Publicação
Joint modelling of longitudinal data and time until premature termination in psychotherapy
| Resumo: | Joint modelling enables the simultaneous study of longitudinal and survival processes, exploiting the association between them. A particular case, adopted in the present work, is the shared random effects model, using a linear mixed effects model to represent the longitudinal process linked to a Cox regression model to represent the survival process. The primary purpose of this work is to briefly review the shared random-effect model methodology, along with independent survival and longitudinal models, and detail its implementation and evaluation, through a real data set. The focus is on the hazard of premature termination in psychotherapy, investigating the effect of two known process variables: therapeutic alliance quality and treatment outcome. Additionally, we aim tho infer which risk factors affect both the hazard of premature termination and these process variables. A data set of 97 clients, along with 12 variables, was collected from a university clinic, over a period of three years. These clients were assigned to the Unified Protocol for transdiagnostic treatment of emotional disorders. The benefits of joint modelling were highlighted through the comparison of joint models and separate survival and longitudinal methods. Results showed that, the therapeutic alliance quality and the treatment outcome mean progression were significantly associated with the hazard of premature termination for these clients. We conclude that independent analysis bring up bias parameter estimates, and an assumption of association between the two processes in a joint model of premature termination data is necessary. |
|---|---|
| Autores principais: | Ferreira, Ângela Cristina Franco |
| Assunto: | Joint modelling Premature termination in psychotherapy Therapeutic alliance Treatment outcome Abandono prematuro da psicoterapia Aliança terapêutica Modelação conjunta Resultados terapêuticos Ciências Naturais::Matemáticas |
| Ano: | 2020 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | Joint modelling enables the simultaneous study of longitudinal and survival processes, exploiting the association between them. A particular case, adopted in the present work, is the shared random effects model, using a linear mixed effects model to represent the longitudinal process linked to a Cox regression model to represent the survival process. The primary purpose of this work is to briefly review the shared random-effect model methodology, along with independent survival and longitudinal models, and detail its implementation and evaluation, through a real data set. The focus is on the hazard of premature termination in psychotherapy, investigating the effect of two known process variables: therapeutic alliance quality and treatment outcome. Additionally, we aim tho infer which risk factors affect both the hazard of premature termination and these process variables. A data set of 97 clients, along with 12 variables, was collected from a university clinic, over a period of three years. These clients were assigned to the Unified Protocol for transdiagnostic treatment of emotional disorders. The benefits of joint modelling were highlighted through the comparison of joint models and separate survival and longitudinal methods. Results showed that, the therapeutic alliance quality and the treatment outcome mean progression were significantly associated with the hazard of premature termination for these clients. We conclude that independent analysis bring up bias parameter estimates, and an assumption of association between the two processes in a joint model of premature termination data is necessary. |
|---|