Written Evidence Submitted by Dr G Barr

(C190056)

“Too little too late”   -  The epitaph for SAGE and advisory groups concerning COVID-19 .

Summary

This  information predominantly relates to  evidence available in March  around  use of face masks in the general population; divided into sections: background , evidence for general use of masks , preventing spread, medical  press articles, modelling, risk in community and need for better quality masks, public health information and compliance, conclusion and  recommendation for the future.  Two incidental areas are mentioned concerning SAGE’S decision making ability concerning anosmia and the statement made twice that efforts to completely suppress the virus would lead to a second wave.  For each section relevant questions that should be  asked are in red with supplementary information in blue. 

 

Introduction

I am submitting evidence  primarily in relation to  the scientific evidence for the use of face masks in the general population. However, I will touch on other areas of concern  as a result of reading the published minutes of SAGE.

I write with a background of 34 years of clinical experience working as an ENT surgeon, having  worked with the clinical  risk of viruses such as  hepatitis, HIV, swine flu , SARS and  prion CJD , including  laser surgery where  is a risk of  papilloma virus infection, requiring FFP3 masks with a 0.1 micron filtration. I  was extremely concerned  by the policy  produced in relation to  general non-use of masks  from the beginning of March which seemed to defy logic and have published two papers on why the general population  should be wearing masks, both forwarded to the Dept. of Health and Scottish government, one at the beginning of  April and the other at the beginning of May. The WHO also need to be strongly criticised  for contradicting its own 2019 document concerning  influenza  which states  regarding  influenza ,‘Face masks worn by asymptomatic people are conditionally recommended in severe epidemics or pandemics, to reduce transmission in the community.’  The government could take comfort in that they were broadly following WHO guidelines  concerning masks in the population, however, this was not  ‘following  the science.’

Reference - WHO Non-pharmaceutical public health measures for mitigating the risk and impact of epidemic and pandemic influenza . 2019. Licence: CC BY-NC-SA 3.0 IGO. https://apps.who.int/iris/bitstream/handle/10665/329438/9789241516839-eng.pdf

In the UK  compulsory face covering in public transport  was only from 15th June    which is   three months later than needed.

 

 

SAGE did not advise  all precautions and measures to protect the British population .

In the minutes there is very little  evidence of scientific discussion in the meetings for a policy that is  described as, ‘following the science.

From the minutes of SAGE meetings on two occasions, the 13th March and repeated  again on 19th, a point was made which is  of concern:

SAGE was unanimous that measures seeking to completely suppress spread of Covid19 will cause a second peak. SAGE advises that it is a near certainty that countries such as China, where heavy suppression is underway, will experience a second peak once measures are relaxed.’  

This implies that the  directions from SAGE were such that  not all possible measures  to reduce COVID-19 infections in the UK were considered or advised.

Question 1 -   What evidence is there that completely suppressing Covid-19 would cause a second wave and did this  underly the  low key approach to supressing Covid-19 in the UK ?

Question 2 -  From a policy of not fully suppressing Covid-19 how was  the  risk of excess deaths  calculated as a  trade-off compared to the theoretical possibility of a second wave 

 

If  the numbers are kept low any second wave would be controlled  more easily with general use of masks , social distancing and test and trace .

There has been no evidence of a second wave and particularly in Hong Kong , a country which has relied on social distancing and masks avoiding lock down , with 7 million population there have been 4 deaths   which had not changed  until the end of June .

In the Spanish Flu  pandemic 1918-19  which spread extensively  throughout the world , although there were some attempts at  measures such as social distancing, the virus  spread more or less unchecked  and unsuppressed  but there were 3 waves .

 

Question 3 -  Did the  statement/conclusion  concerning theoretical risk of a second wave  by too much suppression  lead to the  bizarre  policy statement that Covid-19 would be allowed to spread throughout the population  with an estimated 80% of the population being infected and resultant herd immunity ?

The only concern seemed to be of  reducing the peak , spreading the cases over a more prolonged period  so that the health service could cope .  With  a 1 % mortality this would mean approximately 500,000 deaths.  Was this policy acceptable ?

 

Question 4 – Relating to lack of suppression, was this related  to the fact that 1000’s of people were  allowed to return from endemic areas in Italy largely unchecked ?  Should  all those returning from   endemic areas  at border  control  have been supplied with  masks and asked  to self-isolate for two weeks  ?

 

ANOSMIA   -  evidence of SAGE’s inability to reach timely decisions

The first point is that the evidence shows that  SAGE  under guidance from NERVTAG appears   to have significant shortfalls   in relation to making decisions.

