The Architecture of Containment: Getting to Gold

Covid-19

The Architecture of Containment: Getting to Gold

Paper
Posted on: 16th May 2020
By Multiple Authors
Chris Yiu
Executive Director
Benedict Macon-Cooney
Head of the Science & Innovation unit
Ryan Wain
Strategic Advisor
Daniel Sleat
Special Advisor
Andrew Bennett
Policy Analyst

    A Foreword by Tony Blair

    A Foreword by Tony Blair

    Britain has to have a coherent strategy for living with Covid-19, which will last until there is a vaccine available that is able to be scaled for the whole population. That is the stark reality. But the economic and health costs of lockdown are so immense that we need at the same time to get as many people back to work and have as much of “normal” life return as possible. 

    The only viable solution is to build a containment infrastructure that is comprehensive and that embodies the best technology and practice. A gold standard, if you like. Despite the progress which has been made, Britain is still significantly short of attaining that standard. Whether we reach it depends on the decisions taken now. In this paper we set out the different elements of such an infrastructure. 

    Lockdown won’t eliminate the virus on its own. It does, however, give us time to prepare and to build so that we can open up as widely as possible and as safely as possible. Otherwise we will find that the economic, social and health devastation of the lockdown has an impact as serious or, in some ways, more serious than the disease itself.    

    –Tony Blair, Former Prime Minister of Great Britain and Northern Ireland and Executive Chairman of the Tony Blair Institute for Global Change

    Overview

    Overview

    Lockdown was never a strategy. It was a necessary step to buy the government time to suppress the virus and build critical infrastructure to move the country to a posture of containment. 

    What does this infrastructure look like and how has the government fared in building it?

    Health-care capacity has increased over the last two months and this should be rightly applauded. However, this paper argues that the government should have set out a clear containment strategy by now that includes an ambitious “gold standard” of containment architecture. Within this is mass testing, tracing and mask-wearing, and these measures should be supported by a reorganisation of government that allows for the cultivation of innovative technologies, greater transparency and clear communication. 

    Yes, the government has taken significant steps to prepare for containment by increasing testing capacity. But there is still too substantial a distance between where we are and where we need to be. Its strategy for charting this path has to, with urgency, move up several gears and invest fully in the tools, technologies, infrastructure and systems required to open the country up safely whilst containing the virus. 

    Incrementally adding new announcements and introducing targets fundamentally misses the point. Targets are not strategies. 

    Until a viable vaccine is developed, the only feasible strategy is to contain Covid-19. This will enable citizens to become economically and socially active, reducing the health harms of the virus and prolonged lockdown. The effectiveness of this strategy will be measured in an R number below 1, improved health outcomes and improved economic output. 

    This paper sets out:

    1. The gold standard of containment architecture 
    2. What the government must do to “get to gold” on its containment strategy 

    Snapshot: Progress So Far

    Snapshot: Progress So Far
      What the Government Is Doing What the Government Needs to Do
    Test
    • Has largely met its target of administering 100,000 test per day
    • Aims to conduct 200,000 tests per day by end of May
    • Has validated two lab-based antibody tests (manufactured by Roche and Abbott) but no rapid point-of-care tests
    • Set out a clear strategy on who to test and how
    • Lower the threshold of accuracy required for rapid serology tests and bring these online asap
    • Build capacity to test large parts of the population daily, with both antibody and PCR tests
    Trace
    • Roll-out of Covid-19 contact-tracing app expected following trial in Isle of Wight
    • Centralised approach with take-up of around 65% of the population needed for efficacy
    • Supported by manual tracing efforts with government target of 18,000 contact tracers by mid-May; only 1,500 have been recruited so far 
    • Self-reporting likely to result in significant false positives and lower compliance
    • Switch to Apple/Google exposure notification platform for auto-deployment to phones
    • Recruit up to 100,0000 contact tracers, leveraging NHS volunteer army of 700,000 volunteers
    • Create a software system to allocate tasks and send notifications and an advisory system for follow-up and alerts
    • Use anonymised location data from cell towers and other sources to identify hotspots of transmission and forecast future trends on transmission.
    Protect
    • Encourages public to wash hands regularly
    • Promotes social-distancing measures
    • Recommends mask-wearing
    • Mandate use of masks by all citizens in public where social distancing is not possible, including public transport (this should begin with basic face coverings but as soon as possible move to surgical masks)
    • Onshore production of surgical and N95 mask capacity to ensure significant domestic supply
    Identify
    • Validated a 100% accurate ELISA antibody test
    • Use mass antibody tests to inform an individual's Covid credentials, allowing those with proven immunity to re-enter normal life
    Treat and Vaccinate

