Blog | Thursday, June 18, 2020

Death counts from COVID-19 underestimate actual numbers!


I have recently become aware that some people are mistakenly getting suspicious that COVID-19 death counts are an overestimate. That is probably because the numbers are awful and very hard to visualize. Nearly 110,000 deaths have been documented so far in the U.S., which is undoubtedly significantly below the actual number who have died from the disease (see my previous blog on the subject, from very early on in the pandemic.)

Our freshman U.S. representative in Idaho, who serves a dark red district of a very conservative state, recently wrote a letter to the CDC concerned that the guidance they have given regarding filling in death certificates will lead to a falsely elevated number of people who have died of the disease. I just wrote Mr. Fulcher a letter explaining why this is wrong.
“I saw your letter in your recent email encouraging the CDC to count Covid deaths accurately. I think you may not know some of the nitty gritty of documenting cause of death from a doctor's standpoint. After over 30 years of completing death certificates for patients who die in hospitals, nursing homes and their own homes, I have some familiarity with this. The bottom line is that, when we finally have all the data, our present estimates of deaths due to COVID-19 will be a gross underestimate.
The reasons for this are:
1. 20% of people die at home. For most of these, there is no autopsy and a cause of death is assumed to be something on their known problem list, such as heart attack or stroke or COPD or cancer. Most people who die at home with Covid don't get coronavirus testing and so won't be counted as Covid deaths.
2. From knowing what is happening in nursing homes that are infected with COVID-19, death from that disease looks like this: residents start getting fevers and coughs. They stop eating and getting out of bed. They die. It's often the increase in deaths that clues people in that there is Covid in the nursing home. So most of those people who died from Covid won't have been tested and won't have it on their death certificates.
3. In hospitals for months there was not enough testing kits to go around so when there was clearly an epidemic going on they stopped testing patients with obvious Covid symptoms. Everyone with cough and low oxygen levels had COVID-9. There was no point wasting testing and exposing nurses to more risk of infection by swabbing patients. Doctors don't necessarily feel comfortable documenting those deaths as Covid if they didn't get testing. Many of these were misclassified.
Also—you are asking the CDC to call Covid deaths only if the person died directly of Covid. This is the thing about the comorbidities. When a person has Covid and comorbidities they die of the combination. They wouldn't have died without the Covid, so it is the cause of death. Or at least should be. Some of these patients are being classified as having died due to the heart attack or the COPD exacerbation caused by Covid so they aren't on the official count.
It generally takes a couple of years for the accurate numbers of influenza deaths to get counted by the CDC because they look at more than just the death certificates. With influenza also, most deaths are in patients with comorbidities and if we just counted the ones that said “influenza” on the death certificate it would be a gross underestimate.
Right now we are trying to evaluate the impact on deaths due to Covid by looking at death rates now vs in previous years. That will underestimate the numbers of Covid deaths as well because with lockdown people didn't go out and get killed in car accidents, they didn't mix with others and die of other infectious diseases and they didn't suffer other trauma as much. So without Covid we would have been much healthier in lockdown.
I would encourage you not to worry about the death count from Covid being inaccurately high. The CDC is really good at figuring this stuff out having done it for years with influenza. What you are seeing is falsely low numbers because that is the best we can do right now.”
Just thought I would share.

What I didn't mention to Mr. Fulcher, who in my one experience of him, at a town hall meeting, would not be very interested, is that death rates in the developing world are likely even less accurate than our own. In Tanzania, for instance, they have decided that people will be happier if they don't know what is happening with COVID-19, so after initially testing a few people they have stopped. In Brazil, they are reporting 35,000 deaths so far with over 600,000 cases. But 20% of Brazilians live in poverty and so probably die without receiving medical care. In many developing countries it may be years before there is a way to estimate the numbers of infections and deaths. India, which has very advanced medical care for those who can access it, also has huge numbers of people living in extreme poverty, many of them in crowded conditions with multigenerational families. Over 80% of Indians die at home. I'm not sure how they can accurately document causes of death there.

Much like the influenza pandemic of 1918-1919, we will not know the accurate numbers for many years. And like that influenza outbreak, we are far from being done with coronavirus in June of 2020.

Janice Boughton, MD, FACP, practiced in the Seattle area for four years and in rural Idaho for 17 years before deciding to take a few years off to see more places, learn more about medicine and increase her knowledge base and perspective by practicing hospital and primary care medicine as a locum tenens physician. She lives in Idaho when not traveling. Disturbed by various aspects of the practice of medicine that make no sense and concerned about the cost of providing health care to every American, she blogs at Why is American Health Care So Expensive?, where this post originally appeared.