Category Archives: COVID-19

1. The need to ramp up testing (for many purposes)

The need to ramp up testing is a recurring theme, in which respondents describe low capacity in the beginning, and continuous problems with ‘ramping up’ capacity, then reflect on the difference it could have made in key areas (during more reflective discussions in June).

The figures reported in oral evidence were 2000 per day (Harries, Deputy Chief Medical Officer, 5.3.20: q54), 4000 (Vallance, Chief Scientific Advisor, 17.3.20: q78-84), 7000 (Doyle, Medical Director, Public Health England, 26.3.20: q196), and 50000 (Hancock, Secretary of State, 17.4.20: q312), but increasingly in the absence of a definition of testing, which became important when the UK government began to treat all tests sent out as part of meeting its 100000 per day target.

In March, Vallance (17.3.20: q78-84) described capacity as 4000 reliable tests per day, and noted the lack of accuracy of the larger-scale of tests available in the market, to argue that the UK was one of the most frequent testers at the time. Such arguments dated very quickly. Hancock (17.4.20: q306) then describes ‘ramp-up testing’ as the fourth element of the government’s ‘battle plan’ but often with low clarity on how, and who would be responsible. This confusion is apparent when Doyle (26.3.20: q196) describes separate responsibilities, in which PHE would be responsible for testing NHS staff and patients (target 25000 tests per day), while the Office for Life Sciences would take forward the 100000 tests per day pledge.

In most cases, there is broad agreement on the negative impact of the limited testing capacity, including:

Surveillance and modelling

Vallance (17.3.20 and 5.5.20: q435) reflects on the need for more testing to aid initial surveillance, suggesting that they could have acted very differently if the testing capacity was higher. This issue will run and run. The absence of data affected the ability of advisory bodies to model the estimated ‘peak’ of infection (based partly on its doubling-rate) that became so central to the UK government’s initial lockdown strategy. Further, there is much debate on the adequacy of UK modelling. For example, listen to the Radio 4 series More or Less (10.6.20, ‘Antibody tests, early lockdown advice and European deaths’), which argues that, during the 1st press UK government conference (12th March), Vallance misjudged the UK as being 4- rather than 2-weeks behind Italy. Compare with Vallance (5.5.20: q390).

NHS staff and services

NHS bodies describe their reduced ability to operate effectively (throughout the session on 17.3.20 which included Stevens, chief executive of NHS England). Nagpaul (British Medical Association) (26.3.20: q203-4) describes the combination of (a) advice to self-isolate with symptoms, and (b) an absence of testing, as a potential cause of a 10% shortage of NHS staff during the crisis, and this problem is a continuous theme in this day’s evidence. Green (Care England) (26.3.20: q267) and Bullion (Vice-President, Association of Directors of Adult Social Services) (26.3.20: q279) make the same case for a social care sector already at low capacity. Kinnair (Royal College of Nursing) and Pittard (Faculty of Intensive Care Medicine) describe low testing (and limited PPE) as major worries for staff (17.4.20: q297)

Discharges to care homes

One key example – still to be explored fully – is the absence of routine testing of NHS patients during the push to discharge 15000 people from hospital to social care beds in England (Stevens, 17.3.20: q122-3; Green, 26.3.20: q274). Green (19.5.20: q470 and q478) notes that the UK Government prioritised the NHS at the expense of social care, prompting NHS discharges to care homes before proper testing was in place, while knowing that care homes are ill-suited to isolation measures. Note that the NHS was already under capacity pressure before the crisis (Stevens, 17.3.20: q165), and redeploying medical and nursing care from care homes, while Willet (Director for Acute Care, NHS England 17.3.20: q165) describes an already fragmented system of 12500 care homes in England.

Overall, the absence of sufficient information – from routine testing for the virus, and proper analysis of care home capacity – combined with a huge drive to favour NHS care and move people to care homes, contributed to a disproportionately large coronavirus problem in care homes.

This experience compares with many other countries that addressed care homes more effectively. In Germany, patients were not discharged to care homes unless they could quarantine (Halletz, Chief Executive Officer, AGVP (Employers’ Association – Care Homes) 19.5.20: q455). In South Korea, people were taken from care homes to be quarantined (Comas-Herrera, Assistant Professorial Research Fellow, Care Policy and Evaluation Centre, LSE, 19.5.20: q447).