Regarding anosmia  ENT-UK informed the  Public Health England  that this is  a prevalent symptom  on 24th of March. The ENT-UK position paper advised that people  with anosmia should self-isolate. Anosmia  in absence of  rhinological  disease    is extremely rare .  The evidence is that  no  practical action was taken other than ‘looking into it ,’ for two months  giving the appearance of  an academic exercise.

Chief Scientific Adviser Sir Patrick Vallance admitted at the  briefing on Monday 30th March  that "Loss of taste and smell is something that can happen with other respiratory viruses as well. It does seem to be a feature of this from what people are reporting and it is obviously something that people should take  into account as they think about their symptoms." We would argue that there is now enough evidence to take this symptom much more seriously.
 

Professor Van Tam : 18th May

‘And that’s why we have taken our time in this country because we wanted to do that, again, painstaking and very careful analysis before we jumped to any conclusions. And even if it was  obvious that anosmia was part of this, we wanted to be sure that adding it to cough and fever, as opposed to just listing it, adding it in formerly into our definition, was the right thing to do. And based on advice from NERVTAG, we have made that decision.’

Question 4 - Why was SAGE’s decision making  so slow  and were they aware of the urgency?

This 2 month delay  is with the backdrop of 600 to 1000 people dying per day .  

 

It does  appear that   the advisory committees were overburdened by  scientists and academics when decisions were needed   often without any existing evidence base or protocols.

Question 5 - Should there have been more full-time actively acute practising clinicians  such as  intensive care specialists, infectious disease consultants and clinical virologists   in SAGE/NERVTAG ?

 

 

Face Masks

 

Background

It important to be clear   that  masks work and are highly effective,  otherwise they would not be used in hospitals  and hazardous environments  such as those dealing with  biological or neurotoxic agents,  noting that a respirator is simply a more advanced form of mask. Masks will act at the three  aspects of transmission  to greater or lesser degree for aerosol or airborne particles, all will reduce larger droplets, which in turn reduces contamination of fomites that subsequently  might be touched leading to infection .

Concerning aerosol spread,  this  is  a case for better quality masks, in reality  there is not a strict cut off between droplet spread and fine airborne aerosol particles  which has been known since  Wells’s classic paper from  1934 that showed that many of smaller droplets evaporate before  falling and  become airborne  droplet nuclei.  Some individuals produce high amounts of aerosol when speaking and  also breathing (which may account for super-spreading ). The  2 meters  distancing  is based on how far larger droplets can spread  and will have some effect on airborne reduction but is more limited in enclosed areas.

 

 

SAGE 28th January  minutes . Reasonable Worst-Case Scenario (RWCS)  24. There are a number of scenarios that this outbreak could follow, depending on virulence and transmissibility.  25. The current RWCS is similar to an influenza pandemic where no vaccine or specific treatment is available. 26. The RWCS for the UK should be based on a reproductive number of 2.5 (middle of current estimates) and should assume that some of those who have returned from China are infectious.  27. SAGE also agreed that the UK RWCS should be based on pandemic influenza planning.  

For other beta corona viruses,  SARS  and MERS  there were multiple cases subsequently found to be caused by  aerosol spread, which is also  the case with other respiratory viruses such as influenza. In view of this  it is reasonable to make inferences from available literature and applying this to a Covid-19  together with new information such as that  showing the wide distribution of virus in infected patients’ rooms  meaning airborne spread is likely, of  which  a preprint was  published in March.

Santarpia J L, Rivera D N, Hererra V, et al. Transmission Potential of SARS-CoV-2 in Viral Shedding Observed at the University of Nebraska Medical Centre. (accessed 2020). preprint from the University of Nebraska Medical Center.https://www.medrxiv.org/content/10.1101/2020.03.23.20039446v1.full.pdf

It is true that there is a lack of good studies  concerning public wearing of masks in epidemics and more  especially pandemics. Limitations occur such as trial design, and variation in types of mask.   Most studies concerning community wearing of have encountered problems   making  results  inconclusive including the few randomised controlled trials (RCT’s), often only addressing spread in the home setting or in residencies such as student halls, not surprisingly with compliance problems. One study crucially found a significant benefit  for those that were compliant. The overall finding is that on balance the evidence is in favour of a benefit from public wearing of masks, and was the conclusion in of the Public Health England review 2011  and a similar review by the Dept. of Health 2013,  that there is evidence with good compliance in  efficacy in health care setting and in the community. Masks in the community were not recommended, however, which presumably  is related to compliance and cost effectiveness but noting this is in relation to an influenza epidemic situation. The studies analysed were only behavioural and did not consider the substantial   laboratory  studies on efficacy.  In fact,   there was substantially more evidence for use of masks  than lockdown  in February.

