4.3 How do We Study the Spread of COVID-19?
By Harriet Jones
PhD Student. Faculty of Public Health & Policy, LSHTM
This article will introduce you to some of the different people involved in researching the spread of COVID-19.
Who is involved and what do they
You will have heard about the hard work of doctors, nurses during this pandemic, but who else is involved in the COVID-19 response? Research is being carried out by epidemiologists, mathematicians, statisticians, laboratory-based scientists,
behavioural scientists and clinicians, among many others. These people work at universities and public health agencies around the world. Although these specialists work together, and much of their work will overlap, they also play separate roles. Together
they contribute to our understanding the COVID-19 pandemic. In this article, we will take an in-depth look at work done by epidemiologists and specialists in mathematical modelling .
Image Credit: Katemangostar, www.freepik.com/free-photos-vectors/people
Epidemiology is the study of health and disease in populations. Among other things, epidemiologists look at the distribution, causes, prevention and control of disease. Epidemiology plays a key role in public health emergencies such as COVID-19. One aspect
of this is surveillance to track cases of disease. Epidemiologists conduct studies to understand, transmission and who is more likely to be infected or develop severe symptoms. They also evaluate interventions for disease control and prevention.
1.2. Medical Statistics
The application of statistics (the analysis and interpretation of data) to medicine or public health is often known as medical or bio statistics. Medical statisticians work alongside epidemiologists and other public health professionals, developing and
applying statistical methods to health data and interpreting the results.
You will have heard a lot about mathematical modelling during this pandemic. Mathematical modelling is another key part of responding to public health emergencies. Mathematical modellers use data to predict what might happen in the future. They do this
by combining surveillance and clinical data, with biological information about a disease, along with what is known about other infectious diseases. Models are not always right (it is impossible to see into the future!), but they often act as a starting
point for planning and prevention.
In this section we are focusing on a few specific roles in studying the spread of disease, but there’s more to it than this. There are many other professions involved in responding to a pandemic including laboratory and clinical researchers, virologists,
geneticists, pharmacologists, vaccine researchers and behavioural scientists. Lots of different scientists are working on COVID-19, trying to understand the disease better, developing tests, drugs and vaccines against the virus (step 2.4, 3.4, 4.4).
So much global research is going on we couldn’t possibly cover it all here, but we’ve put some links at the end of the article if you want find out more.
What information is needed and
what do these scientists do with it?
go back to the fields of epidemiology and modelling. Where there’s an outbreak
of a new disease, what do these scientists need to know?
Image Credit: Freepik.com
Before we count our case numbers, we have to define what a case is.
definition allows epidemiologists to use surveillance to identify and track
on in the COVID-19 pandemic, an case definition was produced based on symptoms,
and risk factors (e.g. time in spent in Wuhan). The case definition changed a
few times after this, as we found out more about the symptoms, the disease
became more widespread, and rt-PCR diagnostic test was developed. Updates to
case definitions are not unusual, particularly in outbreaks of novel pathogens.
Depending on the information available about a patient, a case may be classed as ‘probable’, ‘suspected’, or ‘confirmed’.
Disease surveillance &
measures of disease infectiousness
Epidemiologists collected, and continue to collect, data from a number of sources to identify
Who is infected?
Where the infections are being transmitted?
The number of secondary cases (R0)
These data can be collected through testing, population surveys, routinely collected data from hospital records and/or death registrations. Epidemiologists, statisticians and mathematical modellers analyse this data to describe where the infections are
happening, both geographically and among which groups of people. Cases can be mapped over time and statistical methods used to understand what the data shows.
As well as case numbers, we need to think about what happens to people who are infected- how many people get sick (morbidity), or die (mortality). This sort of information is sometimes difficult to find out in the early stages of the pandemic. In the
first weeks of the pandemic, most of the cases seen by doctors were people who needed hospital treatment, so initial estimates of mortality were as high as 15%. Milder cases, where hospitalisation was not needed were not identified,
so the proportion of infected people who went on to die appeared very high. Fortunately, once large numbers of mild or asymptomatic cases were identified, the mortality estimate (the case fatality rate, step 4.1) became a lot lower.
Interestingly, one of the early studies to get a better idea of morbidity and mortality came from the cruise ship, the Diamond Princess, which suffered an outbreak of SARS-CoV-2 whilst at sea in late January 2020. An on-ship quarantine was implemented,
and large numbers of the crew and passengers, were tested. This gave epidemiologists an early picture of the number of cases, both severe and asymptomatic cases and the number of people who died. From this information R0 and disease
morbidity and mortality could be estimated.
Following the outbreak in the UK, more detailed information has been collected using NHS data
. This has helped identify which groups are most at risk, and what underlying health conditions are important.
2.4 Predicting the Spread of the Virus
Mathematical models use information on what is known about the biological mechanisms of transmission (how the virus is spread), how long someone is infectious, how quickly the virus spreads, who is becoming infected, contact patterns among individuals,
hospitalisations and deaths to simulate the spread of COVID-19.
will often vary depending on how old you are, so are separated by age group. Mathematical
modelling has also been used to look at how different interventions work, by
simulating various contact patterns that may happen if schools close, people
work from home or stay indoors. Modelling and epidemiology will help us make
decisions about entering or lifting lockdown. We’ll talk more about this later
in section 4.
Cochrane Review, COVID-19 Studies Tracker: an overview of coronavirus research worldwide https://covid-19.cochrane.org/
The models driving the COVID-19 Response: https://www.nature.com/articles/d41586-020-01003-6
Interactive Models from the LSHTM: https://cmmid.github.io/visualisations
WHO case definition (20th March): https://www.who.int/publications-detail/global-surveillance-for-human-infection-with-novel-coronavirus-(2019-ncov)
UK case definition: https://www.gov.uk/government/publications/wuhan-novel-coronavirus-initial-investigation-of-possible-cases/investigation-and-initial-clinical-management-of-possible-cases-of-wuhan-novel-coronavirus-wn-cov-infection#criteria
Lancet – changing case definitions https://www.thelancet.com/journals/lanpub/article/PIIS2468-2667(20)30089-X/fulltext
Rajgor DD, Lee MH, Archuleta S, Bagdasarian N, Quek SC. The many estimates of the COVID-19 case fatality rate. The Lancet Infectious Diseases. 2020 Mar 27.
Mallapaty S. What the cruise-ship outbreaks reveal about COVID-19. Nature. 2020 Apr 1;580(7801):18-.
Williamson E, Walker AJ, Bhaskaran KJ, Bacon S, Bates C, Morton CE, Curtis HJ, Mehrkar A, Evans D, Inglesby P, Cockburn J. OpenSAFELY: factors associated with COVID-19-related hospital death in the linked electronic health records of 17 million adult
NHS patients. medRxiv. 2020 Jan 1.