Abstract
Background/Aims To allow rapid response to FDA data inquiries, Mini-Sentinel Operations Center (MSOC) has developed seven modular programs (MP) for use within FDA’s Mini-Sentinel Pilot, a large, multi-institutional distributed database with 17 data partners (DP) designed to support the agency’s efforts to develop an active surveillance system for monitoring medical product safety in the United States.
Methods MPs are SAS programs designed to run against the Mini-Sentinel Common Data Model executing a set of common instructions for routine questions. MPs address questions commonly asked within the context of medical product safety assessments such as the number of prevalent and incident users of a medical product, the number of dispensings and days supply of specific drug products, the number of patients with a specific diagnosis or procedure, and the number of outcomes identified during exposure person-time. MP output includes standardized summary-level counts and are presented overall and stratified by user-defined age group, sex and calendar year. A set of input forms allow investigators to easily define parameters (e.g., lengths of drug washout period and look back period for pre-existing conditions, minimum treatment episode duration, allowable gap between dispensings). DPs are expected to respond to MP requests within five business days. Results are returned to MSOC via a secure portal and subsequently a summarized report is generated and returned to FDA requesters.
Results Between February and October 2011, MSOC has completed 15 MP queries in response to specific FDA questions for a total of 235 sets of results returned by participating DPs of which 78.3% were completed within the allotted 5 days. The mean response time across all queries was 3.8 days. Average turnaround time for MSOC to create a summarized report for FDA was 7.0 business days.
Discussion A key benefit of MP-based data request system is that simple summarized data are easily available to investigators. The process will be assessed and modified, as use of the system generates lessons and new approaches to improve efficiency. Planned future developments include addition of new data elements to MPs (e.g., lab values) and decision tree tool to match investigators’ question to the appropriate MP.
