with UQ in enhancing model reliability and interpretability, making them highly suitable for critical healthcare applications like chest X-ray imageQ6 classification.

Determinants of measles second dose vaccination dropout among children aged 18-24 months in Ejere woreda, central Ethiopia; unmatched case-control study.

Background: Measles continues to pose a significant public health challenge, especially in low- and middle-income countries. Despite the implementation of national vaccination programs, measles outbreaks persist in some parts of Ethiopia, and the determinants of dropout from the second measles vaccine dose are not well understood. Hence, this study aimed to assess determinants of measles second dose vaccination dropout among children aged 18-24 months in Ejere woreda, central Ethiopia. Methods: A community-based unmatched case-control design was conducted in the Ejere Woreda of the Oromia regional state in Ethiopia between February 14 and April 6, 2023. Data were collected using a pre-tested structured questionnaire. The collected data were coded and entered into Epi-data version 3.1 and then transported to SPSS version 27 for statistical analysis. Descriptive analysis like frequency, mean, and percentage was calculated. Binary and multivariable logistic regression analysis was done. Finally, variables with a p-value <0.05 were considered statistically significant. Result: A total of 446 mothers/caregivers, comprising 110 cases and 336 controls, participated in this study, making the response rate 97.8%. Lack of a reminder for the measles vaccine during postnatal care (PNC) (AOR = 5.19; 95% CI: 2.34, 7.83), having ≤2 antenatal care (ANC) contacts (AOR = 4.95; 95% CI: 2.86, 9.24), long waiting times during previous vaccination (AOR = 2.78; 95% CI: 1.19, 4.38), children of mothers/caregivers without formal education (AOR = 6.46; 95% CI: 2.81, 11.71), mothers/caregivers of children who were unaware of the importance of the second dose of measles (AOR = 8.37; 95% CI: 4.22, 15.08), and mothers/caregivers whose children did not receive at least two doses of vitamin A (AOR = 4.05; 95% CI: 2.15, 8.11) were significant determinants of measles second dose vaccination dropout. Conclusion: Implementing targeted interventions during antenatal care and when mothers visit health facilities for other vaccines can significantly improve the uptake of the second dose of the measles vaccine. These strategies not only enhance overall vaccination coverage but also mitigate the risk of measles outbreaks in the community.

Gray matter volume differences in intimate partner violence perpetrators and its role in explaining dropout and recidivism.

AIM: Psychological instruments that are employed to adequately explain treatment compliance and recidivism of intimate partner violence (IPV) perpetrators present a limited ability and certain biases. Therefore, it becomes necessary to incorporate new techniques, such as magnetic resonance imaging (MRI), to be able to surpass those limitations and measure central nervous system characteristics to explain dropout (premature abandonment of intervention) and recidivism. METHOD: The main objectives of this study were: 1) to assess whether IPV perpetrators (n = 60) showed differences in terms of their brain's regional gray matter volume (GMV) when compared to a control group of non-violent men (n = 57); 2) to analyze whether the regional GMV of IPV perpetrators before starting a tailored intervention program explain treatment compliance (dropout) and recidivism rate. RESULTS: IPV perpetrators presented increased GMV in the cerebellum and the occipital, temporal, and subcortical brain regions compared to controls. There were also bilateral differences in the occipital pole and subcortical structures (thalamus, and putamen), with IPV perpetrators presenting reduced GMV in the above-mentioned brain regions compared to controls. Moreover, while a reduced GMV of the left pallidum explained dropout, a considerable number of frontal, temporal, parietal, occipital, subcortical and limbic regions added to dropout to explain recidivism. CONCLUSIONS: Our study found that certain brain structures not only distinguished IPV perpetrators from controls but also played a role in explaining dropout and recidivism. Given the multifactorial nature of IPV perpetration, it is crucial to combine neuroimaging techniques with other psychological instruments to effectively create risk profiles of IPV perpetrators.

