Education and information are crucial factors in the management of infectious disease. The distinction between the disease and the epidemic is key.
What do we know about the disease?
The Director General of WHO has emphasized the importance of ongoing containment strategies in particular testing and urged “all countries to take a comprehensive approach tailored to their circumstances – with containment as the central pillar” He has advocated testing, testing, testing and has encouraged all countries to learn from places like Hong Kong, China, Singapore, Taiwan and Korea who have all to different degrees had success in containing the epidemic with the central strategies of case identification, contact tracing and quarantine (see the epidemic curves which will be updated regularly).
The disease is spreading more widely and more rapidly. I have explained in a previous article the benefits of delaying epidemic spread. More than 50% of cases are now occurring in Europe which has also had more than 60% of total fatalities.
In an analysis of the first 44,000 cases, 80% of cases cause mild disease, 14% severe disease and 5% critical disease. A recent review in the Lancet looked at outcomes in critically ill patients. The mean age in this cohort was 60 with 67% males. The authors concluded that ‘the severity of SARS-CoV-2 pneumonia poses great strain on critical care resources in hospitals, especially if they are not adequately staffed or resourced1. This message was reinforced in a further paper in the Lancet looking at the Italian experience2. The mean age of death in Italy was 81 years. This varied from 83.4 years in females to 79.9 years in males. There was a 4 to 1 ratio of males to females. Two thirds of fatalities had other medical conditions or were smokers. Between 9-11% of patients admitted to hospital required intensive care treatment.
Dr. Judith Mackay, a respected public health doctor suggested in a response to our initial article that the sex difference in China may be due to smoking. She points out that 50% of males in China smoke in comparison to 3% of females. We would certainly agree that part of the public health response should include advocacy for healthy living including exercise, healthy diet and not smoking. Further research will be needed to quantify these risks but it is another reminder than we must not take data from one cluster of disease and assume that it will always behave in the same way in different circumstances or different populations.
The current mortality in China is 4.0%. In the rest of the World excluding China it is currently 4.6% (22,545/487,465). These figures are almost certainly exaggerated as there are a lot of people who have mild infections who will not be tested. As an example of the difficulty in assessing mortality in mixed clusters, the current mortality in Italy is currently 10.1%. Whilst factors in national health care systems may be a factor, in particular the overwhelming of intensive care facilities, for reasons explained previously this is most likely to represent more widespread, undiagnosed illness in countries with either limited testing capacity or who have made an active decision not to widely test. For example, Korea has 15% the number of cases in comparison to Italy but has performed 1.5x as many tests and has a mortality of 1.5%. The mortality in Germany is 0.5%.
Currently the mainstay of testing is PCR tests to detect the virus. A number of antibody tests (blood tests) are now being used. It is likely that the accuracy of these tests will be better understood over the next few weeks. It is too early to be certain what the final case severity will be but for the reasons previously described, early mortality rates from previous epidemics are frequently downgraded with time. Imagine a condition of a new disease in which 100 people were admitted to hospital and 10 of them died. This would give a hospital case fatality rate of 10%. But now imagine that we develop a test which shows that only one in every 100 people who developed the illness in the community became unwell enough to go to hospital. The other 99 go to bed at home or some even keep working with sore throats or a mild cough. In this situation, there are actually 10,000 people with the illness, 100 who go to hospital and 10 who die. This means that the true mortality rate of this disease is 0.1%. This is the mortality rate of influenza. Exactly what the case fatality rate of COVID-19 will finally be is too early to say. The WHO are estimating a case fatality of 0.3-1% which compares to the initial WHO estimate of 2% on January 29th. The final rate may be higher or lower depending on the number of milder cases. The number may also be higher if health systems become overwhelmed. This has the potential to increase deaths not only from COVID-19 but also from other urgent medical conditions. The rapid spread of the disease is suggestive of a condition which is more likely to be downgraded in severity over time with the caveat that public health measures are able to keep the pressure off health systems such that they can continue to function effectively. This is the reason for the recent enhanced public health regulations.
