Home page







Rory Stewart
The government has not shared with the public any information on the epidemiological models being used to interpret the "evidence" and justify their applying "the science" to the UK Covid-19 outbreak. We therefore have to assume they are applying standard epidemiological models. These are normally somewhat simplistic. They generate projections of likely levels of infection based on an infection rate and size of the population. This provides an estimate of the inoculum potential or likely levels of cross infection or transmission. Fine. But such models provide little guidance on how to manage the epidemic in the interests of the population.

Rory Stewart has highlighted this by pointing out government inaction on some important needs.

Some aspects of the problem surfaced in the BBC Question Time moderated by Fiona Bruce on Thursday 12th March. The government representative was Steven Barclay, Secretary to the Treasury and the person raising some relevant issues was Professor John Ashton. Unfortunately the blind explanation that the government is "following advice" does not get to grips with informing the public what in fact is happening. In response to Professor Ashton raising the issue of allowing sports events to continue as well as lack of preparation for the dangers to old folk siloed in old people's homes, Steven Barclay's response was:

"There's an agreed position on the science and it is the science, it is the view of our top scientists across all four countries of the UK that is driving the position. And therefore respectfully to John, he is at odds with the leading scientists of the UK."

First of all Professor Ashcroft is not at odds with leading experts and what Barclay appeared to mean was that those individuals the government happens to be relying on for advice, appear to concur with the current government decisions on the matter. However, this was also an indication that Barclay was not listening to what Ashton was stating and preferred to apply a diplomatic put-down. Given the nature of the issue facing the country this is an unacceptable as it is a worrying demonstration of the government's mindset.

There is a need to separate advice from decisions. If advice is poor, decisions can be disastrous. Good governance should involve political decision-makers who never slavishly follow advice but weigh up the options based on their understanding of the expert's assumptions, common sense and prudence. If advice is good, decisions shaped in the light of that advice can also be disastrous because the advice depends upon the ability of the models use to emulate reality and generate evidence upon which to base that advice.

All decision analysis models and attempts to emulate reality. Models generate the outputs of interest to decision-makers. Thus depending upon the information output the government has asked for, so models will take different forms unless the models are extremely sophisticated and able to handle any disease and social structure; most unlikely. What Professor Ashton appeared to be referring to was the social aspects of the modeling which can help them become more useful in practical terms.



The main weaknesses in standard models is the absence of information on population clusters within varying population densities i.e. comparing rural with urban environments. Clusters are simply concentrations of people in close proximity where the likelihood of disease transmission is more likely.

The impact of clusters

There has been a news black out on the likely cause of the Northern Italian Covid-19 catastrophe. However, the Venice Carnival went ahead for "commercial reasons" when knowledge of the risks of Covid-19 were known.

There is a high likelihood that the main epicentre of the North Italian case was the result of such transitory clusters which should have been banned.

There is an unfortunate state of affairs that governments are under the control of political parties and these tend to favour lobby interests in their own self-interests over those of the general constituency. Unexplained delays in closing down social events by the UK government is however justified on the basis of following the "science" and "best advice"; this is far from convincing.
There are two types of cluster
  • permanent
  • transitory
Permanent clusters include all groups that do not change such as:
  • tiny clusters of families who remain voluntarily house bound
  • clusters who by the nature of their occupation are in regular contact with individuals not of the cluster
  • large clusters, such as old people's homes and facilities
Tiny house bound clusters are likely to have negligible impact on growth in infections, indeed they are the means of running the rate of inflection down.

Occupational clusters are the staff of such clusters as the NHS, supermarkets, bars, shops and services. Here the main infection risk is to the staff because every member of the public is a potential hazard just as an inflected member of staff is a hazard to those they encounter from outside the cluster in their as part of their normal activity.

However, large permanent clusters such as old people's homes are exceptionally vulnerable because any person within these communities who picks up the disease is likely to infect the rest. With most being in excess of 70 years of age, many are likely to die. So, in this case, the rate of infection, as far as the population is concerned, would be marginal but the consequences of the high death rate would be terrible for those involved and their extended families. This sort of tragedy is surely something society and governance should work to avoid.

Transitory clusters are those that form on a regular basis either daily or weekly or whatever. These include sports events, meetings, entertainment involving stages and audiences, kindergartens, school and university attendance and regular work places as offices and factories. The problem with these types of cluster is that those who attend them return, normally, to their families after having been at the cluster. This means disease transmission is not only more likely within clusters it is also more likely to be transferred back to the home environment and home environments, if they are the source of inflection, can transmit the disease to a cluster via the individual who is a part of that cluster.

Herd immunity

Herd immunity is a concept that assumes that as more people become immune as a result of exposure the number of active transmission links fall so the inoculum potential is reduced leading to the protection of the rest of the population, including those who are not immune.


There has been much reference to the need to build up within the population "herd immunity" (see box on right). The degree to which this is developed depends entirely on the characteristics of the pathogen and there is insufficient experience with Covid-19 to assert this can be relied upon. Many corona-type viruses either mutate of are followed by variants and this is why the content of flu jabs need to be adjusted each year. Herd immunity depends on the whether people do in fact become immune and on whether or not the virus is stable (i.e. doesn't mutate or have variants). So far Covid-19 appears to be stable but this is early days in terms of experience with this virus. It is, therefore, not productive to assert that herd immunity can be relied upon.

The process of building up any herd immunity, if it is feasible in this case, risks the exposure and deaths of the more vulnerable in our society including the elderly and those with other health conditions.



Effective governance involved weighing up the social consequences of an epidemic such as this and on the question of clusters and the notion of herd immunity it is best to apply logic to reduce the risks of infection to spare the vulnerable, not over-expose and over-work medical practitioners in the health service and to avoid a fatalistic approach of allowing nature to take it's course based on models that no one has been able to see. The objective of the government should be to save lives and be seen to be doing so.

It is already late. Things will get worse than expected. So all permanent clusters need to be protected (old people's homes) through specific precautions but also by taking action to reduce the levels of inflection in the general population rather than risking the herd immunity theory that might not work. This means clusters involving public events of any kind need to be banned and schools and universities need to be closed, for now. Not to do so shows a lack of goverment support for the imperative of valuing and saving lives.

20200314