School pupils have had a tough year. For those with examinations, such as A-levels, it has been even tougher. The Department for Education decreed that A-levels and GCSEs should not take place due to the COVID crisis.
This meant that an alternative way of allocating grades to students needed to be found.
To set the context, Ofqual, the Office of Qualifications and Examinations Regulation, regulates qualifications, examinations and assessments in England. It is a Non-Ministerial Government Department constituted under the Apprenticeships, Skills, Children and Learning Act 2009. Now, it is evident that, despite being a non-ministerial department, the Department for Education has a responsibility for setting policy.
Therefore, a Memorandum of Understanding exists between Ofqual and DfE setting out how the two organizations will work together. Section 2 sets out responsibilities:
On 31 March, the Secretary of State, Gavin Williamson, wrote to Ofqual, setting out a ministerial direction. These are the salient points in the Direction:
Firstly, there needs to be a calculation, and secondly ‘as far as possible, the qualification standards are maintained and the distribution of grades follows a similar profile to that in previous years’.
This is a very strong constraint, and to be fair, one that Ofqual met in their initial calculated grade. The problem was it was not fair.
Much debate has been generated as to algorithms, models, and their use in Government (and arm’s length bodies such as Ofqual). But this is not a new problem.
In 2012, the West Coast rail franchise was awarded to FirstGroup. Virgin Trains complained, serious errors were found in the model used by DfT, and the award to FirstGroup was cancelled.
This cost taxpayers £54 million, and, as a result, and report was commissioned to prevent this happening again.
The Aqua Book recommends that there should be a Senior Responsible Owner for each model (for Government departments and their arm’s length bodies, such as Ofqual).
Public Health England has published an analysis of what is known about the Leiecester outbreak. Diagrams are from the PHE report.
The latest daily case numbers are available at coronavirus.data.gov and are shown below. Note that the latest figures in the data download are not complete, as these will exclude specimens in the post. Also note that the number of positive cases detected will be affected by Leicester being in the news, availability of more testing stations, and the functioning of the NHS Test and Trace service (see update below).
The first thing to note is the mismatch between testing that was disclosed to the public (so called Pillar 1 tests) when a potential lockdown was being discussed by politicians and the total number of tests being conducted (Pillar 1 and Pillar 2 tests). I have discussed why this was a problem here. Since writing, the Government has disclosed total positive tests (but not the number of tests taken) for each location, including Leicester.
Firstly, the number of positive tests rose to 23 June (the chart above updates this slide).
PHE Preliminary Investigation into COVID-19 Exceedances in Leicester (June 2020)
Since then, it appears that the number of positive tests may be falling, but this is preliminary analysis, and we shall know for sure on Thursday when PHE release their updated analysis for the whole country.
The current Leicester cases seem to be through working age people and children (this is where Leicester may be unusual – other outbreaks may be in care homes where the population is older).
PHE Report
This is the spatial analysis of where cases have taken place in Leicester (the left map is Pillar 1 testing and the right map is Pillar 2 testing).
PHE Report
And this breaks down the wards in which most cases were located
PHE Report
with the corresponding map here
PHE Report
It is important to note that testing has been increasing in Leicester, so some of the increase in positive cases may be due to this. William makes the comment below that this may be due to the location of walk-in tests making people from those areas more likely to take tests compared to other areas of the city. There is a feedback effect here, where more positive cases means more testing resources allocated to those areas which means more testing of those areas. Without test data (number of tests in each location), it is not possible to see whether the increased case density is as a result of increased numbers of tests, as we don’t know the percentage of positive cases at each location.
The latest Public Health England national report here with results shown below.
Update: The .gov.uk analysis seems to average out the 7-day average as +/- 3 days which is misleading, as the recent specimen date tests may not have arrived.
The threshold for lockdown is not publicly disclosed (and there is unlikely to be an absolute threshold as local considerations such as where the outbreak is taking place (for example in a factory or a care home that can be relatively well contained). However, Germany has set a threshold of 50 cases per 100,000 to consider an ’emergency brake’ and reimpose lockdown-like restrictions.
Taking the population of Leicester as 348,300, this would mean that this threshold of 50 cases per 100,000 in a week would be 50 * (348,300 / 100,000) / 7 = 25 cases per day as a threshold. Although of course, the threshold for entering and leaving lockdown are not the same. And Directors of Public Health and journalists, armed with timely and complete data, are far more able to understand what is happening at a local level.
I will provide an analysis of the Public Health England data on Thursday when it is published. For updates, please come back to duncanrobertson.com or follow me on Twitter @Dr_D_Robertson
The pubs open on Saturday in England, allowing people to mix in confined spaces and potentially transmit COVID-19.
