This is the second in a series of webinars organised jointly by Health Statistics User Group (HSUG) and the RSS Official Statistics Section on topics relating to
Collecting and Reporting Statistics for Covid 19. The focus of these webinars is on sharing experiences and discussing approaches to the methods being used in the UK to produce the statistics for each of these topics. The topic for the second webinar in the series is Modelling.
The webinar will provide an opportunity for:
- People involved in modelling relating to Covid-19 to share their experiences and exchange ideas
- People using the results of modelling to pose questions and make comments. This will help with their understanding and use of these results.
- Members of the audience to contribute insights to the interpretation of the models based on work they are doing.
- People working on modelling to gain an understanding of the queries and comments that users have regarding their models, which will be helpful in improving the explanations that are provided with the results of the modelling
Audience
The intended audience is people whose work involves modelling and/or using the results of modelling relating to Covid-19 in the UK. We would encourage others with an interest in this area to join us to learn and share.
Speakers and Topics Details
The topics that will be covered by each speaker are:
1) Chris White/Vahe Nafilyan, ONS
Contrast in deaths involving COVID-19 by ethnicity and religious group
Will present findings on the contrast in deaths involving COVID-19 by ethnicity and religious group published on 19th June. These articles present provisional analyses of deaths involving the coronavirus (COVID-19) by ethnic group and religion group for England and Wales during the period March to May 2020 using a unique linked dataset containing census and death registrations.
2) Greg Ceely – ONS
Health Projections
ONS has conducted a few scoping exercises for what a model projecting health information could include, such as the aspects of health which are of most interest; the types of model most suitable; and stakeholders’ needs for what the outputs include. Our current plan is to produce a flexible macrosimulation model, capable of reading in data from a wide breadth of health indicators and creating projections based on the associations between dependent and independent variables selected by the user. ONS’s population projections and trends in key health indicators will provide a baseline, against which users can test the impact of interventions in one area on other aspects of health.
3) Paul Birrell, PHE
As part of the response to the 2009 A/H1N1pdm outbreak, a real-time pandemic influenza modelling framework was developed and subsequently developed under an NIHR Pandemic Influenza funding call. In this presentation, I will demonstrate how this model has been re-purposed for use in the public health response to CoVID-19, from early scenario analyses, to situational reporting, and to model outputs feeding into the SPI-M short-term forecasting exercises and the estimation of regionally-stratified reproduction numbers and infection levels.
4) Mel Giarchi, Scottish Government
Modelling of the epidemic in Scotland is undertaken to look at the progression of the epidemic and to inform the logistical response required. Outputs of the modelling help the Scottish Government, the health service and the wider public sector plan and put in place what is needed to keep us safe and treat people who have the virus, e.g. to decide how many Intensive Care Beds (ICU) we need available for Covid patients. In this presentation we will give an overview of the Scottish Government’s modelling process and the data and assumptions used.
5) Andrew Lee, Service Manager, Public Health Scotland
PHS have collaborated with NHS Board, government and academic partners to develop a range of models and tools to support the local and national response to Covid-19. These include a short term forecasting model, a longer term scenario planning tool, a model of the Covid-19 care pathway and models of the treatment/care backlog and how it could be addressed. This presentation will summarise these developments and briefly demonstrate the available tools.
6) Sally Thompson, Senior Analyst, Local Intelligence Support Team, Public Health Scotland
Home care staff absence modelling
When lockdown was announced, a Health & Social Care Partnership (HSCP) wanted to understand the potential impact that Covid 19-related staff absences could have on their ability to deliver home care services. A dynamic tool was developed, and made available to all HSCPs, that gave insight to the HSCP, as to when they would need to consider suspending some aspects of home care services
7) Craiger Solomons, Welsh Government
Will provide an overview of the work of the Covid-19 Welsh National Modelling Sub-group.
Prior to the Webinar
An introduction to the topics being covered by each speaker Is available on the HSUG web pages within StatsUserNet - see
www.hsug.org.uk
It should be noted that, in order to view this information, participants need to register as users of the StatsUserNet web site and then go to the Health Statistics User Group category within StatsUserNet.
Webinar Speakers
There will be presentations from the following speakers.
- ONS (Office for National Statistics) – Chris White / Vahe Nafilyan
- ONS (Office for National Statistics) – Greg Ceely
- PHE (Public Health England) – Paul Birrell
- Scottish Government- Mel Giarchi
- PHS (Public Health Scotland) – Andrew Lee
- PHS (Public Health Scotland) – Local Intelligence Support Team – Sally Thompson
- Welsh Goverment- Craiger Solomons