Making Sense of COVID-19 Data and Why It Is Too Early to Jump to Conclusions

It is too early to conclusively determine the path and mortality rates for COVID-19. We could learn from the experience of the 2009 H1N1 flu pandemic to understand the current global caseload better.

Public health experts warn of a health catastrophe as the burden of COVID-19 swamps the capacity of health systems around the world. Developing countries such as India are especially vulnerable. The virus packs an insidious second punch that is hammering the global economy. The IMF forecasts that a global recession is already underway. As economists and development practitioners can attest, slow or declining growth poses extreme dangers to vulnerable populations, especially in developing countries.

Governments across the world have imposed restrictions, extremely severe in some countries such as India, to slow the spread of the virus. Public health experts have led the calls for tough measures. Many economists have also supported these calls. But there are others, especially those working on economic issues, who worry that such restrictions could lead to severe economic and social consequences down the road. They believe that excessively harsh measures are disproportionate to the official data on the transmission of COVID-19, caseload, hospitalisation and fatalities.

I myself have struggled with making sense of the official data and the even harder task of reconciling it with dire predictions of hundreds of millions of infections and an unimaginable number of deaths. However, it is important that we advance ideas for understanding all the available data so that policymaking is as well informed as possible. It is with this intention that I am writing this piece.

The Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU) reports that 862,234 COVID-19 cases were confirmed as of April 1. Of these, 178,718 (20.7%) recovered and 42,404 (4.9%) died. The rest are active cases. The global case fatality rate (CFR) may be 4.9%, but there is wide variation across countries. Italy has a higher rate (11.4%), India is at 2.7%, the US at 1.8%, and Germany is at only 0.9%.

Also Read: An Inert Molecule Has Our Backs to the Wall

It is clear that we do not know what the actual infection rate is. That is because not all who carry the virus are tested. For the most part, people are either asymptomatic or have mild symptoms. In such cases, testing is unlikely, especially in countries with limited capacity. However, variations in CFR or actual infection rates are not the main concern of this article.

Here’s what I couldn’t figure out and I am certain many are wondering the same. While 862,234 cases sounds like a lot, it does not match with the dire predictions of health experts. A study by Imperial College, London found that without preventative measures, COVID-19 would lead to 40 million fatalities globally. In the US, Dr Anthony Fauci, who has emerged as the face of America’s COVID-19 response, indicated that the country would see “millions of cases” and “…between 100 and 200,000 (deaths)”. Currently, the death toll is over 4,000. Painting a bleak worst-case scenario for India, Ramanan Laxminarayan of Princeton University predicted that without tough measures, India would see at least 300 million infections.

These predictions have been met with a mix of fear, disbelief and even anger. After all, three months into the COVID-19 season, there are less than a million infections globally and the CFR is variable. Writing in the Wall Street Journal, Stanford professors Eran Bendavid and Jay Bhattacharya argue that based on current case fatality rates and estimates of total infections, the mortality rate is actually closer to 0.01%. “one-tenth of the flu mortality rate of 0.1%.” Using this line of argument, Rupa Subramanya, writing for Observer Research Foundation (ORF) argues that the lockdown ordered by Prime Minister Narendra Modi “…perhaps makes sense in a world in which COVID-19 carries extremely high mortality rates. We now know that this is very likely incorrect.” Subramanya’s main concern is that such a harsh lockdown could threaten livelihoods and eventually harm millions of people in India. I share that concern.

Fake blood is seen in test tubes labelled (COVID-19) in this illustration taken March 17, 2020. Photo: REUTERS/Dado Ruvic/Illustration/File Photo

However, it is too early to conclusively determine the path and mortality rates for COVID-19. To understand the current global caseload better, we could learn from the experience of the 2009 H1N1 flu pandemic, which started in the US in April 2009. By June 2009, the WHO raised the worldwide pandemic level to phase 6. At that time, cases were reported in 70 countries. Eventually, over 214 countries and territories reported cases of H1N1. The WHO officially declared the end of the pandemic in August 2010.

Overall, the WHO reported 18,500 confirmed deaths between April 2009 and August 2010. Researchers at the US Centers for Disease Control and Prevention (CDC) noted that the actual H1N1 death toll was likely much higher than the WHO number. Writing in The Lancet, they argued that “an improved modeling approach” resulted in an “estimated range of deaths from between 151,700 and 575,400 people who perished worldwide from 2009 H1N1 virus infection during the first year the virus circulated.” Just for the US, the CDC estimated that there were approximately 60 million cases, 275,000 hospitalizations, and 12,500 deaths. The estimated fatality rate in the US was only 0.02% but varied across countries. A separate study of reports from 19 countries, including India, put the infection rate at 24% of the population.

In comparison, officially confirmed COVID-19 deaths number 42,404. And we are still in only the third month of the pandemic. The final fatality estimates, using a methodology similar to that of H1N1, would yield significantly higher numbers. For example, “estimates of respiratory and cardiovascular mortality associated with the 2009 pandemic influenza A H1N1 was 15 times higher than reported laboratory-confirmed deaths.” Applying a similar standard to current official data, the number of deaths could be almost 550,000 right now (closer to the upper end of the H1N1 estimate). Having said that, we are still too early in disease progression to make any definitive assessments of the route it takes.

On a more positive note, clinical trials for an H1N1 vaccine started in July 2009, within three months of the first reported case. The US Food and Drug Administration (FDA) approved four H1N1 vaccines by September 2009. In October, the first vaccine was administered and by December 2009, 100 million doses of the vaccine were available for ordering. At that pace, we could have a COVID-19 vaccine this year. Clinical trials for a COVID-19 vaccine have commenced. That is cause for cautious optimism.

Salman Anees Soz, formerly with the World Bank, is deputy chairman of the All India Professionals’ Congress (AIPC). Views are personal.