Importantly  there are  differences between Covid-19 and  seasonal influenza in relation to mask wearing :  

  1. Although rapid spread of Covid-19  (SARS-CoV-2 virus) is  similar to other respiratory viruses and influenza   main  point is that the R number unchecked was 3 to 3.5  .

Estimating the Reproductive Number and the Outbreak Size of COVID-19 in Korea          Epidemiol Health 2020;42:e2020011. doi: 10.4178/epih.e2020011. Epub 2020 Mar12

 

  1. The mortality rate  for seasonal influenza  is 0.001 %  and for Covid-19  from initial reports, was 1 to 3%  which is 1000 to 3000 times more .

 

  1.   Compliance is affected by many factors   and can be increased significantly when the individual perception of risk  increases , also guided by  communicating an effective public health message .

 

 

Sim S W, Moey K S, Tan N C. The use of facemasks to prevent respiratory infection: a literature review in the context of the Health Belief Model. Singapore Med J. 2014;55(3):160–167. doi:10.11622/smedj.2014037

     

Evidence for masks in community

 

Successful strategies in other countries

Comparing  London: 9m  population with 6000  cases  by May  (1 in 1,500)  to Hong Kong:  7.4m population  4  cases   (1 in 1.85 million),  just under 1000 times less. 

The minutes  show  consideration of evidence from China  for social distancing  and lockdown being  effective but there is  no mention of masks .

Sage minutes. 25th February.  Measures to limit spread 9. Interventions should seek to contain, delay, and reduce the peak incidence of cases, in that order. Consideration of what is publicly perceived to work is essential in any decisions. 10. SAGE discussed a paper modelling four non-pharmaceutical interventions: university and school closures, home isolation, household quarantine and social distancing, including use of interventions in combination. 11. All measures require implementation for a significant duration in order to be effective.  12. Evidence from social distancing and school closures implemented in Hong Kong, Wuhan and Singapore indicates that these measures can reduce the Covid-19 reproduction number to approximately 1 (a 50-60% reduction). Reduced spread in the UK through a combination of these measures was assessed to be realistic. 13. Any combination of measures would slow but not halt an epidemic. 

Question 6 -  What is the evidence for statement 13  if  using  all the measures used in East Asia ?

There is no evidence to  separate an  individual  component from a successful strategy . The policy of the countries with best control involves  masks or  facial coverings and these  countries have strikingly lower rates of infection particularly with early introduction. A confounding factor may be time to lockdown but Hong Kong , Taiwan  avoided  lockdown.

 

Question 6 - What evidence is there for taking masks out of a successful strategy implemented in East Asian countries ?

 

General use of masks should be implemented before lockdown and would also be  a more cost effective strategy. One immediate problem apparent is that there was no provision by stockpiling of masks for public use. The most effective time for introducing mask wearing is  before  the exponential rise, reducing new cases sufficiently so that contact tracing can work and is not overwhelmed.

 

If use of masks shortened lockdown by only a few days,  the cost of providing a supply to most of the population  would be cost effective.

 

Approximate Cost of 1 week of UK Lockdown v Masks

Cost lockdown - 10.4% fall in UK GDP Feb to April 2020

UK GDP 2019 £2.21 trillion

Weekly fall UK GDP during lockdown (2.21 x 10.4)/ 52 x100 = £4420 million 

Cost of 5 surgical masks £0.5x 5 for 60 million population plus £1 distribution = £210 million

 

 

 

Preventing spread

Super-spreaders may be related to individuals that produce large amounts of aerosol when breathing up to 10,000 particles per litre. Asymptomatic spread of infection was known from work available online in mid-February  and published in March .

Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV2)March 2020 Science 368(6490):eabb3221 DOI: 10.1126/science.abb3221

 

On the 13th Feb.  SAGE concluded that neither travel restrictions within the UK nor prevention of mass gatherings would be effective in limiting transmission.

Patient 31 in  S. Korea was  identified on the 16th of February  infecting over 100 people at a mass gathering .

On the 5th March  SAGE agreed there is no evidence to suggest that banning very large gatherings would reduce transmission.

