Homicide, Sexual Assault Cases On The Line After Crime Lab Discovers Tech Using The Wrong Tools For The Job

from the more-faith-than-science-in-the-forensic-science-community dept

When life and liberty are on the line, law enforcement lab techs are there to turn hard science into a roulette wheel. Once you get past the fact that a lot of forensic investigative techniques are little more than junk science, you run directly into the failures of the humans staffing forensic/drug testing labs.

In the state of Massachusetts alone, more than 30,000 cases are in the process of being tossed due to lab tech misconduct. One lab tech faked most of her work, speeding through her workload by faking tests and test results. Another used the drug lab as her own personal drug stash, using whatever substances she wanted from incoming evidence and replacing it with filler.

Forensic science is plagued with incompetence and overconfidence, which is an incredibly bad combination when people's freedom is on the line. Only in recent years has the DOJ instructed forensic experts to stop overstating the certainty of their findings. But that hardly fixes the problem. Outside debunkings have led to zero changes in law enforcement forensic work -- a fact so disheartening a judge very publicly resigned from a committee seeking to fix these problems when it became apparent the committee wasn't actually supposed to fix anything.

Here comes more bad news on the forensic front, via criminal justice blog Grits for Breakfast.

The Houston Forensic Science Center has fired a crime scene investigator who violated policy by using unapproved equipment that resulted in false negatives for biological evidence in at least two sexual assault cases, officials said Friday.

Lab officials, who fired investigator Tammy Barette Thursday, said they may never know the full impact her actions may have had on other cases.

Well, sure, you may "never know" if you don't go looking. To be on the safe side, you could consider everything this investigator ever touched tainted. That would cover the "full impact" with some to spare. But if this is handled like any other case of lab misconduct, it will take a court to force a full accounting of the damage.

From what's been uncovered so far, it appears the lab tech used her own equipment to identify the presence of bodily fluids that contain DNA. Her personal tool didn't meet the lab's requirements, making any results of hers suspect. A lab supervisor found 19 cases in which the inadequate tool was used. Even in this limited sample, there were serious problems.

Out of 19 cases where Barette used the improper light source, including the case that sparked the investigation, only four had evidence available to retest, agency spokeswoman Ramit Plushnick-Masti said. Two of those were wrongfully marked as negative for the presence of biological fluids, when they should have yielded positives.

15 more cases "might have had a different outcome." That's not very comforting, considering this lab usually handles violent crimes like homicide and sexual assault.

The lab has responded in the most government of ways: by rewriting a policy that was already in place forbidding the use of personal equipment. The language has been toughened up to make it clear the violation that was always a violation is a policy violation. The single addition is the requirement for investigators to write down the model number of the equipment they use to perform tests. That should certainly prevent any future misconduct. [Gathers rolling eyes from under desk.]

The good news is the lab immediately fired the investigator. The bad news is the problem went undiscovered long enough to do some serious damage. Most of the damage is reputational, providing criminal suspects with ammo to challenge lab findings. If one tech can perform unauthorized tests, surely other accidents and misdeeds have been overlooked. When the lab itself can't say how many cases were affected -- or even how many tests might have been handled with inadequate equipment -- the potential fallout could couple the jailing of innocent people with the release of actually guilty prisoners.


Reader Comments

The First Word

Subscribe: RSS

View by: Time | Thread


  • icon
    That One Guy (profile), 26 Nov 2018 @ 1:42pm

    Not a hard problem to fix, or at least vastly reduce

    Run regular, random tests by an outside, unbiased party, with any violations treated harshly to 'you are fired, and at a minimum you will never work in the field again' level penalties, scaling to the severity of the violation.

    We're talking about testing that can be the difference between prison or freedom for people accused of various crimes, where faulty testing can result in innocent people going to jail for crimes they didn't commit and/or guilty people walking for crimes they did commit.

