Análise fatorial multipla e modelagem por equações estruturais: raiva e suas interfaces
Data
2023-08-22
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A raiva é uma zoonose negligenciada e letal, que acomete todos os mamíferos. Sua prevenção pode ser realizada por meio de protocolos de pré-exposição e de pós-exposição com o uso de imunizantes antirrábicos. O objetivo deste trabalho foi investigar as interações entre as variáveis observadas nas fichas de notificação de agravo do atendimento antirrábico do estado de São Paulo no período de 2013 a 2017 e a variável latente potencial de exposição à raiva do paciente que recebeu o tratamento com o uso de análise de Modelagem de Equações Estruturais. Por meio dos dados de 572.889 fichas de notificação do tratamento antirrábico, gerou-se o modelo estrutural, o qual apresentou os índices de ajuste de modelo aceitáveis, com SRMS=0,028, RMSEA=0,061 (IC95% 0,061-0,0616), CFI=0,970 e TLI=0,954. As variáveis antecedentes da variável latente potencial de exposição à raiva com maior coeficiente de regressão padronizado foram condição animal (β=0,627; p <0,00001) e observação do animal (β=-0,349; p<0,00001). Já entre as variáveis consequentes da variável latente, as maiores cargas fatoriais foram tratamento atual (β=0,936; p<0,00001), número de doses (β=0,774; p<0,00001), fim do animal (β=0,575; p<0,00001), e tratamento com uso de soro antirrábico (β=0,546; p<0,00001). Concluiu-se que o modelo estrutural utilizado é valido para a estimação das variáveis que envolvem a profilaxia da raiva humana, demonstrando que há relação entre o perfil do animal agressor com o potencial de exposição ao vírus da raiva, que por sua vez pode ser mensurada pelo tipo de tratamento indicado.
Rabies is a lethal and neglected viral disease that affects all mammals. The disease can be prevented through pre exposure prophylaxis and post exposure prophylaxis using rabies vaccines and immunoglobulins. Our objective was to investigate the interactions between the observed variables involved in aggravating anti rabies notification files in the estate of Sao Paulo from 2013 to 2017, and the potential latent variable of rabies prevention in patients who received treatme nt using SEM analysis. Using generated data collected from filled out files from 572,899 reported cases of notified anti rabies treatments, a structural model was created, with an adjustable ratio of an acceptable model, RMSEA=0.061 (IC95% 0.061 0.0616), C FI=0.970, and TLI=0.954. The variables of the potential latent variable of rabies prevention with the largest standardized regression coefficient were animal condition ( β=0.627; p <0.00001), and animal observation β 0.349; p<0.00001). Among the construct ’s measurable variables, the largest factorial charges were the current treatment β=0.936; p<0.00001), number of doses β=0.774; p<0.00001), animal outcome β=0.575; p<0,00001), and the treatment with rabies immunoglobulin ( β=0.546; p<0.00001). W e conclud ed that the structural model is valid for variabl e estimation involving human rabies prophylaxis, showing that there is a cor relation between the status of the attacking animal and rabies prevention, which can be measured by the employed treatment.
Rabies is a lethal and neglected viral disease that affects all mammals. The disease can be prevented through pre exposure prophylaxis and post exposure prophylaxis using rabies vaccines and immunoglobulins. Our objective was to investigate the interactions between the observed variables involved in aggravating anti rabies notification files in the estate of Sao Paulo from 2013 to 2017, and the potential latent variable of rabies prevention in patients who received treatme nt using SEM analysis. Using generated data collected from filled out files from 572,899 reported cases of notified anti rabies treatments, a structural model was created, with an adjustable ratio of an acceptable model, RMSEA=0.061 (IC95% 0.061 0.0616), C FI=0.970, and TLI=0.954. The variables of the potential latent variable of rabies prevention with the largest standardized regression coefficient were animal condition ( β=0.627; p <0.00001), and animal observation β 0.349; p<0.00001). Among the construct ’s measurable variables, the largest factorial charges were the current treatment β=0.936; p<0.00001), number of doses β=0.774; p<0.00001), animal outcome β=0.575; p<0,00001), and the treatment with rabies immunoglobulin ( β=0.546; p<0.00001). W e conclud ed that the structural model is valid for variabl e estimation involving human rabies prophylaxis, showing that there is a cor relation between the status of the attacking animal and rabies prevention, which can be measured by the employed treatment.
Descrição
Palavras-chave
Profilaxia Pós-Exposição, Vacina Antirrábica, Análise Multivariada, Post-exposure prophylaxis, Rabies vaccine, Multivariate Analysis