Exploring the impact of depression, anxiety, stress, academic engagement, and dropout intention on medical students' academic performance: A prospective study.

BACKGROUND: Depression, anxiety, and stress (DAS) have been linked to poor academic outcomes. This study explores the relationships among DAS, academic engagement, dropout intentions, and academic performance - measured by Grade Point Average (GPA) - in medical students. It aims to understand how these factors relate to each other and predict academic performance. METHODS: Data were collected from 351 medical students (74.9 % female) through an online survey. The average age was 20.2 years. Psychometric instruments measured DAS, academic engagement, and dropout intentions. Structural equation modeling was used to test the relationships between these variables and their prediction of GPA. RESULTS: DAS was negatively associated with academic engagement ß̂=-0.501p<0.001 and positively associated with dropout intentions ß̂=0.340p<0.001. Academic engagement positively predicted GPA ß̂=0.298p<0.001 and negatively associated with dropout intentions ß̂=-0.367p<0.001. DAS had a nonsignificant direct effect on GPA ß̂=-0.008p=0.912. However, the indirect effect of DAS - via academic engagement - on GPA and dropout intention was statistically significant. LIMITATIONS: The study's limitations include the use of a convenience sample and the collection of all variables, except GPA, at the same time point, which may affect the generalizability of the results. CONCLUSIONS: The study supports the important role of DAS in its association with academic engagement and dropout intentions, which can predict GPA. Addressing DAS could enhance academic engagement and reduce dropout rates, leading to better academic performance.

Crowd Counting Using Meta-Test-Time Adaptation.

Machine learning algorithms are commonly used for quickly and efficiently counting people from a crowd. Test-time adaptation methods for crowd counting adjust model parameters and employ additional data augmentation to better adapt the model to the specific conditions encountered during testing. The majority of current studies concentrate on unsupervised domain adaptation. These approaches commonly perform hundreds of epochs of training iterations, requiring a sizable number of unannotated data of every new target domain apart from annotated data of the source domain. Unlike these methods, we propose a meta-test-time adaptive crowd counting approach called CrowdTTA, which integrates the concept of test-time adaptation into the meta-learning framework and makes it easier for the counting model to adapt to the unknown test distributions. To facilitate the reliable supervision signal at the pixel level, we introduce uncertainty by inserting the dropout layer into the counting model. The uncertainty is then used to generate valuable pseudo labels, serving as effective supervisory signals for adapting the model. In the context of meta-learning, one image can be regarded as one task for crowd counting. In each iteration, our approach is a dual-level optimization process. In the inner update, we employ a self-supervised consistency loss function to optimize the model so as to simulate the parameters update process that occurs during the test phase. In the outer update, we authentically update the parameters based on the image with ground truth, improving the model's performance and making the pseudo labels more accurate in the next iteration. At test time, the input image is used for adapting the model before testing the image. In comparison to various supervised learning and domain adaptation methods, our results via extensive experiments on diverse datasets showcase the general adaptive capability of our approach across datasets with varying crowd densities and scales.

The forecasting of surface displacement for tunnel slopes utilizing the WD-IPSO-GRU model.

