Opinions and practices vary around the issue of performing multiple statistical tests in randomised controlled trials (RCTs). We carried out a study to collate information about opinions and practices using a methodological rapid review and a survey, specifically of publicly funded pragmatic RCTs that are not seeking marketing authorisation. The aim was to identify the circumstances under which researchers would make a statistical adjustment for multiplicity.
A review was performed extracting information from articles reporting primary analyses of pragmatic RCTs in one of seven high quality medical journals, in January to June (inclusive) 2018. A survey (Survey Monkey) eliciting opinions and practices around multiplicity was distributed to the 47 registered clinical trials units (CTUs) in the UK.
One hundred and thirty-eight RCTs were included in the review, and survey responses were received from 27/47 (57%) CTUs. Both the review and survey indicated that adjusting for multiplicity was considered most important for multiple treatment comparisons; adjustment was performed for 11/23 (48%) published trials, and 24/27 (89%) CTU statisticians reported they would consider adjustment. Opinions and practices varied around adjustment for multiplicity arising from multiple primary outcomes and interim analyses. Adjustment was considered less important for multiplicity due to multiple secondary outcomes (adjustment performed for 17/136 [13%] published trials and 3/27 [11%] CTU statisticians would consider adjustment) and subgroup analyses (8/85 [9%] published trials adjusted and 6/27 CTU [22%] statisticians would consider adjustment).
There is variation in opinions about adjustment for multiplicity among both statisticians reporting RCTs and applied statisticians working in CTUs. Further guidance is needed on the circumstances in which adjustment should be considered in relation to primary trial hypotheses, and if there are any situations in which adjustment would be recommended in the context of secondary analyses.

OBJECTIVE: Multi-arm non-inferiority (MANI) trials, here defined as non-inferiority trials with multiple experimental treatment arms, can be useful in situations where several viable treatments exist for a disease area or for testing different dose schedules. To maintain the statistical integrity of such trials, issues regarding both design and analysis must be considered, from both the multi-arm and the non-inferiority perspectives. Little guidance currently exists on exactly how these aspects should be addressed and it is the aim of this paper to provide recommendations to aid the design of future MANI trials.
METHODS: A comprehensive literature review covering four databases was conducted to identify publications associated with MANI trials. Literature was split into methodological and trial publications in order to investigate the required design and analysis considerations for MANI trials and whether they were being addressed in practice.
RESULTS: A number of issues were identified that if not properly addressed, could lead to issues with the FWER, power or bias. These ranged from the structuring of trial hypotheses at the design stage to the consideration of potential heterogeneous treatment variances at the analysis stage. One key issue of interest was adjustment for multiple testing at the analysis stage. There was little consensus concerning whether more powerful p value adjustment methods were preferred to approximate adjusted CIs when presenting and interpreting the results of MANI trials. We found 65 examples of previous MANI trials, of which 31 adjusted for multiple testing out of the 39 that were adjudged to require it. Trials generally preferred to utilise simple, well-known methods for study design and analysis and while some awareness was shown concerning FWER inflation and choice of power, many trials seemed not to consider the issues and did not provide sufficient definition of their chosen design and analysis approaches.
CONCLUSIONS: While MANI trials to date have shown some awareness of the issues raised within this paper, very few have satisfied the criteria of the outlined recommendations. Going forward, trials should consider the recommendations in this paper and ensure they clearly define and reason their choices of trial design and analysis techniques.

While current guidelines generally recommend single endpoints for primary analyses of confirmatory clinical trials, it is recognized that certain settings require inference on multiple endpoints for comprehensive conclusions on treatment effects. Furthermore, combining treatment effect estimates from several outcome measures can increase the statistical power of tests. Such an efficient use of resources is of special relevance for trials in small populations. This paper reviews approaches based on a combination of test statistics or measurements across endpoints as well as multiple testing procedures that allow for confirmatory conclusions on individual endpoints. We especially focus on feasibility in trials with small sample sizes and do not solely rely on asymptotic considerations. A systematic literature search in the Scopus database, supplemented by a manual search, was performed to identify research papers on analysis methods for multiple endpoints with relevance to small populations. The identified methods were grouped into approaches that combine endpoints into a single measure to increase the power of statistical tests and methods to investigate differential treatment effects in several individual endpoints by multiple testing.

Reproducible results define the very core of scientific integrity in modern research. Yet, legitimate concerns have been raised about the reproducibility of research findings, with important implications for the advancement of science and for public support. With statistical practice increasingly becoming an essential component of research efforts across the sciences, this review article highlights the compelling role of statistics in ensuring that research findings in the animal sciences are reproducible-in other words, able to withstand close interrogation and independent validation. Statistics set a formal framework and a practical toolbox that, when properly implemented, can recover signal from noisy data. Yet, misconceptions and misuse of statistics are recognized as top contributing factors to the reproducibility crisis. In this article, we revisit foundational statistical concepts relevant to reproducible research in the context of the animal sciences, raise awareness on common statistical misuse undermining it, and outline recommendations for statistical practice. Specifically, we emphasize a keen understanding of the data generation process throughout the research endeavor, from thoughtful experimental design and randomization, through rigorous data analysis and inference, to careful wording in communicating research results to peer scientists and society in general. We provide a detailed discussion of core concepts in experimental design, including data architecture, experimental replication, and subsampling, and elaborate on practical implications for proper elicitation of the scope of reach of research findings. For data analysis, we emphasize proper implementation of mixed models, in terms of both distributional assumptions and specification of fixed and random effects to explicitly recognize multilevel data architecture. This is critical to ensure that experimental error for treatments of interest is properly recognized and inference is correctly calibrated. Inferential misinterpretations associated with use of P-values, both significant and not, are clarified, and problems associated with error inflation due to multiple comparisons and selective reporting are illustrated. Overall, we advocate for a responsible practice of statistics in the animal sciences, with an emphasis on continuing quantitative education and interdisciplinary collaboration between animal scientists and statisticians to maximize reproducibility of research findings.

OBJECTIVE: Multiple hypothesis testing (or multiple testing) refers to testing more than one hypothesis within a single analysis, and can inflate the type I error rate (false positives) within a study. The aim of this review was to quantify multiple testing in recent large clinical studies in the otolaryngology literature and to discuss strategies to address this potential problem.
METHODS: Original clinical research articles with >100 subjects published in 2012 in the four general otolaryngology journals with the highest Journal Citation Reports 5-year impact factors.
METHODS: Articles were reviewed to determine whether the authors tested greater than five hypotheses in at least one family of inferences. For the articles meeting this criterion for multiple testing, type I error rates were calculated, and statistical correction was applied to the reported results.
RESULTS: Of the 195 original clinical research articles reviewed, 72% met the criterion for multiple testing. Within these studies, there was a mean 41% chance of a type I error and, on average, 18% of significant results were likely to be false positives. After the Bonferroni correction was applied, only 57% of significant results reported within the articles remained significant.
CONCLUSIONS: Multiple testing is common in recent large clinical studies in otolaryngology and deserves closer attention from researchers, reviewers, and editors. Strategies for adjusting for multiple testing are discussed.