This is the second of several articles about COVID-19 and related topics by Dr. Jon Crosbie, an ISU alum, sports fanatic and all-around awesome Doctor of Osteopathic Medicine.

Several people have asked me, “How, exactly, is cause of death determined for COVID-19?” and I’m really glad they did. Now, I should tell you on the front end that I called in a favor and got myself a ringer to review this and make changes as he saw fit. My ringer is my cousin Dr. Darius A. Arabadjief, MD, who’s a pathologist/Medical Examiner. He even trained under, like, the Yoda of Forensic Pathology, a guy named Dr. Charles S. Hirsch. Darius was kind enough to review this and correct it as appropriate because my experience with cause of death is extremely limited and only comes from two places:

1. A rotation I did in med school with the Polk County Medical Examiner for two weeks. Now, this rotation was FASCINATING. You can learn an awful lot about the way people live by studying the way they die. The rotation entailed the medical student (me) going to where bodies were found at all hours of the day and night and then participating in the autopsies the next day. I actually started this rotation 11 years ago to the day of writing this.

The Medical Examiner that I rotated with was even a consultant on the CSI television show for Dr. Robbins’ office and jargon and stuff. A medical examiner is NOT a coroner, by the way, so don’t confuse the two. A medical examiner (M.E.) has gone to medical school and has done a combined Anatomic and Clinical Pathology residency that takes 4 years. That means an M.E. has 8 solid years of training as to how the human body works, and usually one of those years is spent specifically on how people die. A coroner can be anybody. A coroner is a public official, appointed or elected. There are police officers who act as coroners. With all due respect to our police force, they studied application and enforcement of law, not how the human body works and how people die.

2. Filling out death certificates. When one of my patients dies and there *isn’t* an autopsy done, I have to fill out the cause of death. One thing I do remember from my rotation is that sometimes an autopsy by an M.E. is required by law and this depends on the circumstances of death. All of this is different throughout the country. My cousin had different requirements when he did his training in New York City, but if I remember my med school rotation right, in Des Moines, if somebody under 60 dies at home without a known medical condition, I think an autopsy is required.

Generally, a medical examiner or coroner is *supposed* to investigate and certify any cause of death that is considered a public health threat. We’ll get back to this, but remember that a coroner doesn’t have the training that a medical examiner has. Hell, I don’t have remotely close to the training a Medical Examiner has, and in the first four years of our careers, an M.E. and I had the same training.

Coming back to the issue of COVID-19 — there are NOT going to be M.E.s doing full autopsies on every COVID-19 patient. One autopsy takes roughly 2-3 hours to do properly and there aren’t enough qualified pathologists to cover the number of people dying. Don’t forget that people are still dying at the same rate and from the same things as they were before. COVID-19 deaths are add-ons, not replacements.

Generally speaking, the COVID-19 death statistics are going to come from two different places: the National Center for Health Statistics and the CDC. When somebody dies of anything at all (heart attack, car accident, whatever), a cause of death certificate is filled out and submitted to the State Death Registration. Those numbers are tabulated given to the National Center for Health Statistics and this is how we get our stats for all deaths, including COVID-19. The other place we get our numbers from is the CDC. COVID-19 deaths, and any other public health threats, are what’s called “reportable” which means you tell the CDC when somebody dies of COVID-19. The CDC keeps track of this and then reports it to the general public. This is meant to better streamline tracking of the disease.

It’s important to note that because the overall numbers are coming from two different places, they might change at the end of the year and that would be a reason why…it’s not a conspiracy or anything, it’s just two different ways that COVID-19 numbers are tracked.

Now where it gets thorny is when the paperwork is filled out. Firstly, you need to know that there’s a Part 1 and a Part 2 to this. Let’s start with Part 1 and if this next paragraph is hard to follow, we’ll try and clear it up in a minute by using the example of the Lady Who Gets Stabbed in the Shower in the movie “Psycho.” Part 1 of the death certificate includes a couple of different “causes” of death – The Immediate Cause of Death and the Underlying/Proximate Cause of Death (COD). The Immediate COD is considered the final disease or injury that caused the death of the individual. The Underlying/Proximate COD is the disease or injury that initiated the sequence of events that led to the death. Now this part is really important because if it’s not filled out right, the Vital Statistics center may not accept it.

The immediate cause of death DOESN’T have to be etiologically specific but the Underlying/Proximate COD does…don’t worry, I’ll explain what that means. Let’s fill out paperwork for The Shower Lady. Immediate COD would be “exsanguination” (she bled out). Underlying/Proximate COD would be “multiple stab wounds to the thorax”. See, exsanguination could happen from getting stabbed in the shower or getting shot or lacerations from a car accident. “Multiple stab wounds to the thorax” is the specific cause of the exsanguination. Again, if this isn’t filled out quite right, the paperwork might get sent back and not immediately counted.

