Original Research Statistical Methods for Clinical Study Site Selection Jianjin Xu, PhD 1 , Lan Huang, PhD 1 , Zhihao Yao, MBA 1 , Zhiheng Xu, PhD 1 , Jyoti Zalkikar, PhD 1 , and Ram Tiwari, PhD 1 Abstract Background: The US Food and Drug Administration conducts on-site inspections and data audits through Bioresearch Monitoring program for assurance of the quality and integrity of data in the pre- and postapproval processes. It is important to inspect the study sites that are different compared with other sites in clinical studies and identify the problems related to those sites. Usually one cannot inspect all the sites in a clinical study because of limited resources, and statistical tools are needed to help in selecting sites for inspection. Methods: We propose two technical approaches, namely Fisher combination approach and likelihood ratio test (LRT) approach, for site selection, with each approach integrating the information obtained from a P value matrix. The proposed approaches produce site rankings, and the sites with highest rankings may be selected for inspection. Results: The application of the approaches is demonstrated through a hypothetical data set reflecting the pattern of the real data in a premarket approval submission for a diagnostic device. The proposed methods are shown, through extensive simulations, to control false discovery rate, while maintaining good sensitivity. Conclusion: The proposed approaches will be useful for site selection process. However, limitations exist when only using the statistical approaches proposed here. In practice, investigators will select the site for inspection by considering the outputs from the statistical approaches along with other important factors. Future research topic is discussed to facilitate practical application of the approaches. Keywords likelihood ratio test, Fisher combination, clinical study, signal detection, site ranking Introduction The US Food and Drug Administration (FDA) conducts on-site inspections and data audits through a Bioresearch Monitoring (BIMO) 1 program to ensure the quality and integrity of data submitted to the agency in the pre- and postapproval processes. The BIMO program was established in 1977 by a task force that included representatives from the drug, biologics, medical device, veterinary medicine, and food areas. It provides regu- latory oversight of clinical investigators, nonclinical labora- tories (animal studies), sponsors, monitors, contract research organizations, and institutional review boards in support of the pre- and postmarket review program. It not only provides pro- tection of the rights and welfare of the thousands of human subjects and animals involved in FDA-regulated studies but also has become a cornerstone in support of new product approval and market applications. The complexity of clinical studies across multiple sites poses challenges to both the companies and FDA for selecting sites for inspection. For example, sponsors/inves- tigators conducting clinical studies in multiple sites face significant management challenges to maintain data quality and integrity, while obtaining and retaining an adequate sample within targeted populations. 2 Since not all the clinical sites and investigators have the same training and experience in coordinating clinical studies across many study sites, the variability in trial management such as protocol implementation, data collection, and data report- ing could pose great challenges to the companies and investigators. The patient-level variability across sites can also present a challenge to determine which sites should be considered for inspection. Usually, investigators select sites for inspection by manually looking at factors of interest, such as the efficacy and safety information, site inspection history, patient characteristics, and site enrollment. For instance, an extremely high success rate of 1 Division of Biostatistics, Office of Surveillance and Biometrics, Center for Devices and Radiological Health, Food and Drug Administration, Silver Spring, MD 20993, USA Submitted 24-May-2018; accepted 29-Oct-2018 Corresponding Author: Jianjin Xu, PhD, Division of Biostatistics, Office of Surveillance and Biometrics, Center for Devices and Radiological Health, Food and Drug Administration, 10903, New Hampshire Avenue, Building 66, Room 2268, Silver Spring, MD 20993, USA. Email: Jianjin.Xu@fda.hhs.gov Therapeutic Innovation & Regulatory Science 1-9 ª The Author(s) 2019 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/2168479018814474 tirs.sagepub.com