DEATH CERTIFICATION

Introduction

This lecture will function primarily as a companion to my introduction to cause and manner of death. Writing a death certificate is usually fairly straightforward, but can get very confusing in complex cases. Sometimes it can seem like you are being forced to distill an individual’s entire social and medical history into a single line on a single page, which seems a little unfair. It’s a thing that a lot of doctors struggle with, mainly because there tends not to be a great deal of information out there on how to do it correctly. My sincere hope is that this lecture can provide some helpful information. Just be forewarned that this lecture may contain a bit more medical information than the last one as we will go into a bit more detail here.

In most states, death certificates are reviewed by medical examiners in some capacity. For example, in Florida, a death certificate needs to be reviewed by a medical examiner before that person can be cremated. If we find something egregiously wrong, we can follow-up and take steps to clear up any discrepancies. If you happen to be a doctor and are reading this, also know that you can always contact your local medical examiner for advice on signing a death certificate, even if it’s not a case that falls under our jurisdiction. Most importantly, our advice is free!

I joke, but accurate and timely death certification is extremely important. When someone dies, their family members deserve to know why. This can be necessary for closure. More practically, when someone dies, some other agency (usually family) must intervene and take control of any assets in order to properly execute a will or any last wishes that the individual may have had. There may also be property and insurance benefits involved. In order to start this process, the manager of an estate needs proper legal documentation that the individual in question is, in fact, deceased. The best and easiest way to provide this is a death certificate. If a death certificate is delayed or not filed for any reason, it is usually only the family that suffers.

The Death Certificate

Here we’ll take a look at the death certification process. To start, I will begin by reviewing two terms from the first lecture:

• CAUSE OF DEATH: This refers to the event or sequence of events that directly lead to the death of an individual. 
  
• MANNER OF DEATH: This can best be thought of as the summation of the circumstances surrounding the death.

I will also add one more definition:

• MECHANISM OF DEATH: This refers to the biological processes that facilitate death. Think of it as the steps that are taken along the way from the inciting event to expiration. 
  

I like defining mechanism of death as such because it implies that the mechanism gives you a lot of information, but not the whole story. Essentially, you’re missing the beginning if you rely totally on the mechanism. This will make more sense with some examples.

To start, let’s take a look at a sample death certificate:

Take a moment to examine the death certificate above and you’ll notice a couple of things. The top portion is mostly demographic information, including (obviously) the decedent’s name, sex, social security number, date of birth, place of birth, address, and lots of other such information. What we’re primarily concerned with is Part I (Cause of death) and Part II (other significant conditions):

You’ll notice that the cause of death section has multiple lines. It is not strictly necessary to use all of them. If you do, the entries should read logically: A due to B due to C due to D. You’ll find that most medical examiners tend to restrict themselves to one or two lines whereas most primary care doctor tend to use all four, but either option is acceptable so long as it is done correctly. Also take note of the “approximate interval” box next to each cause of death line. This is just an rough estimate for record keeping purposes. It doesn’t require an exact value, but exact time intervals can be put there. We typically use just general operatives like “seconds,” “minutes,” “hours,” etc. 

The most common mistake that we encounter when we review death certificates is the, admittedly confusing, difference between a mechanism and a cause. A  death certificate that includes only mechanisms of death is insufficient. Let’s illustrate this with two examples:

Example #1: Heart Disease

A 65 year old man with a history of coronary artery disease experiences chest pain whenever he walks more than two blocks. This pain is usually relieved by resting or taking his prescribed nitroglycerin. Despite this, he continues to smoke daily, as he has for the past 35 years. One day he is at home and begins to experience crushing chest pain while watching television. He is brought to the emergency department. Work-up includes an electrocardiogram that documents ST segment elevations and highly elevated troponins. Despite aggressive resuscitation, he expires after 21 minutes in the emergency department. Let’s fill in part I of the death certificate. Here is an example of a common way that this might be certified:

Myocardial infarction is just a fancy word for a heart attack. The underlying diagnosis that caused his heart attack would be the presence of atherosclerosis, or calcification and plaque build up in the coronary arteries. Also note that the manner of death has been checked as “natural.” Let’s focus on the first line for now. This is a mechanism, not a cause of death. It’s a mechanism because it only describes the process that had occurred without giving you a reason why that happened. If a person stops breathing, and their heart stops, they are effectively dead. That can happen for basically any reason. Standing by itself, it’s pretty meaningless. Personally, I try to avoid using mechanisms at all on death certificates and stick only to more general causes of death. As I stated before, you’ll find a lot of MEs doing this. Some people prefer to have a more logical sequence of events, and that’s perfectly fine as long as there’s more information than just “cardiorespiratory arrest.” This is a bit of a silly example when I spell it out like this, but we see death certificates on a fairly regular basis that only include terminology like that, so I realize that there is definitely a lot of confusion out there on this topic.

