ABSTRACT
Scientific advice to the UK government throughout the COVID-19 pandemic has been informed by ensembles of epidemiological models provided by members of the Scientific Pandemic Influenza group on Modelling. Among other applications, the model ensembles have been used to forecast daily incidence, deaths and hospitalizations. The models differ in approach (e.g. deterministic or agent-based) and in assumptions made about the disease and population. These differences capture genuine uncertainty in the understanding of disease dynamics and in the choice of simplifying assumptions underpinning the model. Although analyses of multi-model ensembles can be logistically challenging when time-frames are short, accounting for structural uncertainty can improve accuracy and reduce the risk of over-confidence in predictions. In this study, we compare the performance of various ensemble methods to combine short-term (14-day) COVID-19 forecasts within the context of the pandemic response. We address practical issues around the availability of model predictions and make some initial proposals to address the shortcomings of standard methods in this challenging situation.
Subject(s)
COVID-19 , Influenza, Human , COVID-19/epidemiology , Forecasting , Humans , Influenza, Human/epidemiology , Pandemics , UncertaintyABSTRACT
BackgroundThe novel Coronavirus Disease 2019 (COVID-19) pandemic is having a profound impact on global healthcare. Shortages in staff, operating theatre space and intensive care beds has led to a significant reduction in the provision of surgical care. Even vascular surgery, often insulated from resource scarcity due to its status as an urgent specialty, has limited capacity due to the pandemic. Furthermore, many vascular surgical patients are elderly with multiple comorbidities putting them at increased risk of COVID-19 and its complications. There is an urgent need to investigate the impact on patients presenting to vascular surgeons during the COVID-19 pandemic. Methods and AnalysisThe COvid-19 Vascular sERvice (COVER) study has been designed to investigate the worldwide impact of the COVID-19 pandemic on vascular surgery, at both service provision and individual patient level. COVER is running as a collaborative study through the Vascular and Endovascular Research Network (VERN) with the support of numerous national (Vascular Society of Great Britain and Ireland, British Society of Endovascular Therapy, British Society of Interventional Radiology, Rouleaux Club) and an evolving number of international organisations (Vascupedia, SingVasc, Audible Bleeding (USA), Australian and New Zealand Vascular Trials Network (ANZVTN)). The study has 3 Tiers: Tier 1 is a survey of vascular surgeons to capture longitudinal changes to the provision of vascular services within their hospital; Tier 2 captures data on vascular and endovascular procedures performed during the pandemic; and Tier 3 will capture any deviations to patient management strategies from prepandemic best practice. Data submission and collection will be electronic using online survey tools (Tier 1: SurveyMonkey(R) for service provision data) and encrypted data capture forms (Tiers 2 and 3: REDCap(R) for patient level data). Tier 1 data will undergo real-time serial analysis to determine longitudinal changes in practice, with country-specific analyses also performed. The analysis of Tier 2 and Tier 3 data will occur on completion of the study as per the prespecified statistical analysis plan. Ethical ApprovalEthical approval from the UK Health Research Authority has been obtained for Tiers 2 and 3 (20/NW/0196 Liverpool Central). Participating centres in the UK will be required to seek local research and development approval. Non-UK centres will need to obtain a research ethics committee or institutional review board approvals in accordance with national and/or local requirements. ISRCTN: 80453162 (https://doi.org/10.1186/ISRCTN80453162) Ethical Approval: 20/NW/0196 Liverpool Central, IRAS: 282224