Exploring the FDA’s Prospective Routine Observational Monitoring Program Tools (PROMPT) for Safety Surveillance

Exploring the FDA’s Prospective Routine Observational Monitoring Program Tools (PROMPT) for Safety Surveillance

In 2007, partly due to several high profile drug safety issues, the federal government imposed the FDA Amendments Act (FDAAA). The FDAAA mandated that the FDA augment existing US drug safety systems by using large repositories of patient healthcare data to proactively monitor potential issues in marketed pharmaceutical products.

Subsequently, the Mini-Sentinel Pilot Project was launched in 2009 as a part of the FDA’s Sentinel Initiative. The Mini-Sentinel Project was formed to develop the active surveillance system as described by the FDAAA above. The project currently includes more than 15 distributed data partners, encompassing healthcare data for more than 125 million patients.

In the past four years, the Mini Sentinel project focused on two main types of distributed data analyses in support of active surveillance: protocol based assessments (traditional epidemiology) and Modular Programs providing rapid, standardized analyses of drug exposure characteristics in near real time. Recently, however, members of the Mini Sentinel Project Methods and Protocol Core working groups presented a new area of analytic focus -- semi-automated, prospective routine surveillance of specific products/events of interest.  The Prospective Routine Observational Monitoring Program Tools (PROMPT) project was introduced during a Brookings Roundtable Webinar by Jennifer Nelson (Washington University) and Elizabeth Chrischilles (University of Iowa).

During the webinar, several goals were outlined for PROMPT with the overall aim to identify potential risk in newly marketed products, allowing for near real time follow-up. 

A key feature of the PROMPT project is its standardized, modular approach to drug safety monitoring in a distributed database system, enabling rapid, efficient surveillance to be performed almost immediately.  This includes standardization of several key components of the surveillance activity, including definitions for health outcomes, the selection of analysis cohorts and confounders, and a choice of alternative analytic modules based on methodological design.  

Health Outcome Definitions:  Previous Mini Sentinel research focused on the identification and validation of health outcome definitions.   The published results include recommended definitions for multiple outcomes, including Pancreatitis, Venous Thromboembolism, Cardiac Arrhythmias, Seizures, and many others  

Analysis Cohort Selection:  PROMPT cohort selection allows analysis populations to be selected based on parameters that specify characteristics such as baseline history, indication, risk factors, and subgroups. Standard definitions for common cohorts will be created and saved. 

Modular Analysis Approach:  The PROMPT approach is to develop parameter driven, pre-programmed modules that can be activated quickly across the distributed data network.  The characteristics inherent to a specific drug/outcome pair can affect the methodological design and analytic choice; therefore, multiple modules will represent different methodological designs.  Initial module designs include  Self Controlled, Exposure matching by Propensity Score (cohort design), Regression (cohort design), IPTW Regression (cohort design)

The Mini-Sentinel project and PROMPT have crafted comprehensive plans to the FDA’s mandate that aim to make strides in increasing patient safety.

UBC’s SÆfetyworks® software, a unique offering within the pharmaceutical industry, implements a standardized and modular approach to active drug safety surveillance that is very similar to the FDA’s.  SAEfetyWorks enables our clients to perform the same type of proactive monitoring and signal refinement activities that are currently ongoing within FDA and other regulatory agencies on large healthcare databases. As a thought leader in this area, UBC has been actively involved in research and development of standardized analytics and have been committed to maintaining alignment of our software and service offerings with the rapidly changing landscape of Active Surveillance.