TEDE Communidade: SISTEMA DE PUBLICAÇÃO ELETRÔNICA DE TESES E DISSERTAÇÕES

Avaliação do impacto da educação médica no conhecimento médico e na qualidade assistencial em hiperglicemia e diabete no contexto hospitalar

2025-08-18T00:00:00Z

Título: Avaliação do impacto da educação médica no conhecimento médico e na qualidade assistencial em hiperglicemia e diabete no contexto hospitalar Autor: Lopes, Jivago da Fonseca Primeiro orientador: Böhlke, Maristela Abstract: Diabetes mellitus directly impacts hospitalization rates and length of hospital stay. Knowledge about the disease and the performance of specialized diabetes teams are crucial factors in reducing adverse hospital outcomes. This study aimed to evaluate the effects of a medical educational intervention, both in-person and remote, on physicians' knowledge of hospital hyperglycemia, the quality of their prescriptions, patient care, and resulting clinical outcomes. The study design included a multicenter, open-label, randomized controlled trial (RCT), preceded by a non-randomized before-and-after pilot study. Internal medicine residents from four university hospitals were recruited and randomly allocated to intervention and control groups. The intervention group received a remote lecture on hyperglycemia and diabetes in the hospital setting, complemented by active consultations via WhatsApp for one month. The control group received no intervention. Physicians who participated in the pilot study, members of endocrinology teams, and those on vacation or leave were excluded from the RCT. Evaluated outcomes included medical knowledge (measured by a validated questionnaire), length of hospital stay, time in target glycemic range, occurrence of hyperglycemia and hypoglycemia, antimicrobial use, and the adequacy of medical prescriptions (including capillary glucose monitoring, oral antidiabetics, and insulin therapy). In the pilot study, involving 63 physicians, the overall mean score on the knowledge questionnaire increased from 6.9 (±1.7) to 8.8 (±1.5) after the intervention (p<0.001). Notably, the rate of patients with inadequate glycemic control was significantly reduced, from 12.6% to 2.6% (p<0.001). The incorrect prescription of regular insulin using a "sliding scale" showed a non-significant reduction (from 2.2% to 0%, p=0.147). In the RCT, before the intervention, the median scores were 5.5 (5–6) in 10 the control group and 6 (5–7) in the intervention group (p = 0.383). After the intervention, a significant improvement was observed in the intervention group, with a median score of 8 (8–9), compared to 6 (5–8) in the control group (p = 0.005). In the intervention group, the median gain was 2.5 points (p < 0.001). Analysis of 149 post-intervention medical records revealed a non-significant increase in NPH Insulin prescription (from 33.3% to 51.5%, p=0.011). The intervention group had a longer median length of stay (19 days) compared to the control group (13 days) (p=0.009). Our study demonstrated that a structured medical educational intervention on hospital hyperglycemia, utilizing telemedicine and WhatsApp, was effective in promoting a significant and reproducible improvement in the knowledge of medical residents in university hospitals. Although not all clinical outcomes reached statistical significance, a possible improvement in care practices was observed, such as improved insulin prescription practices and the elimination of crucial errors. The variability in length of hospital stay among centers suggests a probable influence of the institutional context. Despite the inherent design limitations and logistical conditions, this study provides robust evidence that simple, scalable, and low-cost educational strategies can enhance the quality of care for patients with hyperglycemia and diabetes in the hospital setting, reinforcing continuing medical education as an essential tool for patient safety and excellence in care. Instituição: Universidade Catolica de Pelotas Tipo do documento: Tese

Pelo direito à vida: uma análise do neoconservadorismo na disputa pelo direito ao aborto