Many institutions stopped mass gatherings before government advice .

Super-spreaders and asymptomatic spread are additional factors mandating the general use of masks early on.

 

Media coverage

On the 9th April  BMJ editorials  concerning mask use  and  cloth masks as alternative including the paper by  Professor Greenhalgh   et al.  mentioning the precautionary principle .

Greenhalgh G , Schmid M B , Czypionka T, Face masks for the public during the covid-19 crisis BMJ 2020; 369 doi: https://doi.org/10.1136/bmj.m1435 ( April 2020)  Cited  as: BMJ 2020;369:m1435.

Question 7 - Why was the first  consideration of masks by SAGE  not until 14th April  and only after SAGE policy being questioned in the medical press ? 

The precautionary principle  is when there is an inadequate evidence base for making a decision but the decision to do nothing risks considerable harm.

A similar situation  would be  crossing  a road. There is no high level evidence  that keeping your eyes open is  beneficial but there are anecdotal  reports of people not looking being knocked down.  Do you need a high level evidenced study to conclude you are better  crossing a road with your eyes open compared to being closed when there is no disadvantage from keeping your eyes open and  it is easy to do ?

Concerning masks there is no proven disadvantage . The only valid point is that deaf people will not be able to lip read  but needs to be put in the context of risk to life .

Hand  hygiene issues,  such as touching the mask with unclean hands, means   public educational  information is needed and similarly advice on fitting (which has been lacking in the UK) . Concerning  other  proposed disadvantages   such as increase in risky behavior,   there is no evidence of this   in countries  that have high mask usage.  The argument is similar  to the compulsory wearing of seat belts  where there is  no evidence  of  increased risky driving behaviour.  

There are some individuals with respiratory disease that would find using a mask difficult  but the more people using masks will provide more protection for them and if necessary powered respirators are available. Lazzarino’s  proposed increased respiratory rate secondary to CO2 retention from the mask  causing increased dispersal of respiratory particles to those unable to wear mask, for instance with COPD, can be disproved by taking a person with COPD  and no mask  and respiratory rate of 12 litres/min,  together with  an infectious person wearing an FFP3 mask  the respiratory rate would have to increase more than 4 times to be more infectious than normal breathing and no mask. A mask would not be tolerated for long if causing a fourfold increase in respiratory rate.

One misquoted study shows cloth masks  allow more infections compared to medical masks in the healthcare setting  and  that cloth masks were worse than the control group but as most of the control group used medical masks there is no study showing that cloth masks are worse than no mask. ( A cluster randomised trial of cloth masks compared with medical masks in healthcare workers C Raina MacIntyre  2015 BMJ Volume 5, Issue 4).

 

As viruses cannot multiply on a mask there is no exposure situation where wearing a mask can be worse than not wearing one, although it does make sense to keep the mask clean and dry to maintain efficiency and  prevent bacterial or fungal colonisation.

Protection  against  eye risk  is  not a case against masks. Eye risk is considerably lower compared to inhalation, taking the surface area of the lungs (the size of a tennis court) to the surface area of the eyes. Eye contamination is  more likely when  close to someone from droplets and  in  this situation  a minimum of everyone using  a basic mask and hand hygiene, similar to  precautions needed in commercial food preparation, will help to reduce eye contamination risk. 

Higher efficiency masks can be more difficult to tolerate for long periods especially in humid environments, however, surgical masks are easy to wear for long periods and the combined effect of everyone wearing one giving a 17 times reduction in infection risk is not insignificant. 

Barr, G.D. (2020). A simple model to show the relative risk of viral aerosol infection and the benefit of wearing masks in different settings with implications for Covid-19 . medRxiv.

 

 

Modelling

The models published by the Dept. of Health did not mention masks .

Tracht’s model showed that for Influenza, N95 masks at 50% compliance would give over 30% reduction in cases . Early intervention before the  exponential rise  is more beneficial  and at this juncture the effect of general mask wearing  would be even  more marked. This together with other measures such as social distancing can keep numbers low enough to allow contact tracing and isolation to work. It is possible with 70% compliance and good quality masks  50%  or more cases could be avoided.

 

Tracht model 29. Tracht S M , Del Valle S Y , Hyman J M  . Mathematical Modelling of the Effectiveness of Facemasks in Reducing the Spread of Novel Influenza A (H1N1PLoS One. 2010; 5(2): e9018. doi: 10.1371/journal.pone.0009018

Mniszewski S M, Del Valle S Y, Priedhorsky R, et al.  Understanding the Impact of Face Mask Usage Through Epidemic Simulation of Large Social Networks. Theories and Simulations of Complex Social Systems 2013;52:97-115. Published 2013 Oct 27. doi:10.1007/978-3-642-39149-1_8

 

Question 8 - Why did the modelling used not  include use of masks ?

