Pharmacovigilance: NLP Helps Drive Drug Safety
February 17, 2021, 2 Minute Read
The Pharmacovigilance (PV) function was borne from a famous tragedy. A bit more than 60 years ago, a drug called Thalidomide was marketed. Originally, this medicine was thought to cure sleeplessness and morning sickness in pregnant women. But as it turns out, this drug caused severe deformities in babies and also caused immune response suppression. It took more than 4 years to establish causality between the drug and adverse events. The FDA, EMA, and other agencies today require monitoring, expedited/aggregate reporting, and attention to suspected ADR’s. Ultimately, Pharmacovigilance today is laden with regulatory requirements, but when executed well, Pharmacovigilance saves patient lives.
What if I told you that modern technologies like social media listening, patient-centric web and mobile interfaces, big data analytics, and Natural Language Processing technologies can enlarge your view into your patient population, arm yourself with adjudication data and Phase III/IV clinical trial design parameters, and drive PV spontaneous reporting events? The long-range benefits could include: gain a closer relationship with your patient population, make stronger risk-benefit assessments, be more responsive to regulatory agencies, and improve healthcare provider communications.
Pharmaceutical companies generally conduct Pharmacovigilance reporting using four methods: 1) spontaneous reporting events, 2) intensified ADR reporting, 3) targeted reporting, and 4) cohort event monitoring. Methods to conduct reporting using methods 2, 3, and 4 are typically controlled and designed for high statistical power; these methods are often strong studies but are not early warning systems for Pharma companies. The opportunity lies in method 1 – broader monitoring through spontaneous reporting events.
Lymba wants to work with you to develop a process to help Pharma make use of natural language processing platforms. The goal will be to better quantify drug risks and benefits to a broader population of patients using Pharma’s IND or post-marketing drugs. The process will include centralizing knowledge base inputs for spontaneous reporting (such as social media commentary, questionnaire forms, longitudinal questionnaires, and more). The process will allow for patient-centric communications and engagement, and gives researchers broader access and crave-able information such as longitudinal data to adjudicate adverse event reporting is possible.
Talk to us today to begin a discussion about your Pharmacovigilance plans, and how Lymba can help extract critical information for your 3500A forms, longitudinal data collection, adverse event adjudication, and more.
Have any questions or want to keep the conversation going?