5 Ways AI & Automation in Pharmacovigilance are Driving Big Results for the Pharma Industry
Life Sciences | 6 min READ
    
Last year, healthcare providers, pharmaceutical professionals, and virologists were grappling to understand the nature of the novel coronavirus and its mutants during the pandemic outbreak. Back then, there was no known specified cure for the SARS-CoV-2 was known. Governments were fighting to procure a steady stockpile of hydroxychloroquine. And vaccine trials were far from the stage of emergency regulatory approval, automation in pharmacovigilance (PV) came to the rescue.
Sreenivas Reddy
Sreenivas Reddy

Global Program Director

Life Sciences

Birlasoft

 
For lack of a better-known alternative, over-the-counter antiviral drugs, like azithromycin, were being used to treat the symptoms of the coronavirus disease. The sales of the antibiotic medication shot through the roof. This chaos was concerning as the efficacy of the drug to treat COVID was far from known. It is then that a centralized, web-based, and automated portal of randomized evaluations enlightened the medical fraternity that the addition of azithromycin to the standard of care does not improve clinical outcomes of patients with severe COVID-19. This clinical trial in Brazil set a precedent for the COVID-19 standard of care globally.
What is AI & Automation in Pharmacovigilance
Automation in pharmacovigilance entails using cognitive technologies such as machine learning (ML) and advanced analytics to transform legacy data compilation processes and information gathering for regulatory approval. The use of new-age technology in PV is modeled around improving practices of a drug’s risk-benefit profile assessment, methods sought to select optimal treatments, and increase overall patient safety through product quality.
Stay Ahead
Visit our Pharmacovigilance page
Benefits of Automation in Pharmacovigilance
Improved Employee Productivity
While the business case for automation in pharmacovigilance primarily centers around gaining process efficiency, a more significant benefit is the freeing of resources to allocate the same for performing value-added tasks. Monitoring and evaluating adverse drug reactions, real-world evidence analysis, and signal investigation forms the basis of functions executed by resources freed up due to automation in pharmacovigilance. An additional consequence of automating drug safety programs is improved quality assurance, accuracy and consistency in testing cycles, and reduced PV costs. A 2020 McKinsey study reports the potential of automating as much as 70% of PV-related tasks, the business value for which can be realized within three years of implementation.
By enabling robotic intelligence for case automation, one can speed up overall case processes and reduce the administrative burden on employees due to repetitive activities like duplicate searches of cases and reporting. In addition, the Robotic Process Automation technology offers numerous privacy and security benefits by eliminating the chance of human error or misuse. For example, RPA can be deployed to archive sensitive clinical cases of ICSR detection without ever necessitating human involvement.
Case Intake in Pharmacovigilance
Most pharmaceuticals and medical devices organizations spend an overwhelming part of their PV budgets on case processing. Further, to make things complicated, case volumes grow at an incremental rate year on year. Naturally, driving cost out of case processing is often the primary goal for leaders in the pharmaceutical industry. By automation, they can take advantage of scale and generate cost savings per individual case safety report (ICSR). Productivity drivers to this end include native automation and “bolt-on” technologies that reduce the manual effort required to carry out duplicate checks, speed up coding functions, and streamline case writing. Automation also helps in generating proofs of concept across the entire PV value chain.
5 Benefits of AI & Automation in Pharmacovigilance
While optimizing functions at the level of ICSR can help pharma companies deal with overall case intake, the aspect of ensuring regulatory requirements and enhancing patient safety is entirely dependent on their ability to automate more and more of their PV activities.
Improvements in Drug’s Risk-Benefit Profile
Legacy methods of assessing a drug’s risk-benefit profile entail the use of signal detection. Most pharmaceutical companies still use traditional investigative models to track individual reports of adverse events, mine databases, and opportunities for intervention in clinical trials. While some findings may indicate a better safety profile or therapeutic benefit of a drug, others may reveal side effects, making the drug unfavorable.
Considering the lack of time to react to adverse events, leaders in the pharma industry are now migrating towards predictive signaling. This calls for short-term investments in dashboards for real-time visualization of clinical trials and a long-term focus on data integration technologies. Automation essentially improves the quality and consistency of data generated during a clinical trial. Safety information can be used far more effectively with reduced gaps in the discovery phase of a drug’s risk-benefit profile.
Faster Hypothesis to Testing Cycles
Automation in PV testing cycles brings forth a significant opportunity to formulate faster hypotheses. A 2018 pilot study conducted by the American Society for Clinical Pharmacology and Therapeutics (ASCPT) takes cognizance of the same. The said study tested the feasibility of using Robotic Process Automation (RPA) and Artificial Intelligence (AI), among other new-age technologies, to automate the processing of adverse event reports. Proposed solutions by three commercial vendors who participated in this pilot study were simultaneously tallied against Pfizer.
The outcomes confirmed that AI-enabled automotive tools expedite the deduction of adverse events from source documents. Further advancement of technology in this domain, including leveraging Natural Language Processing (NLP) technology, Big Data Analytics, and data extracted from social media, would pave the way for complete migration of manually-intensive methods of source document annotation, which would effectively be replaced by total reliance on data fields of safety databases. While the benefits of such an overhaul can only be studied in the long term, what has been accepted universally by pharmacovigilance professionals is that the ability to differentiate between vendor capabilities and identify the best candidate in a testing cycle for the discovery phase analysis is an assured benefit of process automation.
Cost Savings
As a natural result to improved productivity in pharmacovigilance case intake and optimizing resources allocated in the process flow, pharma companies can save up majorly on costs. To put this into perspective, let us take a look at Birlasoft’s very own collaboration with a pharmaceutical major. The client, which has a support staff of 5000 people working round the clock, scanning 400,000 literature abstracts annually to identify potential safety signals, was faced with process inefficiency. Only 50% of the documents were worthy of scanning, and upon analysis, a mere 5-8% reportable adverse events would be cited on an average. Their legacy methodologies required a complete overhaul. With the right automation tools, the pharmaceutical company could reduce its compliance cost and handle a 25% increase in reviewable data volume with its existing set of resources.
Benefits of Automation in Pharmacovigilance
Benefits of Automation in Pharmacovigilance

Click to zoom in

Cognitive AI & Automation technologies of automation and advanced analytics are providing immense opportunities to transform the PV ecosystem. Since all stakeholders within the life cycle of drug development share responsibilities towards ensuring patient safety, the onus lies on all of them to set a motion in which the industry moves forward, both through transformational and incremental change. To that end, the pandemic has taught us the biggest lesson of all.
Since quarantine and social distancing norms have limited the scope of traditional face-to-face patient-physician care models, telehealth is gradually becoming accepted as the standard of care. While this is only a start, one must wait and watch for AI-enabled decision support to transform triaging, clinical care, and home monitoring. As far as one knows, the adoptive roadmap will only evolve and for the better.
 
 
Was this article helpful?