A carregar...
Publicação
Deteção de simulação : validação do Coin in hand-extended version com adultos idosos
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.03 MB | Adobe PDF | |||
| 134.63 KB | Adobe PDF |
Orientador(es)
Resumo(s)
O presente estudo teve como objetivo validar o Teste de Validade de Desempenho Coin
in Hand-Extended Version (CIH-EV; Daugherty et al., no prelo) com grupos de adultos idosos
saudáveis e com demência, segundo um paradigma de simulação. Um dos grupos, de idosos
saudáveis não-simuladores (n=30) foi instruído a ter o melhor desempenho possível, e outro
grupo, de idosos saudáveis simuladores (n=29) foi instruído a simular um défice mnésico típico
de uma demência, com o objetivo de obter algum tipo de pensão, subsídio ou reforma
antecipada. Além destes, incluiu-se um grupo clínico (n=17), com diagnóstico de demência.
Os resultados mostraram que os grupos não-simulador e clínico tiveram um desempenho mais
preciso no CIH-EV do que o grupo simulador, ainda que o grupo clínico tenha apresentado
tempos de resposta superiores. A precisão e o tempo de resposta não diferiram entre os três
níveis de dificuldade do teste nos três grupos. Através da análise das curvas ROC, determinou
se o ponto de corte (≤26) para adultos idosos saudáveis a partir da comparação entre os grupos
não-simulador e simulador, traduzindo 100% de sensibilidade e 97% de especificidade. A
comparação entre os grupos simulador e clínico sugeriu o ponto de corte ≤17 para idosos com
demência, representando 94% de sensibilidade e 97% de especificidade. O CIH-EV apresentou
boa validade convergente com as diferentes medidas de validade de desempenho utilizadas
(TOMM, Rey-15IMT e RDS) e mostrou-se insensível às variáveis sociodemográficas e ao
funcionamento neurocognitivo em todos os grupos. Além disso, apresentou uma reduzida taxa
de falsos positivos relativamente aos outros testes e, em geral, foi o instrumento com maior
precisão de diagnóstico nos três grupos. Assim, o presente estudo corrobora a utilidade do CIH
EV como um instrumento eficaz na deteção de simulação de défice cognitivo em adultos idosos
saudáveis e em idosos com demência.
The aim of the current study was to validate the Performance Validity Test Coin in Hand-Extended Version (CIH-EV; Daugherty et al., in press) in groups of healthy older adults and older adults with dementia, according to a simulation paradigm. One of the groups, the non-simulating healthy older adults (n=30) was instructed to have their best performance, and the other group, the simulating older adults (n=29) was instructed to simulate a typical dementia memory deficit, with the purpose of obtaining allowance, financial aid or early retirement. Furthermore, a clinical group (n=17), diagnosed with dementia, was included. The results showed that the non-simulator and clinical groups have performed more accurately than the simulator group, although the clinic group have shown superior response times. There were no significant differences in accuracy and response time between the three test difficulty levels, in all groups. Through ROC curves analysis, the cut-off point (≤26) for healthy older adults was determined from the comparison between non-simulator and simulator groups, showing 100% of sensibility and 97% of specificity. The comparison between the simulator and clinical groups suggested a cut-off point ≤17 for older adults with dementia, showing 94% of sensibility and 97% of specificity. The CIH-EV demonstrated good convergent validity with all performance validity measures used (TOMM, Rey-15IMT and RDS) and was insensitive to sociodemographic variables and neurocognitive functioning in all groups. Besides, it showed reduced rate of false positives when compared to the other tests and, generally, it was the most accurate diagnostic test in the three groups. Thus, the present study corroborates the CIH-EV utility as an effective instrument in detecting the simulation of cognitive deficits in healthy older adults and older adults with dementia.
The aim of the current study was to validate the Performance Validity Test Coin in Hand-Extended Version (CIH-EV; Daugherty et al., in press) in groups of healthy older adults and older adults with dementia, according to a simulation paradigm. One of the groups, the non-simulating healthy older adults (n=30) was instructed to have their best performance, and the other group, the simulating older adults (n=29) was instructed to simulate a typical dementia memory deficit, with the purpose of obtaining allowance, financial aid or early retirement. Furthermore, a clinical group (n=17), diagnosed with dementia, was included. The results showed that the non-simulator and clinical groups have performed more accurately than the simulator group, although the clinic group have shown superior response times. There were no significant differences in accuracy and response time between the three test difficulty levels, in all groups. Through ROC curves analysis, the cut-off point (≤26) for healthy older adults was determined from the comparison between non-simulator and simulator groups, showing 100% of sensibility and 97% of specificity. The comparison between the simulator and clinical groups suggested a cut-off point ≤17 for older adults with dementia, showing 94% of sensibility and 97% of specificity. The CIH-EV demonstrated good convergent validity with all performance validity measures used (TOMM, Rey-15IMT and RDS) and was insensitive to sociodemographic variables and neurocognitive functioning in all groups. Besides, it showed reduced rate of false positives when compared to the other tests and, generally, it was the most accurate diagnostic test in the three groups. Thus, the present study corroborates the CIH-EV utility as an effective instrument in detecting the simulation of cognitive deficits in healthy older adults and older adults with dementia.
Descrição
Tese de mestrado, Psicologia (Área de Especialização em Psicologia Clínica e da Saúde - Psicologia da Saúde e da Doença), Universidade de Lisboa, Faculdade de Psicologia, 2019
Palavras-chave
Validação de testes Simulação Idosos Demência Teses de mestrado - 2019