    Ensuring that the testing is actually accurate should be the absolute minimum that's done.

    reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 26 Nov 2018 @ 2:01pm

      Re: Not a hard problem to fix, or at least vastly reduce

      Every test should require testing by 3 different labs and a single disagreement between the labs should force a retest. If all 3 labs cannot agree on the result then it should not be admissible in court. Oversight should be in place to prevent the labs working together to "agree" on a result, to include prevention of law enforcement persuading them or otherwise affecting the outcome.

      I'm all for putting criminals away but not if the innocent are rounded up in the process.

      reply to this | link to this | view in chronology ]

      • identicon
        hegemon13, 26 Nov 2018 @ 2:06pm

        Re: Re: Not a hard problem to fix, or at least vastly reduce

        I like this idea. Even two separate labs would help. And the distribution should be completely blind until after the results are returned. Neither law enforcement nor the individual labs should know which labs received the samples for testing. This should help prevent the kind of collusion you mention, which would definitely become a problem.

        reply to this | link to this | view in chronology ]

      • icon
        madasahatter (profile), 26 Nov 2018 @ 3:41pm

        Re: Re: Not a hard problem to fix, or at least vastly reduce

        If the samples are sent blind to the labs selected randomly from a list that would make collusion much less of a problem. The other is to have an FDA type list of bad apples who are banned from the industry.

        reply to this | link to this | view in chronology ]

        • icon
          JoeCool (profile), 26 Nov 2018 @ 6:24pm

          Re: Re: Re: Not a hard problem to fix, or at least vastly reduce

          The issue is that there seems to be a LOT more bad apples than thought, but no one's looking for them. They just kinda stumble across them randomly when they do something incredibly stupid and someone is actually paying attention for a change. Labs need to be certified, then recertified every year. Certification should be by a neutral third party with labs picked at random with no warning. It should be more stringent that any corporate random drug testing of employees.

          reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 26 Nov 2018 @ 4:15pm

      Re: Typical government bureaucrats

      the Massachusetts government couldn't even competently hire & supervise its very own lab workers -- so how is it gonna find and supervise outside labs that are unbiased, diligent and highly competent ?

      The incentives for both inside & outside labs are to work slowly and cut corners -- the paychecks look the same no matter what they do, as long as there is a superficial appearance of competence.


      Core problem is that government prosecutors/cops control all the primary evidence in a criminal trial ... and the defense is barred from equal & timely access to the evidence. The government wants convictions, not acquittals -- and the whole government justice system operates on that basis.
      Sloppy forensics labs are fine, as long as they provide adequate testimony for convictions -- the labs know what prosecutors want.

      reply to this | link to this | view in chronology ]

    • identicon
      Bruce C., 27 Nov 2018 @ 5:13am

      Re: Not a hard problem to fix, or at least vastly reduce

      Good luck with that.

      How do you distinguish a violation from honest error or false positive/negative due to equipment or process limitations? For that matter, how can you tell if the error isn't on the third party side?

      And what constitutes an "outside, unbiased party"? The crime lab in a different state?

      A good portion of these errors are due to laziness or case overload, not bias on the part of the lab tech. Heck, even in this case we don't know if she started using her own equipment just for convenience or because the official tool was broken, required an appointment or had other hurdles that reduced her case throughput unacceptably.

      The best solution is adequate funding of the crime labs: for equipment, employee retention and for training. The same applies to the courts and so many other areas of government. One of the big reasons government is broken is that we expect more out of it than we're willing to pay for.

      reply to this | link to this | view in chronology ]

      • icon
        That One Guy (profile), 27 Nov 2018 @ 10:13am

        Re: Re: Not a hard problem to fix, or at least vastly reduce

        And what constitutes an "outside, unbiased party"? The crime lab in a different state?

        That would be one option, yes.

        When I said testing I meant of the 'double-blind' sort, where a sample that is known to either contain or not contain a particular substance is sent in. If the lab flags it incorrectly then it would be clear that they are doing it wrong, intentionally or not.

        How do you distinguish a violation from honest error or false positive/negative due to equipment or process limitations? For that matter, how can you tell if the error isn't on the third party side?