There have been two large epidemiological studies which have received significant publicity. A study from Imperial College London suggested an accelerated epidemic surge with high potential mortality in the absence of significant public health measures aimed at suppression of the epidemic3. This paper was reported as a factor in the change of strategy in both the UK and US. More recently a paper from Oxford has predicted more widespread, milder disease which may suggest that herd immunity will be achieved in a shorter time4. The incidence of mild and asymptomatic cases remains a key piece of missing data which will help to determine which model is closer to the actuality on the ground
The passengers on the cruise ships remain interesting epidemic clusters. 712 passengers were infected on the Diamond Princess. The cruise ship clusters have been the subject of a recent academic paper from the CDC5. It is important not to assume that these clusters would be representative of different situations. Cruise ships tend to have older people with increased risk of other diseases. There are factors related to air quality, sanitation and central cooking facilities. The current case fatality in the Diamond Princess cruise ship cluster of mostly older patients is 1.4% (10/712). These passengers were generally elderly, had a higher incidence of coexisting illness and were in an enclosed space with significant numbers of infected people suggesting potential exposure to a higher viral load. For these reasons, mortality in this cluster is likely to overestimate the case fatality rate for a ‘normal exposure’ risk of an average younger population. Of interest 46.5% of positive cases in the cluster had no symptoms at the time that they tested positive and 17.9% of cases never developed symptoms.
Highest fatality rates are for people over 80 which is the same for all infections. Under 40 years of age the mortality is 0.2% many of whom have co existing illness. The higher mortality in health care workers early in the epidemic, is likely to be explained by high viral load. Health care workers have always died in epidemics. They are working with the sickest people and we know that in many infections the amount of virus which an individual is exposed to is related to the severity of the disease that they suffer. A recent report by the WHO suggests that since the understanding of the spread of the disease improved in late January the incidence of illness in health care workers reduced significantly in China3.
The international clusters will give very interesting information about the infectivity and severity of COVID-19. The epidemic curves of the international clusters are regularly updated by the CHP6. Italy introduced significant social distancing measures on March 9th then 11th 2020. Watching the epidemic curves of Italy, France and Spain over the next two weeks will give some information about the likely impact of European public health measures in comparison to the stricter regimes in Hong Kong, China and Singapore. An epidemiological model has been reported in the Lancet as showing some early and positive signs in comparison to non-intervention6. This suggested some benefit within 3-4 days of the regulations. Obviously it will be important if this continues to have impact.
An important recent study of 9 pregnant women suggested that the pregnant women in the later stages of pregnancy had a disease severity similar to non-pregnant women4. The infection did not transfer to their babies. Subsequent studies suggest no increased risk of COVID-19 in pregnancy nor evidence to suggest impact on foetal development. This will require further study.
There are currently 454 cases in Hong Kong. The majority of recent cases are from one of three sources. Individuals returning from overseas, people in quarantine or direct contacts of a known case. There are some community acquired cases but they remain relatively rare.
There is increasing evidence that this infection is spreading via milder cases. It seems that individuals are most infectious over the first couple of days that they are symptomatic. The role of people without symptoms and those who are not yet symptomatic is not yet clearly understood. These factors will tend to make containment strategies more challenging but equally would suggest a milder disease on a case by case basis.
There is no doubt that this is an important illness and some people will have severe complications. A notable feature of this illness is that some cases deteriorate and develop severe breathing difficulties. A disease with a low mortality can still cause significant impact if widespread. It can cause severe strain on health systems and this is the reason for the public health response.
As I have previously explained. Questions about the disease relates to the question what happens if something bad happens to me or my family? These are What If? questions. In order to understand the risk of this bad thing happening we need to replace these questions with What is the risk of catching COVID-19? To answer this question we need information. When we are dealing with uncertainties it helps to have an anchor for risk. In order to assess this risk we need to understand the epidemic.
The Epidemic Process
In order to understand the likelihood of an illness spreading it is necessary to understand the factors which influence this spread.
What determines whether a virus spreads?
The main factors influencing the evolution of an epidemic of illness are:
The Mode of Spread: This illness is spread by droplets and close personal contact. There have been some reports of the virus spreading via the stool. One feature of this illness is the tendency for family clusters. Epidemiological records in China suggest that up to 85% of human-to-human transmission has occurred in family clusters6.
- Isolation of infected individuals and social distancing such as closing schools
- Hand washing and personal hygiene
We wish to stress our support for these appropriate health measures.
The Incubation Period: This refers to the time between an individual being infected and when they first develop symptoms. The WHO reports the mean incubation to be 5-6 days though in some cases may be up to 14 days. The mean epidemic doubling time is 6-7.4 days. Influenza by comparison is typically two days. This means that every week that passes we have twice as much data as the previous week.
The Infectivity in the Incubation: We do not yet know for certain when this virus becomes infective. We now have definite evidence of person to person transmission in multiple countries, including within Hong Kong. The current consensus is that viral shedding (infectiousness) peaks within a couple of days of symptoms. The increase in incidence certainly suggests a virus which is likely to be reasonably infectious. It is important to understand that infectivity and severity may be inversely related. The more cases that are occurring means that the disease severity (the Case Fatality Rate) generally reduces.
