The dilemma of the death certificate

by Carolyn Thomas     @HeartSisters 
In a good old-fashioned murder mystery, we know that the plot starts to heat up when the dead body is discovered and the cause of death determined. But in real life, most of us will not die quite so dramatically.
If we live with one or more chronic illnesses, in fact, the name of at least one of those diagnoses will probably be listed on our official death certificates someday. (We could also get run over by a bus long before then, but let’s face it, chronic diseases cause 70% of deaths worldwide).
It struck me recently that, had I died during what doctors call my widowmaker heart attack in 2008, the official cause of death would have likely read “myocardial infarction”. But that would have been wrong. The actual cause of my death would have been that I was misdiagnosed with acid reflux and sent home from the Emergency Department.
Obviously, I did survive my debilitating textbook cardiac symptoms long enough to be able to return to Emergency (the second time, happily, to a correct diagnosis and appropriate urgent treatment). I did not die. I’m still alive. But let’s consider what might have logically happened had I died.
Because there is no mandatory reporting of diagnostic error in most healthcare systems, it’s entirely possible that after my death, the fact that I’d been earlier misdiagnosed and sent home might never have been reported, never reviewed at hospital rounds, never discussed with the medical director, never used as a case study with future med students, never considered as an important teaching tool that might help prevent this type of cardiac misdiagnosis from happening to other women in the future – and certainly never recorded on a death certificate as the cause of my death.
And what a lost opportunity that would have been.
In fact, because there is still no requirement to report misdiagnoses, the autopsy was actually described a century ago as the “gold standard for detecting diagnostic error.”(1) 
In other words, you had to die before your true diagnosis was officially revealed post-mortem on the autopsy table.

Aside from not having to report any diagnostic errors, modern regulators do tend to be quite picky about what does in fact go into an official death certificate.

Depending on the jurisdiction, cause of death is generally determined by a medical examiner, a coroner (who may or may not be a physician), a “pronouncing” physician at the scene, or a hospice nurse.

Here’s an example of a case study from Dr. Douglas Campos-Outcalt, published in the Journal of Family Practice:(2)

A 68-year-old woman is admitted to the ICU because of acute chest pain. She has a history of Type 2 diabetes, high blood pressure, obesity and angina. Over the next 24 hours, an acute myocardial infarction (heart attack) is confirmed. Heart failure develops, but improves with  medications. The patient then experiences a pulmonary embolus, confirmed by a lung scan and blood tests. Over the next two hours, she becomes unresponsive and dies.

Question #1: What should be written on the death certificate as the immediate and underlying cause of death?

Correct Answer:  Pulmonary embolus due to acute myocardial infarction due to atherosclerotic heart disease.

Question #2: What should be listed as conditions contributing to death but not directly causing death?

Correct Answer:  Type 2 diabetes, obesity, hypertension and heart failure.


Dr. Campos-Outcalt includes some handy definitions of those important terms. For example:

  • Immediate cause of death: the final disease or injury causing the death
  • Intermediate cause of death: a disease or condition that preceded and caused the immediate cause of death
  • Underlying cause of death: a disease or condition present before and leading to the intermediate or immediate cause of death. It could be present for years before the death.
  • Manner of death: The circumstances leading to death—accident, homicide, suicide, unknown or undetermined, natural causes.

This seems pretty clear cut to me, but as Dr. Campos-Outcalt explains:

“Death certificates are important official records used for personal, legal, and public health purposes, yet they are frequently filled out inaccurately.”

In fact, many studies have found death certificate errors in cause and/or manner of death ranging from 33% to 41% of cases, with disproportionate over-representation of cardiovascular causes of death.(3)  Mortality data worldwide are coded according to international classifications and published by the World Health Organization, so imagine how inaccurate those global statistics are if over one-third of the official findings are not correct?

One of the most common inaccuracies, Dr. Campos-Outcalt warns, is in determining that first part: the immediate cause of death:

“This should be a disease, complication or injury that directly caused the death. A common error is to list a mechanism of death rather than a disease.”

Consider, however, this comparable example:

A pet dog dies who has not been adequately fed or cared for by his owner. The dog’s cause of death is multi-organ failure due to starvation.