        For the 'intentional/accidental' violation question, frequency would probably make a pretty good metric to use. Did they make this sort of mistake once, or were they doing it on a regular basis? Once could be an accident, nineteen times, as was the case here, was most certainly not.

        As for how you check the accuracy of the testing of the lab people have been coming up with various ideas, to which I would add 'just like you check the accuracy of other forms of measurement': Multiple data points from multiple sources, checking how well they match. If lab 1 says the sample is positive, and 2 says it's negative, then there's a problem that needs to be addressed.

        A good portion of these errors are due to laziness or case overload, not bias on the part of the lab tech. Heck, even in this case we don't know if she started using her own equipment just for convenience or because the official tool was broken, required an appointment or had other hurdles that reduced her case throughput unacceptably.

        Irrelevant. Whether someone is screwing up the tests intentionally or not, the end result(innocent people in jail, guilty people walking) is the same, and as such both deserve to be cracked down on harshly. If a lab tech doesn't have the tools they need for whatever reason then they can make this known to their superiors, and if that doesn't work take it public('I do lab testing of evidence used in trials, and yet our superiors don't care if it's accurate' would be sure to get plenty of attention). What they should never do is treat their jobs as more important than the freedom or lack thereof of those on trial, and I have absolutely zero sympathy for someone who would come to that point and make the wrong choice.

        The best solution is adequate funding of the crime labs: for equipment, employee retention and for training. The same applies to the courts and so many other areas of government. One of the big reasons government is broken is that we expect more out of it than we're willing to pay for.

        Funding sufficient to do their jobs properly would be of great help, sure, but ultimately useless if they're not using it properly via making sure that the tests they perform are accurate. The most amazing gear and highly trained lab techs are no different than junk scrounged from a garbage bin and a bunch of kids if they aren't making proper use of the tech and training.

        reply to this | link to this | view in chronology ]

  • icon
    That Anonymous Coward (profile), 26 Nov 2018 @ 2:20pm

    Its just the lives of of 'bad people' that got screwed up so it isn't a big deal.

    We deserve better & much more accountability.
    We talk about it being better that 100 guilty men walk rather than an innocent put in jail for a day... now we have 100 guilty 'men' abusing the system to stick someone in jail & pat themselves on the back for a quick closure rather than the right outcome.

    reply to this | link to this | view in chronology ]

    • icon
      stderric (profile), 26 Nov 2018 @ 2:45pm

      Re:

      We talk about it being better that 100 guilty men walk rather than an innocent put in jail for a day...

      I can never let a Blackstone/Voltaire/Franklin reference go without mentioning the Dick Cheney Revision:

      I'm more concerned with bad guys who got out and released than I am with a few that, in fact, were innocent.

      reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 26 Nov 2018 @ 3:42pm

      Re:

      Its just the lives of of 'bad people' that got screwed up so it isn't a big deal.

      Actually, this time, it's quite the opposite. By generating false negatives, the technician may have caused acquittals or dismissals in cases where the government would have secured a conviction if the evidence were tested properly. Due to the double jeopardy rules, an acquittal based on the forensic report stating "No match" would protect the defendant, so the state cannot go back and retry the case with the proper evidence, even if they have incontrovertible proof that the test, properly administered, shows the defendant to be guilty. Thus, the "bad people" were rewarded, and the people whose lives are screwed up are the victims who will not get justice. If the perpetrator reoffends after this erroneous release, we could speculate that those victims might have been spared if the perpetrator had instead been convicted and incarcerated.

      reply to this | link to this | view in chronology ]

  • icon
    Coyne Tibbets (profile), 26 Nov 2018 @ 3:13pm

    The Houston Forensic Science Center has fired a crime scene investigator who violated policy by using unapproved equipment that resulted in false negatives for biological evidence in at least two sexual assault cases, officials said Friday.

    It rather seems to me that they are far quicker terminate someone for false negatives then they are for someone who yields false positives.

    I mean in this case, she was terminated "immediately" for using unapproved equipment that yielded false negatives. But my recollection is that previous investigations into false positives have taken months, years even, and involved legal system foot-dragging at every inch of the way. In the Annie Dookhan case for example, it was said that their first response to the discovery that she was falsifying tests, yielding false positives, was to give her even more tests to falsify.