In Friday’s Number 10 briefing, the Chief Medical Officer said “The biggest risks are when lots of people from completely different households are brought together in close proximity indoors. And whether that’s in a pub or a cricket pavilion that is a high risk activity. And that’s the reason why the really quite onerous social distancing guidelines that are going to cause a significant change to pubs and cause difficulties for many publicans, and we all recognise that, are so essential. There is no doubt these are environments whose principle job it is to bring people together. That’s a great thing to do socially, but it’s also a great thing from the virus’s point of view. And therefore we do have to have a really clear and really disciplined approach to trying to maintain social distancing whilst also enjoying pubs, and this would be true in any other environment”
Which regions of the country are particularly risky? Leicester for a start, where drinking in pubs is banned. Kirklees, Bradford, Blackburn, Rochdale, Rotherham, and Oldham have the next highest cases.
But there are vast swathes of the country that exceed the US Centers for Disease Control threshold for re-opening (10 cases per 100,000 people in a two-week period). While not equivalent, I have used 5 cases per 100,000 people in one week as a cut-off.
I have analyzed the latest Public Health England data to work out which parts of the country exceed these thresholds and plotted them on the map below.
This of course does not mean that other parts of the country are risk-free.
This article was updated on 4 July with the latest data for PHE specimen date for the week to 3 July 2020 inclusive.
Public Health England has today released the second tranche of data for COVID tests. This is the most comprehensive data we have for tests as it includes Pillar 1 tests (those conducted by PHE and NHS hospitals) and Pillar 2 tests (those conducted by private companies under NHS Test and Trace).
Some journalists had been using data from coronavirus.data.gov – but this was dangerously misleading as it only revealed Pillar 1 tests – which are now a relatively small proportion of tests. This had led to erroneous league tables based on Pillar 1 data only until the site was updated late on 2 July.
We want to be able to identify possible regions that have the potential to have remedial action taken, for example local ‘lockdowns’, in the future. This does not mean that these areas will be locked down, more that they should be investigated by Directors of Public Health and local journalists. Without specific local knowledge, here are the criteria I have used to identify regions of interest:
HIGH INCIDENCE REGIONS (RED): Greater than 50 cases per 100,000 individuals. While the Joint Biosecurity Centre has not issued public guidelines for this threshold in the UK, Germany has defined 50 cases per 100,000 inhabitants in a week as the number of cases required for a region to apply an ‘emergency brake’ and reimpose restrictions. Areas meeting these criteria could indicate that there is sustained transmission in this area (but see the caveats below).
WATCHLIST REGIONS (AMBER): Between 40 and 50 cases per 100,000 individuals. (The threshold of 40 is chosen based on a qualitative comparison of Barnsley and Bradford in the PHE Leciester epidemiology report.)
RECOVERING REGIONS (GREEN): In a high incidence or watchlist region last week and fewer than 40 cases per 100,000 this week. It is important to bear in mind that no area is recovered from COVID-19, as outbreaks can recur in any region at any time.
A more general point needs to be made about the paucity of publicly available data. Without timely, complete, accurate data available to the public, there are several issues:
Other towns may see that they have relatively high case levels on the Coronavirus data service, causing unnecessary alarm;
Having data at a coarse geographical area (currently Upper Tier Local Authorities) does not allow outbreaks in towns and villages to be identified;
Unitary authorities (such as Leicester – where the city council performs the role of district and county councils) are separated on the maps, but cities such as Oxford (where there is a city council and a separate county council) are included in the data for Oxfordshire, where urban and rural data is evened out, hiding outbreaks in cities.
There are several caveats that need to be emphasized:
The number of tests carried out: when there are low number of tests, there are necessarily low numbers of detected cases. We do not currently have information for the number of tests carried out in each region, so cannot take account of this – it is possible that high cases per 100,000 is due to particularly high levels of testing in that region.
More local testing in locations with known cases: As local outbreaks are detected, extra testing resources may be allocated to towns such as Leicester, with mobile testing stations being set up. There is a feedback loop here meaning that extra cases will be detected – this does not necessarily mean that there is a higher incidence, just that the cases are being detected.
There is a time lag before this data becomes available: The latest data (published today, 2 July 2020) is for cases detected up to 28 June 2020 – so is not in real-time. In addition, there is a further delay between individuals becoming infected and a case being capable of being detected.
Outbreaks in care homes, hospitals and prisons: These need to be treated independently, and are currently included in the data. We know that there are outbreaks in these locations, and PHE report on these (but we don’t know where these outbreaks are taking place). So Pillar 1 and Pillar 2 data by itself does not indicate that there is community spread – this may be confined to these special locations
Local Directors of Public Health are the experts in their local areas: These professionals are experts, know their communities well, and understand the dynamics of transmission far more than can be ascertained by looking at figures in a database. There have been delays in getting this data to local authorities and issues with data quality, but the Prime Minister has promised in the House of Commons that the data is now getting through.
