A room without a view

If you imagine that the UK Government’s strategy for dealing with novel corona virus is primarily to save lives, think again. The only real strategic objective is preventing a catastrophic NHS meltdown.

When the current restrictions are eventually lifted there will be an inevitable steep rise in both cases and deaths unless there is a fundamental change in strategy. In less than a month this increase would create new pressure to re-impose restrictions.

Mass testing alone will not substantially reduce the number of deaths over the coming year. Experience from countries which have restricted the growth of cases demonstrates that contact tracing is key to saving lives. Achieving a successful contact tracing programme will require careful planning and this should already have begun.

In his second article James Urquhart asks the questions that go begging at every daily news conference:

• What is the exit strategy?
• Stop and go lockdown: herd immunity by stealth?
• Testing alone not enough?
• Testing and tracking should start when? 
• If UK can’t do it, can Scotland go it alone?

————————————————————————————————

Situation room

As with every major crisis, Covid-19 requires its own command centre, the situation room which receives reports from the field and coordinates an emphatic response.

So, what do we know about the Covid 19 operational headquarters?  Is it a hive of focussed activity? Is it the place where cutting edge science informs cohesive administration? Does a fully functioning command centre even exist?

What are the essential requirements for a successful command and control operation? First, of course, you need some strategic objectives Next you will define a series of tactical objectives, each of which will contribute to the desired strategic outcome. Intelligence will be essential – information, data analysis and data interpretation.  Our control room will also be the place where people balance difficult choices. Standing ready will be the team who draw up the operational plans and ensure their implementation.

Strategic Objectives

What are the strategic objectives of the putative Covid 19 Command Centre? These have not been set out in any Government document. Our only sources are the statements made by Ministers and their advisors.

Surprisingly, perhaps, nowhere will you find a clear statement that reducing the total number of deaths figures large on the UK Government’s agenda. In fact, the only true strategic objective held up to scrutiny at successive press conferences is: preventing the NHS from being overwhelmed. This is summarised in the slogan “Protect the NHS Stay at Home Save Lives”.

Now there can be no doubt that preventing the NHS from being overwhelmed is a vital strategic objective. If we allowed the healthcare system to collapse under the weight of coronavirus cases it would catastrophic not just for people who become very ill with the virus but also for the tens of thousands who become seriously ill each month with other conditions. For most people, protecting the functioning of the NHS would top a list of strategic objectives. But the issue does not arise because this is a list of just one.

Tactical Objectives

The belated decision to institute the clampdown measures on March 23^rd was taken primarily to keep hospital admissions at a sustainable level. Other tactical objectives to reduce pressure on the NHS included discouraging people from reporting “mild” symptoms to the help line and cancelling routine procedures and appointments. Little consideration was given as to how to ease the burden of the virus in other parts of the care sector.

Of course, as an indirect benefit from the March 23rd restrictions there will be a considerable reduction in the potential number of deaths during the period of clampdown.  But the benefit will be a time limited unless steps are also taken to reduce the number of deaths in the period following the relaxation of restrictions. It is this uncertainty that underlies the increasing demand for answers about the so-called exit strategy.

Intelligence

 On April 7^th the UK Government’s Chief Scientific Advisor assured us that the data on hospital utilisation by patients with Coronavirus were “moving in the right direction” It is curious that the data are not being made readily available on a daily basis given that would provide some basis for assessing the success of the Government’s sole strategic objective: protecting the NHS.  Perhaps the number of confirmed cases is meant as some sort of proxy for hospital admissions. But these data provide little insight into how hospital resources are actually being utilised.

Increasingly now, we are witnessing “the war of the models”. The model developed by the Imperial College team has been challenged on the one hand by a group in Seattle claiming  that it drastically underestimates future UK deaths and on the other hand by a group in Oxford surmising , on an entirely theoretical appraisal, that half the population may already have contracted the virus. And the Imperial College team have been inundated by requests for their source code and detail of their assumptions to allow other groups the opportunity to second-guess their work. Should we be concerned? Or can we ignore these dissenting voices?

It must be stressed that the Imperial College team rendered an invaluable service to us all by releasing their model when they did.  Without their intervention, we would still be pursuing a vague aspiration of “protecting the economy” and the NHS would be facing meltdown. But with increases in available data an elaborate model may be less essential to the formulation of new strategic objectives. Indeed, dwelling on the complexity of the models perhaps provides a place for the politicians to hide. And they are supported by advisors who say we still have insufficient data – but insufficient for what?

Take for example the key question of what will happen when the current restrictions are relaxed. Since the Imperial College model was first developed literally a world of data has become available. Using these data it is possible to make relatively simple empirical estimates of the effect of relaxing restrictions that do not rely on models.

Suppose you want an assessment of what will happen just three weeks after restrictions are relaxed. From the available data in different countries it is possible to estimate a rough and ready statistic to describe how the number of cases increases over 21 days when there are only modest restrictions. To create this 21-day multiplier you look at a period during which there were few restrictions and calculate the number of days required for the number of cases to double. In the UK the Government’s advisors have stated that this is about 3 to 4 days. Using this statistic you can calculate the UK case multiplier for a 21 day period which about 65.