    Isn't that odd?

    reply to this | link to this | view in chronology ]

    • icon
      madasahatter (profile), 26 Nov 2018 @ 3:47pm

      Re:

      Not really surprised given most prosecutors want to win cases not find justice. There have been a few who cared about justice to the point of not prosecuting someone because they did not believe the defendant was guilty but the circumstantial evidence seemed to point to guilt.

      The old Dana Andrews movie "Boomerang" is based on a 1924 unsolved murder in Bridgeport, CT where the DA (Homer Cummings in real life) refused to prosecute the defendant. He spent 90 minutes debunking the state's case much to the disgust of many. The case was never solved.

      reply to this | link to this | view in chronology ]

    • icon
      ShadowNinja (profile), 27 Nov 2018 @ 6:59am

      Re:

      Not to mention this stuff can be used to wrongly execute people, like Cameron Todd Willingham of Texas.

      Cameron Todd Willingham's case is the reason I went from on the fence about the Death Penalty to strongly opposing it. He was sentenced and executed purely based on junk science for arson cases.

      reply to this | link to this | view in chronology ]

  • icon
    Killercool (profile), 26 Nov 2018 @ 6:18pm

    Since no-one has mentioned it...

    What massive defect did this lab have, to let slide the use "personal equipment" in the first place? I'd say the "reputational damage" was well deserved, and probably not nearly severe enough, by the time it's over.

    I've done lab work myself, albeit for a much less important but more lucrative industry. What sort of idiot hires the sort of idiot that would use their own tools instead of those supplied by the lab? You're supposed to periodically test and calibrate everything. Even at my job, using improper equipment wouldn't just get you canned, there was a possibility of real legal consequences. Filling out documentation with those numbers made your documents fraud. For a crime lab, it would seem to be fraud AND perjury.

    And let me tell you, the tech's coworkers almost definitely knew what was going on. The sort of person who prefers their personal equipment over supplied equipment isn't quiet about it. They complain to anyone who'll lend them an ear, and typically act proud of the fact that they've "got one over" the supervision.

    It's one thing doing quality control for the petrochem industry, but it's a whole different matter when justice for victims and liberty for the falsely accused are at stake. The fact that "only" 19 cases are in question is ALMOST DEFINITELY a lie. This sort of bad behavior is usually systemic, and has a lot to do with the management tolerating otherwise fire-able offenses.

    In my professional opinion (literally), everything the lab touched is poisoned. Unless they can prove otherwise, any evidence that went through them should be suspect. Frankly, if the DA/PD/whoever keeps using that lab, it's just evidence that they like the kind of false results the lab usually sends them.

    reply to this | link to this | view in chronology ]

    • icon
      stderric (profile), 26 Nov 2018 @ 9:26pm

      Re: Since no-one has mentioned it...

      I used to work at a research facility, and this guy Ted would always bring in his personal UHPLC system to use at the end of the week. His reasoning was that "Dude it's, like, casual Friday, right? And my home gear's got, you know, a wicked awesome skull sticker on it... way laid back and cool, see what I'm sayin'?"

      My boss decided to let him keep doing it, telling the rest of us that "Ted's logic is impeccable, so if I stopped him we'd all have to go back to wearing ties five days a week."

      reply to this | link to this | view in chronology ]

      • identicon
        anon, 27 Nov 2018 @ 4:41am

        Re: Re: Since no-one has mentioned it...

        @iconstderric
        https://dilbert.com/strip/2000-12-19
        Transcript
        The Boss says to Catbert, "Casual Dress Day is hurting our productivity. We need to cancel it."
        Catbert says, "Is it possible that our real problems are caused by irrational management?"
        The Boss says, "No, I think comfortable pants are the problem."
        Catbert says, "Sounds right."

        reply to this | link to this | view in chronology ]

        • icon
          stderric (profile), 27 Nov 2018 @ 10:04am

          Re: Re: Re: Since no-one has mentioned it...

