| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.33 MB | Adobe PDF |
Advisor(s)
Abstract(s)
Introdução: A avaliação do risco de viés é uma etapa fundamental na pesquisa científica, especialmente em revisões sistemáticas e meta-análises, devido ao seu impacto significativo na interpretação e confiabilidade dos resultados.
Partindo de uma revisão de revisões sistemáticas publicada anteriormente, efetuou-se uma avaliação secundária dos métodos de análise do risco de viés.
Materiais e métodos: O presente estudo representa uma análise secundária de dados obtidos durante a execução de uma revisão sistemática de revisões sistemáticas. O critério de exclusão de ter obrigatoriamente de conter uma meta-análise tornou-se irrelevante.
Preparou-se uma tabela predefinida para extrair os dados necessários de cada revisão sistemática elegível, incluindo nela: a identificação do estudo (autores e ano), o número de estudos incluídos na meta-análise, o tipo e número de estudos incluídos, a(s) condição(ões) oral(ais) avaliada(s), a(s) condição(ões) sistémica(s) avaliada(s), a ferramenta de qualidade metodológica utilizada, o tamanho do efeito, o intervalo de confiança de 95% e a informação sobre financiamento.
Resultados: Num total de 726 revisões, os estudos observacionais predominaram. Os estudos experimentais foram menos frequentes que os estudos observacionais. O instrumento de risco de viés mais utilizado foi Newcastle–Ottawa scale. PRISMA foi a guideline mais referenciada nos artigos.
Conclusão: Verificou-se que alguns instrumentos de risco de viés utilizados não eram apropriados para o tipo de estudos que estavam a ser avaliados, comprometendo, assim, a confiabilidade dos resultados.
Aim: The assessment of the risk of bias is a crucial step in scientific research, particularly in systematic reviews and meta-analyses, due to its significant impact on the interpretation and reliability of results. Based on a previously published review of systematic reviews, we conducted a secondary evaluation of the methods used for risk of bias analysis. Material & Methods: The present study represents a secondary analysis of data collected during a systematic review of systematic reviews. The exclusion criterion requiring the presence of a meta-analysis became irrelevant. We prepared a predefined table in order to extract the necessary data from each eligible systematic review, including: study identification (authors and year), number of studies included in the meta-analysis, type and number of included studies, oral condition(s) evaluated, systemic condition(s) evaluated, methodological quality assessment tool used, effect size and 95% confidence interval, and funding information. Results: Of a total of 726 reviews, observational studies were predominated. Experimental studies were less frequent than observational studies. The most commonly used risk of bias instrument was the Newcastle–Ottawa scale. PRISMA was the most referenced guideline in the articles observed. Conclusion: It was found that some of the instruments used to measure the risk of bias were not appropriate for the type of studies being evaluated, which compromised the reliability of the results.
Aim: The assessment of the risk of bias is a crucial step in scientific research, particularly in systematic reviews and meta-analyses, due to its significant impact on the interpretation and reliability of results. Based on a previously published review of systematic reviews, we conducted a secondary evaluation of the methods used for risk of bias analysis. Material & Methods: The present study represents a secondary analysis of data collected during a systematic review of systematic reviews. The exclusion criterion requiring the presence of a meta-analysis became irrelevant. We prepared a predefined table in order to extract the necessary data from each eligible systematic review, including: study identification (authors and year), number of studies included in the meta-analysis, type and number of included studies, oral condition(s) evaluated, systemic condition(s) evaluated, methodological quality assessment tool used, effect size and 95% confidence interval, and funding information. Results: Of a total of 726 reviews, observational studies were predominated. Experimental studies were less frequent than observational studies. The most commonly used risk of bias instrument was the Newcastle–Ottawa scale. PRISMA was the most referenced guideline in the articles observed. Conclusion: It was found that some of the instruments used to measure the risk of bias were not appropriate for the type of studies being evaluated, which compromised the reliability of the results.
Description
Dissertação para obtenção do grau de Mestre no Instituto Universitário Egas Moniz
Keywords
Saúde oral Medicina baseada na evidência Revisão sistemática Risco de viés