Now as an example – say at the time restrictions are lifted – we have achieved a situation where we have one or two deaths and 200 new cases a day.  Application of the UK’s 21-day case multiplier to these data suggests that three weeks later if no other major factor has changed there will be perhaps 1300 total new cases a day and – say, a week later still -there might perhaps be more than 60 deaths a day. You could instead use a 21-day case multiplier of 45 or 85 instead of 65 it really doesn’t make any difference. On any scenario after a month the number of cases and deaths will be at a level where the re-imposition of restrictions will have to be considered.

And then what? Five weeks later perhaps when the levels have fallen again the restrictions can be lifted. This stop-go cycle, the possibility of which is also suggested in the Imperial College report, would have to repeat perhaps six or seven times in the next 15 months.

The application of the stop-go process is certainly compatible with the strategic objective of ‘Protecting the NHS’ because it keeps hospital admissions below the critical level. But there will continue to be substantial additional deaths during each cycle. Effectively this is a herd immunity strategy by stealth.

Inspection of world data throws up some important comparisons.  Take for example Singapore where the first indigenous case was reported some three weeks before the UK. Singapore has a population of similar size to Scotland but confined in an area slightly larger than East Lothian. Given its population density it might be expected to be facing problems of mass transmission of the virus. Yet the 21-day multiplier factor for Singapore has been less than ten compared with the 65 estimated for the UK. As of April 8, the number of daily reported cases has not risen above 125 and the number of deaths per day has not risen above two. By implication, Singapore has adopted a strategic objective of keeping the number of deaths below a very low figure. It has achieved this by testing anyone that might have the disease and vigorous contact tracing.

On April 9 Singapore reported a surge in cases, in its migrant worker population. This in no way changes the core message: We should be learning from the Singaporean and South Korean experience and, if we do not, it should be a source of shame.

Operational plan

Assume that once restrictions have been lifted there is a new strategic objective to keep further deaths to an absolute minimum. Assume also it has been recognised that mass testing and contact tracing programmes are essential. What are the parameters of the operational plan?

Based on the Singapore experience contact tracing must have the following features.

• As many cases as possible must be identified whether they are in the community or hospital. This means reversal of existing policy and people must be encouraged to notify symptoms to a central point. We know that 111 could not cope with this influx so we would need a new fully staffed contact centre

 

• All identified cases must be interviewed to identify people with whom the patient has had significant contact. This interview is a skilled procedure which carries risk for the interviewer

 

• To supplement the interview there is detective work to establish other significant contacts. This can include CCTV, and phone records. In Singapore this was carried out by police because they had the necessary expertise. People have been asked to keep their taxi receipts for a month

 

• Testing capability must be sufficient to cover all possible cases and their contacts. South Korea has open access testing accessible to all

 

• All identified significant contacts must go into quarantine for 14 days. And to recognise their sacrifice on behalf of the community as a whole they must be given generous financial support which compensates for all losses

 

• There must a free movement of relevant data across agencies. Discussions about the Caldicott principles of patient confidentiality would need to be compressed into a very short time frame

When the Singapore measures were first announced it was said they would never be acceptable in a western democracy. But this concern looks increasingly questionable given the sacrifice of civil liberties that people have already made in the UK. In Singapore and South Korea, the restaurants are open and there is relative freedom of movement.

Implementation

On April 5th  Matt Hancock stated that “It is too early to be making decisions on an exit strategy”. Was he right? Well that depends on the decision you are actually required to make

To be effective, symptom reporting, contact tracing and mass testing would need to be in place at least two and probably three weeks before restrictions were lifted. The new staff for the reporting call centre would need at least two weeks for training and recruitment from among the volunteer force would take at least a week before that.

So, adding it all up, if restrictions were to be lifted, say, on June 1^st detailed planning for contact tracing and mass testing should have already been announced and begun.

Command and Control

A depressing feature of the daily press conference is a collective failure to ask the right questions. Yes, it is appropriate to pursue the question of mass testing, but mass testing without contact tracing will achieve very little in preventing deaths. The question which should be uppermost is: are you planning a programme of contact tracing and if not why not?

Recent events and pronouncements give little reassurance that effective and urgent thought is being given to developing a strategy with the objective of keeping deaths at a defined and relatively low level. Nor can there be much optimism that there is sufficient flexibility at the centre to develop an effective exit strategy from the current lockdown. The Command and Control Centre is a room without a view.

Perhaps now is the time to offer a challenge to the devolved administrations. If the Westminster-based COBRA machinery is too cumbersome and sclerotic to offer innovative solutions, are you prepared to develop your own strategies? Of course, there are advantages in pursuing a common policy across the four home nations. But if all that is on offer in a month’s time from the UK government is a stop-go lockdown strategy are you willing to accept that deaths will continue on a rising trajectory (as set out above)?

Scotland has the devolved power to follow its own strategic objectives. The question is: does it have the courage and skills to start planning now for an exit strategy which wholeheartedly and comprehensively embraces contact tracing? The answer firmly rests with all our politicians and their advisors. We should be hearing from them now.

Main image Dominic Raab at daily briefing April 7 courtesy of Number 10 Flickr CC BY-NC-ND 2.0

Singapore image via Getty

Further reading: Scientists slow to sound alarm, Reuters

Prof Devi Sridhar, Four exit scenarios, the Guardian

Craig Dalzell, Testing, testing, 1, 2…Source

David Herszenhorn/Sarah Wheaton, How Europe failed the corona test, Politico

…and viewing: Citizens TV with Paul Mason/Afua Hirsch plus experts https://youtu.be/05-jbrHRmrs