Digital Technology

Digital technologies may help perform a wide range of functions in pandemic management, as summarized in Table 1. ^{Reference Whitelaw, Mamas, Topol and Van Spall2} Health authorities in some countries have attributed their disease control successes to early adoption of such technology. ^{Reference Wang, Ng and Brook3-5}

Table 1.

Applications of Digital Technology in Pandemics

Source: Adapted from Whitelaw. ^{Reference Whitelaw, Mamas, Topol and Van Spall2}

Abbreviations: AI, artificial intelligence; PDA, personal digital assistant.

Nonetheless, major issues have arisen on data oversight, privacy, protection of personally identifiable health information, and redress for errors that implementing jurisdictions must confront. The scientific community has seen retractions of papers published in The Lancet and The New England Journal of Medicine due to post-publication concerns over provenance of proprietary data and inconsistencies from electronic health records. ⁶ This experience is cautionary about the importance of data source/origin, data audit trails, and oversight in managing such datasets.

Core Databases

Core Databases for PBM are characterized below in three overriding domains:

1. 1. Background Information (Table 2) describes the affected population pre-pandemic—demography, access to essential services, health status, and socioeconomic status.

2. 2. Clinical Case and Virological Data (Table 3) encompasses case definitions, case identification and outcomes, and clinical dynamics (case-fatality rates, death rates, and time course to death).

3. 3. PH Control Measures and Consequences (Table 4) encompasses details of contact tracing, quarantine measures, travel restrictions, and all epidemic modelling.

Table 2.

Datasets and Visualizations Pre-Pandemic

Table 3.

Datasets and Visualizations Intra-Pandemic

Table 4.

Datasets and Visualizations Intra-Pandemic

The Database Challenge

The technical standards for disease surveillance and surveillance systems comprise an extensive literature. The Centers for Disease Control and Prevention (CDC; Atlanta, Georgia USA) published its first guidelines in 1988, which were re-issued in 2001. ^{Reference Klaucke, Buehler and Thacker7,Reference German, Lee and Horan8} The World Health Organization (WHO; Geneva, Switzerland) Program on Disease Control in Humanitarian Emergencies published numerous adaptations of communicable disease and surveillance guidelines for epidemic-prone diseases in natural disasters and complex emergencies through the 2000s. Recent emerging infectious diseases such as Ebola have prompted additional reviews of data needed for outbreak evaluation culminating in descriptions of an emerging science of outbreak analytics. ^{Reference Cori, Donnelly and Dorigatti9,Reference Polonsky, Baidjoe and Kamvar10}

Taking into account the rate of global population growth, it was predicted in 2019 there would be 1.7MB data created every second for every person on Earth, bringing forth new data problems, “most notably where and how to store it.” ^{Reference Cori, Donnelly and Dorigatti9} With unprecedented movement of migrants and refugees from the Middle East, compounded by massive relocations of urban populations because of COVID-19, are only two entities that will impact data collection critical to future infectious diseases. The era of data-centric computing is here, rate of product change accelerating, storage options expanding, and already digital transformation has taken center stage as a key driver. Outbreaks, epidemics, and pandemics are driving streaming data, larger memory servers, video, artificial intelligence (AI), and machine learning (crucial to guarantee data security), and need for unprecedented “big data and analysis.” ^{Reference Harris11} Multi-disciplinary AI questions to the proposed Global Public Health (GPH) Database will be increasingly common, as it should be in pandemics. All these demands require new training and professions. Cost effective and resilient Cloud storage infrastructure can adapt to rapid workload requirements such as transient or spiking demands of world-wide PBM requirements. With the advent of “fast and dense memory,” the intelligent storage can be trained to be largely self-managing, responding to detect anything from race conditions or access to care changes. Harris concludes that the era of data-centric computing is here with the potential to bring top-priority to the demands of the proposed PBM system. ^{Reference Harris11}

The Economist is more detailed in their assessment in managing future pandemics accelerated by population growth and globalization that have markedly increased the spread of infectious diseases: ¹²

• Data analysis can pinpoint outbreaks and reveal where they may unfold next.

• Can show where people were moving within 24 hours, right from the initial outbreak, and could distribute that information widely, especially PBMTs.

• Can take advantage of diverse skill sets in tracking new, less-understood pandemics, favoring the proposed multi-disciplinary focus of their teams to include: “physicians, veterinarians, ecologists, data scientists, epidemiologists, environmental health specialists, geographers, designers, and software developers,” to help make sense of raw data and AI requests and critical mobility-related information. ¹²

• The WHO concludes that “working together with data firms and health agencies (ie, PBMTs) before a disaster strikes mitigates searching for data at the start of an outbreak, which is already too late, reiterating a multi-stakeholder collaboration archives more than working in isolation.” ¹²

Artificial Intelligence in support of the GPH Database

Expectations were high for AI to help fight COVID-19. However, before AI tools can make an impact, global collaboration and high-quality data and model sharing are needed. Nature Machine Medicine emphasized that early warning and alerts, prediction and detection of outbreak of diseases, real-time disease monitoring world-wide, analysis and visualization of spreading trends, prediction of infection rate and infection trend, and rapid decision making to identify the effectiveness of management. ¹³ Other authors emphasize challenges that need to be addressed before AI can have a beneficial global impact. ^{Reference Naudé14-Reference Vaishya, Javaid, Khan and Haleem16} They point out that a first challenge is knowing where to start with developing AI tools that can be most effective, which requires close cooperation with practitioners at the health care frontline, those that manage the PDMTs. The best solutions may involve adapting already validated systems rather than building new tools from scratch. Furthermore, Hu, et al described how clinical needs are evolving as the pandemic is moving through different stages, from early detection and anticipation, to containment and mitigation, and finally eradication. During these transitions, the specific types of AI models may need to change too.Ideally, the AI system itself becomes a smart, intelligent, and adaptive system. ¹³

Education and Training for PBM and the GPH Database

The origins of this pandemic can be attributed to climate change and the emergence of complex GPH crises such as climate extremes, biodiversity loss, emergencies of scarcity, rapid unsustainable urbanization, migrant and refugee surges, domestic and international terrorism, the civilianization of war and conflict, and the global rise of resistant antibiotics that have resulted in an unprecedented rise in direct and indirect mortality and morbidity, among others. ^{Reference Benis, Notea and Barkan17} In 2019, The Tohuku Journal of Experimental Medicine recognized the emergence of complex GPH crises that are beyond the current decision-making and operational capabilities of traditional disaster management and its providers, most of who are community-level practitioners representing every discipline.

The 1930s “disaster cycle” concept described a phase-related approach to meeting the strategic, operational, research, educational, and training components required of disasters; this presents an opportunity for the structured development of a Health Crisis Management Framework to oversee the phase-related strategic and operational requirements for prevention, preparedness, response, recovery, and rehabilitation challenges of major GPH crises. ^{Reference Burkle18} No longer will physicians, for example, focus on the response phase alone. Future PBMTs will be trained across the entire disaster cycle. Operational knowledge on prevention and preparedness will lessen the number of victims more than any other action. The Japanese Editorial launched academic interest in training multi-disciplinary Health Care Crisis Managers and Scientists to fill the full- and part-time roles required of PBM, PBMTs, and those managing the GPH Database. For example, the university-centered CRIMEDIM (Research Center in Emergency and Disaster Medicine) in Navaro, Italy has worked diligently to offer both Masters and PhD degrees leading to training Health Crisis Managers and Scientists from many contributing disciplines to fill these proposed PBMT positions. A global effort is recommended post-COVID of multi-disciplinary educational programs to help prepare the potential PBMT and GPH Database experts that this study recommends.