Thursday, August 21, 2014

"Partial Autopsies Yield Partial Answers"

"Partial autopsies yield partial answers" Dr. Charles Hirsch taught me as a young doctor during my forensic pathology fellowship training in New York City, in 2001-2003. This was one of scores of aphorisms we called "Hirschisms," which he employed to instruct us about forensic pathology. Another one was, "Don't confuse the autopsy with the death investigation."

Both come to mind this week in the wake of the release of the findings of the second Michael Brown autopsy, conducted by Shawn Parcells and Dr. Michael Baden. The body diagram they released to the press in the New York Times has now been altered into the "Hands up! Don't Shoot!" pose and is trending on Twitter.

The wounds are clustered closer together, so it must be "realistic," right? 

No. 

This is click bait—a picture that you can retweet easily without thinking about it much. This "tweaked" diagram is a grossly inaccurate portrayal of the victim's body position. The horizontal graze wound near the elbow is missing entirely; the top of the head wound doesn't make sense unless the victim is leaning forward; and the diagnonal graze wound on the right thumb would not occur if the right hand were up in this way—the graze wound has to be in line with the gun barrel, which means the hand was possibly extended forward toward the officer. That's assuming you trust the diagram.

And that's where my experience as a forensic pathologist and death investigator makes me ask a few questions:

1. Why was this diagram from the second autopsy—this evidence of the wounds on the body—even released to the public? The original autopsy diagram, the one prepared during the first autopsy by the St. Louis County medical examiner's office, is still sealed. So are the photos, scene data, and other evidence the district attorney will rely on to decide whether to prosecute the police officer. In all homicide cases, evidence such as this diagram is kept sealed so as to not influence potential witnesses who might see it in the press and change their testimony. Why release crucial evidence that could scuttle that prosecution? 

2. How reliable is this diagram? As I wrote in my recent OpEd on CNN.com, gunshot wound interpretation is not always easy. Parcells and Baden have opined that there were "at least 6 gunshots"—but without knowing the number of rounds fired, the number of casings and bullets recovered, the condition and location of those bullets, and the possible witnessed positions the victim was in, how can they say that with any certainty? What if the same bullet grazed the thumb and then re-entered the body, causing another wound? This diagram alone doesn't tell the whole story.



The diagram reproduced in the New York Times and elsewhere has both Parcell and Baden's signatures at the bottom—but which one of them made the diagram? And why are there no other details about the individual wounds on it? Why have they not noted the location from the top of the head/right of midline, wound size and shape? Where are their notes about soot, or stippling, or other trauma besides the gunshot wounds? The body diagrams I prepare during the course of a forensic autopsy are a whole lot more detailed than this; I need those details in order to accurately dictate my report after I finish the autopsy and get out of the morgue.

3. How reliable is the second autopsy?
There have been recent reports in the press that forensic technician Parcells has no credentials or accreditation, and misrepresents his experience.  Following allegations that Parcells did the autopsy alone, a doctor/blogger in PathologyBlawg.com interviewed Parcells. Parcells affirmed that he alone examined the body on 8/15 before it was embalmed; Dr. Baden was not yet in Missouri at that time. Two days later, after the body had been embalmed, Dr. Baden performed the second autopsy.

There is a big difference between the examination of Michael Brown's undisturbed body during the first, legally-mandated autopsy, performed by the St. Louis medical examiner, and the follow-up examination done days later on his washed and embalmed cadaver. In the embalming process, preservative fluids are injected into the arteries and organs using a sharp tool called a trocar. The trocar pokes holes in the organs. The preservative fluid in the blood vessels pushes the blood ahead of it to the site of any injuries. These changes, which we call "embalming artifact," can exaggerate the size and shape of injuries. 

Even if Dr. Baden, a board-certified forensic pathologist, looked at photos of the injuries taken prior to the embalming, the orientation and quality of the photos taken by the technician would influence his interpretation of the findings. Autopsy means "see for yourself"—and there is no substitute for seeing the undisturbed body for yourself if you are going to be offering opinions with legal ramifications.

Monday, August 11, 2014

Expert Mistakes

"An expert is someone who knows some of the worst mistakes which can be made in a very narrow field."
Niels Bohr

When I review others' reports and find mistakes I always try to learn from them because it is always easier to learn from others' mistakes than to make and learn from your own.

No one is immune to mistakes. It is what makes us human. The question is how do we, as scientists and experts, deal with our errors? Do we ignore them? Deny them? Or do we delve into the reasons why they occurred and make a change? And can we accept that by making a change in policy or procedure we will be opening ourselves up to future attacks by attorneys who will use the change as a basis to invalidate our previous opinions?

What are some of the worst mistakes? The worst I've seen are the result of arrogance. As an expert and legal consultant one needs to be confident and project that confidence when testifying. But knowing the facts of your case and showing proficiency in analyzing and conveying those facts is different from insulting or tearing down an opposing expert, criticizing the person instead of their opinion (ad-hominem attacks). I've been on the receiving end of those with opposing counsel making fun of me for having gone to Harvard or for my dress ("fancy"). Generally I know that if they are attacking me personally it is because they don't know how to attack me based on the facts of the case. I also know that the more obnoxious they get the more they will alienate the decision makers - the jury.

I have seen world-renowned experts, confronted with inconsistencies between their previous and current testimony, or between their testimony and a recently published article get defensive and even aggressive. They respond to legitimate questions with bluster and arrogance. Nothing turns off a jury more. The best way to deal with a direct attack is to address the specific issues at hand and simplify it for the jury. Explain to them why the case the attorney is asking about is different from the current case and how interpretations in science can vary based on these crucial differences. By the time the expert is done explaining, the jury will have either forgotten the attorney's challenge or gotten so wrapped up in the explanation of the facts of the current case that they will be right back along agreeing with the expert.

But how does an expert learn to keep her ego in check? The best way is to hang around people who know a lot more than you do. By teaching residents and medical students and working alongside staff in a university setting you are constantly barraged with questions that force you to challenge your assumptions and stay up on the scientific advances that drive the peer-reviewed medical literature. Take challenging consult cases: by sparring with attorneys on high-profile cases you are going to be confronted with sharp criticism and you'll find that you can't just rely on your experience and training - you need to stay current and sharp. And finally, go to professional meetings. Nothing humbles me more than attending an AAFS or NAME meeting, and sitting in lectures about the cutting-edge research others are conducting, or the challenging cases that others have successfully investigated. I can't sit for more than 10 minutes before having that "shoulda coulda woulda" feeling about some of my own cases. Yet at the same time, when I leave the conferences, I feel invigorated. Forensic science can be incredibly isolating, especially if you are the only doctor in a small rural Coroner's office. Lunching and dining with colleagues makes you realize that there is camaraderie and support; that we may not always agree on the best way to interpret an injury, or certify a death, but we can come together, break bread and do what scientists do best: collaborate.

The next NAME meeting is: Sept 19-23, 2014 in Portland, OR
The next AAFS meeting is: Feb 16-21 in Orlando, FL