Platt R
Channing Laboratory, Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts.
Rev Infect Dis. 1991 Sep-Oct;13 Suppl 10:S810-4. doi: 10.1093/clinids/13.supplement_10.s810.
Many issues affect the design, conduct, and analysis of clinical evaluations of perioperative antibiotic prophylaxis. The most important of these are the selection of appropriate procedures for evaluation, the choice of meaningful outcomes, the comparison of outcomes among groups of patients who are really comparable to one another, the unbiased assessment of outcomes, the assessment of a number of patients sufficient to provide adequate statistical power for the detection of meaningful differences, and the generalizing of results. In evaluating perioperative prophylaxis, investigators often strive to identify small absolute differences among the risks of infections with different prophylactic regimens. For many specific procedures and drug regimens, such evaluations are limited by the difficulty of studying a sufficient number of individuals. Although randomized clinical trials are the standard for clinical evaluation, observational data may also prove useful under some circumstances.
许多问题会影响围手术期抗生素预防的临床评估的设计、实施和分析。其中最重要的是选择合适的评估程序、选择有意义的结局指标、在真正可比的患者组之间比较结局、对结局进行无偏倚评估、评估足够数量的患者以提供足够的统计效力来检测有意义的差异，以及结果的推广。在评估围手术期预防措施时，研究人员常常努力确定不同预防方案在感染风险方面的微小绝对差异。对于许多特定的手术和药物方案，此类评估因难以研究足够数量的个体而受到限制。尽管随机临床试验是临床评估的标准，但在某些情况下观察性数据可能也会有用。