2025-07-28T00:00:00Z

Título: Pelo direito à vida: uma análise do neoconservadorismo na disputa pelo direito ao aborto Autor: Miranda, Raíssa Ferreira Primeiro orientador: Souza, Aknaton Toczek Abstract: This dissertation analyzes neoconservative political-legal discourses on abortion and their effects on the production of law in contemporary Brazil. The research is based on the understanding that the criminalization of voluntary pregnancy termination goes beyond the legal field, functioning as a power device that regulates bodies, subjectivities, and social norms, disproportionately affecting women — particularly Black, poor, and marginalized women. Drawing on Foucauldian genealogy, the investigation is structured around three main axes: the historical reconstruction of abortion regulation in Brazil; the analysis of Justice Rosa Weber’s opinion in the ADPF 442 case; and the mapping of neoconservative discourses disseminated on digital platforms, with a focus on Instagram. From Justice Weber’s vote — a legal and symbolic milestone in the defense of sexual and reproductive rights — the dissertation examines the constitutional arguments presented and the subsequent technopolitical reactions, showing how social media function both as tools for conservative mobilization and as arenas of dispute over the authority to speak the law. Through document analysis and digital ethnography, the research identifies how performative, affectively mobilizing, and moralizing discourses are used to legitimize exclusionary practices and hinder institutional progress. Ultimately, it is argued that the right to abortion in Brazil is contested not only in courts or in parliament, but also within digital spaces, where the boundaries between morality, politics, the body, and justice are constantly negotiated. Instituição: Universidade Catolica de Pelotas Tipo do documento: Dissertação

iHorus: um modelo para detecção de transtornos depressivos

2024-12-05T00:00:00Z

Título: iHorus: um modelo para detecção de transtornos depressivos Autor: Zoppo, Gabriel Härter Primeiro orientador: Nascimento, Chiara das Dores Abstract: Depressive disorders are among the leading causes of work absenteeism, suicides, and result in billions of dollars in annual costs. The literature presents various approaches for detecting Major Depressive Disorders (MDD), with most studies using textual data from users, where machine learning methods achieve moderate results in terms of F1-score. However, there is a scarcity of studies that utilize non-textual data. In this context, the development of a depres sion detection model called iHorus is proposed, which employs five machine learning tech niques: Logistic Regression, Support Vector Machine (SVM), Decision Tree, Random Forest, and Naive Bayes. The methodology of iHorus is structured into three main stages: preprocess ing, model construction, and result analysis. During preprocessing and modeling, a correlation analysis was performed between the dataset variables and four attributes related to depressive disorders, revealing that the strongest correlations involve other mental and mood disorders. After training and testing the models, Random Forest demonstrated the best performance, with SVM and Decision Tree achieving F1-scores close to 0.8. In contrast, Naive Bayes and Logistic Regression had the poorest performance, with results just above 0.7, which is consistent with the literature that also highlights Random Forest and SVM as the most effective methods, while Logistic Regression and Naive Bayes often show inferior results. Instituição: Universidade Catolica de Pelotas Tipo do documento: Dissertação

Perfil dos usuários do Centro de Atenção Psicossocial Infanto Juvenil (CAPSi) do Município do Rio Grande durante o ano de 2023

2024-12-05T00:00:00Z

Título: Perfil dos usuários do Centro de Atenção Psicossocial Infanto Juvenil (CAPSi) do Município do Rio Grande durante o ano de 2023 Autor: Soares, Silvana Pereira Primeiro orientador: Matos, Mariana Bonati de Abstract: Introduction: Psychosocial Care Centers (CAPS) were established as alternative services to the asylum model, aiming to propose a new psychosocial care approach. Among the CAPS modalities are the Child and Adolescent Psychosocial Care Centers (CAPSi), which aim to address mental health demands focused on children and adolescents. Objective: To identify the profile of CAPSi users in the municipality of Rio Grande, RS, during the year 2023. Method: This descriptive study aimed to identify the characteristics of CAPSi users through access to their physical medical records and online records when necessary. Statistical analyses were performed using the SPSS software, with results presented through univariate analysis, using absolute and relative frequencies. Results: Of the 251 CAPSi users admitted in 2023, the majority were male (55.0%), with a predominance in the age group of 9 to 14 years (38.6%). Additionally, 79.5% of the patients did not have a prior diagnosis, and among them, 30.3% exhibited aggressiveness as the most frequent symptom in their complaints. Other findings included Autism Spectrum Disorder (ASD) was identified as the most persistent subsequent diagnosis (64.5%), and there was a higher prevalence of users referred to CAPSi by health-related services (42.9%). Conclusion: The findings of this study underscore the importance of discussing issues related to child and adolescent mental health, assessing the adequacy of the services provided by CAPSi, and implementing improvements where necessary. Instituição: Universidade Catolica de Pelotas Tipo do documento: Dissertação