Artificial Intelligence in Pharmacovigilance and COVID-19

Area of Science:

• Pharmacovigilance and Drug Safety
• Public Health Surveillance
• Computational Pharmacology

Background:

• The history of pharmacovigilance traces back 169 years, with a pivotal shift following the 1961 Thalidomide tragedy.
• Systematic, spontaneous, and regulated adverse drug reaction reporting became essential for public health safety.

Observation:

• Signal detection from reported cases is a primary goal of pharmacovigilance.
• The Spontaneous Reporting System, while widely used, suffers from under-reporting limitations.
• International collaboration through the Uppsala Monitoring Centre (established 1978) and its databases (VigiFlow, VigiBase) facilitates global drug monitoring.

Findings:

• Computational methods like Bayesian Framework and E-Synthesis, alongside social media, enable early drug safety surveillance.
• India's National Pharmacovigilance Program and amendments to Schedule Y (2005) highlight regulatory advancements.
• Artificial intelligence (AI), emerging in pharmacovigilance since 2017, significantly improves data quality and accuracy.

Implications:

• Integrating Information Technology and pharmaceutical companies can accelerate AI adoption in pharmacovigilance.
• Enhanced pharmacovigilance systems are crucial for ensuring the ongoing safety of marketed drugs.
• Proactive drug safety monitoring through advanced techniques protects public health effectively.