But isn’t the true cause of death the fact that his owner did not provide the care that could have saved the dog’s life?

So let’s go back to the imaginary worst case scenario of my own theoretical death certificate, and consider how we might accurately fill it out:

  • – The immediate cause of death would have been the actual heart attack itself.
  • – The underlying cause of death would have been the coronary artery disease that ultimately caused that heart attack. We know that heart attacks are years, even decades in the making, so this seems to fit.
  • – We also know that women who suffer serious pregnancy complications (like my own preeclampsia diagnosis while pregnant with my first baby, Ben) have a significantly higher risk of heart disease, so preeclampsia also qualifies as an underlying cause of death.
  • – The failure of the Emergency staff to provide the care that could have saved my life qualifies as another underlying cause of death, or at the very least (as in the above Question #2), misdiagnosis should be included on the death certificate as one of the conditions contributing to death.

There’s a gap between immediate and underlying descriptors on a list that’s basically limited to either disease or injury – but clearly omits the absence of appropriate care for either due to diagnostic error.

Similarly, after being misdiagnosed and untreated in mid-heart attack (“You are in the right demographic for acid reflux!” is what the Emergency physician had pronounced confidently before sending me home), I now suggest that it’s entirely reasonable to conclude that absence of appropriate care can cause death.

So in my imaginary demise, was it the heart attack that would have officially killed me –  or the fact that I did not receive appropriate guideline-based treatment for that heart attack?

UPDATE: Since the beginning of the COVID-19 pandemic in March 2020, many people experiencing cardiac symptoms were either too afraid of catching the corona virus to seek emergency treatment or too reluctant to burden already over-burdened healthcare staff – often with fatal results. Should those who died because of treatment-seeking delay during a medical emergency have #COVID19 as the underlying cause of death on their death certificates?  See also: Empty Beds: When Heart Patients are Afraid to Seek Help


1. D Campos-Outcalt, MD, MPA: “Cause-of-death certification: Not as easy as it seems.” J Fam Pract. 2005 February;54(2):134-138
2. DR Lakkireddy et al, “Death certificate completion: how well are physicians trained and are cardiovascular causes overstated?” Am J Med. 2004 Oct 1; 117(7):492-8.
3. RC Cabot, “Diagnostic pitfalls identified during a study of three thousand autopsies. JAMA. 1912;59:2295–2298.

Q: If you were in charge of determining the cause of my fictitious death, how would you decide?

NOTE FROM CAROLYN: I wrote much more about surviving misdiagnosis in my book A Woman’s Guide to Living with Heart Disease (Johns Hopkins University Press, 2017). You can ask for this book at your local bookshop, or order it online (paperback, hardcover or e-book) at Amazon, or order it directly from Johns Hopkins University Press (use their code HTWN to save 30% off the list price when you order).

See also:

Deep thoughts about death and heart disease

Skin in the game: taking women’s cardiac misdiagnosis seriously

Why doctors must stop saying: “We are all patients” (my guest post in the British Medical Journal (BMJ)

“It’s not your heart. It’s just _____” (insert misdiagnosis)

Misdiagnosis: is it what doctors think, or HOW they think?

Seven ways to misdiagnose a heart attack

Misdiagnosis: the perils of “unwarranted certainty”

Cardiac gender bias: we need less TALK and more WALK

Unconscious bias: why women don’t get the same care men do

When you fear being labelled a “difficult” patient

The sad reality of women’s heart disease hits home

How can we get heart patients past the E.R. gatekeepers?

10 thoughts on “The dilemma of the death certificate

  1. Yet again your story illustrates more than one of one of the “pitfalls” of many current healthcare systems; the issue of death certificates is a complex one and the current standards do not help epidemiologists collect accurate data – but the throwaway comment by your initial diagnosing doctor is also a significant source of poor data. You may well have been “in the right demographic for reflux” but for one thing you were also “in the right demographic” for MI and for another, diagnosis by demographic should be a hanging offence!

