Harmonization of SEND Implementation to Enable Historical Control Data Analysis
The SENDIG allows for considerable flexibility in how data is represented; interpretation and application of these guidelines has led to significant variability in the approach to SEND dataset creation (Carfagna et al., 2020)). The scope of this project is to evaluate and recommend approaches for SEND harmonization to better enable analysis of historical control data. SEND harmonization strategies include creation of new variables, controlled terminology, preferred terms, reference lists and/or analysis strategies (scripts) to enable cross study analysis. Developing a solution framework for variability that includes representation of stakeholders involved in the creation and use of SEND datasets will enable more efficient routine analysis of warehoused SEND data. Harmonizing the implementation and use of SEND is expected to benefit all involved stakeholders and to ultimately contribute to the goal of increased productivity in nonclinical research.
Examples of areas identified as needing harmonization for historical control searching include standardized histopathology finding names (i.e., normal and unremarkable), vehicle name, identification of negative control animals, animal supplier name and site, and animal age. These data are all considered important to execute historical control data searches. Many existing structures and terminology in SEND are sufficient to perform searches with a high level of confidence. However, harmonization of some fields is needed to confidently extract historical control data. SEND harmonization is an important key to enabling high confidence when performing database searches. Harmonized data sets will provide information to improve the design and interpretation of toxicology studies (Mihalcik et al., 2016; Briggs, 2017).
The planned activities for this project are as follows. The first deliverable will be a prioritized list of SEND domains/variables that require harmonization for query of historical control data. Prioritization will be based on an evaluation of anticipated value, feasibility and effort needed to achieve harmonization. A workshop will be held at the 2020 CSS meeting to obtain prioritization input from the nonclinical topics group. The team will consider publication of white papers for domains/variables that are particularly complex and/or need additional input. Based on the prioritization, variables/domains will be selected for harmonization and recommendations will be created. Approximately 2-4 harmonization projects will be selected for this initial effort. The timing of the final recommendations will enable incorporation into the next version of the SEND IG. The recommendations will be included in a PHUSE publication or white paper, and presented at the annual CSS meeting.
|Kevin Snyder||Project Lead||FDAfirstname.lastname@example.org|
|Mark Carfagna||Project Lead||Eli Lilly/BioCelerateemail@example.com|
Objectives and Timelines
|Prioritized list of domains/variables required for harmonization||Q2 2020 – Q3 2020|
|Develop harmonization recommendations for selected domains/variables and create white paper(s) for specific domains/variables as needed||Q4 2020 – Q2 2021|
|Poster presentation and update of current working group progress at CSS||Q2 2021|
|PHUSE Publication or white paper||Q2 2021|
Vehicle Components Survey
Draft White Paper