Social and academic integration variables have shown to be relevant for the understanding of university dropout. However, there is less evidence regarding the influence of these variables on dropout intention, as well as predictive models that explain their relationships. Improvements in this topic become relevant considering that dropout intention stands as a useful measure to anticipate and intervene on this phenomenon. The objective of the present study was to evaluate a predictive model for the university dropout intention that considers the relationships between social and academic variables, during the first university semester of 2020. The research was carried out using a cross-sectional associative-predictive design, with a convenience sampling (n=711) due the restrictions of pandemic period. The results showed a good fit of the proposed hypothetical model that explains 38.7% of dropout intention. Both social support and perceived social isolation predicted the sense of belonging, and through it, engagement. Previous academic performance predicted early academic performance, and through it, engagement. The set of variables predicted the intention to quit, through engagement. These results are a contribution both to the understanding of the phenomenon and to guide potential interventions in the early stages of the university experience.

Pediatric obesity management strategies suffer from a high rate of dropout and persistence of weight excess, despite the use of new tools, such as the automated mobile technology (MT). We aimed to compare the efficacy of two personalized MT protocols with/without monthly in-presence recalls in terms of better adherence to follow-up, and improved anthropometric and lifestyle parameters. MT contacts consisted in three not automated messages per week, inserted between three-monthly in-presence regular visits with (PediaFit 1.2) or without (PediaFit 1.1) monthly in-presence recalls. The sample included 103 children (mean age 10 years, range 6-14) recruited in the Pediatric Obesity Clinic between January 2017 and February 2019, randomized in Intervention group (IG) (n=24 PediaFit 1.1; n=30 PediaFit 1.2) and Control group (CG) (total n=49). Both IGs achieved significantly better results than the CGs for all considered parameters. Comparison of the two IGs at the 6th month showed that IG 1.2 had a statistically significant lower drop-out rate (10% vs. 62%), along with improved body mass index z-score, systolic blood pressure, sleep duration and physical activity. The study suggests that the hybrid association of messaging through personalized/not automated MT plus monthly in-presence recalls may be considered for a favorable outcome of pediatric obesity programs.

Research is still lacking regarding the question as to how programs to promote healthy aging should be organized in order to increase acceptance and thus effectiveness. For older adults, ecological factors, such as physical distance to program sites, might predict participation and retention. Thus, the key aim of this analysis was to examine these factors in a physical activity intervention trial. Adults (N=8,299) aged 65 to 75 years were invited to participate and n=589 participants were randomly assigned to one of two intervention groups with 10 weeks of physical activity home practice and exercise classes or a wait-list control group. Response, participation, and dropout data were compared regarding ecological, individual, and study-related variables. Kaplan-Meier curves and Cox regression models were used to determine predictors of dropout. In total, 405 participants completed the study. Weekly class attendance rates were examined regarding significant weather conditions and holiday periods. The highest rates of nonresponse were observed in districts with very high neighborhood levels of socioeconomic status. In this study, ecological factors did not appear to be significant predictors of dropout, whereas certain individual and study-related variables were predictive. Future studies should consider these factors during program planning to mobilize and keep subjects in the program.

Alcoholism is attributed to regular or excessive drinking of alcohol and leads to the disturbance of the neuronal system in the human brain. This results in certain malfunctioning of neurons that can be detected by an electroencephalogram (EEG) using several electrodes on a human skull at appropriate positions. It is of great interest to be able to classify an EEG activity as that of a normal person or an alcoholic person using data from the minimum possible electrodes (or channels). Due to the complex nature of EEG signals, accurate classification of alcoholism using only a small data is a challenging task. Artificial neural networks, specifically convolutional neural networks (CNN), provide efficient and accurate results in various pattern-based classification problems. In this work, we apply CNN on raw EEG data, and demonstrate how we achieved 98% average accuracy by optimizing a baseline CNN model and outperforming its results in a range of performance evaluation metrics on the UCI-KDD EGG dataset. This article explains the step-wise improvement of the baseline model using the dropout, batch normalization, and kernel regularization techniques, and provides a comparison of the two models that can be beneficial for aspiring practitioners who aim to develop similar classification models in CNN. A performance comparison is also provided with other approaches using the same dataset.