    The government has taken all necessary steps to achieve this at the current moment in time, including:

    • Investing in internationally coordinated efforts for treatments and vaccines
    • Running large-scale, expedited randomised controlled trials for promising drugs

    The Architecture of Containment

    The Architecture of Containment

    Containing the spread of the virus while reopening the country requires a combination of measures. The government has fallen short in three areas of execution. First, they have tried to optimise individual measures when they should be optimising the portfolio as a whole. Second, they have underestimated the scale of the problem by at least one order of magnitude. Third, they have been behind the decision-making curve from the outset, failing to put the right testing infrastructure in place early on and not moving quickly to onshore mask production to ensure the country has the masks it needs. 

    Figures 1 and 2 below give an overview of the current standard of containment infrastructure versus what is needed to reach the “gold standard”. It paints a stark contrast. Getting to “gold” is not impossible, but it is hard. It means trekking unfamiliar territory for a government and thinking beyond the immediate political pressures. That said, progress needn’t be slow. There are things the government could be doing now which will have an immediate impact on containment, and groundwork that, laid now, will materialise in the coming months to dramatic effect.

    Figure 1 – UK government containment infrastructure to date

    UK Government containment infrastructure to-date

    Figure 2 – The “gold standard” of containment infrastructure

    The Gold Standard of containment infrastructure

    Optimising the whole of the containment strategy matters because the different individual measures all interact. For example, if masks are worn routinely by most people, then this affects the risk thresholds for contact tracing. If you can test everyone routinely, then you don’t need to ration public transport. If wearables can detect signs of infection, people can then choose to self-isolate accordingly. 

    Policymakers must face up to the scale of the problem. Population-scale vaccination is not an option for the foreseeable future, meaning we have to face up to the reality of what it will take to ease the lockdown via containment. Even if this is very expensive, it will always be more cost effective than lockdown.

    Getting to Gold

    Getting to Gold

    Landing on the right exit strategy is the easy part. The hardest side to governing will always be getting things done. Therefore, the government must pursue a dual approach of using its current infrastructure to mitigate Covid-19 while simultaneously reorganising itself and scaling up to reach an unparalleled gold standard of mitigation. This is akin to renovating the house while living in it.

    There are many examples of good practice from around the world. We should learn from these, but no country has yet fully mastered the crisis. The UK should not therefore seek only to emulate those who have gone before but instead to solve the problem for the long term.

