Patient safety is at the heart of healthcare. Yet, while clinical trials remain the gold standard for assessing the safety and efficacy of drugs, they have inherent limitations. Their controlled conditions don’t accurately reflect real-world settings well and could exclude patients with complex health profiles, for instance.
It is here that real-world data (RWD) becomes an important and powerful ally in the quest for safer and more effective drugs. That’s assuming this data can be harnessed reliably and efficiently.
RWD helps life sciences organizations and healthcare providers understand how patients use and respond to an approved drug. Drug developers can access RWD from many sources, including claims and billing activities, electronic health records (EHRs), product and disease registries, and social media. That’s in addition to patient-generated data collected through mobile devices and wearables.
Today, however, only about half of life sciences organizations use RWD to drive research and development. That means a lot of organizations are missing out on a rich source of existing data.
When harnessed reliably, RWD can greatly benefit patients through its help for pharmacovigilance (PV) teams in:
Speeding up risk and safety signal detection
PV teams need high-quality data to do their jobs effectively, ensuring patient safety while ideally providing invaluable feedback to those responsible for drug development. Overreliance on case-by-case signal data can lead to fragmented information that is difficult to scale, causing delays in uncovering important safety insights.
Incorporating new RWD sources significantly improves safety signal identification and trend monitoring by making it more timely, comprehensive, and proactive.
The potential resources that PV teams can draw on are significant. They include:
- Foundational datasets to understand treatment patterns, disease incidence, and prevalence.
- Highly focused disease-specific data linked with data at the patient level for additional context.
- Curated fit-for-purpose data to identify a cohort, intervention, or outcome of interest.
Technology, including advanced AI/cognitive computing in the form of large language models that can distill and deliver intuitive insights from vast volumes of data, has an important role in extracting value from RWD. It can help cut through any noise, for instance, to enable more timely identification of patterns and trends, such as negative or positive drug interactions, patterns within patient populations, or other behaviors affecting drug efficacy.
RWD empowers PV teams to take a more proactive approach to safety rather than reacting to adverse events as they’re detected. The right technology combined with the right data equips PV and safety teams to perform analyses on the fly, without the intervention of epidemiologists or data specialists, and without a pre-defined hypothesis.
Developing personalized medicine
A more fluid and responsive approach to signal detection is much more aligned with personalized medicine, too. A drug’s impact and efficacy is rarely uniform. It is influenced by lifestyle, health history, genetic factors, and treatment adherence. For example, although non-steroidal anti-inflammatory drugs (NSAIDs) are shown to be beneficial for patients suffering from pain and inflammation, they can be dangerous for patients at higher risk of developing bleeding stomach ulcers.
Personalized medicine accounts for these variables, aiming to deliver the right treatments to the right patients at the right time. Leveraging RWD allows for a better assessment of social determinants of health and biological factors when making healthcare decisions. Smart analysis of RWD can assist by creating patient profiles and isolating risk factors to inform medication prescribing and identify alternative treatment options.
RWD analysis helped identify that anticholinergic (ACH) medications can cause cognitive decline in elderly patients. This kind of discovery paves the way for more optimal medication usage.
Identifying additional benefits
Smart application of RWD can also help identify and address previously unknown benefits of and additional potential applications for existing drugs once they are on the market. Just as aspirin — originally developed to treat pain and inflammation — has been found to prevent heart attacks and strokes in some patients. And antihistamines can help with disturbed sleep.
With reliable technology to perform the detection and analysis, RWD sources and spontaneous reporting signals can provide important insights into unanticipated outcomes. By harnessing data mining, clustering, and knowledge graphs, life sciences organizations have an opportunity to uncover patterns or relationships that may indicate a correlation between a drug and a positive event. RWD also enables long-term disease and outcome tracking, enabling a deeper understanding of drug impacts, treatment opportunities, patient needs, and disease courses over time.
The right tools can help PV teams leverage RWD to revolutionize patient safety. Up to now, capitalizing on RWD has proved challenging due to the volumes of raw data and its complexity, which can quickly become overwhelming. Next-generation technology is helping to address this. AI-powered pharmacovigilance software automates data collection, management, and analysis, while cognitive computing identifies deeper insights in less time than manual review alone.
All of this frees up PV and safety teams to apply medical reasoning to the findings and create more robust benefit-risk and treatment safety profiles — driving positive change and improving patient care.
Elizabeth Smalley is the senior product manager at ArisGlobal. With over 15 years of experience across data analytics, Elizabeth has a passion for exploring the intersection of data science and human reasoning, with experience bringing AI-powered software to market to drive safety and clinical outcomes for patients and clinicians. In her current role, she leads the teams managing the Data Platform, LifeSphere Clarity and LifeSphere Signals, and Risk Management product lines.