Gandhi Monica, Ameli Niloufar, Bacchetti Peter, Sharp Gerald B, French Audrey L, Young Mary, Gange Stephen J, Anastos Kathryn, Holman Susan, Levine Alexandra, Greenblatt Ruth M
Department of Medicine, University of California, San Francisco 94122, USA.
AIDS. 2005 Nov 4;19(16):1885-96. doi: 10.1097/01.aids.0000189866.67182.f7.
Applicability of randomized controlled clinical trial (RCT) results to 'real world' situations is dependent on the comparability of trial participants to general patient populations. A full disclosure of criteria employed for trial enrollment is necessary for clinicians to assess generalizability. We sought to assess both the impact on generalizability and the disclosure rate of enrollment criteria for 32 major HIV RCTs in the AIDS Clinical Trial Group (ACTG) and Community Programs for Clinical Research on AIDS (CPCRA) trial networks.
Eligibility criteria were compared in complete protocols to criteria listed in publications from these 32 NIH-funded HIV RCTs. We then applied these criteria to the Women's Interagency HIV Study (WIHS), the largest cohort study of HIV-infected women in the US.
When applied to WIHS, eligibility criteria from protocols excluded 0-67.6% (median 42%) of WIHS participants (50.6% excluded from ACTG trials). Eligibility criteria in publications excluded 0-62% (median 19.6%) of WIHS (21.2% excluded from ACTG trials). The number of women in WIHS seemingly ineligible for trial participation per enrollment criteria listed in publications averaged only 60% of those actually excluded based on the protocols.
We found that HIV RCT eligibility criteria excluded a large proportion of a representative cohort of HIV-infected women from trial participation. Furthermore, trial publications are not fully reflective of protocols in terms of disclosing eligibility criteria. Standardization and full disclosure of trial methodology will allow clinicians and researchers to more fully assess the generalizability of findings to their patient populations.
随机对照临床试验（RCT）结果在“现实世界”情况中的适用性取决于试验参与者与一般患者群体的可比性。临床医生要评估结果的可推广性，就必须全面了解试验入组所用的标准。我们试图评估艾滋病临床试验组（ACTG）和社区艾滋病临床研究项目（CPCRA）试验网络中32项主要HIV RCT的入组标准对可推广性的影响以及披露率。
将完整方案中的纳入标准与这32项由美国国立卫生研究院资助的HIV RCT发表文章中列出的标准进行比较。然后将这些标准应用于美国最大的HIV感染女性队列研究——女性机构间HIV研究（WIHS）。
应用于WIHS时，方案中的纳入标准排除了0 - 67.6%（中位数为42%）的WIHS参与者（ACTG试验排除了50.6%）。发表文章中的纳入标准排除了0 - 62%（中位数为19.6%）的WIHS参与者（ACTG试验排除了21.2%）。根据发表文章中列出的入组标准看似不符合试验参与条件的WIHS女性人数平均仅为根据方案实际排除人数的60%。
我们发现，HIV RCT的纳入标准将很大一部分具有代表性的HIV感染女性队列排除在试验参与之外。此外，在披露纳入标准方面，试验发表文章并未完全反映方案内容。试验方法的标准化和全面披露将使临床医生和研究人员能够更全面地评估研究结果对其患者群体的可推广性。