A second example will illustrate why this is a problem.

Example #2: A Delayed Death

A 62 year-old woman is driving down a two lane, suburban highway. She is wearing a seat belt and is not intoxicated. She has a history of chronic obstructive pulmonary disease and recently had a right total knee replacement for osteoarthritis. A driver in the oncoming traffic lane swerves across the center line for unknown reasons. A head-on collision results. Both individuals are transported to the local emergency department. The woman is noted to be hypotensive. Laboratory investigations reveal highly elevated troponins. A mesenteric laceration is suspected and confirmed on imaging studies. However, she expires before she can be brought to the operating room to repair the injury.

The emergency room doctor takes it upon himself to sign this death certificate. He submits it as follows:

And that’s it. Since we know more about the background in this case, this is clearly inadequate because it omits the exact reason that brought this woman to the hospital: the head-on collision. In a case like this, we will usually contact the doctor that filled out the death certificate and get a little more information. If this fell through the cracks, it would mean that something that should have been reported to the medical examiner was not. As we noted before, accurate classification and certification of deaths is necessary for the collection, reporting, and analysis of vital statistics. And speaking of vital statistics, note that the manner in this case has been left as natural, which we know from the last lecture is inappropriate.

Now, this is an extreme example that does not happen very often. It does, however, illustrate why we tend to treat vague death certificates with a little more scrutiny. Most MEs tend to have a list of “red flags” in their heads that will prompt them to seek out a little more information. Basically, these are diagnoses that people die with rather than of. A common theme is that they tend to be things that occur when one has a prolonged hospital stay, which may mean that their hospital admission is related to trauma, poisoning, or some other non-natural means. Oftentimes, the event that precipitated the hospital stay is overlooked, and that’s where the error occurs. We do our best to catch these. Here’s a few examples:

Abscess	Organ or organ system "failure"	Bowel obstruction	Decubitus ulcers	Broncho- pneumonia	GI hemorrhage
Fat embolism	Intoxication	Sepsis/shock/septic shock	Pulmonary embolism	Intracranial hemorrhage	Aspiration
Pnemonia	Peritonitis	Deep venous thrombosis	Disseminated intravascular coagulation	"Respirator brain"	Paraplegia/ quadriplegia
Necrotizing fasciits	Urosepsis	Toxicity	Allergic reaction	Hematoma	Dehydration

So let’s close the loop and fix the death certificate that we examined before:

We added a couple of lines to flesh out the sequence of events and changed the manner to accident. One additional section that is not included on our abbreviated death certificate is section 43, labeled “Describe how injury occurred.” This is required on all death certificates that have a non-natural manner. It’s just a short description of the event that occurred. I prefer to use complete sentences, but it’s not required. In this case, I’d fill it out like this:

One last word before we move onto the next topic. The distinction between cause of death and mechanism of death is not well delineated, so how can you be confident that what you put down on the certificate will be valid and you won’t receive a call from your local medical examiner? I think a good rule of thumb is to reserve established medical diagnoses using standardized medical terms for your death certificates. This will clear up most of the easy problems. But, I bet a lot of people are scratching their heads right now because aren’t things like “myocardial infarction” and “pneumonia” medical diagnoses? It’s true that they are, but they are still pretty general. It’s not that you can’t use them, but as we saw above, more information is generally necessary to let us know that it isn’t linked to some kind of injury. For example, “pneumonia” can be a lobar pneumonia acquired in the community, or a nosocomially-acquired bronchopneumonia that occurred as the consequence of a devastating brain injury. A more specific descriptor is extremely helpful. For example, “Community acquired pneumonia” is an acceptable death certificate entry that can stand alone. As for “myocardial infarction,” can be a traditional ST segment elevation infarct that occurs as a consequence of coronary artery disease, or a non-ST segment elevation infarct that can be due to many reasons, including hemorrhagic shock due to catastrophic blood loss. A little extra information can help avoid our phone calls. 

These last two examples will handle most common death certification problems. Now that that is settled, let’s change directions a little bit. Let’s talk now about a couple of little quirks in death certification that are considered consensus practice. In order to do this, we’ll focus on another field on the death certificate, part II:

Part II of the death certificate deals with medical or circumstantial factors that contributed to expiration, but in such a way that is is difficult to ascertain their direct contribution. Language in this section can be a little looser than in Part I because you are not DIRECTLY attributing a death to things that find their way there, but you are acknowledging their role. Let’s take a look at an example.