To quickly assess slope stability based on field displacement monitoring data, this paper constructs a hybrid optimization model that predicts surface displacement during tunnel excavation in base-overburden slopes. The model combines Wavelet Decomposition (WD) with a Gated Recurrent Unit (GRU), and the GRU's hyperparameters are optimized using an Improved Particle Swarm Optimization algorithm (IPSO). The specific steps are as follows: First, the Wavelet Decomposition (WD) technique is applied to decompose the raw displacement data, extracting features at different time-frequency scales. Next, the Dropout technique is incorporated into the GRU model to prevent overfitting. Additionally, nonlinear inertia weight ω improved cognitive factor c1, and social factor c2 are introduced. The PSO algorithm is improved by integrating crossover and mutation concepts from genetic algorithms. Finally, the IPSO is used to optimize the number of neural units hN, HN, LN and dropout rates D1 and D2 in the GRU network architecture. After constructing the WD-IPSO-GRU model, a comprehensive comparison is made with various swarm intelligence algorithms and state-of-the-art models. The experimental results demonstrate that the WD-IPSO-GRU model significantly improves the prediction accuracy of surface displacement in slopes during tunnel excavation. Compared to directly using raw data for prediction, the introduction of the WD preprocessing technique improved the prediction accuracy at measurement points 01 and 02 by 28% and 45.9%, respectively. Additionally, with the model optimized by IPSO, the prediction accuracy at measurement points 01 and 02 increased by 76% and 56.7%, respectively. The WD-IPSO-GRU model effectively addresses the challenges of extracting features from univariate displacement time-series data and determining the parameters of the GRU network. It improves the prediction accuracy of surface displacement in base-overburden type slopes and demonstrates excellent generalization ability and reliability. The research results validate the potential application of the model in geotechnical engineering and provide strong support for assessing slope stability during tunnel excavation.

Meibomian Gland Dysfunction and Dropout in Diabetic Patients with Non-Proliferative Diabetic Retinopathy.

This study aims to compare meibomian gland (MG) dropout and MG dysfunction (MGD) between patients with diabetes mellitus (DM) with moderate-severe non-proliferative diabetic retinopathy (NPDR) and patients with no diabetes (NDM). This prospective, transversal, age, and gender-matched case-control study included 98 DM and 106 NDM eyes. Dry eye disease (DED) and MGD evaluations were performed, including meibography (Keratograph 5M®). The objective MG dropout percentage was obtained by analyzing meibography images with ImageJ software (v. 1.52o, National Institutes of Health, Bethesda, MD, USA) and was subsequently graded with Arita's meiboscore. The DM duration was 18 ± 9 years. The mean meiboscore (3.8 ± 0.8 vs. 3.4 ± 1.0, p = 0.001), meiboscore severity (p = 0.016), and MG dropout (45.1 ± 0.1% vs. 39.0 ± 0.4%, p < 0.001) were greater in DM than in NDM. All patients showed MG dropout (meiboscore > 1). Lower eyelids showed greater MG dropout in both groups. A correlation with age (r = 0.178, p = 0.014) and no correlations with DM duration or gender (p > 0.005) were observed. Patients with diabetes showed greater corneal staining (1.7 ± 1.3 vs. 0.9 ± 1.1; p < 0.001), reduced corneal sensitivity (5.4 ± 1.1 vs. 5.9 ± 0.4; p < 0.001), lower MG expressibility (3. 9 ± 1.6 vs. 4.4 ± 2.1; p = 0.017), and worse meibum quality (1.9 ± 0.8 vs. 1.7 ± 0.5; p = 0.019). Tear breakup time, osmolarity, MMP-9, Schirmer, and the Ocular Surface Disease Index showed no significant differences. In conclusion, patients with DM with NPDR have greater MG dropout and meiboscore, as well as more severe MGD and DED parameters than persons with NDM.

Cluster Analysis in Victims of Perceived Cyberbullying at the Higher Education Level.

A number of studies have demonstrated the prevalence of cyberbullying in university settings. The objective of this research is to conduct a cluster analysis to categorize victims according to the nature of the behavior they have received and to examine the relationship between gender and intention to drop out. To this end, the Online Victimization Questionnaire was administered to a sample of 800 first-year students at a university in northern Spain who had opted to participate in the study. All analyses were conducted using the SPSS statistical software, version 27.0. Results indicate the presence of four clusters: Cluster 4 (73.625%) exhibited no instances of cyberbullying behaviors. Cluster 1 (21.875%), which exhibited low scores across all cyberbullying behaviors except identity manipulation, was the most prevalent. Cluster 2 (3.125%) demonstrated high scores for public aggression and social isolation. Finally, Cluster 3 (1.375%) exhibited high scores for all cyberbullying behaviors. Furthermore, gender differences play a significant role in the formation of these clusters. It is therefore evident that there are various profiles of cyberbullying victims, which both public policies and educational programs should be aware of in order to adapt their prevention strategies. This is also a factor that affects university dropout prevention programs.