Committee update, 12th June 2020: Figures confirming discharges of hospital patients into care homes, responding to the National Audit Office report Readying the NHS and adult social care in England for COVID-19. The NAO press release states:

‘Patients discharged quickly from hospitals between mid-March and mid-April were sometimes placed in care homes without being tested for COVID-19. On 17 March, hospitals were advised to discharge urgently all in-patients medically fit to leave in order to increase capacity to support those with acute healthcare needs. Between 17 March and 15 April, around 25,000 people were discharged from hospitals into care homes, compared with around 35,000 people in the same period in 2019. Due to government policy at the time, not all patients were tested for COVID-19 before discharge, with priority given to patients with symptoms. On 15 April, the policy was changed to test all those being discharged into care homes. It is not known how many patients discharged to care homes had COVID-19 at the point they left hospital’

Testing and contact tracing

The issue of testing in relation to initial contact tracing is less straightforward. On several occasions, PHE and health department respondents note that there came a point when the number of infections, and rate of infection, ruled out the effectiveness of contact tracing and testing in favour of UK-wide lockdown measures: ‘several weeks ago that ship had begun to sail. Nevertheless, we kept going until mid-March, until we were absolutely sure that contact tracing in that way would not work’ (Doyle, 26.3.20: q198; see also q232 on testing representing only one of many necessary measures).

There is a continuous discussion in multiple sessions on why the UK stopped contact tracing after it became clear that the rate of transmission was very high (e.g. Harries 5.5.20: q415 on the shift from ‘contain’ to ‘delay’ on 12th March, q418 on using limited tests in hospitals where most needed, and q425-8).

Costello (Professor of Global Health and Sustainable Development, UCL, 17.4.20: q303-4) agrees with this approach in relation to London and several other cities where prevalence and transmission were unusually high, but argues that in many places there were very few cases (fewer than 10 cases in 50 local health areas until mid-March) and that they could have been contained. This issue will run and run too (see the discussion on contain/ delay in the run up to the first peak).


Filed under COVID-19

COVID-19 policy in the UK: oral evidence to the Health and Social Care Committee (5th March- 3rd June 2020)

This series of posts describes the key themes and issues to arise from oral evidence to the House of Commons Health and Social Care Committee on COVID-19. It is the first committee on my to-do list.

When possible, I have (or will) connect them to some other sources of information, such as the minutes from NERVTAG, the not-yet-read-by-me minutes by SAGE, and the 8000-word paper that I am writing (which is currently 20000 words, and based initially on this unwieldy blog post). The result is a very long read, which I have broken down into a collection of 9 reads. One unintended consequence is that you may not see a respondent’s full title in some posts, because originally I only listed it the first time on the full document. Fortunately, there is a simple solution: read all the posts from 1-8.

Two issues often seem to dominate the oral evidence to the Health and Social Care Committee in multiple sessions from March to June 2020:

  1. Limited testing: antigen testing to detect the virus now, then antibody testing to detect if someone had COVID-19 in the past.
  2. Shortages of personal protective equipment (PPE), initially for NHS staff, followed by concerns about availability in social care and other sectors.

These issues connect to a series of knock-on issues, such as the discharge of patients from NHS hospitals to care homes without being tested.

They also intersect with broader policy themes which include how to:

  1. Define coronavirus as a policy problem, such as with reference to the oft-abused phrase ‘herd immunity.
  2. Act despite uncertainty, or a lack of information on which to give advice and make choices.
  3. Define different stages of intervention, including contain, delay, research, mitigate, and suppress.
  4. Describe the relationship between science advice and policy, to project the sense that policy is evidence-informed but that elected politicians are responsible for choice.
  5. Identify the many changes to policy and practice that would otherwise receive minimal attention (in other words, they are low salience but high importance).
  6. Address the links between health inequalities and race and ethnicity.

These sessions generally relate to activity for England, but with few indications that the actions or issues are markedly different in Northern Ireland, Scotland, or Wales. Indeed, (a) there is frequent reference to UK-wide cooperation and coordination, and (b) issues such as NHS hospital discharges to care homes without testing or quarantine measures seem UK-wide (albeit with variations in practice). A proper focus on devolved government is also on the to-do list.

