1 Introduction
Severe acute respiratory syndrome-coronavirus 2 (SARS-CoV2) emerged in Wuhan, China, in November 2019. By October 2020, this virus had resulted in more than 41.5 million cases of COVID-19 and 1 million deaths. The majority of COVID-19 cases currently being reported are predominantly in developed countries, such as the United States, Brazil, Russia, and France. Despite the increasing number of models and stud- ies on COVID-19, there is very little information available to inform humanitarian response interventions, the need for which may be unprecedented, particularly where infrastructure is lacking to effectively prevent spread of transmission and treat affected patients.
Download Report: https://bit.ly/34JCCiT
In late 2019, the United Nations (UN) Office for the Coordination of Humanitarian Affairs (OCHA) Centre for Humanitarian Data created a new workstream for predictive analytics. This was based on demand from OCHA’s leadership to “use data, and especially the tools of predictive analytics to get ahead, to be more an- ticipatory, to predict what is about to happen and to trigger the response earlier.” This ambition aligns with the overall goal of the Centre, which is to increase the use and impact of data in the humanitarian sector. The COVID-19 pandemic has brought into stark focus the need for data and the value of models to inform response strategies. Anticipatory action is no longer an abstract idea but something populations are actively doing by staying home and increasing the number of hospital beds to protect the most vulnerable populations.
Epidemic forecasting is one tool through which we can gain an understanding of the final outbreak size and indicators of when the COVID-19 epidemic peaks in a country. This provides decision-makers with the capability to plan, surge, and manage resources during a pandemic. UN OCHA and the Johns Hopkins Uni- versity Applied Physics Laboratory have therefore established a partnership to inform COVID-19 strategies for humanitarian interventions by both national authorities and the humanitarian community in selected high-priority countries, resulting in increased technical capacity to predict new and compounded humani- tarian needs, and use of data science to arrive at interventions to mitigate them.
This partnership developed a series of adjustments to a novel COVID-19 model (JHUAPL-Bucky) that in- corporates different vulnerability factors to provide insights on the scale of the crisis in priority countries at national and sub-national levels, how different response interventions are expected to impact the epi- demic curve, and the duration of the crisis in specific locations. The resultant model (OCHA-Bucky) strati- fies COVID-19 dynamics by age and population vulnerability. Input to the model consists of geographically distributed COVID-19 cases and deaths, as well as attributes such as inter-regional mobility, population vul- nerability, nonpharmaceutical interventions (NPIs) , and social contact matrices. Model output consists of future projections of these same quantities, as well as severe cases (defined as a proportion of total cases). The model considers both inter-regional mobility of the population and time-varying NPIs. OCHA-Bucky has been used to provide weekly projections to six OCHA country offices: Afghanistan, the Democratic Republic of Congo, Iraq, Somalia, Sudan, and South Sudan.
The results of this model, when applied in the context of the United States, have been included in the Centers for Disease Control and Prevention (CDC) COVID-19 Mathematical Modeling Forecasting Ensemble [1]. Here, we detail the modifications of the JHUAPL-Bucky model used to shift results to the context of countries in receipt of humanitarian aid.
A number of critical complementary components must be modified in order to perform accurate disease modeling within this context. Namely, these include:
• Estimating disease parameters;
• Acquiring data sources that accurately reflect both the current and historical states of the outbreak;
• Estimating mobility within a country;
• Estimating the impact of secondary/tertiary factors on the vulnerability of a given population;
• Estimations of the effects of measures taken by individuals and governments to curb the spread of the disease; and
• Estimations of the effects of the above at both the sub-national and national levels.
The following sections describe the OCHA-Bucky model in detail. Section 2.1 gives an overview of the model and its components. Further details related to parameter estimation, data sourcing, and model initialization are given in Sections 2.2 and 2.3, respectively. Section 3 details the model input and output. Lastly, this model is publicly available; details about how to access the model as well as the corresponding documentation are given in Section.
Distributed by APO Group on behalf of Office for Coordination of Humanitarian Affairs (OCHA).