Attrition of older adults in web-based health interventions: Survival analysis within an observational cohort study.

To identify demographics and personal motivation types that predict dropping out of eHealth interventions among older adults. We conducted an observational cohort study. Participants completed a pre-test questionnaire and got access to an eHealth intervention, called Stranded, for 4 weeks. With survival and Cox-regression analyses, demographics and types of personal motivation were identified that affect drop-out. Ninety older adults started using Stranded. 45.6% participants continued their use for 4 weeks. 32.2% dropped out in the first week and 22.2% dropped out in the second or third week. The final multivariate Cox-regression model which predicts drop-out, consisted of the variables: perceived computer skills and level of external regulation. Predicting the chance of dropping out of an eHealth intervention is possible by using level of self-perceived computer skills and level of external regulation (externally controlled rewards or punishments direct behaviour). Anticipating to these factors can improve eHealth adoption.

DGD-CNet: Denoising Gated Recurrent Unit with a Dropout-Based CSI Network for IRS-Aided Massive MIMO Systems.

For the deployment of Sixth Generation (6G) networks, integrating Massive Multiple-Input Multiple-Output (Massive MIMO) systems with Intelligent Reflecting Surfaces (IRS) is highly recommended due to its significant benefits in reducing communication losses for Non-Line-of-Sight (NLoS) conditions. However, the use of passive IRS presents challenges in channel estimation, mainly due to the significant feedback overhead required in Frequency Division Duplex (FDD)-based Massive MIMO systems. To address these challenges, this paper introduces a novel Denoising Gated Recurrent Unit with a Dropout-based Channel state information Network (DGD-CNet). The proposed DGD-CNet model is specifically designed for FDD-based IRS-aided Massive MIMO systems, aiming to reduce the feedback overhead while improving the channel estimation accuracy. By leveraging the Dropout (DO) technique with the Gated Recurrent Unit (GRU), the DGD-CNet model enhances the channel estimation accuracy and effectively captures both spatial structures and time correlation in time-varying channels. The results show that the proposed DGD-CNet model outperformed existing models in the literature, achieving at least a 26% improvement in Normalized Mean Square Error (NMSE), a 2% increase in correlation coefficient, and a 4% in system accuracy under Low-Compression Ratio (Low-CR) in indoor situations. Additionally, the proposed model demonstrates effectiveness across different CRs and in outdoor scenarios.

No medication prescription and residential distance from the hospital are important factors associated with nonsurgical weight-loss treatment discontinuance in Japanese patients with high-degree obesity: a retrospective study.

BACKGROUND: Although the percentage of the population with a high degree of obesity (body mass index [BMI] ≥ 35 kg/m2) is low in Japan, the prevalence of obesity-related diseases in patients with high-degree obesity is greater than that in patients with a BMI < 35 kg/m2. Therefore, treatment for high-degree obesity is important. However, clinical studies have reported that 20-50% of patients with obesity discontinue weight-loss treatment in other countries. The circumstances surrounding antiobesity agents are quite different between Japan and other countries. In this study, we investigated the predictors of treatment discontinuation in Japanese patients with high-degree obesity. METHODS: We retrospectively reviewed the medical charts of 271 Japanese patients with high-degree obesity who presented at Toho University Sakura Medical Center for obesity treatment between April 1, 2014, and December 31, 2017. The patients were divided into non-dropout and dropout groups. Patients who discontinued weight-loss treatment within 24 months of the first visit were defined as "dropouts." Multivariate Cox proportional hazards regression analysis and Kaplan-Meier survival analysis were performed to examine the factors predicting treatment withdrawal. RESULTS: Among the 271 patients, 119 (43.9%) discontinued treatment within 24 months of the first visit. The decrease in BMI did not significantly differ between the two groups. No prescription of medication and residential distance from the hospital exceeding 15 km were the top contributors to treatment discontinuation, and the absence of prescription medication was the most important factor. The dropout-free rate was significantly higher in patients with medication prescriptions than in those without and in patients who lived within 15 km of the hospital than in those who lived farther than 15 km from the hospital. CONCLUSIONS: No medication prescription and longer residential distance from the hospital were associated with treatment dropout in Japanese patients with high-degree obesity; therefore, the addition of antiobesity medications and telemedicine may be necessary to prevent treatment discontinuation in such patients.