PS I also left out some issues because they seemed unresolved by June:

  • Test, track, and trace (Hancock, 17.4.20: q325-7; Vallance and Harries, 5.5.20: q425-8; Chen, 3.6.20: 492-504; Fraser, Professor of Pathogen Dynamics, University of Oxford and Harding, Executive Chair of NHS Test and Trace programme, 3.6.20: 510-52)
  • When to have border restrictions (Cooper, 17.4.20: q344)
  • Testing for a vaccine (Van-Tam, Deputy Chief Medical Officer, 17.4.20: q366)
  • Who to learn from, in relation to comparability (Vallance, 5.5.20: q435; see also the dedicated session 19.5.20 on South Korea, Hong Kong, Germany).

The full series of posts:

  1. The need to ramp up testing (for many purposes)
  2. The inadequate supply of personal protective equipment (PPE)
  3. Defining the policy problem: ‘herd immunity’, long term management, and the containability of COVID-19
  4. Uncertainty and hesitancy during initial UK coronavirus responses
  5. Confusion about the language of intervention and stages of intervention
  6. The relationship between science, science advice, and policy
  7. Lower profile changes to policy and practice
  8. Race, ethnicity, and the social determinants of health


Filed under COVID-19

Summary of NERVTAG minutes, January-March 2020

NERVTAG is the New and Emerging Respiratory Virus Threats Advisory Group, reporting to PHE (Public Health England).

It began a series of extraordinary meetings on the coronavirus from 13th January 2020 (normally it meets once per year), summarized in Table 1.

In January, it agreed with PHE that the risk to the UK population was ‘very low’, rising to ‘low’ (by this stage, the rate of human-to-human infection was unclear). It focused primarily on (a) developments in the city of Wuhan (population: 11m) and then other parts of China, and (b) advice to UK travellers to China, then (c) giving advice for the NHS on how to define a case of COVID-19 in relation to symptoms (primarily fever) and a history of travel to an affected area. From the end of January, it began to discuss personal protective equipment (PPE) frequently, without describing the need to modify PHE advice significantly (and was not responsible for securing supply).

In February, it agreed (on the 21st) that the risk to the UK population was ‘moderate’. It responded to questions from COBR (Cabinet Office civil contingencies committee, convened to discuss national emergencies) on the most effective public preventive efforts, prioritizing frequent and effective hand washing and advising against face masks for members of the public with no symptoms. In response to questions from the Department of Health and Social Care (DHSC), it described a ‘Reasonable Worst Case’ in the UK (to inform scenario modelling) as an 85% infection of the population, with half of those affected showing symptoms, then suggested that an estimate of 4% (of those with symptoms) needing hospital care ‘seems low’, while 25% (of the 4%) requiring respiratory support ‘seems high’.

In March, it advised that voluntary self-isolation should be 7-14 days after ‘illness onset’, depending ‘on desired balance between containment and social disruption at the particular stage of the epidemic’. It should be longer during the ‘containment’ phase (‘In the current situation NERVTAG would prefer this period to be towards the longer end of the range’) but could be shorter when transmission is so widespread that someone infected represents a decreasing share of the infected population (‘an increased proportion of people may still be infectious when they end self-isolation but they will constitute a decreasing proportion of all infectious people’, 6.3.20: 2).

Throughout, members of NERVTAG focused quite heavily on what seemed feasible to suggest, informing initial thoughts on:

  1. Handwashing advice. Initially it warned against too nuanced messages to the public, such as on the amount of time to wash.
  2. Face mask use. It identified (in multiple discussions) the unclear benefits if someone is well, plus the unlikely widespread public compliance, coupled with limited public training in their hygienic use and disposal (and the possibility that mask use in the UK ‘may add to fear and anxiety’ – 28.1.20: 8)
  3. Voluntary self-isolation. It expressed uncertainty about public compliance, and the difficulty of knowing when the illness begins and infectiousness ends.
  4. Port of entry screening, assuming a low impact since it would miss most cases.

