Distributed, Collaborative Intelligent Agents for Proactive Post-Marketing Drug S

Wayne State University, Detroit MI

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Healthcare systems and insurers nationwide regularly make decisions regarding which drugs to include or exclude from their formularies based on evidence concerning benefits, risks, and costs of the medications. A major barrier to effective drug selection is the lack of sufficient published information on the safety of drugs, particularly new drugs. A computerized system operating at healthcare facilities that could provide continuous, active surveillance and timely identification of potential safety issues following the introduction of a new drug to a formulary is highly desirable. Such a system could lead to safer drug use policy, more cost-effective formulary decisions, better healthcare, and earlier detection of adverse drug reactions (ADRs). The implications of such technology for improving a national drug surveillance system will be apparent because ADRs can complicate the patient's medical condition, increase morbidity, and result in death (about 7,000 deaths per year in the U.S. were attributed to ADRs). At present, evidence on safety issues that would inform these decisions is generated primarily by the FDA's post-marketing surveillance system MedWatch(tm). MedWatch(tm) is a passive system that depends on voluntary, spontaneous reports. Because the system is limited by low reporting rates and the slow accumulation of sufficient events to enable a critical analysis, delays occur in the identification and withdrawal of problematic drugs from the market or labeling them with black box warnings. These delays have resulted in unnecessary mortality, morbidity, and costs of healthcare. We propose to develop an innovative team-based agent system, named ADRMonitor, for actively monitoring and detecting signal pairs implicating anticipated or potential ADRs at a healthcare facility. Each ADRMonitor user (e.g., physicians and drug safety officers) will have his/her own software agent that is accessible via the Internet and plays two roles -- assisting the user in his/her decision-making, and collaborating with agents of other team members. A key feature of the proposed approach is that the agents will continuously and autonomously collaborate with one another. They anticipate information needs of their teammates and share information proactively so that the users can be alerted timely about signal pairs. To demonstrate the feasibility, we plan to develop a prototype of ADRMonitor in this two-year pilot project, which will be undertaken collaboratively by our multidisciplinary team. Our preliminary design and analysis show the proposed methodology to be promising. The proposed effort represents a critical first step toward a subsequent development of a more comprehensive ADRMonitor in later phases of this research endeavor that would use the signal pairs to detect ADRs and expand the resultant system to cover healthcare in a region or across the country. The proposed methodology is general in nature and can be adapted for other important applications such as bioterrorism surveillance. PUBLIC HEALTH RELEVANCE: A computerized system operating at healthcare facilities that could provide continuous, active surveillance and timely identification of potential safety issues following the introduction of a new drug to a formulary is highly desirable. Such a system could lead to safer drug use policy, more cost-effective formulary decisions, better healthcare, and earlier detection of adverse drug reactions (ADRs). The implications of such technology for improving a national drug surveillance system will be apparent because ADRs can complicate the patient's medical condition, increase morbidity, and result in death (about 7,000 deaths per year in the U.S. were attributed to ADRs).