Rates and predictors of service disengagement in first episode psychosis: Results from a 2-year follow-up study in an Italian real-world care setting.

AIM: Service disengagement is a major problem for "Early Intervention in Psychosis" (EIP). Understanding predictors of engagement is also crucial to increase effectiveness of mental health treatments, especially in young people with First Episode Psychosis (FEP). No Italian investigation on this topic has been reported in the literature to date. The goal of this research was to assess service disengagement rate and predictors in an Italian sample of FEP subjects treated within an EIP program across a 2-year follow-up period. METHODS: All patients were young FEP help-seekers, aged 12-35 years, recruited within the "Parma Early Psychosis" (Pr-EP) program. At baseline, they completed the Positive And Negative Syndrome Scale (PANSS) and the Global Assessment of Functioning (GAF) scale. Univariate and multivariate Cox regression analyses were carried out. RESULTS: 489 FEP subjects were enrolled in this study. Across the follow-up, a 26 % prevalence rate of service disengagement was found. Particularly strong predictors of disengagement were living with parents, poor treatment adherence at entry and a low baseline PANSS "Disorganization" factor score. CONCLUSION: More than a quarter of our FEP individuals disengaged the Pr-EP program during the first 2 years of intervention. A possible solution to reduce disengagement and to facilitate re-engagement of these young patients might be to offer the option of low-intensity monitoring and support, also via remote technology and tele-mental health care.

Modelling a self-defined CNN for effectual classification of PCOS from ultrasound images.

BACKGROUND: Polycystic Ovary Syndrome (PCOS) is a medical condition that causes hormonal disorders in women in their childbearing years. The hormonal imbalance leads to a delayed or even absent menstrual cycle. Women with PCOS mainly suffer from extreme weight gain, facial hair growth, acne, hair loss, skin darkening, and irregular periods, leading to infertility in rare cases. Doctors usually examine ultrasound images and conclude the affected ovary but are incapable of deciding whether it is a normal cyst, PCOS, or cancer cyst manually. OBJECTIVE: To have access to the high-risk crucial PCOS and to detect the condition and the treatment aimed at mitigating health hazards such as endometrial hyperplasia/cancer, infertility, pregnancy complications, and the long-term burden of chronic diseases such as cardiometabolic disorders linked with PCOS. METHODS: The proposed Self-Defined Convolution Neural Network method (SD_CNN) is used to extract the features and machine learning models such as SVM, Random Forest, and Logistic Regression are used to classify PCOS images. The parameter tuning is done with lesser parameters in order to overcome over-fitting issues. The self-defined model predicts the occurrence of the cyst based on the analyzed features and classifies the class labels effectively. RESULTS: The Random Forest Classifier was found to be the most reliable and accurate among Support Vector Machine (SVM) and Logistic Regression (LR), with accuracy being 96.43%. CONCLUSION: The proposed model establishes better trade-off compared to various other approaches and works effectually for PCOS prediction.

Who stays? Who goes? Motivation and tendency to drop out in music schools.

