| Proper Completion of Death Certificates | ||
 |
||
|
|
CLINICAL VIGNETTE Proper Completion of Death CertificatesJerome Greenberg, M.D.Case Report An elderly gentleman with metastatic pancreatic carcinoma is about to die any day at home. He was hit as a pedestrian by a car 4 months earlier. He sustained facial/orbital fractures and was hospitalized for 5 days. Computed tomographic scans of his chest and abdomen showed no sign of trauma internally; however, the scan of the chest showed an anterior mass and the abdomen showed a pancreatic mass. He saw a gastroenterologist in follow-up and was advised that this looked like a malignant or pre-malignant cyst and he should have surgery. He declined and then saw a thoracic surgeon for the mediastinal mass, who deferred further work-up pending work-up of the pancreas mass. All this time he was having some chest and abdominal pain which he says was the result of the accident. He was previously asymptomatic. Over the next 2 months, he started losing his appetite and weight. The pancreas mass and liver were biopsied and he was found to have metastatic adenocarcinoma. Soon after, he entered home hospice and stopped taking in anything by mouth. His wife stated she believed that the accident did not cause this, but she strongly felt that it weakened his immune system and the stress caused the cancer to spread. I received a call from their lawyer, who became involved as a result of the initial accident. He said that he did not want to tell me what to do, but asked if I was planning to list the motor vehicle accident somehow in the death certificate, saying: "If so, it could sure help his wife out." DiscussionOne would think that the finality of one's life, an event recognized in almost all cases without ambiguity, would lend itself to a document which certifies such an event with equal certainty. Nothing could be further from the truth. Any physician who has been in practice long enough to follow an aging population through to death knows that completing a death certificate accurately and with specificity is a challenge. Death certificates serve 2 main purposes: (1) they serve as an important source of epidemiologic data used for many reasons, including obtaining disease prevalence, signaling unexpected health risks, and helping to apportion funds for public health programs and research1; and (2) they serve as important legal documents, screening for causes of death that may warrant further investigation or for purposes such as settling life insurance claims and to allow one to be buried. The importance of death certificates in public health is the demonstration that about half of the decline in United States deaths from coronary artery disease from 1980 to 2000 may be attributable to reduction in major risk factors and approximately half to evidence-based medical therapies.2 Death statistics for this study were obtained from the National Vital Statistics System of the National Center for Health Statistics which are from data abstracted from death certificates. Numerous studies have demonstrated the inaccuracy and lack of specificity when completing death certificates.1,3,4 In one U.S. academic medical center, the most common mistakes made by clinicians included the following:
Death certificates have 4 lines available with the topmost line being the immediate cause of death (the final disease or condition resulting in death). The next 3 lines are conditions that have led to whatever is written above it, in order of causation. Not all lines have to be completed. In fact, because atherosclerosis and neoplasm are the 2 most common causes of death that do not always lead to interim conditions before death, it is acceptable to list either one as the immediate cause of death. This would be the case especially if the ultimate cause of death (beyond those conditions) is not known. Here are 3 examples of single diagnoses with time intervals:
The following are examples of immediate causes of death along with underlying causes:
An example of not qualifying non-specific processes would be listing an immediate cause of "pulmonary hemorrhage" due to "multiple organ system failure" due to "hepatic failure." Here conditions are listed which fail to denote a specific underlying cause of death.1 Other non-specific mechanisms such as "cardiopulmonary arrest", "hypotension", "renal failure" or "respiratory failure" should be avoided. Here are additional points to remember:
In the section "Other Significant Conditions," one should only list conditions that are unrelated to the underlying cause. Examples might include diabetes, alcoholism, smoking, carcinoma of the prostate or congenital heart disease (if none of these conditions were related to the cause of death). It should be noted that certain causes listed on death certificates may trigger reporting of the death to the coroner's office. There are the more obvious examples of homicides, suicides, and poisonings, but some - such as aspiration - create ambiguity. A common error is to list "aspiration" when it is an agonal event rather than an unexpected cause, i.e. an accident. Unnecessarily listing an event associated with an accidental manner of death may prompt a coroner's review. Similarly, in the case of procedure-related complications, such as someone dying during a bronchoscopy when they had a potential to die due to a serious underlying condition, it would not be considered an accident. Thus, if a terminal event were an expected risk of a procedure, it would still be a natural death.5 Indeed, it is not always easy to determine a cause of death, especially given that one will almost always complete a death certificate before an autopsy is done, and that most patients will not ever undergo an autopsy. One has to use the best clinical judgment to postulate a cause, if one is not known. In fact, there are frequent disagreements between the death certificate entries and the autopsy report, but that is a subject for another article. The case of the patient presented in the beginning juxtaposes the legal and clinical issues of completing death certificates. Putting aside the ethics of the issue raised by the lawyer, the physician must determine whether there was a link between the motor vehicle accident and the patient's demise. It was my best clinical judgment that there was not; that, in fact, the accident merely brought to light the already existing conditions. While it did affect his overall health for a short time after the accident, I determined it did not contribute to his death. The certificate was completed as simply adenocarcinoma of the pancreas, metastatic. REFERENCES
Submitted on July 4, 2007 |
