by A.C. Thompson
“Gone Without a Case” was part of Post Mortem, a collaborative effort to look at America’s death investigation system, the more than 2,300 coroner and medical examiner offices responsible for probing sudden and suspicious fatalities. ProPublica – my employer – has been working with NPR, PBS Frontline, and other news organizations on the series since 2009, producing web stories, TV documentaries, radio features, and a database.
For “Gone Without a Case,” ProPublica teamed with Carl Byker, a Frontline producer with a particular interest in aging issues. The web piece – a 4,100-word story – provided us with the roadmap for a television documentary that will air this year.
We came to the subject matter organically: Our reporting for the Post Mortem series took us to morgues around the country and introduced us to forensic pathologists, prosecutors, defense attorneys, judges, detectives, and academics. Over and over, our sources told us the death investigation system was missing cases in which older Americans died of neglect, abuse, or outright homicide. It was a recurring subject in our conversations.
So we started working.
Our goal was, admittedly, quite ambitious. We wanted to go beyond anecdotes and quantify, nationally, how many suspicious senior deaths had slipped by the authorities, or could possibly have slipped by.
We obtained five years of multiple cause-of-death mortality data from the U.S. Centers for Disease Control and Prevention. Our plan had three basic parts: Identify deaths of people older than 65. Identify deaths that could’ve been caused by neglect, abuse, or homicide – for example, pressure sores and malnutrition are two conditions that can be caused by neglect, so those would be deaths we’d look at. Then we’d check to see if these seniors received an autopsy. Failure to autopsy such cases, we thought, was highly suggestive that the case had eluded the scrutiny of the medical examiner or coroner.
With all that in mind, one of our news app gurus, Krista Kjellman Schmidt, used MySQL to sift roughly 9.6 million death records.
We eventually abandoned this approach. After much consultation with epidemiologists and geriatricians, we concluded that we couldn’t say anything meaningful or accurate using that methodology.
There were a couple of reasons. While we’d made a list of causes of death that we considered suspicious, all of those causes of death could also have been the result of innocuous circumstances. For example, a nursing home patient who died of malnutrition may have been killed because staffers failed to adequately care for the patient. Or the patient may have suffered from a fatal disease that made it impossible to eat and digest food. With the CDC’s records, it’s impossible to tell what really occurred.
This setback pushed our reporting in a different direction.
Using Nexis, court searches, interviews and requests for information on email listservs, we’d gathered details on roughly three dozen suspicious senior deaths that went undetected – at least initially – by coroners and medical examiners. We took this small universe of cases and looked for patterns.
They quickly became apparent. Death certificates, we realized, were key. If a doctor declares that a person has died of natural causes, there will rarely be any further investigation. Because it’s a natural death, the coroner or medical examiner won’t take jurisdiction.
What we found, though, was that doctors repeatedly certified that patients had died of natural causes, when, in fact, they had been murdered.
So how did the docs get it so wrong?
In some of the cases, the certifying physicians didn’t actually see the decedents – and thus never saw the obvious evidence of foul play, like extensive bruising. In some cases, the physicians hadn’t seen the patient for months before he or she died.
These practices seemed lax to us. and we wondered what sort of standards governed death certification. We checked out the laws in all 50 states – a Google Documents spreadsheet shared among reporters was helpful for this – and contacted state health officials to make sure we were properly interpreting the laws. In most states, we needed to inspect both the laws governing vital records and the laws setting out the duties of coroners or medical examiners.
Nearly every state allows doctors to certify a death without physically examining the body, we learned. And most laws allow doctors to certify a death even if they haven’t seen the patient in the weeks or months before his or her death.
Not surprisingly given this looseness, a host of academic research has shown death certificates to be widely erroneous.
We also scrutinized autopsy rates. Using the mortality data we’d obtained from the CDC, we calculated autopsy rates for older Americans. Our calculations squared with the CDC’s own research, which it published last year in a report that went largely unnoticed by the media. Autopsies of seniors have become increasingly rare even as the population age 65 and older has grown. Between 1972 and 2007, the CDC found, the share of U.S. autopsies performed on seniors dropped from 37 percent to 17 percent.
The autopsy numbers are low, partially because of the flaws in the death certification process. But there is another factor at work – one made quite clear by Catherine Hawes, a Texas A&M professor who studies elder abuse.
In anonymous interviews for a U.S. Department of Justice study, Hawes talked to 40 coroners and medical examiners. They expressed a deep reluctance to autopsy seniors, and, in many cases, had little knowledge about the signs of elder abuse or neglect.
A.C. Thompson is an investigative reporter for ProPublica, a nonprofit newsroom based in New York. His work focuses on crime, poverty and human rights.
