Every digital process and social media exchange produces Big Data. It is being generated around us all the time, and we have seen the knowledge derived from Big Data positively impact almost every industry. Market leaders are using Big Data to gain a better understanding of actionable customer insight, supply chain optimization and how to improve operational efficiencies. In the banking and insurance industries, Big Data trends, risk performance and measurement, and regulatory compliance are the driving market forces behind investments in business analytics. Organizations are using advanced statistical algorithms to understand risk and manage it. So, in this era—where the benefits of Big Data are being capitalized on, where regulatory bodies pose increasingly complex questions, and where the volume of data being collected is growing exponentially—why are we not embracing Big Data to manage risk and improve the operational efficiency in every clinical trial?

It is widely known that clinical trial sponsors still rely heavily on 100% source data verification (SDV) and on-site monitoring. Today, however, as the cost of drug development continues to rise, this intensive SDV method is losing its relevance. In a recent study, Transcelerate Biopharma used operational data available through the Medidata Clinical Cloud to explore the relative contribution of SDV—the process by which data within a case report form (CRF or electronic CRF) is compared to the original source of information—to overall clinical data quality. Results revealed that only 1.1% of total site-entered eCRF data are corrected as a result of SDV, confirming that SDV has a minimal impact on overall data quality. This finding supports the conclusion that SDV should not be the primary data quality control method used in trials.

Nonetheless, clinical teams continue to depend on 100% SDV to review all data points. This shows us that clinical teams don’t know where to look, so they look at everything. That’s why it’s important to draw attention to another concept introduced in TransCelerate’s paper: source data review (SDR), a process that checks the quality of the source data, reviews protocol compliance and ensures critical processes and source documentation are adequate. When SDR is targeted at specific areas of risk, it has a more important impact on the quality of trial conduct by resolving the primary source of the issue.

TransCelerate’s paper on risk-based monitoring (RBM), published in November 2014, indicates that a shift away from conventional on-site monitoring methods is warranted. This shift is predicated on the belief that centralized, dedicated resources will be better able to identify emerging risks through Big Data analytics. A centralized process for risk detection is more effective because resources look at real-time data rather than on site traditional monitoring activities. This function will be responsible for identifying key areas of risk and working with field monitors to adjust activities for SDV and SDR based on identified risks.

Through the use of Big Data, we can now isolate the truly relevant data and not only spot erroneous or poor quality data documented in a trial, but also use predictive models to alert ourselves to the clinical trial sites or protocol design elements that are most likely to generate poor data in the future. Also, with predictive modeling, Big Data allows us to intervene in certain cases before questionable data is collected and while processes can still be adjusted or reversed, thus improving the overall quality of clinical trial outcomes.

Using robust statistical algorithms, we can conduct a comprehensive scan of a clinical trial database for inconsistencies across data domains, sites and patients. Automated processes can be setup so that every clinical trial can have a quality score calculated for data and other important study parameters. The quality score can be a measure of overall data consistency, and be compared across sponsors, studies, indications and disease areas. Individual site scores can be used to measure data quality within the site and identify studies and sites at high risk for procedural problems and data errors.

While compliance, quality, visibility and insight are key benefits, the monetary return on such an investment cannot be overlooked. Big Data-driven RBM holds the promise of reducing the time it takes to analyze large volumes of data—without reducing quality or accuracy—thereby distinguishing itself as the cost-effective solution.

So what are we waiting for? Why not benefit from real-time, Big Data by routing clinical trial information to a centralized, cloud-based repository that can be used to effectively conduct RBM? The advantages are compelling—a reduction in staffing and travel costs, the ability to assess and prioritize risks for maximum safety and data quality, and source data errors caught early, before they negatively impact trials.

It’s time to embrace technology in pursuit of enabling safer, faster and more insightful and cost-effective clinical trials.