Coronavirus Antibody Testing Provides Hope as Mortality Rate Appears Lower than Early Indications
The potential for hope might seem misplaced as the world faces down the start of another month battling the novel coronavirus.
It may be hard to understand the full context of how far we have come in the last several months. Our newsfeeds and televisions have seen a steady flow of articles, opinion pieces and news conferences detailing the latest information and data surrounding this worldwide pandemic.
In March, early models out of the Imperial College London forecasted a grim scenario for the United States of 2.2 million dead. As we began to digest images and data out of Italy and China, Americans became increasingly concerned that fatality rates for the infected might be 10% or more and an underlying concern that some unknown number could carry the virus undetected led governments around the world to take sweeping action on a scale not seen by the modern world.
Toilet paper, cleaning products, hand sanitizer, bread, meat, rice, beans and other assorted grocery products disappeared from shelves. Americans were told masks were unnecessary. Then they were told masks were mandatory for anyone without a medical exemption. Businesses and workers, told to stay home, waited anxiously for unemployment checks to arrive. Some are still waiting.
All these unknowns and more have left Americans stranded in a sea of confusion, as well as political & social division, with available data offering little solace. The confirmed positive tests and fatalities linked to the virus have given Americans little room for hope or positivity in the weeks since America’s first outbreaks. Projected mortality rates based solely on confirmed infections and presumptive fatalities have continued to paint a grim picture of numbers as high as 10% or more in some regions.
In the weeks since, death estimates have been revised from millions to hundreds of thousands and now hover around 75,000 by August. Despite this reassuring trend, the countless tragic stories pouring in from around the country maintain the concerns of many politicians and constituents about the veracity of the coronavirus and the disease it causes, Covid-19.
The concerns of citizens, desperate for the economy to reopen and those still grappling with the apocalyptic view presented by mounting presumptive fatality & confirmed infection totals may be granted a reprieve in the coming weeks as the tide of data coming in begins to take a reassuring turn.
Antibody testing in several places in the United States and around the world is beginning to shed light on many of the coronavirus’ darkest corners. Most recently, extensive testing has been compiled in New York, Boston (Chelsea) and Miami-Dade County. Other testing was also completed in two separate California studies (Santa Clara/Stanford & LA County/USC). While numbers are not exhaustive, they do give us the clear indication that our infection totals, and thus previous estimates of the virus’ mortality, are off by some degree. In fact, if recent findings are remotely accurate, conservative estimates put infection totals off by a factor of 15-20x depending on the region of the country.
While that news might be shocking and perhaps even terrifying, there is reason for optimism. As the picture of the infection denominator comes into view and a more accurate depiction of infection spread is known, the rate of deaths per infection necessarily falls. That change in data means significant reductions in estimates of the coronavirus’ mortality rate.
In the University of Miami study, researchers determined with reasonable probability through antibody testing that infection rates were anywhere between 12-22x higher than the county’s confirmed total of approximately 11,500. The findings led the team reach a median rate of 16.5x more infections than recorded through previous testing. The data, in effect, would cause Miami-Dade county’s infection totals to jump to just under 190,000. With that change in estimated infections, the mortality rate of Miami-Dade county would plummet from 2.6% to .16% of those infected with the virus. So while early fears regarding the virus’ ability to be carried with little to no symptoms provided serious cause for concern, the prevalence of those relatively unaffected by the virus may have skewed the earliest available data towards the most ill carriers by well over an order of magnitude.
New York City antibody results paint a very similar picture. Testing done in America’s largest hotspot showed that as many as 1 in 5 have had the virus. Adjusting for this indicator means a full-scale reduction in the mortality rate to around .5%. While this rate is still considerably higher than the mortality we have come to accept with seasonal influenza outbreaks each year, it is important to note that it falls well below the reported mortality of a staggering 10.7%. That would put NYC’s official tally off by 21.5x the projected mortality according to previous testing for the virus.
New York Governor Andrew Cuomo announced on April 29 that 17.1% of FDNY/EMT and 10.5% of NYPD tested positive for antibodies to the virus. According to Cuomo, the entire downstate antibody average is 18%. Another test of NYC transit workers is also being conducted this week. The National Institute of Health has also recently announced a test of 10,000 healthy individuals to gauge possible antibody spread across the nation.
A much smaller scale study in the Chelsea neighborhood outside of Boston found 1 in 3 had antibodies for the virus. The neighborhood has been visibly hit by the virus with 39 fatalities and 712 confirmed cases, which totals just 2% of the population and 5.5% mortality rate. However, antibody indicators could point to the area being farther along the outbreak curve than previously believed and could mean that confirmed cases are off by a factor as high as 16x. That possibility means the viral mortality in Chelsea could drop from 5.5% to just .34%.
The approximate 15-20x factor of antibody indicators to confirmed cases in Miami-Dade, Chelsea and NYC has shown to be a consistent trend across the country and even in international studies.
This should come as no surprise, as data has been available for weeks giving glimpses at just how contagious and sneaky the virus can be. In a review of cruise ship outbreaks, the CDC determined that as many as 46.5% of positive tests were asymptomatic at time of testing. Available statistical models also led the CDC to suggest that the virus never developed symptoms of any kind in 17.9% of those infected. The CDC has estimated that approximately 25% of cases among the general population carry the virus asymptomatically. These numbers are even more staggering given the much higher than average median age of those embarking on cruises. Both numbers seem to support the growing evidence that the outbreak is far more widespread than testing has shown and gives substantial credence to the numbers coming out of ongoing antibody research.
Antibody testing, like the live viral testing going on around the globe, is not without its limitations. Early tests out of Oxford showed that there is a margin of error in both false positives and false negatives. To combat this, researchers have factored some amount of this into their findings by generating deriving the most conservative models from their findings and have got to work on formulating more specificity in future testing.
Despite these limitations, several researchers, speaking with Bloomberg News gave reasons to be optimistic. Natalie Dean, a biostatician with the University of Florida, explained, “The results change a lot depending on what assumptions you put in…I think there is value in getting an answer very quickly, even if it’s not exactly right. As long as it kind of gets us in a ballpark.”
John Ioannidis, professor of medicine and epidemiology with Stanford also supported this point. “No study, no matter how well done, will be able to give the definitive answer alone…But at last we have data, real data, because until now we have mostly relied on speculation.”
These antibody studies, coupled with testing of outbreaks among contained populations (Ohio Department of Rehabilitation & Correction, several cruise ships and the USS Theodore Roosevelt), show that mortality rates fall far below figures that were initially believed in the early days of the virus. Though each of the above contained populations are not representative samples of the general population, the statistics are encouraging when tested against results shown in antibody studies.
While some critics of the government response have pointed to potentially inflated death statistics being linked to the virus, it is becoming increasingly clear that infection totals being amended according to new data will have a bigger impact on our understanding of the risk of this virus to the public at large. It is clear the virus is a significant threat, but Americans should be encouraged that our most dire predictions and indicators of the deadliness of this illness fall far below what we previously believed.