Proceedings of the University of Pennsylvania eighth annual conference on statistical issues in clinical trials: Statistical issues in pragmatic clinical trials.

This issue of Clinical Trials contains the proceedings of the eighth annual symposium on statistical issues in clinical trials held at the University of Pennsylvania in April 2015. This symposium considered special statistical issues in the design, conduct, and analysis of ‘‘pragmatic’’ clinical trials. Interest in pragmatic trials is rapidly increasing in recognition of the fact that many important public health questions cannot be feasibly addressed using more traditional, highly resourceintensive approaches. Although such trials relate back to the ‘‘large, simple trials’’ conducted in the 1980s, there are many aspects of pragmatic trials that are far from simple. For example, cluster-randomized designs may be optimal to address management and behavioral intervention questions, but raise important and difficult statistical and ethical issues. Large pragmatic trials conducted in community settings pose special challenges regarding retention of subjects and handling of missing data. Pragmatic trials may be an appealing approach to address safety concerns because such trials often need to be quite large, but regulatory considerations may limit the ability to move toward the ‘‘pragmatic’’ end of the trial spectrum. Conference speakers addressed a variety of issues related to pragmatic trials, including derivation of trial data from electronic health records, use of ‘‘platform’’ trials embedded in health care systems, novel approaches to informed consent, and issues in cluster-randomized trials. As in prior conferences, panel discussants and members of the audience offered additional interesting perspectives on the issues. All papers presented at the conference are included in this issue, as well as edited transcripts of the panel and general discussions (Table 1). The Penn series of 1-day conferences on statistical issues in clinical trials is intended to provide intense focus on an area of statistical methodology of continued and/or emerging interest related to the design, conduct, and analysis of clinical trials. Topics have ranged from ‘‘proof-of-concept’’ trials to comparative effectiveness studies; the 2016 conference addressed current issues and current thinking about adaptive clinical trials. All conference proceedings except the first have been published in Clinical Trials.