[Note: please use the PDF if the tables look a bit weird below]

NERVTAG table 1aNERVTAG table 1b


Filed under COVID-19, Uncategorized

Coronavirus and the ‘social determinants’ of health inequalities: lessons from ‘Health in All Policies’ initiatives

Many public health bodies are responding to crisis by shifting their attention and resources from (1) a long-term strategic focus on reducing non-communicable diseases (such as heart diseases, cancers, diabetes), to (2) the coronavirus pandemic.

Of course, these two activities are not mutually exclusive, and smoking provides the most high-profile example of short-term and long-term warnings coming together (see Public Health England’s statement that ‘Emerging evidence from China shows smokers with COVID-19 are 14 times more likely to develop severe respiratory disease’).

There are equally important lessons – such as on health equity – from the experiences of longer-term and lower-profile ‘preventive’ public health agendas such as ‘Health in All Policies’ (HIAP).*

What is ‘Health in All Policies’?

HIAP is a broad (and often imprecise) term to describe:

  1. The policy problem. Address the ‘social determinants’ of health, defined by the WHO as ‘the unfair and avoidable differences in health status … shaped by the distribution of money, power and resources [and] the conditions in which people are born, grow, live, work and age’.
  2. The policy solutions. Identify a range of policy instruments, including redistributive measures to reduce economic inequalities, distributive measures to improve public services and the physical environment (including housing), regulations on commercial and individual behaviour, and health promotion via education and learning.
  3. The policy style. An approach to policymaking that encourages meaningful collaboration across multiple levels and types of government, and between governmental and non-governmental actors (partly because most policy solutions to improve health are not in the gift of health departments).
  4. Political commitment and will. High level political support is crucial to the production of a holistic strategy document, and to dedicate resources to its delivery, partly via specialist organisations and the means to monitor and evaluate progress.

As two distinctive ‘Marmot reviews’ demonstrate, this problem (and potential solutions) can be described differently in relation to:

Either way, each of the 4 HIAP elements highlights issues that intersect with the impact of the coronavirus: COVID-19 has a profoundly unequal impact on populations; there will be a complex mix of policy instruments to address it, and many responses will not be by health departments; an effective response requires intersectoral government action and high stakeholder and citizen ownership; and, we should not expect current high levels of public, media, and policymaker attention and commitment to continue indefinitely or help foster health equity (indeed, even well-meaning policy responses may exacerbate health inequalities). 

A commitment to health equity, or the reduction of health inequalities

At the heart of HIAP is a commitment to health equity and to reduce health inequalities. In that context, the coronavirus provides a stark example of the impact of health inequalities, since (a) people with underlying health conditions are the most vulnerable to major illness and death, and (b) the spread of underlying health conditions is unequal in relation to factors such as income and race or ethnicity. Further, there are major inequalities in relation to exposure to physical and economic risks.

A focus on the social determinants of health inequalities

A ‘social determinants’ focus helps us to place individual behaviour in a wider systemic context. It is tempting to relate health inequalities primarily to ‘lifestyles’ and individual choices, in relation to healthy eating, exercise, and the avoidance of smoking and alcohol. However, the most profound impacts on population health can come from (a) environments largely outside of an individual’s control (e.g. in relation to threats from others, such as pollution or violence), (b) levels of education and employment, and (c) economic inequality, influencing access to warm and safe housing, high quality water and nutrition, choices on transport, and access to safe and healthy environments.

In that context, the coronavirus provides stark examples of major inequalities in relation to self-isolation and social distancing: some people have access to food, private spaces to self-isolate, and open places to exercise away from others; many people have insufficient access to food, no private space, and few places to go outside (also note the disparity in resources between countries).

The pursuit of intersectoral action

A key aspect of HIAP is to identify the ways in which non-health sectors contribute to health. Classic examples include a focus on the sectors that influence early access to high quality education, improving housing and local environments, reducing vulnerability to crime, and reforming the built environment to foster sustainable public transport and access to healthy air, water, and food.

The response to the coronavirus also appears to be a good advert for the potential for intersectoral governmental action, demonstrating that measures with profound impacts on health and wellbeing are made in non-health sectors, including: treasury departments subsidising business and wages, and funding additional healthcare; transport departments regulating international and domestic travel; social care departments responsible for looking after vulnerable people outside of healthcare settings; and, police forces regulating social behaviour.