Based on self-determination theory, this study examined the extent to which the satisfaction of the basic psychological needs for autonomy, competence, and social relatedness in instrumental lessons explain the quality and quantity of motivation, which are responsible for persistence and dropout in music schools. This study also investigated whether parental involvement contributes to dropout. A total of 140 music students from Austria (37.16% male, 62.1% female, 0.8% diverse) were surveyed using a quantitative questionnaire. The central variables are the tendency to dropout (dependent variable) and, as predictors, the motivational regulation styles, the satisfaction of basic psychological needs in the classroom and parental involvement. The results of a structural equation model indicated that satisfaction of basic needs in class and parental involvement, mediated by motivation, predicted dropout tendencies. Autonomous motivation in lessons is negatively associated and controlled motivation is positively associated with the tendency to drop out of music schools. Satisfaction of basic psychological needs during lessons and parental involvement predicts autonomous motivation. However, basic psychological needs cannot predict controlled motivation but parental involvement can predict controlled motivation to a limited extent. Finally, this study emphasizes the practical importance of need satisfaction and parental involvement in motivation and continuing to play a musical instrument.

Medical Professionalism and Its Association with Dropout Intention in Peruvian Medical Students during the COVID-19 Pandemic.

BACKGROUND: The COVID-19 pandemic introduced unprecedented challenges to medical education systems and medical students worldwide, making it necessary to adapt teaching to a remote methodology during the academic year 2020-2021. The aim of this study was to characterize the association between medical professionalism and dropout intention during the pandemic in Peruvian medical schools. METHODS: A cross-sectional online-survey-based study was performed in four Peruvian medical schools (two public) during the academic year 2020-2021. Medical students, attending classes from home, answered three scales measuring clinical empathy, teamwork, and lifelong learning abilities (three elements of medical professionalism) and four scales measuring loneliness, anxiety, depression, and subjective wellbeing. In addition, 15 demographic, epidemiological, and academic variables (including dropout intention) were collected. Variables were assessed using multiple logistic regression analysis. RESULTS: The study sample was composed of 1107 students (390 male). Eight variables were included in an explanatory model (Nagelkerke-R2 = 0.35). Anxiety, depression, intention to work in the private sector, and teamwork abilities showed positive associations with dropout intention while learning abilities, subjective wellbeing, studying in a public medical school, and acquiring a better perception of medicine during the pandemic showed a negative association with dropout intention. No association was observed for empathy. CONCLUSIONS: Each element measured showed a different role, providing new clues on the influence that medical professionalism had on dropout intention during the pandemic. This information can be useful for medical educators to have a better understanding of the influence that professionalism plays in dropout intention.

Analyzing Dropout in Alcohol Recovery Programs: A Machine Learning Approach.

Background: Retention in treatment is crucial for the success of interventions targeting alcohol use disorder (AUD), which affects over 100 million people globally. Most previous studies have used classical statistical techniques to predict treatment dropout, and their results remain inconclusive. This study aimed to use novel machine learning tools to identify models that predict dropout with greater precision, enabling the development of better retention strategies for those at higher risk. Methods: A retrospective observational study of 39,030 (17.3% female) participants enrolled in outpatient-based treatment for alcohol use disorder in a state-wide public treatment network has been used. Participants were recruited between 1 January 2015 and 31 December 2019. We applied different machine learning algorithms to create models that allow one to predict the premature cessation of treatment (dropout). With the objective of increasing the explainability of those models with the best precision, considered as black-box models, explainability technique analyses were also applied. Results: Considering as the best models those obtained with one of the so-called black-box models (support vector classifier (SVC)), the results from the best model, from the explainability perspective, showed that the variables that showed greater explanatory capacity for treatment dropout are previous drug use as well as psychiatric comorbidity. Among these variables, those of having undergone previous opioid substitution treatment and receiving coordinated psychiatric care in mental health services showed the greatest capacity for predicting dropout. Conclusions: By using novel machine learning techniques on a large representative sample of patients enrolled in alcohol use disorder treatment, we have identified several machine learning models that help in predicting a higher risk of treatment dropout. Previous treatment for other substance use disorders (SUDs) and concurrent psychiatric comorbidity were the best predictors of dropout, and patients showing these characteristics may need more intensive or complementary interventions to benefit from treatment.