Up until recently, many diseases had very little information about signs, symptoms, and causes online, yet tons of data on the medication to treat it. This gap in information left many patients unaware of how to identify symptoms associated with a diagnosis, but equally, which of those symptoms they could attribute to adverse effects of the medication itself.
In recent years, patients and HCPs have been taking to dedicated social media groups and community boards to share experiences, ask questions, and give advice on how to manage illnesses and the adverse effects of certain medications. This new effort is part of a broader existing practice called pharmacovigilance.
In this article, we’re going to discuss how social media is now directly tied to pharmaceutical companies’ pharmacovigilance practices, its importance, and the challenges and benefits associated with social media pharmacovigilance practices.
What is Pharmacovigilance?
Pharmacovigilance is the phase in drug creation that includes monitoring for long-term safety outside of a controlled group once medication or vaccines hit the market. The World Health Organization notes that pharmacovigilance should detect unknown adverse reactions, detect increased frequency of known reactions, and the analysis of benefits and risks.
This practice dates back to the Thalidomide Tragedy in the UK during the 1960s. Thalidomide is a sedative that was produced in Germany and used to treat morning sickness in pregnant women. It was recommended after only testing in pregnant animals and humans which did not present any relevant concern. However, the rise in malformations in fetuses prompted the reexamining of the medication and safety laws. Since the creation of the Committee on the Safety of Drugs, protocols have been developed to ensure that there is thorough testing and long-term monitoring of medications.
Social Media’s Role in Pharmacovigilance
Background
Pharmaceutical companies can tap into social media pharmacovigilance practices to provide medical professionals with curated content and streamlines the identification of long-term adverse reactions.
Part of this process is secondary data mining on existing social platforms. Using secondary data from the more typical platforms below serves as a starting point for brands interested in pharmacovigilance efforts on social media. Through secondary data, medical professionals can widen their datasets and offer more context to traditionally collected adverse drug reactions.
General platforms can provide a greater opportunity to finetune the purpose of the group and provide a focused view of particular adverse effects associated with a particular disease in a subset of users. For example, a search for eczema on Reddit will offer options for patients specifically looking for a cure, and communities who are trying to understand the science behind the skin condition as well as associated skin concerns.
And then there are more niche sites:
- Patients Like Me
- WebMD Communities
- Daily Strength
These platforms dedicated to medical concerns are more controlled, with moderation efforts that are tailored to patient protection and care. They typically include blog posts and articles that can provide more insights to users. They provide the option to include relevant healthcare information such as gender, age, condition, and medication.
Applications of Social Media Pharmacovigilance
Traditionally, pharmacovigilance efforts are mostly based on reactive reports from patients to their doctors. These reports can possibly be brushed off or misunderstood by medical professionals or can simply be neglected to be reported through the FDA. Insufficient information poses a major risk to patients and pharmaceutical companies. While there have not been any recorded deaths or other widespread issues, traditional pharmacovigilance practices are still insufficient in gaining insights.
Social Media Pharmacovigilance is believed to offer a fuller picture as it isn’t filtered through the eyes of a doctor and does not include the typical barriers of adverse drug reaction reporting via healthcare providers. The use of social media listening to understand potential adverse reactions includes extracting the data from the chosen social media platform, analyzing the language and colloquialisms used, and removing any irrelevant personal information.
The Process
Relevant content could be populated by utilizing Named Entity Recognition and Topic Modeling. Topic Modeling is a method of data gathering that is typically abstract and useful when unsure of what adverse reactions are being looked for. It includes automated, related keyword searches. Named Entity Recognition is a type of information collection that sorts raw data and classifies it into a category such as personal identifiers, medication name and prescribed amounts, and already-determined adverse effects. This is especially useful for tracking any increase in reaction frequency.
The WEBRADR project uses these ideas and principles for pharmacovigilance with the creation of an app that would allow for in-platform adverse effects reporting. According to the app developers, the Innovative Medicines Initiative, this app fills the gap between areas where reporting is insufficient by adding the information directly into the patient’s chart. This can result in faster recognition of adverse reaction indications and quicker action in providing necessary care.
Opportunities and Challenges of Social Media Pharmacovigilance
While the use of Pharmacovigilance and the associated opportunities are still growing, there are a number of challenges that pharmaceutical companies, brands, and users are facing.
Opportunities
Social media provides a new landscape to approach pharmacovigilance:
- Social media provides anonymity and often, patients will disclose detailed reports (even more so than to their doctors).
- Community board discourse is often rich and recent – adverse events are timely and varied to a global audience
According to the study Harnessing Social Media Data for Pharmacovigilance, there is an opportunity for direct reporting in social media as well. As users are already comfortable with engaging in social media, prompted reporting within these platforms would be less intrusive and more ethically responsible. Solicited adverse reaction reports allow users to be aware of what the information is being used for and the option to share or withhold information that they may not want to report.
Challenges
The challenges that are creating barriers to the best use of social media include understanding the best practices in safely collecting large amounts of data while protecting patients who aren’t aware that their information is being used for pharmacovigilance. There are studies that suggest frameworks to standardize this method, such as:
- Using public information like usernames to identify who the adverse report is associated with
- Closely monitoring the terms of use for the platform that the information was gathered from
- Considering that these users cannot be interacted with for any purpose including follow-up.
With such limitations, it can be difficult to monitor these reactions or any other developments over time.
Automated social listening tools can be useful in collecting and sorting information quickly however there have to be set rules to ensure reporting is accurate and inclusive. This includes:
- Accurate tags to identify which adverse effect category the reported symptoms fall under and how the user’s medical history, lifestyle, and environmental factors affect the report.
- Technology suited for language variance. The technology would need to be able to adjust to the changes in language to describe the medication and the adverse effect over time.
The Future of Social Media Pharmacovigilance
Social media pharmacovigilance is still growing and has yet to be considered a standard practice. Computing methods are not readily available to fully and accurately extract and analyze the data needed to make advancements in this area. The future of social media pharmacovigilance includes the use of artificial intelligence, mobile application, and other technological advancements to detect useful information, develop measurable benchmarks, and turn them into actionable steps for pharmaceutical companies and medical professionals. With several methods of data gathering employed, medical providers can gain comprehensive insights into adverse reactions.
Conclusion
Artificial intelligence solutions can sift through data and identify details of adverse reactions, however only when coupled with the human touch of understanding sentiment, context, and usability of the raw data, do they become valuable and actionable insights. Moderation and tracking methods enacted by ICUC’s team can increase the efficiency of understanding these adverse reactions. Our strategists and insight analysts will work as an extension of your team to streamline the process of protecting patients. Learn more by booking a meeting with us today.
English
UK English
German
Dutch