A researcher from Harvard University called after I responded to a survey sent to members of the National Association of Medical Examiners. He wanted to know what criteria medical examiners use to conclude that a drug overdose caused death rather than some other factor, such as heart disease.
The caller, who sounded young and idealistic, seemed shocked when I said that medical examiners and coroners don’t have universal criteria for making drug-death determinations. Our certifications vary according to individual practice philosophies.
It sounds, the researcher said, as if medical examiners “jump to conclusions.” That’s an unfair characterization, but I understand why he found my answer disturbing. He had asked a reasonable question, and he expected my answer to reflect a degree of scientific certainty that forensic medicine often can’t deliver.
Consider the most obvious criterion medical examiners use to decide whether an opioid drug caused death: the postmortem drug level.
When the level is either miniscule or off the charts, certification is easy. But most cases fall somewhere in between.
Individuals vary widely in their susceptibility to drugs. To derive reliable scientific data on lethality, we’d have to give increasing amounts of drugs to thousands of people of different ages, races and genders with different drug-use histories and degrees of pre-existing physical incapacity until they all die. Nobody would volunteer for, and no ethics board would approve, such a project.
We use lethal ranges derived from individual case reports: This dead person had this level of drug in his system. That dead person had some other level. Add a bunch of cases together and publish a range.
For most drugs, ranges are broad. Innocuous and lethal levels overlap. When considering a specific case, it’s reasonable to assume that a younger, healthier, experienced drug user could tolerate a higher level than an older, sicker, novice user. But nobody knows exactly what the relative lethal levels would be. Scene findings, the deceased’s history and a given medical examiner’s training and experience influence determinations.
Add some level of heart disease or another dangerous pre-existing condition to a given drug level, and uncertainties increase. There’s no way to know if heart disease alone or the drug alone would have caused death. There’s no way to prove that “but for” the interaction of both factors, the person would have survived. There’s no way to assign percentages of blame. There’s no way to conduct an experiment to figure it out.
Practice standards aren’t universal, but they enable us to be internally consistent when dealing with cases such as these. For example, most of us believe a non-natural cause of death trumps natural disease. So by tradition, if a person has both heart disease and a high opioid level, the drug is the primary cause of death.
Some put heart disease on the death certificate as a contributory cause; some don’t. Some certify the cause of death undetermined because there’s no way to be sure.
Give me a thousand deaths involving combinations of drugs and diseases and I guarantee that my thousand certifications will not be identical to the certifications of any other medical examiner.
The good news for research is that the Harvard study is looking at trends, not individual cases. Medical examiners haven’t changed their practice standards in the last few years. Their certifications may not be uniform, but they are internally consistent.
Almost all medical examiners are reporting striking increases in the numbers of drug deaths, and that trend is real.
Dr. Carol J. Huser, a forensic pathologist, served as La Plata County coroner from 2003-12. She now lives in Florida and Maryland. Reach her at firstname.lastname@example.org.