Example #3: Too much medical history

A 64 year-old man has a history of hypertension, coronary artery disease, hypercholesterolemia, chronic obstructive pulmonary disease (COPD), and chronic liver disease. He has been prescribed multiple blood pressure medications and has undergone coronary artery bypass grafting in the past. He smokes a pack of cigarettes per day and has for the past 45 years. He also drinks approximately six 12 ounce cans of beer per night as he has for the past 25 years. Two years prior, he felt that his visits to his primary care doctor and specialists were providing him with no improvement in his conditions, so he decided to stop going and stop taking all his medications. One night, he is watching television with his wife and decides to go to bed a bit early. He has no specific complaints, he just says that he feels “a bit tired” and he’s worried that he might be catching a cold. His wife stays up a bit longer before finally retiring as well, In the morning, she notices that her husband is cold and stiff in the bed next to her.

A case like this would probably fall to a medical examiner because it has been quite a while since this individual had seen a doctor. So how should we certify it? It’s clear that this man has multiple medical problems, each of which could probably kill him. So how do we choose what to put on the death certificate? 

A pretty standard practice in cases like this is to assume that the prime mover is the heart disease. Heart attacks are extremely common in the general population. Is it possible that the strain his COPD was placing on his heart precipitated an arrythmia and killed him? Sure. Did his high cholesterol contribute (along with all his other lifestyle choices) to his coronary artery disease. Of course. Did these problems CONTRIBUTE in a hard-to-quantify manner to his cardiovascular compromise? You bet they did. I bet you can figure out how we are going to certify this death. 

A quick note on the first line. We are pretty sure that this individual had a heart attack based on his medical history. However, a true myocardial infarct is an entity that is diagnosed via well-established criteria, namely electrocardiogram changes and cardiac enzymes. Since this man died at home, it’s not information that we are going to get. You’ll find that this terminology used often because, in most cases, a specific type of arrythmia (ventricular fibrillation, specifically) accompanies many heart attacks due to the factors listed on line b. It also provides a bit more of a definitive terminal event, which family members sometimes appreciate. However, it can’t stand alone. Why can’t it? Because it is a mechanism, not a cause of death. In part II we’ve listed his other diagnoses. It’s a natural death, so the only other thing we’d have to check is the box related to smoking (not pictured in the example, but present on the original death certificate, item 36). Smoking definitely contributed, so we’ll indicate that in the appropriate section. 

Cases like this sometimes cause consternation because it can be difficult to properly weight individual diagnoses in people that have extensive medical histories. I prefer the above option because, statistically, cardiac deaths are more common than any other type. However, if a doctor felt that this person’s COPD was debilitating and that was the primary cause of death, that’d be a perfectly fine way to fill out the certificate. How can there be two correct answers in this situation? The answer may come as a bit of a surprise. The level of certainty that is legally required for a death certificate is not very high. In most cases, a doctor’s BEST MEDICAL OPINION is sufficient to certify a death. It’s the 51% rule. We’ll get into the common reasons why doctors may be reluctant to sign death certificates a little later. When we do, remember this point.

Finally, before we get to another example, I think we owe it to the deceased to do the best job we can with these documents. They are important for legal and financial reasons, but they also provide necessary information to family members. One strategy that we encounter all too often is what I like to call the “spaghetti on the wall” strategy. In complex cases, doctors sometimes just list every single medical condition that their patient had in a long series across the fields of the death certificate. They’ll throw the spaghetti at the wall and hope that something sticks so that it’s not rejected by us. As I stated above, a medical examiner will always be able to provide advice in these cases. It’s not wrong enough to get too worked up over, but it’s a pet peeve of mine. 

On that note, let’s talk about a little quirk of certification.

Example #4: Trivial injuries

An 84 year-old woman is driving home on a local road. She has a history of severe congestive heart failure secondary to multiple heart attacks that have been the consequence of coronary artery disease and a severely calcified aortic valve. She has undergone multiple interventions in the past, but is no longer a candidate for any surgical therapy. She has near constant shortness of breath and chest pain and palpitations at rest. Recently, she has had frequent fainting spells. She begins to lose consciousness while driving and veers off the road. She collides with a tree at low speed in front of someone’s house. The occupants of the house call 911 and she is transported to the emergency department. She is diagnosed with a moderately displaced fracture of the right humerus and is admitted. Over the course of several days, her cardiac function declines and eventually expires.