The delay and lack of detail of Pillar 2 results at district council level, or within-local authority breakdowns (as is disclosed for Pillar 1 tests) highlights data weaknesses, compounding the delays in convening the Joint Biosecurity Centre, and the failure of the centralised NHS Test and Trace App. While directors of public health, Public Health England, NHS Test and Trace, the Joint Biosecurity Centre, and the Department of Health and Social care all play their part, the policy for controlling a pandemic rests with central Government and is ultimately the responsibility of the Prime Minister and the Cabinet. Authority can be delegated but responsibility can not.
Our paper in Journal of Simulation ‘How simulation modelling can help reduce the impact of COVID-19‘ setting out how simulation modelling can help in the fight against COVID-19 and subsequent epidemics and pandemics. Click here to access the paper.
ABSTRACT
Modelling has been used extensively by all national governments and the World Health Organisation in deciding on the best strategies to pursue in mitigating the effects of COVID-19. Principally these have been epidemiological models aimed at understanding the spread of the disease and the impacts of different interventions. But a global pandemic generates a large number of problems and questions, not just those related to disease transmission, and each requires a different model to find the best solution. In this article we identify challenges resulting from the COVID-19 pandemic and discuss how simulation modelling could help to support decision-makers in making the most informed decisions. Modellers should see the article as a call to arms and decision-makers as a guide to what support is available from the simulation community.
There has been much talk about using social distancing as a response to the Coronavirus outbreak. Here’s a video of how social distancing can be used:
Close Contact, where the population is infected very rapidly. This is not A Good Thing – the national health service is overwhelmed, and the death rate goes up as a result
Extreme Distancing, where the viral spread takes too long. This is not A Good Thing – the social and economic costs are huge
Optimal Distancing, where the spread is controlled and the human, social, and economic costs are optimized. This is A Good Thing. It is also very tricky to get right.
The United Kingdom is at a crossroads, an ideological battle between natural science and behavioral science. Let’s hope for all our sakes we get this one right.
Boris Johnson, the UK Prime Minister, is facing a dilemma. When do we go from the so-called containment phase for controlling Covid-19 Coronavirus, to the delay phase.
In the medical / natural science corner, is the Chief Medical Officer, Professor Chris Whitty, who has presented himself calmly, reassuringly, as completely on top of his brief. He is a physician and an epidemiologist (as well as a lawyer, and an MBA). His evidence at the newly formed House of Commons Coronavirus Committee was calm, frank, precise. He is exactly the sort of advisor that any government would be proud to have. Flatten the peak. Delay the virus spread. Keep the height of the peak low. Save lives.
Image: New York Times adapted from CDC/Economist https://www.nytimes.com/2020/03/11/science/coronavirus-curve-mitigation-infection.html
In the behavioral science corner is, well, I am not sure who. Maybe it’s the Chief Scientific Advisor, who highlighted the need to take account of behavioral science. Yes, please do. It’s a wicked problem, and please include more complex social modelling.
But what we are now seeing is what the Director General of the World Health Organization (up until now also criticised for its seemingly political response to the issue) could be referring to as ‘alarming levels of inaction’.
I do hope, however, that Boris Johnson is being guided by the science, both behavioral and epidemiological, and not by advisors who profess to be superforecasters. You don’t have to be a superforecaster to forecast that if we get this wrong, many will die unnecessarily.
The 2011 film Contagion, starring the spectacularly ill-fated Gwyneth Paltrow, is a dramatization of a viral pandemic starting in pretty analagous circumstances to the current Wuhan Coronavirus (2019-nCoV) outbreak. It’s a good film, and is a great introduction to the work of Centers for Disease Control (CDCs) that monitor the spread – the epidemiology – of the disease. There are two scenes where R-nought, or R0, are described:
Despite the blogger character in the clip describing the spread, using a R0 of 2, as being a problem you can do on a napkin, it takes a little more thinking about. He also seems a bit confused about R0, talking about growth from 2 to 4 to 16, to 256, to 65,536 each day. That’s not what R0 is – it is not a rate, and actually if the rate was 2, this would mean 2 to 4 to 8 to 16 to 32 etc., each time doubling the number. It is possible that he is thinking that there are two generations each per day, but that’s not whatR0 is.
So, on to the professionals:
The CDC epidemiologist in the clip is more on point (despite having sloppy notation with no subscripts). This is better – it shows the reproduction number for the infection – note again, this is not a rate – no time dimension is involved – it basically shows the number of cases on average each case generates.
This population modelling – so called SIR (Susceptible, Infected, Recovered) system dynamics modelling – is just one of several approaches that can be used to model contagion across a population. My recent paper ‘Spatial Transmission Models: A Taxonomy and Framework’ sets out a review of what they are and the advantages and disadvantages of each. In brief, we can model the population numbers, the individual agents that carry the virus, the network of contacts between infected individuals, or the regions or cells in which individuals are located (city districts, for example). The paper is available to read by clicking on the link here.
