Methods Searchstrategy and selection criteria We undertook a literaturesearch and review process according to a protocol designed before data collection. We aimed to identify studies on theepidemiology of health-care-associatedinfection in developing countries, with a particular focus on the most frequent bacterial infections—urinary-tract infection,surgical-site infection, bloodstream infection, hospital-acquired pneumonia, and ventilator-associated pneumonia. We searchedMedline for reports published betweenJanuary, 1995, and December,2008, with no language restriction. We used a comprehensive list of terms (panel 1),including MeSH terms “crossinfection”, “infection control”, and “developingcountries”, together with individual names of countries of low and middle income, according to the 2008 World Bank classification. We applied the same search strategy to the Cochrane database to identify anypublished reviews, and we searched WHO regional medical databases (panel 1). Furthermore, we consulted Embase but our search did not yield additional publications. 方法 检索策略和选择标准 根据收集资料之前制定的设计方案,我们着手进行了文献检索和审核。我们旨在对发展中国家医院感染的流行病学进行研究,特别关注最常见的细菌感染、泌尿道感染、手术部位感染、血液感染、医院获得性肺炎和机械通气相关肺炎。我们对1995年1月至2008年12月发表的医院感染相关报道通过Medline进行了检索,没有语言限制。我们使用一个全面的术语列表 (表1),包括MeSH 中的“交叉感染”,“感染控制”和“发展中国家”术语及根据2008年世界银行对国家经济收入分类中的中低收入国家的名称。我们对Cochrane数据库实施了同样的检索策略来鉴别任何已经发表的评论,我们也检索了WHO区域医疗数据库(表1)。此外,我们还咨询了Embase,但我们没检索其他的出版物。 |