TOPSHOT - Women stand in a queue as they wait to collect cash of financial assistance through a mobile wallet under the governmental Ehsaas Emergency Cash Programme for families in need during a government-imposed nationwide lockdown as a preventive measure against the COVID-19 coronavirus, in Lahore on April 13, 2020. (Photo by ARIF ALI / AFP) — AFP or licensors

'It's the math, stupid': Why we may be reading the Covid-19 numbers all wrong

We break down for you what each of these numbers imply and why you should care for any or all of them.
Published April 13, 2020

For the past three months, the world has watched in disbelief as health authorities have announced numbers of new patients affected by the novel coronavirus.

One day, the Special Assistant to the Prime Minister on Health Dr Zafar Mirza warns the nation that fatality rate could rise if preventive measures aren’t taken.

Then there is the Sindh chief minister, who fears that an increase in locally transmitted cases is a bad omen.

So what do these numbers mean?

Here, we break down for you what each of these numbers imply and why you should care for any or all of them.

Confirmed cases

The number of persons affected by the novel coronavirus, which causes a disease called Covid-19, has risen past 1.8 million worldwide, as of April 12.

But like with all data, this number too should be read with a pinch of salt. For all we know, this could well be just a fraction of the actual persons infected by the virus.

There are multiple reasons for this:

  • In Pakistan for example, patients are less willing to come forward due to stigma or fear of being quarantined
  • Not enough tests being conducted
  • Deliberate under-reporting by governments
  • Efficacy and accuracy of tests being conducted leading to false negatives
  • False-positives (Rare but certainly a possibility)

Suspected cases

One of the biggest obstacles in the global response to the Covid-19 pandemic is that so little is known about the virus that causes it. Over the past three months, scientists from around the world have put their heads together to figure out the puzzle.

And yet, a new advisory is issued almost everyday based on new evidence from groundbreaking research.

While the scientific community struggles to find an antidote, governments have been struggling to muster the resources required to control the spread of the virus.

What we do know is that the virus causes specific symptoms when it affects the human body. What we also know is that it affects the elderly and the immune-suppressed more than others. This is how authorities can monitor suspected cases.

In Pakistan, for example, you will only be tested if you are running a fever and can prove you have either travelled or met someone who has tested positive for the virus.

However, a significant percentage of the population does not show any symptoms from the virus. This is what skews the data.

What’s more dangerous is that these asymptomatic patients are effectively carriers and can infect others who may be more vulnerable.

Number of tests

For Nate Silver of Five-Three-Eight, an online US-based publication that is in the habit of incorporating big data into intelligible stories, this is the most important number to be looking at.

In his recent report, titled “Coronavirus Case Counts Are Meaningless”, Silver argues that if you’re not accounting for the number of tests when looking at Covid-19 data, your analysis could well be just the opposite of what is actually true.

And that is where governments must be careful. In Silver’s words: “A country where the case count is increasing because it’s doing more testing, for instance, might actually be getting its epidemic under control. Alternatively, in a country where the reported number of new cases is declining, the situation could actually be getting worse, either because its system is too overwhelmed to do adequate testing or because it’s ramping down on testing for PR reasons.”

Case Fatality Rate (CFR)

For many epidemiologists (experts in the branch of medicine that deals with the incidence, distribution, and possible control of diseases), this is the number that holds the key to understanding the prevalence and severity of a pandemic.

The case fatality rate, according to the Dictionary of Epidemiology, is “the proportion of cases of a specified condition that are fatal within a specified time”. In other words, how many of the patients who test positive eventually die within a specified time.

This is the metric used by epidemiologists to gauge the severity of pandemics such as that caused by Ebola or even the various strains of influenza and to guide responses.

In the case of Covid-19, however, instead of relying on the number of infections, or the CFR, researchers are looking at the daily deaths attributed to Covid-19, according to this report on Nature magazine.

Some experts are of the view that total deaths from the virus over time is a better stat to compare how much the outbreak has devastated countries instead of looking at total cases. That is because they argue that total positive cases are heavily correlated to the number of tests done – which in itself varies from place to place depending on the country's resources and strategies employed and because a significant number of infected persons are asymptomatic.

Reproduction ratio or R

In epidemiological models, the ‘reproduction ratio’ or R stands for the number of persons each infected person goes on to infect.

Silver explains it like this: “If a disease has an R of 3, that means each infected person transmits it to three more people. So one initial case becomes three newly infected people, which becomes nine people, which becomes 27 people, which becomes 81 people, and so forth — the very nature of exponential growth is that it gets out of hand quickly!”

The R for Covid-19, established by the WHO from data gleaned from Chinese authorities, was 2.6. That meant every infected would infect at least two other people.

This is why social distancing is so important to curb the spread of the coronavirus. In doing so, authorities around the world, including Pakistan, have been aiming to keep the R under 1.

Locally transmitted cases

The first case of Covid-19 reported on Pakistani soil was that of a person who had just returned from Iran. Over the next few weeks, as the number of patients rose, almost all of them had some travel history.

The government responded by temporarily locking down airports and banning the entry of both Pakistani citizens and foreigners alike.

On April 12, SAPM Dr Zafar Mirza announced that half the Covid-19 cases in the country are now locally transmitted.

This is indeed worrying as it denotes that the virus has now gained a foothold within our communities and is spreading among the local population.

So which number?

The Covid-19 pandemic will possibly be studied by epidemiologists and other researchers for years to come. Plenty will be said about the global response and plenty more about the lack of it.

Right now, however, when we look at these numbers, it is important to remember that they may not be telling the complete picture. In some cases, a combination of these stats may need to be looked at to develop a better picture of the situation on the ground.

That being said, context is key when looking at Covid-19 data and if health experts are able to read into these numbers with a holistic approach, it should give them a fair idea of the virus’ trajectory in the region and guide their response.