Another Forensics Competency Test; More Bad News

Researchers at the University of London used a suction device to create bruises on 11 subjects. They then took sequential photos of the bruises over nine days. They showed the photos to 15 forensic experts and asked them to order the photos chronologically. Forensic experts regularly testify about the age of bruises down to the hour, providing chronologies that allow prosecutors to give a suspect a window to have committed a crime. Or, alternately, to attempt to put a the crime at a time for which the defendant has an alibi.

But these particular 15 experts didn't fare so well. The results:

Lead author, Margaret Pilling, an Honours Medical Student at Barts and the London School of Medicine and Dentistry, said: "The greatest accuracy, from forensic experts, occurred in very fresh bruises (between 0 and 12 hours) however there were still a number of significant misjudgements in this age range.

"The median difference between the estimated age and the real age was 26 hours - a considerable disparity. We conclude that forensic experts' estimates of bruise age from photographs are, at best, unreliable."

More on forensics aptitude here, here, and here.

Comments

[Jordan Elliot]

But they work when they use it on CSI!

Joking aside, is there a real possibility that the unreliability of some forensics will ever be acknowledge by law enforcement and prosecutors? I won't hold my breath and I doubt that it will catch on with the public. Barring some drastic case of inaccuracy.

[wylie]

The same day cops get the personal video recorders on them, and actually become responsible for their abuses of authority.

Read: Definitely don't hold your breath. Unless you're an eco-nut, then by all means hold it. It's good for gaia.

[Zeb]

I am not going to hold my breath, better things to do. But I think that there is a reasonable possibility that enough courts will hear about studies like this and decide some cases in such a way as to force law enforcement and prosecutors to acknowledge the shortcomings of some of the forensic techniques commonly used today.

[TrickyVic]

""Joking aside, is there a real possibility that the unreliability of some forensics will ever be acknowledge by law enforcement and prosecutors? ""

I thought you said joking aside. We are talking about people who won't acknowledge an obvious lie, is a lie when it works in their favor.

[C Harm]

Isn't an estimate always just that? Is the problem with the interpretation and not the actual testimony?

[B?nzli]

"The median difference between the estimated age and the real age was 26 hours - a considerable disparity. We conclude that forensic experts' estimates of bruise age from photographs are, at best, unreliable."
They should have made a comparison study and let 'non-experts' guess the age. Would be interesting to see how good they'd have done.

[J sub D]

I completely agree. They should have use housewives as a control group.

[Groovus Maximus]

Researchers at the University of London used a suction device to create bruises on 11 subjects.

I see Russell Brand has found gainful employment while in the UK.

[virginia]

from photographs

Unfortunately, if someone tries to pull this study out in court the prosecution will say the forensive expert saw the bruise in a lab environ and not in a picture.

[Axman]

Different people bruise more or less easily, depending on platelet count and the meds they're on, and their bruises fade at different rates. Also, deep bruises last longer than skin bruises (like the ones produced by suction) on the same person.

Given the variables, estimating the age of a bruise to an accuracy of a day or so actually seems pretty good. But that's only a median, meaning that any particular estimate could be a whole lot farther off--so far wide of the precision claimed by experts that even pleading that lab examination is better than looking at photos should not restore credibility.

[Jennifer]

I was going to say something similar -- I seem to bruise very easily, and furthermore when I get a bruise it seems to last longer than does a similar bruise on my friends. So what would be a three-day-old bruise on my friend's arm is more like a week-old bruise on mine, so without knowing my medical quirks inside and out, I highly doubt anyone, even an MD, could look at a bruise on me and accurately estimate how old it is.

[Groovus Maximus]

From photographs, no. A tactile examination would be much more accurate. Factors include med HX, amount of edema present, and quality of color and type of contusion. I'm a DO.