| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2 MB | Adobe PDF |
Autores
Resumo(s)
A Meta-análise é uma técnica estatística que permite combinar e sintetizar os resultados de estudos realizados de forma independente, sobre uma mesma questão. É uma técnica aplicada nas mais diversas áreas mas, quando utilizada especificamente na comparação de medicamentos, fica limitada à comparação de pares de medicamentos, chamada comparação direta. Para fazer face a essa limitação, têm sido desenvolvidos métodos de comparação indireta de medicamentos. Estes métodos para além da comparação indireta permitem também a comparação simultânea de mais de dois medicamentos.
Neste trabalho, foi utilizado o método direto e um método indireto - o método MTC (Mixed treatment comparisons), que se baseia num modelo bayesiano hierárquico, de efeitos fixo ou feitos aleatórios - para comparar os medicamentos utilizados no tratamento da Doença Pulmonar Obstrutiva Crónica (DPOC). A DPOC é uma doença do aparelho respiratório, geralmente progressiva e de reduzida reversibilidade e caracteriza-se pela limitação de passagem do fluxo aéreo nos brônquios. Os medicamentos utilizados no seu tratamento têm por objectivo prevenir e controlar os sintomas, reduzir a frequência e a severidade das exacerbações.
Para a comparação dos tratamentos foi considerado, como tamanho do efeito, a razão das chances e, como evento, o número de doentes com pelo menos uma exacerbação durante o ensaio clínico.
Os métodos de comparação foram aplicados após a revisão sistemática da literatura que reuniu um conjunto de 17 ensaios clínicos com dois ou mais ramos de comparação e informação relativa a 10 medicamentos: Tiotrópio18, Formoterol12, Fluticasona500, Salmeterol42 e Salmeterol5, Fluticasona500+Salmeterol50, Indacaterol150, Indacaterol300, Indacaterol600 e Placebo. Apesar de algumas limitações dos dados, foram consideradas as comparações obtidas pelo método MTC que mostraram que o Tiotróipio 18 e o Indacaterol 150 reduzem a chance de ter exacerbação face aos restantes medicamentos.
Meta-analysis is a statistical technique that allows one to combine and synthesize the results of studies carried out independently on one same issue. This technique is applied in several areas but, when specifically used for medicine comparison, it is limited to the analysis of medicine pairs, called direct comparison. To address this limitation, methods have been developed for indirect comparison of medications. These methods allow the simultaneous analysis of more than two medicines, besides the indirect comparison. On the present work there were implemented both the direct method and an indirect one - the MCT method (mixed treatment comparisons), based on a hierarchic bayesian model, with fixed or random effects - to compare the medications used for the treatment of the Chronic obstructive pulmonary disease (COPD). The COPD is a respiratory disease, usually progressive and with a limited reversibility, and characterized by the constraint of the airflow in the bronchi. The medicines used in its treatment intend to prevent and control the symptoms and to reduce both the frequency and severity of exacerbations. To compare the treatments it was considered, as the effect size, the odds ratio and, as the event, the number of patients with at least one exacerbation during the clinical trial. The comparison methods were implemented after a systematic literature review that collected a set of 17 clinical trials with two or more branches of comparison and information on 10 drugs: Tiotropium18, Formoterol12, Fluticasone500, Salmeterol42 e Salmeterol5, Fluticasone500+Salmeterol50, Indacaterol150, Indacaterol300, Indacaterol600 e Placebo. Despite some data limitations, there were considered the comparisons obtained by the MTC method, showing that the Tiotroipium18 and the Indacaterol 150 reduce the odds of occurrence exacerbation, when compared with the other drugs.
Meta-analysis is a statistical technique that allows one to combine and synthesize the results of studies carried out independently on one same issue. This technique is applied in several areas but, when specifically used for medicine comparison, it is limited to the analysis of medicine pairs, called direct comparison. To address this limitation, methods have been developed for indirect comparison of medications. These methods allow the simultaneous analysis of more than two medicines, besides the indirect comparison. On the present work there were implemented both the direct method and an indirect one - the MCT method (mixed treatment comparisons), based on a hierarchic bayesian model, with fixed or random effects - to compare the medications used for the treatment of the Chronic obstructive pulmonary disease (COPD). The COPD is a respiratory disease, usually progressive and with a limited reversibility, and characterized by the constraint of the airflow in the bronchi. The medicines used in its treatment intend to prevent and control the symptoms and to reduce both the frequency and severity of exacerbations. To compare the treatments it was considered, as the effect size, the odds ratio and, as the event, the number of patients with at least one exacerbation during the clinical trial. The comparison methods were implemented after a systematic literature review that collected a set of 17 clinical trials with two or more branches of comparison and information on 10 drugs: Tiotropium18, Formoterol12, Fluticasone500, Salmeterol42 e Salmeterol5, Fluticasone500+Salmeterol50, Indacaterol150, Indacaterol300, Indacaterol600 e Placebo. Despite some data limitations, there were considered the comparisons obtained by the MTC method, showing that the Tiotroipium18 and the Indacaterol 150 reduce the odds of occurrence exacerbation, when compared with the other drugs.
Descrição
Tese de mestrado em Bioestatística, apresentada à Universidade de Lisboa, através da Faculdade de Ciências, 2011
Errata disponível em papel
Errata disponível em papel
Palavras-chave
Meta-análise Comparação direta Comparação indireta Mixed treatment Comparisons Teses de mestrado - 2011