      Current Status Gold Standard Getting to Gold
    Detect
    • New Joint Biosecurity Centre announced by the government
    • NHS data store and public information dashboard
    • Remote monitoring in care homes
    • Testing strategy
    • Infections detected within first two days, including asymptomatic phase
    • Real-time monitoring, including through use of wearables and symptom trackers 
    • Predictive modelling of potential hotspots
    • Rapid, regular testing
    • Greater investment in data collection and monitoring infrastructure
    • Multidisciplinary analytical teams, comprising of epidemiologists,  data scientists and others 
    • Wide-scale deployment of wearables to front-line workers and vulnerable groups
    • New detection mechanisms deployed, including smart thermometers
    • Increased roll-out of remote monitoring systems
    • Oximeters provided to those confirmed as having the virus
    Test
    • Current testing target means 100k lab-based antigen and antibody tests are being carried out per day
    • New target of 200k tests per day set for the end of May
    • As yet only two antibody tests have been approved by the government (made by Roche and Abbott), both are ELISA (lab-based) tests
    • No point-of-care rapid antibody tests have been approved
    • Launch and scale production of rapid antigen and antibody tests
    • Test all people reporting symptoms within 24 hours.
    • Test a significant percentage of the population on a regular basis through contact tracing (around 300,000 per day if 10,000 tested positive with an average 30 contacts per positive case1Del Valle, Sara & Hyman, James & Hethcote, Herbert & Eubank, SG. (2007). Mixing patterns between age groups in social networks. Social Networks. 29. 539-554. 10.1016/j.socnet.2007.04.005.2)
    • The testing strategy would prioritise key workers and use remaining testing infrastructure to undertake mass population sampling in order to identify hotspots and outbreaks and gather deep understanding of transmission and demographic outcomes
    • Test all frontline health workers daily
    • We believe the government should build capacity to be testing at least 1 million people per day3This would provide capacity to test, for instance: 300,000 contact trace tests, 200,000 frontline health staff, 100,000 tests for those self-reporting with symptoms, 400,000 regular community-based tests. 
    • Stretch goal: mega testing, capacity to test large proportion of the population daily
    • Significantly increased lab capacity
    • Validation of at-home/point-of-care tests that are (a) reliable (b) validated
    • This will require reducing the required accuracy level from 98% standard set by MHRA. Even a small reduction (of 1 or 2%) would bring online a range of tests available (including those being used in South Korea, Germany and New York)
    • Provide positive patient samples and test validation infrastructure to promising test suppliers
    • Ramp up of manufacturing of rapid tests
    • Build up logistics capability to deliver tests rapidly after request
    • Anticipate regional demand of home testing kits and ensure they are distributed to regional centres ahead of time
    Tracing
    • NHS Covid-19 app developed and currently trialled in Isle of Wight
    • Centralised approach, with data being collected for secondary use by government
    • App is opt-in and efficacy based on 65% take-up
    • Supported by manual tracing effort, with government target of 18,000 contact tracers by mid-May; currently met 8% of this
    • Self-reporting likely to result in significant false positives and lower compliance
    • Ensure take-up of app at least 65%
    • Ensure integration with NHS services
    • Trace and notify all smartphone users instantly following confirmed diagnosis
    • Trace out from all non-smartphone users manually
    • Augment all automated tracing with manual follow-up
    • Make sure system is interoperable across countries
    • Create automated system for rapid deployment of testing
    • Decentralised plan B ready to deploy, with switch to Apple/Google exposure notification platform for auto-deployment to phones
    • Increase efforts to ensure take-up, thinking through privacy-preserving mechanisms that will ensure this
    • Recruit up to 100,000 contact tracers, leveraging NHS volunteer army of 700,000
    • Build a software system to allocate tasks, send notifications and for rapid deployment of tests i.e. a deeply integrated system
    • Use anonymised location data from cell towers and other sources to identify hotspots of transmission and forecast future trends on transmission.
    Suggested strategy based on current infrastructure
    • Ask people to self-isolate if they exhibit symptoms and, in absence of an app, manually contact anyone they’ve encountered; provide materials for infected households to mark themselves out as such
    Treat
    • £544 million committed to vaccine development
    • Vaccines in development, including government funded ones at the Jenner Institute and Imperial College
    • A government Vaccine Taskforce established
    • A Therapeutics Taskforce established with six drugs in clinical trials
    • Increased collection of biomedical data
    • Increased collection of genetic data
    • Increasing collection of and access to biomedical data, funding projects such as OpenSAFELY platform
    • Making digital dashboard available to show progress in clinical trials
    • Multilateral data-sharing arrangements
    • Adoption of therapeutics learning systems such as one developed by Oracle, linked to electronic health records
    • Greater adoption of open data, increasing access to biomedical data
    • Multilateral data-sharing agreements and research projects, in particular in genomics and other omics; this can include greater use of mechanisms such as P3G2
    • International agreements on interoperability, standardisation and accreditation
    • Increased focus on inclusion and representativeness of data given divergent outcomes witnessed across different groups
    • Introduce online dashboard of ongoing trials
    • Support for broader efforts towards human challenge trials
    • Fast funds for research
    Protect
    • Alert system
    • Advice to public to wash hands regularly
    • Social-distancing measures
    • Recommendation to wear a mask
    • Health-care and other key workers to have access to N95 standard masks and other PPE; importantly, this will include getting the right equipment, in the right quantity, at the right time, to care staff
    • All households to have access to acceptable quality masks; ideally these would be N95, but short to medium term would include custom face coverings and/or disposable surgical masks
    • All venues (offices, shops, etc) to have access to hand sanitiser per footfall per week
    • All public spaces – particularly public transport infrastructure – to be regularly sanitised
    • Onshore surgical and N95 mask production
    • Draw on available options to deliver effective masks in the short term, e.g. Seal-A-Mask tool by Prestige Ameritech
    • Ramp up manufacturing capacity to deliver tens of millions of surgical masks per month
    • Manufacturing capacity to deliver sufficient sanitiser to meet full spectrum of UK needs
    Identify
    • Limited antibody testing infrastructure, only used for population sampling
    • Verifiable credential asserting permissions for international travel or restricted access to buildings
    • Secure, user-friendly digital credential wallet
    • International recognition of travel credentials, via either an international, interoperable standard or bilateral/regional mutual recognition
    • Clear public-health policy on criteria permitting travel, e.g. presence of antibodies, negative PCR test, etc

