Pharmacovigilance and Digital Transformation: Growing Role of Technology

Digital Transformation and Industry 4.0 – these are the topics regularly featured in titles of Life Sciences events and publications recently. While we talk a lot about technologies that disrupt drug development, clinical trials, and pre-clinical stage processes, it is difficult to come across materials dedicated to the new technologies in pharmacovigilance.

A less noticeable discussion around digital transformation in pharmacovigilance can be explained by two main factors: the niche nature of the field and by the highly regulated nature of the field. However, now with the increased focus on the accelerated drug approvals and the massive amount of data generated on a daily basis, it is difficult to imagine that pharmacovigilance will remain in the periphery of digital transformation.

Already today, we can single out major technologies that are influencing pharmacovigilance strategy:

  • Big Data;
  • Natural Language Processing;
  • Cloud;

Big Data

In healthcare, Big Data refers to the vast and growing volumes of computerized medical information available in the form of electronic health records, administrative or health claims data, disease and drug monitoring registries. For years, all sorts of medical information were accumulated without the recognition of its value and potential usage. The development of the new powerful computer tools that can process and analyze big volumes of information has brought Big Data in focus – it can be now used for predictive purposes.

When it comes to pharmacovigilance, Big Data includes such sources as:

  • Signal detection;
  • Substantiation and validation of drug or vaccine safety signals;
  • Online channels and social media.[1]

Post-marketing safety analysis also uses informative systems for suspect adverse reactions spontaneous reporting. Here are just a few of them:

  • The Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS). Already in 2006, FAERS received almost half a million reports, reaching 1.2 million in 2014.
  • Vigibase, the World Health Organisation (WHO) informative system;
  • EudraVigilance by the European Agency for Medicines (EMA);

All these sources contribute to the mass of the data that companies have to sift through in order to improve the safety profile of their products.

Natural Language Processing

Natural language processing (NLP) is a “branch of artificial intelligence that helps computers understand, interpret and manipulate human language.”[2]

Life Sciences companies have to filter the high volumes of noise if they want to stay behind all the postmarketing safety signals that emerge across the Internet and digital resources. If they fail to keep up with it, they will miss up to 17% of adverse events.[3]

In pharmacovigilance NLP can help to identify events/outcomes/risk factors using the following sources:

  • Labels;
  • EMRs;
  • Manufacturers’ web sites;
  • Patients’ forums and other online communities;
  • E-mails;
  • Scientific literature;
  • Social media:
    • Twitter;
    • Public Facebook posts;
    • Blog articles and comments on platforms like WordPress;
    • YouTube;

Right now, the choice of NLP tools that allow filtering safety signals effectively is quite limited. The companies have two main options. First is purchasing a proprietary analytical tool like IBM Watson Analytics and second adapt existing tool, for example, a tool for social listening.

This technology still needs refining and development, especially when it comes to detecting adverse events on social media. Thus, the research done under WEBRADR project shows that broad-ranging statistical signal detection in Twitter and Facebook, using currently available methods for adverse event recognition, performs poorly and cannot be recommended at the expense of other pharmacovigilance activities.[4]

Cloud

Cloud is a technology that allows a company to benefit from the ability to store and analyze huge amounts of data. Among the main drivers for moving into the cloud are:

  • Cost and efficiency: cloud could allow companies to work with a big amount of data from cases without compromising quality, security and data privacy;
  • Scalability: the adverse event case workload for Life Science companies has been rising steadily, with some companies seeing a 50% increase yearly.[5] This rise calls for technology that can quickly accommodate the growing volume of data;
  • Simplicity: cloud usage can simplify the life of companies by allowing them to avoid concerns about module compatibility and scaling up servers.

Such trends as AI, wearables and patient centricity will eventually increase even further the amount and variety of data that has to be analyzed in order to take the most informed decisions about the benefit-risk profiles of drugs.[6] Thus cloud in pharmacovigilance is becoming more and more the reality. In fact, last year Oracle conducted a survey 60% respondents of which either has some or all of their safety solutions in the cloud or are planning to move there within the next two years. [7]

The main obstacles though still remain data privacy and safety and the cost of the implementation of the technology.

Picture 1: Deterrents to Leveraging the Cloud. Source: Oracle Health Sciences

Deterrents to Leveraging the Cloud

Conclusion

In the last years, the variety of safety data sources has grown immensely, allowing pharmacovigilance to develop from a single-source passive system into a complex, holistic, dynamical process. With this change, technology has become a key to compliance, safety and a way to stay behind all the new safety data sources.

Digital transformation in pharmacovigilance for sure has an impact on internal procedures of the department, creating a need to review and possibly restructure certain processes. seQure acts as a niche provider in Pharmacovigilance that helps companies to review internal workflows and ensures compliance of the new structure.

Do you need support in managing digital transformation in your company? Fill in our contact form and we will get back to you to answer your question.

References:

[1] Gianluca Trifirò, Janet Sultana, Andrew Bate: “From Big Data to Smart Data for Pharmacovigilance: The Role of Healthcare Databases and Other Emerging Sources”. Retrieved from: https://link.springer.com/article/10.1007%2Fs40264-017-0592-4

[2] SAS: “Natural Language Processing. What it is and why it matters”. Retrieved from: https://www.sas.com/en_us/insights/analytics/what-is-natural-language-processing-nlp.html 

[3]  Adam Sherlock, Christopher Rudolf: “Artificial Intelligence as an Aid to Pharmacovigilance”. Retrieved from:  http://www.pharmexec.com/artificial-intelligence-aid-pharmacovigilance

[4] Ola Caster, Juergen Dietrich, Marie-Laure Kürzinger: “Assessment of the Utility of Social Media for Broad-Ranging Statistical Signal Detection in Pharmacovigilance: Results from the WEB-RADR Project”. Retrieved from: https://link.springer.com/article/10.1007/s40264-018-0699-2

[5] Oracle Health Sciences: “Pharmacovigilance Research Reveals Momentum towards Cloud and Artificial Intelligence to Reduce Costs and Increase Compliance”. Retrieved from: http://www.appliedclinicaltrialsonline.com/pharmacovigilance-research-reveals-momentum-towards-cloud-and-artificial-intelligence-reduce-costs-a

[6] Oracle Health Science, “Addressing the Data Challenges of Pharmacovigilance”. Retrieved from: http://www.oracle.com/us/products/applications/health-sciences/pharmacovigilance/index.html

[7] Oracle Health Science, “Addressing the Data Challenges of Pharmacovigilance”. Retrieved from: http://www.oracle.com/us/products/applications/health-sciences/pharmacovigilance/index.html

 

CategoryBlog
Write a comment:

*

Your email address will not be published.

©2017 - seQure - All Right Reserved - Privacy Policy - Cookie Policy