Yesterday, the UK government announced it was moving into the delay phase of the national plan for managing the COVID-19 outbreak. Unlike in many—but not all—other European countries, schools across the UK will remain open. Scotland has banned large gatherings of over 500 people “to free up emergency services, including police and ambulance crews, to deal with the coronavirus outbreak,” but Scottish schools will remain open for now. Communicating to the public and media what evidence underlies the government’s decisions on these tough calls will be critical to maintaining public trust in government decisions in the difficult weeks ahead.
The UK government’s position, so far
The UK government’s approach has been criticised by many, including the former health secretary and the editor of the Lancet (via twitter), amongst others. The government has defended its position with the country’s most senior medical and scientific advisors—the chief medical officer and the chief scientific advisor—claiming that big gatherings are not that critical in disease transmission, and that closing schools would negatively affect the National Health Service by pinning down healthcare worker parents whose kids would have to be at home. Additionally, the government cites behavioural science that suggests people tire of overly restrictive measures and may start to ignore advice. They urge that the timing of restrictive measures is crucial to their success.
While not well enough communicated, the Government seems to be saying that more draconian social distancing measures will materialise eventually but are not high-impact now. Already the government signalled it will be asking the elderly to self-isolate and it will ban gatherings in due course. Earlier today it announced the cancellation of local elections. And perhaps by the time this blog is posted, it will have moved to emulate the example of most of its European neighbours.
At 798 cases and 11 deaths (as of 13 March) the UK has fewer cases than similarly sized European countries such as Spain, Germany, and France and even much smaller ones such as the Netherlands and Denmark. The consensus is, however, that the real number of infected individuals is much higher. At the same time, the country’s NHS is scaling up testing to up to 10,000 per day, a significant expansion which will give experts a better idea of actual numbers of the infection in the community and help better tailor the response. At over 25,000 tests (as of 11 March) the UK seems to be performing well against other EU countries and the US, where the ability to test has been severely undermined by a fragmented private insurance dominated system, supply shortages, and lack of coordination.
Is the government doing the right thing—or are critics correct in calling for more aggressive measures?
What is the evidence underpinning the UK government’s response, and could the government communicate the tough trade-offs better to the general public and its critics? The short answer is, we don’t really know. For instance, the evidence on the impact of school closures on the outbreak trajectory is limited and conflicting. While some are in favour, others cite evidence closures can do more harm than good. This lack of clarity is reflected perhaps in the varied response across the EU, with most countries opting for full or regional closures and others—such as the UK, the Netherlands, Romania, Finland, and Sweden—opting for keeping their schools open and running for now. In Asia, Singapore was widely praised for its COVID-19 response and is currently considering implementing school closures—however, it has not yet done so given the wider implications of such a move. As our colleagues at CGD have argued, timing is of the essence, as is considering the full costs of shutting down schools for the children and society.
What does the evidence on effectiveness and cost-effectiveness of alternative interventions to tackle pandemics suggest?
Again, things are hardly clear-cut. First, there is very limited evidence to start with. In considering both (net) costs and benefits of alternative options, we would expect governments to use a health technology assessment framework (akin to what NICE uses) in deciding on priority measures as part of their emergency response plan. Yet a quick search of the scientific literature yields fewer than 50 studies from the past 10 years mentioning HTA or priority setting in relation to outbreaks such as Ebola, Zika, SARS, or flu. None of them considered economic outcomes. A relatively recent systematic review of interventions for pandemic influenza (H1N1)* suggests school closures are not good value for money—a function of how well they work and what their net economic cost is. This conclusion would be reversed for a disease with high mortality rates amongst children, however COVID-19 does not seem to be causing severe disease amongst children nearly as much as H1N1. So far fewer than 0.9 percent of cases are children, and there have been no fatalities in children under the age of 10.
Writing in the Lancet last week, a group of UK and Dutch researchers concluded that “school closure, a major pillar of the response to pandemic influenza A, is unlikely to be effective given the apparent low rate of infection among children, although data are scarce.” They also highlighted the importance, at least “in western democracies” of individual behaviour and personal as opposed to government action as more definitive factors in how the outbreak develops.
What is the role of the global health community in such situations?
Given the overall uncertainty and significant economic costs of any government decision, we can’t say whether following in the footsteps of most European countries is the right or wrong move. Can we use the analytical tools in our disposal to assess what is currently happening and what is (or is not) working in the UK and overseas in order to course-correct, including through dynamic modelling exercises which are currently informing government decisions? Can we make sure, if this is not the case already, that such models build in the economic and social aspects of any measures from the outset as opposed to epidemiology alone as it has done before (e.g. see here for the modelling approach underpinning NICE guidance for mad cow disease) through an open evidence and data driven process, albeit not in a pandemic crisis setting? And can government and its advisors try and quantify uncertainty and the cost of getting it wrong given the high stakes, and attempt to communicate all this (however hard) to the press and the people?
At a time of great uncertainty, it is essential for the government to communicate the evidence that supports its plan. Claiming evidence underpins the government’s actions is important and somewhat reassuring; better communicating this evidence and the uncertainty surrounding it together with the trade-offs and tough choices facing the government and each one of us in how we go about our daily routines, is also of the essence to maintain public trust and support. In the battle against COVID-19, the government’s ability to communicate and explain the reasoning behind its decisions may end up being even more important than the limited and uncertain evidence driving the decisions.
*Another, older analysis of influenza-targeting interventions is inconclusive though seems more favourable towards school closures as part of a combined response with other clinical and social distancing measures.