Ignoring the group in group-level HIV/AIDS intervention trials: a review of reported design and analytic methods.

摘要: Objectives: Studies evaluating the efficacy of HIV/AIDS interventions often involve random assignment groups participants or treatment in groups. These studies require analytic methods that take within-group correlation into account. We reviewed published to determine extent which was dealt with properly. Design: group-randomized trials (GRTs) and individually randomized group (IRGT) general public health journals 2005-2009. Methods: At least two authors each article, recording descriptive characteristics, sample size estimation methods, judgments about whether took intraclass account Results: Of those articles including sufficient information judge were correct, only 24% used appropriate for dealing correlation. The percentages differed substantially GRTs (41.7%) IRGT (8.0%). Most (69.2%) also made no mention a priori estimation. Conclusion: A majority our review reported analyses ignoring This practice may result underestimated variance, inappropriately small P values, incorrect conclusions effectiveness interventions. Previous analyzed incorrectly need be re-analyzed, future should designed methods. Also, journal reviewers editors aware special requirements design analysis quality reporting on such according standards.