    A diagnosis should be reached by going through a formal process of examination of evidence from the patient, not from a graph! Too many medics now diagnose by statistical probability rather than on an individual basis – this fails those who make up the minority in the statistics.

    Liked by 1 person

    1. Hi Eva – I’m not a fan of capital punishment, but might make an exception for diagnosis by demographic! Might be tough to get expert witnesses to testify against their peers in the trial, though… 😉

      You are so right: that WAS a “throwaway comment” from that Emergency physician. Imagine diagnosing a male patient based only on his ethnic or socioeconomic background or age or height or what school he went to. The doc could do that while sitting in another room just reading the man’s chart notes…

      And I too was surprised to learn that so many death certificates are WRONG, which can significantly skew mortality data everywhere.


  2. Carolyn,
    I’m so glad YOU were not a “lost opportunity” for “an important teaching tool that might help prevent this type of cardiac misdiagnosis from happening to other women in the future.”

    I do understand your point but I hope “old age with complications of orneriness” is listed on my death certificate.

    Liked by 1 person

    1. Ha! I love that cause of death. Reminds me of my grandmother who outlived three husbands, did all her own gardening and canning (why do old Ukrainian Babas do so much canning when they live alone?!) and died instantly falling down the stairs – in her 90s! Not a bad way to go.

      But your ornery comment reminded me of this wonderful essay on “Grumpy Old Man Syndrome” that I heard on our CBC national radio yesterday…. Read (or even better listen!) – I think you’ll like this!


  3. I am a survivor of many cardiac events. In October I went to Emergency because my left eye had been increasingly irritated for weeks. I was told it was just irritated and to use drops.

    Not satisfied, 3 days later I went to my ophthalmologist without an appointment. He examined me with his specialized equipment, diagnosed squamous cell cancer, and referred me to a sub-specialist on an urgent basis. The next day I was seen in an office and then a surgical suite where a biopsy was taken and a debulking of the tumor was done.

    I have had extensive, successful surgery. Now I am cancer-free. I am a survivor. But this is a cancer that spreads quickly if it is in the head an neck. If it spreads, survival rates at 3 years are less than 50%.

    Had I died I presume squamous cell cancer would have been listed as the cause of death. But like you, misdiagnosis on first hospital visit would have been a major contributor.

    Liked by 1 person

    1. Thank you for sharing this unique perspective, Jenn. The thing that struck me while reading it was that your initial misdiagnosis (“just irritated, use drops”) was very likely never reported when you were ultimately diagnosed with the correct diagnosis of cancer.

      Many physicians still consider even hearing the word “misdiagnosis” as being “doctor-bashing”, so there’s precious little tolerance for talking about the subject, never mind mandatory reporting so that the next poor schmuck in that ER won’t be sent home with the same misdiagnosis.

      And because they refuse to report it, we don’t even know how truly widespread diagnostic error is.

      I’m sure glad you were “not satisfied” and insisted on that second opinion!


  4. There are so many of us that were misdiagnosed several times over the years. I am absolutely amazed that I lived 20 plus years before getting a correct diagnosis, finally!

    Once diagnosed, you can rest assured that that will probably be noted as cause in my death certificate. No autopsy necessary. My Mother had CABGx3, AFIB, Congestive Heart Failure and Dementia. She passed away in 2017. No Autopsy. It was clear that at least one of those was cause of death. But, WHICH ONE? I would have appreciated knowing for sure!

    Liked by 1 person

  5. Interesting and important distinctions, Carolyn. Particularly relevant when one thinks about the always ongoing need to track disease patterns in human populations. One question – I take it there would be no room for heredity in either standard or enlarged death certificates? It’s clear from my own heart problems and those of both my parents – and their parents and siblings – that heredity made us all prone to cardiac disease. Something you can’t change, by the way.


    Liked by 2 people

    1. Good question, Judy. You’d expect significant risk factors that resulted in a diagnosis to be at least the underlying cause of that immediate cause of death. (Some risk factors don’t, of course – we can have a family member diagnosed with an inherited disease yet not carry that specific gene ourselves).

      The surprising fact to me was that so many of the official documentation is WRONG!


Your opinion matters. What do you think?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s