So what about deaths involving COVID-19? Let’s say that somebody came from a nursing home. They were 74 years old, and had COPD/Emphysema and poorly controlled diabetes. Let’s say this person became infected with COVID-19. Six days into the course of the symptoms, they suddenly developed shortness of breath and passed away at the nursing home. Now, it’s very likely that an autopsy WOULD NOT be done on this patient, but a Death Investigation WOULD BE. This could be dealt with over the phone, and often times is.

This might sound half-assed, but think about it…do you really think that there’s enough time in the universe to do a full, proper autopsy on every situation like this? People are still getting murdered and otherwise dying suspiciously and, quite honestly, people like my cousin need to take their talents to things like murders and other deaths for which the nuance of the cause is critical.

If we were filling out Part I of the death certificate for our above COVID-19 patient, it would be very reasonable to do the following: The Immediate Cause of Death would be Acute Respiratory Failure. The Underlying/Proximate COD would be Infection/Complications to COVID-19. Makes sense, right?

That’s Part I. When we fill out Part II, we list the conditions that contributed to but didn’t overtly cause the death of the patient. In our above COVID-19 patient, Part II (other significant conditions), we’d list the conditions the patient had that contributed, in some way, to the death. For my medical friends who may be reading this, this isn’t a list of everything they ever had. This is a list of things that contributed to the death — were these conditions not present, perhaps the patient wouldn’t have died. Filling out Part II for our patient would include COPD/Emphysema and Type 2 Diabetes Mellitus, poorly controlled.

We could debate some of these points. Did COPD/Emphysema belong in Part 1 or Part 2? Was it directly involved in the chain of events? It’s a fair question, but never mind that for a second. Take a deep breath, because I’m about to throw a reeeeeeally complicated COVID-19 death at you.

Let’s say we have a little old lady with high blood pressure, chronic bronchitis from smoking, and heart disease living at home. She’s on the phone with her daughter and goes into a coughing spell, gets dizzy and falls. The phone goes dead and the daughter hangs up to call 911. By the time the ambulance gets there, the patient is unconscious. On the way to the hospital, her heart stops. They get her heart restarted, but she’s intubated when she gets to the emergency room. When she gets to the ER, testing there reveals she had a heart attack. She stays intubated for several weeks and then develops a fever. She’s tested for COVID-19 and it’s positive. Unfortunately, she continues to decline and she dies.

What’s the Cause of Death?

Like I said, take a deep breath. Part I – Immediate COD: Bronchopneumonia. Underlying/Proximate COD: Anoxic Encephalopathy due to myocardial infarct due to atherosclerotic heart disease.

Part II – COVID-19 infection, Chronic Bronchitis
See, the thinking here is that the Chronic Bronchitis certainly added to the problem but wasn’t the main contributor. And this stuff is subject to a lot of interpretation. People disagree on this – even M.E.s disagree amongst themselves.

(For the record, Darius helped walk me through that second scenario)

And since you’re almost certainly wondering, yes, any death that has COVID-19 in either part 1 or part 2 would be included in the COVID-19 death toll.

Now here’s the thing…many doctors DON’T get any extra training on this stuff and almost none have the benefit of a Forensic-Pathology-Obi-Wan-Kenobi Cousin they can call up. Nor do they have the time, quite frankly, to get this stuff exactly right. Is that good? Of course it isn’t. Is it the way it is? Yeah, I’m afraid so. I have the benefit of working in an academic institution and have the time and the flexibility and the support from my bosses to look into these sorts of things.

So are our COVID-19 numbers artificially inflated? Given the scenario I wrote above with the little old lady who fell at home, I could certainly understand why it would seem that way, but I don’t actually think that. Personally, I think they’re underestimated. Why’s that? Let’s complicate things a little bit more (God, really?). What if our first nursing home patient never had a positive test? What if…

1. The patient never had a fever and wasn’t tested initially, then died, and a week later one of the nursing home staff tested positive for COVID-19 which is, in fact, what killed the patient?
2. What if the nasal swab was done incorrectly and the test was a false negative — the patient who died of respiratory failure really did have COVID-19 all along?

Both scenarios are currently happening. And #2 is happening way more than you’d like to know about. There’s one other thing you should know…there’s no oversight here. Think about this…if I had a patient who was in hospice, who contracted COVID-19 while in hospice, and who died and I filled out the death registration form with COVID-19, but the patient actually died of organ failure because of cancer, nobody is going to check my work.

Again, if nobody checks my work, it seems like we’d be overestimating the COVID-19 deaths. However, I think that, on the whole, we’re underestimating them because of all the scenarios in which a COVID-19 death may not be counted as such mostly because of the lack of testing. This all assumes, by the way, that we actually *can* determine the right answer. Sometimes that’s not even possible because the case is really hard.

The point of this isn’t to rip into the system or criticize anybody. The simple fact is that this is a really hard problem to solve and, understandably, medical resources get aimed at people who are alive, rather than people who are dead. HOWEVER, for my medical friends, it’s really important that we get this right as right as we can because research dollars are allocated based on Death Certificates.

You can learn a lot about how people lived by studying how they died. If you don’t understand the way they die, the process tends to repeat itself.

Dr. Jon Crosbie, D.O., is a Physician and Assistant Professor at Des Moines University Medical School