Over the years, pharmacovigilance (PV) processes have involved PV professionals manually going through high volumes of data to identify, assess and report adverse event (AE) information. This set-up will face increasing challenges, as more and more work needs to be done in fixed amounts of time. The options for addressing the challenges can either be labor intensive or technology enabled. Technology-enabled processes have the advantage of being sustainable over time, with the added benefit of potential savings in time and costs.

Within PV, scientific and medical literature review lends itself to the introduction of natural-language-understanding technology due to the orderliness of the data and the low ratio of AE information found to volume of journal articles reviewed. A simple, three-step approach has been developed to speedily and accurately search through journal articles to identify the ones with reportable AE information using a technology solution. An initial test of the technology solution showed that this new approach could pick up all AE terms with no false-negative results and go on to deliver further benefits.

Addressing the growing PV burden

In 1796, Edward Jenner, a local doctor in Gloucestershire, tested the theory that milkmaids did not get smallpox (a potentially fatal disease) because they had suffered from cowpox (a much milder disease). He put pus from a cowpox postule into an incision in the arm of an eight-year old boy. He went on to carry out the same experiment on other people and submitted his results to the Royal Society. This experiment was the basis of vaccination, a cornerstone of modern healthcare.

Thankfully, things have evolved considerably over the years from a patient-safety perspective. Although researchers all around the world continue to test medical theories to discover new ways of treating diseases, robust pharmacovigilance (PV) now exists to make sure the public is safeguarded both at clinical trials and after the medication has been given marketing authorization. Pharmacovigilance is defined as “the science and activities relating to the detection, assessment, understanding and prevention of adverse events (AEs) or any other drug-related problem”1. Traditional sources of reported AEs are clinical trials, spontaneous reports from health professionals or patients, and literature reviews of scientific and medical publications. PV functions have the responsibility to collate, evaluate and report adverse events to healthcare regulators.

It is generally accepted that many AEs are not reported. A systematic review of published data from 37 studies worldwide found the median under-reporting rate of adverse events to be 94 percent in spontaneous-reporting systems2. As patients become more empowered and avenues such as social media are increasingly used to discuss responses to medication, we should expect increases in the reporting of adverse events. This means the burden on PV departments will increase, while their regulatory obligated reporting timelines are unlikely to change.

In response to the increasing AE-reporting burden, PV organizations can take a number of approaches to meet their obligations: (i) increase the workload on existing PV staff, (ii) increase the resourcing of the PV department (directly or via outsourcing), or (iii) employ smart working approaches enabled by technology. Of the options above, (i) and (ii) are not sustainable in the long term. That leaves the technology-enabled option, which is often discussed, but has not been explored to any reasonable extent. Injecting technology into the PV process will potentially offer significant benefits over labor-intensive approaches.

PV literature review: a starting point for technologyenabled approaches

In searching for a “beach head” in PV for technology to demonstrate its capabilities, literature review of scientific and medical publications is a good starting point. A PV literature report review involves monitoring a set of journals at least once a week as mandated by regulation to see if any reportable AEs can be identified for a particular drug. This area of PV lends itself to the involvement of technology for a number of reasons; firstly, the articles are written by professionals using standard medical
terminology. Secondly, the articles exist in text format and are therefore relatively easy to search. Thirdly, the ratio of AEs found
to volume of articles reviewed is very low. These factors make it possible for technology-enabled approaches to show significant
benefits in the review of literature reports.