Fatores preditores de abandono de tratamento de tuberculose: uma metanálise

• Autor(a) principal: Brasil, Pedro Emmanuel Alvarenga Americano do
• Data de Publicação: 2006
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/4243
• topic: Tubeculosis; Meta-analysis; Meta-analysis [publication type]; Patient dropout; Tuberculose; Metanálise [tipo de publicação]; Matanálise; Desistência do paciente; CNPQ::CIENCIAS DA SAUDE::SAUDE COLETIVA::EPIDEMIOLOGIA

Tuberculosis (TB) has been as a world emergency by World Health Organization in 1993. Nowadays, the tools available for TB control are early diagnosis and efficient treatment. However, TB treatment default is a concerning problem in the world and may vary between 3% and 80%, So, the identification of factors that may predict TB treatment default may help to develop better strategies to control TB. The objective of this work is, through a meta-analysis, to pool a summary estimate of associations of each factor (a) related to the health services, (b) to clinical condition and therapy and (c) to the individuals and TB treatment default. The remote electronic search strategy to find relevant publications was specifically developed for the different databases considered important, wich are: LILACS and MedLine (Pubmed). Searches for cross reference and search at COCHRANE systematic reviews database were also made. Investigations were included if they were observations or experimental that studied risk factors or predictive factors of the outcomes of interest (TB treatment default) through comparing two or more groups and the data could be extracted. Two independent reviewers classified and extracted the data and the discordances were solved in later meetings. More than 190 full texts were apt to be combines. From these texts, 40 exposures were extracted and combined. From these, 19 were shown in this investigation. 13 of 19 shown exposures demonstrated associations and could be considered as predictors ( male gender, alcoholism, HIV infection/AIDS, illicit drug use, foreign nationality, illiteracy, retreatment, positive sputum smear, previous default, short course therapy, easy access to health unit and adherence training extrapulmonary TB) and six did not show associations (unemployment , adverse effects, drug resistance, need for hospitalization, delay in start treatment, long wait for attendance). However these associations must considered very in a conservative way due to high level heterogeneity found in all exposures. Only five exposures have partially explanation and one have total explanation to the heterogeneity. Publication bias was detected in only two of the 19 exposures studied.