    The Role of Testing in Containment

    The Role of Testing in Containment

    As we have set out in previous papers, mass testing will be a vital element in exiting lockdown and pursuing a policy of containment. 

    We welcomed the government’s target of reaching 100,000 tests per day by the end of April. We have been clear, however, that this target was not, and is not, a strategy. Since meeting that target, the government approach has appeared increasingly rudderless.

    The right strategy seeks to identify clearly who to test and why, and then works capacity back from that.

    We believe the government must build capacity to enable:

    • Rapid testing of those identifying with symptoms (results within 24 hours).
    • Testing of any individual who has come into contact with someone who has tested positive for Covid-19  (on a presumed 30 contacts4Del Valle, Sara & Hyman, James & Hethcote, Herbert & Eubank, SG. (2007). Mixing patterns between age groups in social networks. Social Networks. 29. 539-554. 10.1016/j.socnet.2007.04.005. per positive case), using a combination of digital and manual contact tracing.
    • Regular testing for a sizeable part of the population with both antigen and antibody testing, with those proving immune removed from the cycle. This would utilise the STIR model, set out in our recent paper on mass testing (Figure 3 below).
    • Daily testing for frontline health-care staff.

    This requires the government to conduct a significantly higher level of testing per day, utilising both antibody and antigen tests. We believe this would involve, in the medium term, having the capacity to test at least 1 million people per day. The ultimate ambition, through rapid antigen and antibody testing, will be to have the capacity to test a significant part of the population. 

    Figure 3 – The STIR testing model

    the STIR testing model

    A key element of this strategy is antibody, or serology, testing. This test shows if someone has had the virus and is therefore immune, at least in the short to medium term.

    So far the government has been conducting antibody testing at its Porton Down lab in order to understand the spread of the virus across the UK. This acts in much the same way as an opinion poll.

    PHE has validated lab-based (ELISA) antibody tests made by Roche and Abbott, which are highly accurate. This is welcome and will be critical for tests where the purpose is to prove an individual’s Covid credentials, such as them having had the disease. Here 100 per cent accuracy is needed. This also sets a benchmark for other antibody tests where the purpose is to show credentials (note, not for sampling where lower accuracy requirements suffice). Community roll-out and rapid versions of antibody tests serving this purpose are also needed, and the government must continue to support innovators in this field. There are a number of promising examples around the world.

    For rapid point-of-care antibody tests where the purpose of testing is to understand the spread of Covid-19, we believe the current accuracy requirements of 98 per cent should be lowered. Even a small drop in this requirement would make it possible to bring onstream a number of antibody tests currently available and being used elsewhere in the world. As we know, containment is more than just a series of measures. It is a calculation about the true risk people face. This means collecting as much data on Covid-19 as possible and understanding its transmission and outcomes – mass antibody testing is critical for this. The government should work with major international suppliers, as well as the widest range of smaller firms in the UK, for instance BioPanda Reagents, whose rapid test is already being used elsewhere in Europe.

    Rapid point-of-care antigen testing is key to the STIR model and mass testing. The gold standard of containment would mean a future where anyone, anywhere in the country, could be tested to see if they are currently infected. This means antigen tests free from the shackles of laboratories. We are aware of the development of these accurate point-of-care tests elsewhere in the world, and the government should actively ensure we have capacity to develop these under licence once validated. They must work with promising suppliers to make validation as seamless as possible – including saliva-based testing providers.

