Fergusson Dean, Hebert Paul, Shapiro Stan
Department of Epidemiology and Biostatistics, McGill University, Montreal, Quebec, Canada.
Transfus Med Rev. 2002 Oct;16(4):296-303. doi: 10.1053/tmrv.2002.35211.
To ascertain the effectiveness of an intervention, a randomized controlled trial (RCT) is considered the gold standard. An obstacle to conducting an RCT only rarely discussed is the universal implementation of an intervention. Universal implementation clearly precludes the feasibility of conducting an RCT. Thus, the most attractive alternative study design in such instances becomes the before/after study. This article describes important methodologic considerations in undertaking a before/after evaluation. The methodologic considerations to be discussed are (1) threats to internal validity, (2) precision and (3) generalizability. Two before/after studies evaluating the potential effectiveness of universal leukoreduction serve as examples. Because of the universal application of many transfusion interventions, one has to consider carefully the methodologic rigor as to which of these interventions are evaluated. We have outlined the major methodologic issues one must consider when undertaking a before/after study design. When properly conceived, conducted, and analyzed, such a before/after study design can yield informative associations.
为确定一项干预措施的有效性,随机对照试验(RCT)被视为金标准。开展随机对照试验时一个很少被讨论的障碍是干预措施的普遍实施。普遍实施显然排除了开展随机对照试验的可行性。因此,在这种情况下最具吸引力的替代研究设计就成了前后对照研究。本文描述了进行前后对照评估时重要的方法学考量。将要讨论的方法学考量包括:(1)内部效度的威胁因素、(2)精确性和(3)可推广性。两项评估普遍白细胞滤除潜在有效性的前后对照研究作为示例。由于许多输血干预措施都已普遍应用,人们必须仔细考虑评估这些干预措施时的方法学严谨性。我们概述了在进行前后对照研究设计时必须考虑的主要方法学问题。如果构思、实施和分析得当,这样的前后对照研究设计能够得出有益的关联结果。