Influence of COVID-19 on the Pharmacovigilance Workforce of the Future.

In the middle of difficulty lies opportunity.—Albert Einstein

Amid 2020's trifecta of pathos—societal and political instability, economic disruption, and the pandemic death and disability that have followed the novel coronavirus 2019 (COVID-19)—there have been unexpected opportunities recognized by innovators, researchers, and especially health care leaders, who have found themselves in the depth of caring for patients with COVID-19. Pharmacovigilance (PV), too, may well be the beneficiary of unexpected opportunities in these times. During the past 1 to 2 decades, PV has taken shape as it has settled into three core disciplines—case management, signal management, and benefit-risk management—supported by professionals with distinct and gradually defined knowledge bases and skill sets. Furthermore, PV has played increasingly important roles in other domain-facing disciplines of pharmaceutical research and development, including nonclinical and clinical pharmacology, regulatory affairs, manufacturing, and medical affairs. These evolving dynamics have been occurring against a backdrop of a long-term demographic slowing of population growth and ever-tightening availability of talent in all market sectors, despite the significant impact of COVID-19's forces on the business environment during the better part of the past year.

The knowledge bases and skill sets that are needed to support PV activities at these intersections are identified in the Table 1 . At the top level, these knowledge bases and skill sets fall into 2 basic types: content-focused, which are generally performed by subject matter experts, and process-focused, which are generally performed by operational experts. At the next level, developments in knowledge and skills have occurred in 2 parallel and mutually reinforcing components: a quantitative component, with greater emphasis on statistically based approaches, and a technological component, with increased emphasis on software platforms and artificial intelligence. Beninger and Ibara have previously discussed the potential contributions of the latter.

Table 1

Scope of PV domain-facing activities in the context of knowledge bases and skill sets needed to support these activities.

Domain-Facing Activity	Knowledge Bases and Skill Sets
	Content-Focused:Subject Matter Experts	Process-Focused:Operational Experts
Core PV Disciplines
Case management	Analytical, medical	Project management database management
Signal management	Systems analysis, medical, TSA, social media	Data management, social media
Benefit-risk management	Risk: concepts, analysis, communication	Risk: communication, project management
Other PV Domain-Facing Disciplines
Animal pharmacology	Animal physiology: adverse event profiles, PK/PD properties	Project management
Clinical pharmacology	Human physiology: adverse event profiles, PK/PD properties	Project management
Regulatory/labeling	Global regulatory structures, risk communication, negotiation	Regulations, risk communication
Clinical trials	Bioethics: consent, IRB, DMC, trial design
Manufacturing	Manufacturing process; quality: FMEA; SPC; security
Database management	Software, artificial intelligence
Epidemiology/biostatistics	Quantitative methods (including analytics)
Medical affairs	Communication

DMC = data monitoring committee; FMEA = failure mode effects analysis; IRB = institutional review board; PK/PD = pharmacokinetic/pharmacodynamic; PV = pharmacovigilance; SPC = statistical process controls; TSA = time series analysis.

Large-scale change, historically driven by innovative new technologies and regulatory pressures, has generally occurred incrementally. However, the deeply felt effects marked by COVID-19 are forcing change at a jarring, quantum level.

Enter Hauben and Hung in this issue with their analysis entitled Global and Country Level Time Series Analyses of the Effect of the COVID-19 Pandemic on Spontaneous Reporting, which addresses the question, “Has COVID-19 negatively influenced spontaneous reporting of adverse drug events?” Although a reasonable, intuitive hypothesis, the results actually reflect a more nuanced outcome of these uncommonly applied methods to pharmacovigilance: global total reporting of serious and nonserious adverse events decreased, but global reporting of serious adverse events did not unexpectedly decrease. Japan, alone among country-level analyses, had a significant decrease in reporting, and Taiwan had a statistically unexpected increase in reporting.

The outcomes follow from time series analyses, a method with a surprisingly broad degree of application, including economics and finance, physics and engineering, weather forecasting and earthquake prediction, but with limited prior application in PV. For example, Whalen et al recently described a use of time series analysis through the application of an autoregressive integrated moving average–based algorithm to large-scale screening of databases for the detection of outliers in signal detection. Statistical process control, better known as the Shewhart control chart, is another method with an established pedigree going back nearly 80 years in the manufacturing sector that has recently begun to find application in PV, likewise in the detection of outliers that suggest potential signals. Recognizing that these are but 2 examples of cross-applications of established methods from other disciplines to PV, it would not be surprising that other methods are similarly awaiting discovery in the near future.

If the knowledge bases and skill sets described in the Table are viewed from a risk management perspective, most of the focus is on the front end, that is, on the detection/identification and analysis of pharmaceutical risks, activities that pharmaceutical companies can control. However, moving the focus to the back end, that is, on mitigation and communication of these risks, makes it clear that a significant shift is needed from quantitative and technological methods to soft skills, such as teamwork, collaboration, conflict resolution, and negotiation, all areas that require strong communication skills far beyond the immediate control of the pharmaceutical company, and into the pharmacy, clinic, and the patient's home environment.

This change in focus may open the way for the application of behavioral economics and other soft skill approaches that may have a more direct impact on improving patient outcomes and reducing patient harms. This may be "the middle of difficulty" in which lies the next opportunity.

Disclosures

In addition to receiving a salary as a faculty member in the department of Public Health & Community Medicine at a private medical school, the author has received compensation as a topics editor for a medical journal since 2016, as a subject matter expert for the American Association for the Advancement of Science in 2019, and as a training consultant for a pharmaceutical company in 2020. The author has indicated that he has no other conflicts of interest regarding the content of this article. The author is the sole contributor in the preparation of this manuscript. The author did not receive monetary support from any source for the preparation of this manuscript.