    We are supportive of regional-based testing programmes, provided the results feed into a central testing strategy and the STIR testing model is followed. Recent proposals by Norwich5https://www.uea.ac.uk/about/-/covid-19-population-study-proposal to test its 140,000 citizens on a weekly basis are particularly welcome and feasible if the government ramps up testing capacity to the scale a gold standard demands. This is not impossible. In Wuhan, plans were drawn up for 11 million citizens to be tested over a ten-day period – on 15 May, these plans became reality6https://www.nytimes.com/2020/05/14/world/asia/coronavirus-testing-china-wuhan.html.

    The Role of Masks in Containment

    The Role of Masks in Containment

    In our recent paper The Importance of Masks in Exiting Lockdown, we recommended the government mandate the usage of masks in public where social distancing is not possible, including  public transport. 

    Eighty-two per cent of the world’s population now live in countries that either require or recommend the usage of masks in public. Emmanuel Macron, for instance, committed to providing all French citizens with masks this month.

    Figure 4 – Countries recommending or requiring mask-wearing

    map showing the countries requiring masks

    Source: https://masks4all.co/what-countries-require-masks-in-public/

    Trisha Greenhalgh, professor of primary health care sciences at the University of Oxford, and Jeremy Howard, co-founder of Masks4All, have written in detail on the value of masks in inhibiting transmission of the virus. 

    They draw on modelling by HKBU that suggests that if most citizens wear a mask in public, the transmission rate of the virus (R) can be kept below 1, stopping the spread of the disease.

    Figure 5 – Impact on transmission rate of Covid-19 at different levels of mask usage

    Diagram showing impact of mask usage

    Alongside other measures like social distancing, mass testing and contact tracing, we believe masks can play a vital role in ending lockdown and in containing the virus, helping the economy return to normal and mitigating the risk of a second wave.

    We recommend that those working in health-care settings are given N95 masks and that the rest of the public are asked to wear a form of face mask while in public and around large groups of people. This will have a positive impact on the R rate.

    In the short term, we recommend use of either a custom face covering or a disposable surgical mask.

    In the medium to long term, we recommend the government commits to onshoring the manufacturing of both disposable medical masks and high-grade N95 masks with an ambition to develop enough stock to equip the entire nation. We have discussed this with several global manufacturers supplying the UK but with no manufacturing base here, and there is huge interest in doing so. 

    One major international mask and PPE manufacturer told us they are ready and willing to manufacture in the UK. They could produce 200+ million medical-grade masks per month starting in just four months and millions of N95 masks per month by the end of the year. These would be manufactured on British soil. Engagement must start now, and the government should be prepared to commit to advance-purchasing contracts over several years.

    In parallel, the government must explore all possible options to find ways to enhance the protective capability of masks that are accessible and work with innovators to scale the manufacture and distribution of these. One example we cite is “Seal-A-Mask” by Prestige Ameritech, which provides an add-on to a basic surgical mask to bring it up to close to the level of an N95 mask. This innovation was detailed in a previous paper.

    The Role of Technology in Containment

    The Role of Technology in Containment

    The lack of agility in the British response to date has been due to a long-standing failure to reform government for the 21st century. However, many barriers have been eroded as the result of this crisis and NHSX is beginning to make headway that would have previously been unimaginable. There will be a threshold in the capacity of the system to absorb too much, too soon, but it has underlined the need for far greater focus on technology across the spectrum. 

    As we set out in a previous papers, there are a number of domains where this can happen in fighting the virus and in cushioning the impact. Across all of these, three principles should guide government policy: 

    1. Turn the networked public into an advantage
    2. Lean in to innovation and experimentation
    3. Be more transparent with the public 

    Rather than think of technology as a series of individual programmes, government needs to think in a unified manner, which essentially encompasses a new theory of state. If not, it will not be able maximise the potential interdependencies and opportunities of using technology in tandem. Key to this will be building the internet-era infrastructure, which can aggregate information and provide real-time actionable insights to government. There have been positive steps in this regard, such as the NHS Platform and the Joint Biosecurity Centre, but this needs to go further. Without a clear evidence base and understanding, the only viable policy option has been lockdown. Opacity and a lack of containment measures has meant that other choices are essentially a gamble, a Russian roulette, compounded by states’ inability to adequately prepare. 