 

 

Argument for better quality masks for public use

SARS  and MERS  were found to  spread by aerosol transmission, both the government advisors and the WHO  acknowledged that there is an aerosol risk in hospitals. The virus is the same within and without hospitals,  if  aerosol spread can happen in hospital it can happen anywhere ; there is simply a higher concentration of risk in certain situations.

As the virus is the same there is no reason a mask should be effective in  hospitals but not in the community.  45000+ people  did not  die  simply because they did not  wash their hands,  but  each one was in a situation exposed to  significant risk.

 

Question  9 -  How many could have been saved by others and themselves wearing masks  ?

Early introduction  is likely to have saved somewhere between 1000 and 22000 deaths, with better quality masks  being nearer the higher range .

In the absence of  good quality masks being available for the general public  cloth masks or facial coverings  are a second best, however, the latter are low cost without disadvantage  although  their limitations need to be emphasised.

 

Public Health information and compliance

 

The first advert in  Scotland mentioning face coverings  was on the 5th June  noting Scotland has generally been ahead of the rest of the UK in relation to masks. 

The public health information appeared weak from the start. In answer to repeated questions from journalists about general mask wearing  the reply was consistently that  the public do not need masks . There are some that would take  this to mean Covid-19 can only be contracted by poor hand hygiene.

Compliance has been a problem in western countries   even as far back as the Spanish Flu pandemic on the other hand masks have developed significantly since then. The health belief model shows perception of risk of death and to health are important factors in compliance and also when livelihood becomes affected.

In Hong Kong during  the SARS epidemic  the public mask usage was around 65% , but for asymptomatic individuals as  low as 21.5%  for H1N1 [41]  rising to over 95% for Covid-19 .

A strong and consistent public health message is needed  such as the campaign in  Czechia, with  visual media messages, “ I protect you, you  protect me,”  also introducing the compulsory public wearing of masks in March. This compares to the USA and UK where for weeks even after the exponential rise in cases and hospitals being overwhelmed  the official message was that masks were not needed. Not surprisingly this has to led to a relatively low compliance.

Protection by face masks against influenza A(H1N1)pdm09 virus on trans-Pacific passenger aircraft, 2009.  Emerg Infect Dis. 2013;19 (9) :1403‐1410. doi:10.3201/eid1909.121765 

Question 10 - Why were masks not advised on public transport or anywhere social distancing was not possible, including care homes  and other confined areas at the start of  lockdown?  

Images were seen on national news  of   key workers crammed into  the London underground or buses and shops with queues  of people not distanced.  Care homes are a prime example of close contact with carers needing protection  and other high risk situations such as cleaners of toilets . This would explain the prolonged peak of Covid-19 cases in the UK with little fall in the number of daily new cases for 6 to 7 weeks despite lockdown.

In addition to a strong public health campaign provision of masks for the general population  is also a way of increasing compliance.

Question 11 -  What   advice was followed to have no  stockpiles of masks for the general population?

 

Conclusion

The default position given the evidence available in  February  should have been  that masks in the general population should be advised  unless evidence appears to the contrary, which has not been the case.

Question 12 - Face coverings or  cloth masks  although less effective are   easy to implement and a risk free  alternative.  Why  were facial coverings advised on public transport only from the 15th June and not the beginning of  March  when the same evidence was available ?

 

Future

Encouraging mask use such as in  influenza epidemics for  public transport, in shops and care homes could reduce morbidity and loss to the economy from sickness absence .

Stockpiles of surgical masks for  the population are necessary  for epidemics or pandemics with  significant morbidity or mortality and need to be implemented early on. Logistically this could be achieved by  gradually building up a national stockpile which supplies the Health Care Sector’s needs with the amount  used each year being  replenished.  Manufacturers  extending  the shelf life to 7 years or longer  from 5 years would aid this. In addition, this could be supplemented by encouraging as many people to have good quality masks in their households  once supplies are back to normal, through efficient and sustained public health information, similar to advice on smoke alarms. This at first might seem excessive, but if carried out correctly could be done in a cost efficient manner. Any  reluctance can be countered with the “What if “ question , what if the next pandemic has a 90% mortality ?

 

G Barr  July 2020