As coronavirus has become the new reality for every health worker and life science company to fight in the front, it is time to relook the approach being taken by the industry towards pharmacovigilance. Health agencies around the world are evaluating the impact of the virus, clinical trials, and pharmacovigilance systems. The rapidly changing situation with the outbreak of the virus has compelled agencies to be more vigilant and proactive in determining the effects of the pandemic for better patient care. Pharmacovigilance is detection, assessment, understanding, and prevention of adverse effects of the therapeutic product on the patient. A thorough pharmacovigilance strategy is essential to ensure patient safety and protect the reputation of pharma and medical device companies. Pharmacovigilance is the report card that helps us identify the successful patient care of the medicine post its clinical trial.
During the preparation of a drug, it goes through different clinical tests to get approved for patients. Pharmacovigilance helps in knowing about the adverse effect of the therapeutic product after it has been launched in the market. It is an important part, like clinical trials, for the product lifecycle. Pharmacovigilance involves the collection of a large amount of data and analysing the same for better patient care and adverse effects.
The increasing number of data sources used to identify adverse events (AE): call centres fielding reports from patients and healthcare practitioners (HCPs); safety databases such as ARGUS, ARISg, electronic health records (EHRs); clinical trial data; medical insurance claims; contract research organisations (CROs); post-marketing safety (PMS) studies; scientific literature; regulatory and NGO databases; and legal cases play a major role in pharmacovigilance.
Moreover, the process of finding reportable adverse events in such a plethora of information requires significant manual effort. It is also prone to human error, resulting in missed safety signals, very high skilled labour costs and non-compliance risk. The problem of data explosion worsens when you consider pharma and medical device companies operating in multiple geographies, receiving information about suspected adverse drug reactions (ADRs) in various languages across disparate channels. WHO’s Vigibase which is a global dataset that has all the data on adverse drug reactions from across all the major countries works as a key support system for the healthcare industry. While every country has its way to stop the spread of the virus, the healthcare industry must share the success of clinical trials, new symptoms and effects of the virus.
To put that in perspective, pharmacovigilance staff may have to scan through 144,000 to 240,000 abstracts every year to find the literature of interest, which may require anywhere between 8,000 to 20,000 human hours to execute. Out of those abstracts, 72,000 to 120,000 pieces of literature may be of interest, and then only 3,600 to 9,600 may be reportable. All this review requires close to 30,000 human hours to sort out. Assuming the number of abstracts being collated increases every year by 15 per cent, it becomes apparent that legacy pharmacovigilance workflows and systems need a massive overhaul to ensure organisational efficiency and reduce the cost of compliance. Minimising operating costs is particularly of interest to pharma and medical device industries which have an eye on costs driven by the focus on value-based healthcare, and the dwindling blockbuster drug pipeline, which would, in the past, generate global sales of at least $1 billion annually for companies.
Today, the pharma and medical device industries are driving collaboration among industry participants, HCPs, regulators, patients, and academia for boosting the efficacy and safety of their therapeutic products.
Collaborations can demonstrate the value of real-world data, queries of spontaneous databases combined with real-time observational studies using data from big data networks gathered via these collaborations with other scientific originations can allow for more efficient evidence generation to explore any medicines safety issues. Achieving that requires robust integration and standardisation of data sets, enabling data transparency, accessibility, and reliability. Moreover, there are additional challenges due to ongoing M&A activity, which requires the integration of data sources and data management processes across companies to drive watertight compliance and pharmacovigilance outcomes. Timely review of incoming data and real-time signal detection can provide important safety information for healthcare providers.
Interestingly, the expectations of the millennial, technology-enabled consumer are changing the way pharma and medical device companies manage and respond to pharmacovigilance events, considering heightened public awareness and scrutiny. The new-age consumer is empowered and has a voice that can be leveraged by pharma and medical device companies to accelerate product improvements and drive innovation.
A next-generation approach to pharmacovigilance
While in some cases – such as when a patient is put on a new medication or develops a medical condition during the drug therapy, or there are significant changes in the patient’s lifestyle – it may be more straightforward to understand ADRs. However, finding possible issues with drug safety by reviewing scientific literature, medical journals, online publications, clinical trial data, and social media posts in huge volumes can be particularly daunting. Pharmacovigilance solutions are now adding artificial intelligence (AI) as a tool which can help in the analysis of big data in a faster manner as well as in-depth processing which can help practitioners in evaluations, cross-checking patterns of data. The way companies report those pharmacovigilance events could also lead to misinformation and affect drug safety initiatives. A recent study by the University of Colorado Cancer Center found that significant variation in reporting of drug side effects can lead to some drugs appearing safer than they are. Applying innovative technology automation tools and processes to pharmacovigilance can help overcome these issues as it helps in high performance and scalability by automating tracking and monitoring of data. Leveraging Natural Language Processing (NLP) which self-learns through experience, to make predictions based on patterns observed in large volumes of data with the help of machine learning (ML) eventually helping human decisions.
By finding hidden correlations between adverse events from various data sources can emphasise the real-world efficacy of drugs and uncover in-depth insights into multivariant patterns and correlations of undiscovered potential threats to public health and safety concerning a patient population consuming a specific dosage or type of medication. Those correlations cannot be mined from large volumes of data by humans, and there must be increased reliance on intelligent technologies, such as artificial intelligence, natural language processing and machine learning. Those data insights will not only help with understanding the adverse effects of medication but also bring to light the potential benefits that may have been missed in the past.
The pandemic underlines how advanced big data analytics and cognitive technologies like AI, machine learning and natural language processing will radically improve how pharmacovigilance solutions will be able to play a vital role in the success of public health initiatives all over the world. The industry work will be appreciated and found living in the pages of next generations history literature. While multiple vaccines are under developments some going through clinical trials, the world’s life sciences companies have an eye to always improving the safety of patient care.