The Immunity of Individuals: This is influenced by many factors including age and the presence of other illness. An individual can also enhance factors which influence their immunity. This may include healthy diet, exercise, adequate rest, avoiding stress, maintaining a positive psychological outlook in addition to immunization against other infectious illnesses including influenza.
The way in which these factors subsequently impact the size of an epidemic is discussed here.
Many media stories confuse these two distinct concepts. They refer to the number of deaths (mortality) or severity of the illness and then conflate this with infectivity. The importance of this illness to our population will be determined by the evolution of the epidemic. This will become much clearer as we accumulate data from contacts of confirmed cases. SARS affected over 8,000 people with 744 deaths in 2003. Influenza kills up to 650,000 people annually just from respiratory complications alone (WHO estimate). The CDC estimates that there have been between 23,000-59,000 deaths from influenza in the USA so far this Winter compared to the 22,538 currently reported from COVID-197.
This is not to reduce the importance of this new virus nor the appropriate public health responses but to place the current data in context. We are not suggesting that COVID-19 is the same as or even a type of influenza, it is not. Accepting the limits of projections made early in an epidemic this illness currently seems to be more severe than seasonal influenza. Exactly how severe will take longer to establish but early indicators suggest this disease will be less severe than first thought. The impact it will have on the population will be determined by the evolution of the epidemic.
All infectious illness has greatest impact on the young, the old, the poor and the immunocompromised. The WHO has declared COVID-19 to be a pandemic. One of the greatest worries in population terms is the seeding of this disease into developing communities. A recent study made the following statement ‘the management and control of COVID-19 importations heavily rely on a country's health capacity8.
Hong Kong has a world class public health system and quite rightly public health measures have been introduced in order to protect the most vulnerable members of the community.
We will continue to update the article as data becomes available and information changes.
1. Yang, X., Yu, Y., Xu, J., Shu, P. H., & Xia, P. J. (2020, February 24). Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study. Retrieved February 26, 2020, from https://www.thelancet.com/journals/lanres/article/PIIS2213-2600(20)30079-5/fulltext
2. Remuzzi, P. A., & Remuzzi, P. G. (2020, March 13). COVID-19 and Italy: what next? - The Lancet. Retrieved from https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30627-9/fulltext
3. Imperial College COVID-19 Response Team. (2020, March 16). Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand. Retrieved from https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf
4. Fundamental principles of epidemic spread highlight the immediate need for large-scale serological surveys to assess the stage of the SARS-CoV-2 epidemic. (n.d.). Retrieved from https://www.dropbox.com/s/oxmu2rwsnhi9j9c/Draft-COVID-19-Model (13).pdf?dl=0
5. Public Health Responses to COVID-19 Outbreaks on Cruise Ships - Worldwide, February–March 2020. (2020, March 23). Retrieved from https://www.cdc.gov/mmwr/volumes/69/wr/mm6912e3.htm
6. (2020). Retrieved 2 March 2020, from https://www.chp.gov.hk/files/pdf/statistics_of_the_cases_novel_coronavirus_infection_en.pdf
8. Preliminary In-Season 2019-2020 Flu Burden Estimates. (2020). Retrieved 5 February 2020, from https://www.cdc.gov/flu/about/burden/preliminary-in-season-estimates.htm
9. Gilbert, M., Pullano, G., Pinotti, F., Valdano, E., Poletto, C., & Boëlle, P. P.-Y. (2020, February 20). Preparedness and vulnerability of African countries against importations of COVID-19: a modelling study. Retrieved February 26, 2020, from https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30411-6/fulltext
10. Wu, J., Leung, K., & Leung, G. (2020). Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study. Retrieved 5 February 2020, from https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30260-9/fulltext
11. Huang, C., Wang, Y., Li, X., Ren, L., Zhao, J., & Hu, Y. et al. (2020). Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The Lancet.
12. World Health Organization. (2020, February 24). Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19) . Retrieved March 2, 2020, from https://www.who.int/docs/default-source/coronaviruse/who-china-joint-mission-on-covid-19-final-report.pdf
13. Chen, H., Guo, J., Wang, C., Luo, F., Yu, X., Li, P. J., et al. (2020). Clinical characteristics and intrauterine vertical ... Retrieved February 14 2020, from https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30360-3/fulltext
14. Wang, F., & Zhang, C. (2020). What to do next to control the 2019-nCoV epidemic?. Retrieved 5 February 2020, from https://www.thelancet.com/pb-assets/Lancet/pdfs/S0140673620303007.pdf