How are we going to certify this death.

On one hand, we have an elderly individual that suffered a traumatic injury sustained during a minor car accident. But it seems a little silly to blame a death entirely on a broken arm, especially considering that this woman had a severely compromised heart. However, what if we apply the “but-for” principle? Would this individual have died at this particular time but for the car accident? We can’t rightly answer “no” to this question, either. So it comes down again to properly weighting individual diagnoses. In this, case, I think that her death is primarily due to her cardiovascular compromise. However, I also believe that the broken arm interacted adversely with her natural disease. Pain, stress, and blood loss into the fractured area can all contribute to an increased cardiac demand that most people have no trouble compensating. However, an individual like the woman in this example may not be able to. So here’s how we are going to express this idea on the death certificate:

So you can see that we are considering the heart disease the primary cause of death and contributing the minor injury in part II. The next thing we will do is assign a manner of death, and here’s where it gets a little confusing. We are blaming her death on her “natural” diseases, but there is a non-natural contribution to her death. If any circumstance of a death is considered “non-natural,” then the manner of death cannot be natural. So in this case, because a car accident played a role in her death, the manner will be ACCIDENT. Finally, this is a non-natural death certificate, so the “how injury occurred” box must be filled out as well. 

And that’s pretty much it. With this short primer, you will be able to confidently sign most death certificates. If you happen to have had the misfortune of encountering one in your life, I hope it will make a little more sense. Before we conclude this discussion, I have just a little bit to say about one special topic related to death certification.

A Word on Difficult Doctors

As medical examiners, part of our expertise is death certification, however, that doesn’t mean that we bear this responsibility in all cases. Unfortunately, we sometimes have to field calls from doctors in our jurisdictions that refuse to sign death certificates for their patients. Jurisdictional issues can be tricky (and sometimes state-specific), but for the sake of this discussion, we will assume that the responsibility is unequivocally that of the doctor who presided over a patient’s care. The reasons for refusal can be varied. Sometimes a doctor may feel that poor followup hampered the doctor-patient relationship and so they may feel that they don’t have enough information. Other times doctors may feel that they can’t rule out some suspicious cause of death. I have also heard concerns about “implied liability” when it comes to signing a death certificate. We feel that the signing of a death certificate is a part of patient care and that it is a doctor’s final responsibility to their patient. We are certainly sympathetic when a doctor feels mystified by a death and we sometimes hear that a particular doctor didn’t know a decedent “that well.” To that I would say that we know even less as we have never had a chance to interact with a person while they were alive. When it comes to suspicions of foul play, police usually investigate deaths that occur in the community and relay this information to medicolegal death investigators that work in our offices. If we don’t feel that anything suspicious has occurred, a doctor can also rest assured that nothing did. In the rare case that something suspicious was overlooked, that responsibility would fall on us, not the primary care doctor. Like we stated above, a doctor’s best medical opinion is appropriate, even if he or she is not 100% certain. 

The question of liability is a bit of an interesting one. The argument usually goes like this: I treated a patient for a particular condition; if that condition is listed on the death certificate, I am implying that I did not treat that condition correctly and it ended up killing this patient; therefore I will not sign this death certificate and expose myself to liability. To my knowledge, there has never been a medical malpractice or negligence lawsuit that has hinged upon a death certificate. True violations of standards of care and dereliction of duty are actually pretty rare. Such things probably won’t come to light based on how a death certificate was filed. In any case, it’s the United States, people can sue you for whatever they want.

If uncertainty exists, I would implore doctors to contact a local medical examiner and we can help you resolve the issue.

The only people that suffer when death certificates are delayed or filed incorrectly are family members. Let’s take a look at what might happen:

• The probate process will be delayed as assets cannot be claimed
• Insurance benefits will not be paid
  
• Services oftentimes cannot be completed (depends on the state)
• Lack of closure for family members

We sometimes take jurisdiction over these cases if primary care doctors are obstinate as a courtesy to the family. However, this is not a perfect solution, either:

• Overall, it delays things as the body will need to be transported to the ME office
• Funeral homes often charge for this transport to and from the office
• The family may also incur storage fees from a funeral home if a significant amount of time passes before the death certificate is filed
  

As you can see, significant harm can be done if death certificates are not filed in a timely manner. We’re always here to help in difficult cases.

I think that is enough written about the death certification process. Leave a comment below if you think I’ve left anything out, or you have a specific question. I look forward to hearing from you.