    Building a dynamic rather than static comprehension of what is happening requires a reset to the operating model of government, with technology at the heart of it. System updates need to be regular and based on feedback loops with the user. It will also allow the government to deliver operational effectiveness at scale. 

    In particular, it can accelerate the hard science, such as the delivery of rapid at-home tests including those that operate earlier in the viral cycle than PCR or antibody. Initiatives such as the OpenSAFELY platform and further efforts to collect genetic sequencing have the potential to increase the rate of drug discovery and efficacy. Software systems can also support automated and manual contact tracing, while a decentralised, private and secure digital identity can underpin everything.

    This will require government to work closely with partners at the forefront of technological development, seeking out and adopting the best solutions that the market has to offer. There is a huge collective will to solve the crisis right now, and innovation is going to be fundamental. The government needs to be less risk-averse, less concerned with politics, and driven entirely by purpose. That means a relentless focus on building a new architecture for government delivery, which removes friction and is mission-driven. The aim is to modernise, to address inequalities, and to harness the power of new technologies to advance the greater good.

    Conclusion: Reactive to Proactive

    Conclusion: Reactive to Proactive

    We need to achieve two things. First, at any point in time we need the most coherent combination of measures (the balance of these will change over time as capabilities evolve). Second, over time we need to achieve the greatest possible scale. If we achieve both, then our overall stance shifts from reactive to proactive, and government will retake the initiative.

    What needs to happen:

    1. Allocate a specific government fund for advance purchase of rapid tests, masks, wearables, pulse oximeters, etc. 
      By contracting with companies to purchase large quantities at pre-agreed prices, the government creates a strong incentive for companies to move fast and scale aggressively. To support this, the government would underwrite impact bonds issued by companies to private investors to fund the ramp up of R&D and production. The only downside risk is over-incentivising production, but at the scale we are talking about the most likely outcome is still being gated by supply.
    2. Establish a containment delivery unit to optimise overall performance of the containment portfolio. 
      This requires a wholesale re-orientation of the systems of government in order to move quickly and deliver at scale. The unit should take a technology-enabled approach to ensuring all aspects of the response operate efficiently at scale and in response to real-time data – everything from achieving maximum utilisation of volunteer contact tracers through to ensuring tests get to the places where they are most needed. Where necessary it should invest in infrastructure, e.g. distribution hubs or thermal imaging networks. It should report directly to the PM and key areas would be headed up by a senior minister (for instance, The Minister for Testing).
    3. Build and deploy new digital identity infrastructure.
      This infrastructure would be for the express purpose of supporting the containment effort. Existing NHS apps for services such as symptom tracking and requesting a test should continue, but separately the government should at least have the option to issue secure credentials so that in specific, appropriate settings such as airports people can credibly assert that, for example, they have had Covid-19 or have tested negative in the last 24 hours. For citizens this means showing a verified QR code, likely secured by biometrics, while for businesses this should provide reassurance that certain environments are safe. Three critical guardrails must underpin government’s approach to secure public trust:
      1. Strict separation with proximity-based contact tracing and other digital surveillance (so switch to Apple/Google for that and do in smartphone operating system)
      2. Primary legislation to specify when credentials can be asked for (if/where mandatory, governments should be clear about this extraordinary measure and provide opportunities for robust scrutiny and accountability)
      3. A viable alternative for people who don’t have a smartphone, who still need access to services

    A shift from reactive to proactive government will allow the UK to build a gold standard of infrastructure. Unprecedented, this operation would mark the biggest peacetime mobilisation of people, technology and resources in our country’s history. Such scale should not intimidate but inspire. It’s imperative that we set out clearly what is needed over the months and possibly years ahead, galvanising public and political will to turn that vision into reality. As unlikely as it may seem at this moment in time, the UK really could lead the way in building the architecture of containment to defeat Covid-19. 

    Find out more
    institute.global