Mostrar registro simples

dc.contributor.advisorRodacki, André Luiz Félix, 1965-pt_BR
dc.contributor.otherVillarejo Mayor, John Jairo, 1981-pt_BR
dc.contributor.otherUniversidade Federal do Paraná. Setor de Ciências Biológicas. Programa de Pós-Graduação em Educação Físicapt_BR
dc.creatorWong, Benny Wai Chunpt_BR
dc.date.accessioned2024-04-24T14:21:07Z
dc.date.available2024-04-24T14:21:07Z
dc.date.issued2023pt_BR
dc.identifier.urihttps://hdl.handle.net/1884/87653
dc.descriptionOrientador: Prof. Dr. André Luiz Felix Rodackipt_BR
dc.descriptionCoorientador: Prof. Dr. John Jairo Villarejo Mayorpt_BR
dc.descriptionDissertação (mestrado) - Universidade Federal do Paraná, Setor de Ciências Biológicas, Programa de Pós-Graduação em Educação Física. Defesa : Curitiba, 19/12/2023pt_BR
dc.descriptionInclui referênciaspt_BR
dc.description.abstractResumo: As quedas representam o principal acidente entre idosos. Um terço dos idosos experimentam uma queda ao ano e sua prevalência aumenta com a idade. Diante dessa perspectiva, é importante identificar idosos com propensão a quedas. Nesse sentido, o objetivo do estudo foi verificar se o teste Time up and Go (TUG) e o conjuntos de variáveis extraídas desse teste conseguem discriminar o idoso com e sem histórico de queda. O presente estudo tem delineamento transversal e foi composto por 131 idosos da comunidade de Curitiba. A avaliação foi realizada em uma sessão. O histórico de quedas e as características antropométricas foram avaliadas por meio da anamnese; a função cognitiva foi avaliada pelo teste de MoCA; e foi realizado o TUG com sensor inercial na velocidade habitual e acima da usual (TUGVUS). Foram extraídos um conjunto de variáveis dos sinais de aceleração e velocidade angular das fases do TUG por meio de uma rotina específica desenvolvida no Matlab. A análise dos resultados foi feita por meio da estatística descritiva, média e desvio padrão. Foi realizada o teste T de amostras independentes para comparar a performance do TUG, suas subfases e suas variações entre idosos caidores e não caidores. Foi realizada a Análise de Componentes Principais (ACP) no conjunto de variáveis extraídas das fases do TUG e posteriormente o teste T de amostras independentes. Foi desenvolvido um modelo de regressão logística com as variáveis extraídas das fases do TUG e feito a análise da curva de característica de operação do receptor (ROC) e a área sob a curva (AUC) para detectar a acurácia do modelo em discriminar caidores e não caidores. O nível de significância adotado foi de p<0.05. Os resultados evidenciaram que não houve diferenças na performance do TUG, nas suas fases e suas variações entre idoso caidor e não caidor. Na análise da ACP, foi observado diferenças nos componentes principais da fase de levantar, fase da primeira caminhada, fase do segundo giro e fase de sentar do TUG entre os grupos. O modelo de regressão logística apresentou uma boa capacidade em discriminar não caidores e caidores (AUC = 0.821, sensibilidade = 42%, especificidade = 96%), enquanto a performance do TUG apresentou uma baixa capacidade preditiva (AUC=0.493). Com esses resultados, conclui-se que o teste de TUG, suas fases e suas variações apresentam limitações e não são recomendas como ferramenta para avaliar o risco de queda, enquanto o modelo de regressão baseado em variáveis de aceleração e velocidade angular das fases do TUG apresentou boa capacidade em discriminar idosos com e sem histórico de quedas.pt_BR
dc.description.abstractAbstract: Falls represent the main accident among older adults. One third of older adults experience a fall per year and its prevalence increases with age. Given this perspective, it is important to identify older adults who are prone to falls. In this sense, the objective of the study was to verify if the Time up and Go (TUG) test and the set of variables extracted from this test can discriminate between older adults with and without a history of falls. The present study has a cross-sectional design and was composed of 131 older adults from the community of Curitiba. The assessment was carried out in one session. History of falls and anthropometric characteristics were assessed through anamnesis; cognitive function was assessed using the MoCA test; and subsequently, the TUG was performed with an inertial sensor at the usual speed and above the usual speed (TUGVUS). A set of variables was extracted from the acceleration and angular velocity signals of the TUG phases through a specific routine developed in Matlab. Analysis of the results was carried out using descriptive statistics, mean and standard deviation. The T-test of independent samples was performed to compare the performance of the TUG, its subphases and its variations between fallers and non-fallers. Principal Component Analysis (PCA) was performed on the set of variables extracted from the TUG phases and subsequently the T test for independent samples. A logistic regression model was carried out with the variables extracted from the TUG phases and the analysis of the receiver operating characteristic curve (ROC) and the area under the curve (AUC) was used to detect the accuracy of the model in discriminating fallers and non-fallers. The level of significance adopted was p<0.05. The results showed that there were no differences in the performance of the TUG, in its phases and its variations between between fallers and non-fallers. In the PCA analysis, differences were observed in the main components of the standing phase, first walk phase, second turn phase and sitting phase of the TUG between the groups. The logistic regression model showed a good accuracy to discriminate non-fallers and fallers (AUC = 0.821, sensitivity = 42%, specificity = 96%), while the TUG performance showed a low predictive accuracy (AUC=0.493). With these results, it is concluded that the TUG test, its phases and variations have limitations and are not recommended as a sensitive tool to identify elderly people at risk of falling, while the model based on acceleration and angular velocity variables of the TUG phases are indicated to discriminate older adults with and without history of falls.pt_BR
dc.format.extent1 recurso online : PDF.pt_BR
dc.format.mimetypeapplication/pdfpt_BR
dc.languagePortuguêspt_BR
dc.subjectEnvelhecimentopt_BR
dc.subjectQuedas (Acidentes) em idosospt_BR
dc.subjectEducação Físicapt_BR
dc.titleAssociação entre o histórico de quedas em idosos através do teste time up and gopt_BR
dc.typeDissertação Digitalpt_BR


Arquivos deste item

Thumbnail

Este item aparece na(s) seguinte(s) coleção(s)

Mostrar registro simples