However, most (relevant) HIAP studies identify a general lack of effective intersectoral government action, related largely to a tendency towards ‘siloed’ policymaking within each department, exacerbated by ‘turf wars’ between departments (even if they notionally share the same aims) and a tendency for health departments to be low status, particularly in relation to economic departments (also note the frequently used term ‘health imperialism’ to describe scepticism about public health in other sectors).  Some studies highlight the potential benefits of ‘win-win’ strategies to persuade non-health sectors that collaboration on health equity also helps deliver their core business (e.g. Molnar et al 2015), but the wider public administration literature is more likely to identify a history of unsuccessful initiatives with a cumulative demoralising effect (e.g. Carey and Crammond, 2015; Molenveld et al, 2020).  

The pursuit of wider collaboration

HIAP ambitions extend to ‘collaborative’ or ‘co-produced’ forms of governance, in which citizens and stakeholders work with policymakers in health and non-health sectors to define the problem of health inequalities and inform potential solutions. These methods can help policymakers make sense of broad HIAP aims through the eyes of citizens, produce priorities that were not anticipated in a desktop exercise, help non-health sector workers understand their role in reducing health inequalities, and help reinforce the importance of collaborative and respectful ways of working.

An excellent example comes from Corburn et al’s (2014) study of Richmond, California’s statutory measures to encourage HIAP. They describe ‘coproducing health equity in all policies’ with initial reference to WHO definitions, but then to social justice in relation to income and wealth, which differs markedly according to race and immigration status. It then reports on a series of community discussions to identify key obstacles to health:

For example, Richmond residents regularly described how, in the same day, they might experience or fear violence, environmental pollution, being evicted from housing, not being able to pay health care bills, discrimination at work or in school, challenges accessing public services, and immigration and customs enforcement (ICE) intimidation … Also emerging from the workshops and health equity discussions was that one of the underlying causes of the multiple stressors experienced in Richmond was structural racism. By structural racism we meant that seemingly neutral policies and practices can function in racist ways by disempowering communities of color and perpetuating unequal historic conditions” (2014: 627-8).

Yet, a tiny proportion of HIAP studies identify this level of collaboration and new knowledge feeding into policy agendas to address health equity.

The cautionary tale: HIAP does not cause health equity

Rather, most of the peer-reviewed academic HIAP literature identifies a major gap between high expectations and low implementation. Most studies identify an urgent and strong impetus for policy action to be proportionate to the size of the policy problem, and ideas about the potential implementation of a HIAP agenda when agreed, but no studies identify implementation success in relation to health equity. In fact, the two most-discussed examples – in Finland and South Australia – seem to describe a successful reform of processes that have a negligible impact on equity.  

A window of opportunity for what?

It is common in the public health field to try to identify ‘windows of opportunity’ to adopt (a) HIAP in principle, and (b) specific HIAP-friendly policy instruments. It is also common to try to identify the factors that would aid HIAP implementation, and to assume that this success would have a major impact on the social determinants of health inequalities. Yet, the cumulative experience from HIAP studies is that governments can pursue health promotion and intersectoral action without reducing health inequalities.

For me, this is the context for current studies of the unequal impact of the coronavirus across the globe and within each country. In some cases, there are occasionally promising discussions of major policymaking reforms, or to use the current crisis as an impetus for social justice as well as crisis response. Yet, the history of the pursuit of HIAP-style reforms should help us reject the simple notion that some people saying the right things will make that happen. Instead, right now, it seems more likely that – in the absence of significantly new action** – the same people and systems that cause inequalities will undermine attempts to reduce them. In other words, health equity will not happen simply because it seems like the right thing to do. Rather, it is a highly contested concept, and many people will use their power to make sure that it does not happen, even if they claim otherwise.

*These are my early thoughts based on work towards a (qualitative) systematic review of the HIAP literature, in partnership with Emily St Denny, Sean Kippin, and Heather Mitchell.

**No, I do not know what that action would be. There is no magic formula to which I can refer.

See also: Tired of science being ignored? Get political by @DrMaryTBassett


Filed under COVID-19, Prevention policy, Public health, public policy