On 2018 Jan 09, David Keller commented:

"The most common cause of death was PD"?

It is notoriously difficult to get physicians to complete death certificates properly, especially when it comes to concisely and accurately identifying the cause of death. Some clinicians list the cause of death as "cardiopulmonary arrest" in nearly all cases, because (they argue) the cessation of pulse and breathing is one of the final events in most deaths, and thus is the most proximate cause of death. This type of error in identifying the cause of death is akin to "missing the forest for the trees", and interferes with research that would benefit from a cause of death that really answered the question: "what disease entity was the main reason this patient died on this admission?"

On the other hand, we have doctors who are too vague and imprecise in identifying the cause of death, as in this study. Assigning "Parkinson disease" (PD) as the cause of death tells us little in most cases. The fact that a patient had advanced PD when he died does not mean that he died from the direct effects of PD, which is unusual. It would be better to identify a more proximate cause of death, such as stroke associated with recent onset of atrial fibrillation (for example) and list "advanced Parkinson disease" as an "other illness present at death". This would give researchers who use the death certificate a more accurate designation of the proximal prime mover in the patient's death at this time, and also allow for investigations into associations between that cause of death, and the patient's underlying PD.

On 2018 Jan 09, David Keller commented:

"The most common cause of death was PD"?

It is notoriously difficult to get physicians to complete death certificates properly, especially when it comes to concisely and accurately identifying the cause of death. Some clinicians list the cause of death as "cardiopulmonary arrest" in nearly all cases, because (they argue) the cessation of pulse and breathing is one of the final events in most deaths, and thus is the most proximate cause of death. This type of error in identifying the cause of death is akin to "missing the forest for the trees", and interferes with research that would benefit from a cause of death that really answered the question: "what disease entity was the main reason this patient died on this admission?"

On the other hand, we have doctors who are too vague and imprecise in identifying the cause of death, as in this study. Assigning "Parkinson disease" (PD) as the cause of death tells us little in most cases. The fact that a patient had advanced PD when he died does not mean that he died from the direct effects of PD, which is unusual. It would be better to identify a more proximate cause of death, such as stroke associated with recent onset of atrial fibrillation (for example) and list "advanced Parkinson disease" as an "other illness present at death". This would give researchers who use the death certificate a more accurate designation of the proximal prime mover in the patient's death at this time, and also allow for investigations into associations between that cause of death, and the patient's underlying PD.