          The more I comment, the more I realize that my ideas are just bargain-basement reformulations of material from Dilbert, xkcd, and Gravity's Rainbow.

          reply to this | link to this | view in chronology ]

  • icon
    Mononymous Tim (profile), 26 Nov 2018 @ 7:26pm

    I have a black light from Walmart and I'm not afraid to use it!

    reply to this | link to this | view in chronology ]

  • icon
    Bamboo Harvester (profile), 27 Nov 2018 @ 5:40am

    There's more to this...

    ...Looking at the news reports, she was apparently using a limited frequency UV flashlight at crime scenes rather than the spec'd ALS unit.

    The real question is... WHY?

    Did the lab not have ALS unit(s) available?

    Did she just not feel like lugging a heavy, bulky, ALS unit around?

    I have found that when employees use their own, personal, tools, it's usually because management isn't supplying the correct tools in the required quantities.

    Also, she doesn't appear to be a Lab Tech, but a Crime Scene Investigator. So did she use the UV flashlight at the crime scene to look for bodily fluids, or at the lab on bagged evidence?

    While either case reflects badly on the lab, the latter would mean a complete lack of supervision in the lab itself.

    This one is a rare exception to the usual bad lab work - criminals may have gone free, rather than innocents being wrongfully convicted.

    Which excuses nothing.

    reply to this | link to this | view in chronology ]

  • icon
    tom (profile), 27 Nov 2018 @ 8:40am

    She used the unapproved gizmo at crime scenes and no one there noticed the wrong device? What does this say about the other investigators on scene?

    It is very likely her supervisor knew and had tacitly approved her using the thing as a way to get the job done because purchasing wouldn't approve a new approved gizmo. Of course, once the crap hit the fan, said supervisor all of a sudden forgot said verbal approval. If the supervisor didn't know, we have to wonder what else is out of whack in that lab.

    reply to this | link to this | view in chronology ]

    • icon
      Bamboo Harvester (profile), 27 Nov 2018 @ 1:10pm

      Re:

      The six or so articles I read on the case just say "ALS". I used one about 25 years back, pre-LED revolution, it was as big, heavy, and clunky as a sewing machine. I can't see anyone taking such to "the average" crime scene.

      A true ALS maybe (most likely is) much smaller and lighter.

      I've got a couple of LED "flashlights" that throw multi-band UV and IR. They'll light up just about any biological residue.

      I'm wondering if such "flashlights" ARE considered an ALS (technically, of course, they are), and she was simply using a cheaper model.

      There's mention of missing three positives out of eight tested. Which makes me think that she was using a short-wave UV device that turned up blood residue just fine but not saliva or semen. They didn't mention what the eight samples were *of*.

      Pointless in this case, but in the usual run of things this exemplifies why the Defense should ALWAYS have any lab-tested items (including fingerprints) independently tested at a different lab.

      reply to this | link to this | view in chronology ]

  • icon
    Tanner Andrews (profile), 1 Dec 2018 @ 9:23am

    why the Defense should ALWAYS have any lab-tested items (including fingerprints) independently tested

    Sometimes that is not practical. EIther the state has conveniently used up/lost the samples, or there is no budget for it.

    Remember that, for budget purposes, the state has the full budget of state government behind it. Few defendants have the same access to funds. That assumes that samples can even be had for testing. Many times the state is very reluctant to allow defense testing.

    Also, there is essentially no cost-shifting: defendant wins, that's nice, but he does not get his fees or costs. Interestingly enough, the state automatically gets to add ``investigation costs'' to pretty much any judgment and sentence.

    reply to this | link to this | view in chronology ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Use markdown for basic formatting. HTML is no longer supported.
  Save me a cookie
Follow Techdirt
Techdirt Gear
Shop Now: I Invented Email
Advertisement
Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Advertisement
Report this ad  |  Hide Techdirt ads
Recent Stories
Advertisement
Report this ad  |  Hide Techdirt ads

Close

Email This

This feature is only available to registered users. Register or sign in to use it.