Reason.com

Free Minds & Free Markets

A Rogues' Gallery of Bad Forensics Labs

Despite improvements in DNA matching and reliability, forensics labs across the country still continue to train and monitor technicians improperly.

Despite improvements in DNA matching and reliability, forensics labs across the country still continue to train and monitor technicians improperly, resulting in shoddy work and tainted cases.

Austin, Texas

The Austin Police Department's DNA lab closed last June after state officials discovered it had been using outdated statistical methods for years, which may have led it to overstate its confidence in DNA matches. New, more reliable methods were established in 2010 and eventually adopted by every accredited DNA lab in the country—except Austin's.

"One would think it would have been picked up," Lynn Garcia, general counsel for the Texas Forensic Science Commission, told the Austin American-Statesman.

There were other problems as well, including freezer failures. It will cost $7–$14 million to review the roughly 4,000 cases suspected to be impacted by the lab's substandard performance.

Broward County, Florida

The Broward County Sheriff's Office crime lab temporarily suspended some of its DNA testing last July. Accreditors found it had been using faulty procedures that could misidentify inconclusive results as positive matches.

Local public defenders and DNA experts had filed an official complaint with the American Society of Crime Lab Directors (ASCLD), claiming the lab was using incomplete or mixed DNA evidence. "The DNA identification process used in the crime lab and relied upon by your office is tantamount to junk science," Broward Public Defender Howard Finkelstein wrote in a scathing letter to the state attorney. "Your office should immediately halt the use of DNA mixture results generated by the crime lab in all pending criminal prosecutions to prevent wrongful and unjust convictions."

After an investigation by the ASCLD reached similar conclusions and the group threatened to strip the lab's accreditation, the sheriff's office vowed to adopt more stringent procedures.

Houston, Texas

The Harris County District Attorney's Office sent letters to local defense attorneys in 2014 notifying them that a Houston DNA lab technician had tampered with official evidence, possibly affecting up to 185 criminal cases, 51 of them murders. It was the latest in a string of scandals for the local forensics laboratory.

The Houston cops' DNA testing lab had already been closed twice—once in 2002 and again in 2008—for "serious management, employee and structural problems," the Houston Chronicle reported, "including a leaky roof that for years dripped water on stored evidence."

In 2005, a probe by an independent investigator found still more problems. According to the Chronicle, the lab had bumbled along for 15 years "as employees failed proficiency tests, botched analyses and taught themselves scientific technique by reading books at home."

DNA testing later overturned three wrongful convictions resulting from the gross mismanagement of the Houston lab.

Prince George's County, Maryland

Prince George's County police suspended a DNA lab employee in April and launched an audit of the facility.

"The ongoing county review uncovered neglected DNA profiles that should have been entered into a national database, lags in notifying investigators of DNA profile matches and the use of outdated methods to calculate the individuality of profiles," The Washington Post reports.

According to that paper, the county began the investigation after learning the employee in question had certified the Austin Police Department's now-shuttered DNA lab while serving as an auditor for the ASCLD.

San Francisco, California

The San Francisco Police Department announced in March 2015 that it was reviewing 1,400 criminal cases potentially tainted by a lab technician and her supervisor, both of whom later failed a DNA testing proficiency exam and were barred from processing evidence, according to documents obtained by the San Francisco Chronicle.

The duo's misconduct came to light during a child molestation trial, when it was revealed that the lab tech filled in missing data—with approval from her supervisor—from two low-quality DNA samples and submitted them to investigators as complete genetic profiles.

The defendant was convicted.

Washington, D.C.

In 2015, a major accreditation board stopped all DNA testing at the Washington, D.C., crime lab after the U.S. Attorney's Office said it discovered numerous errors in the lab's work. The audit revealed "insufficient and inadequate" procedures at the lab, leading to the resignation of its director and several top staff.

The U.S. Attorney's Office then ordered a review of 182 cases possibly tainted by the shoddy work, at a cost of hundreds of thousands of tax dollars.

Photo Credit: 18percentgrey/Dreamstime.com

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  • Scarecrow Repair & Chippering||

    What annoys me most about these "revelations" is that they are so chancy. How many other labs are just as shoddy but not yet uncovered?

    Seems to me the only real fix is to get rid of government testing altogether. Farm out everything to independent for-profit labs who depend on their reputation for staying in business instead of political favors. Send multiple samples to different labs selected randomly from a short list, make sure the defense gets to do the same, and raise holy hell for discrepancies. Put those discrepancies in their Yelp reviews.

    Same old story, government fucking things up. They only do two things well: fuck things up, and expand.

  • Jerryskids||

    Imagine the reaction you'd get to a proposal that the public defender's office operate the crime labs. Why? Aren't we all part of the criminal justice system, all of us officers of the court, all of us fair and impartial neutral parties dedicated to seeing that justice is done? Why would there be some presumption that the public defender's office would be biased but the crime labs operated by law enforcement are strictly neutral? Bullshit - they're only looking for incriminating evidence, exculpatory evidence is returned with an "inconclusive results" tag, it's no evidence at all.

  • Lester224||

    You've left out the Annie Dookan fiasco in MA.

  • Jerryskids||

  • Jerryskids||

    The money quote, emphasis added:

    Indeed, according to Business Insider, "In many jurisdictions, crime labs receive money for each conviction they contribute to, according to a 2013 study in the journal Criminal Justice Ethics. Statutes in Florida and North Carolina mandate that judges provide labs with remuneration "upon conviction" and only upon conviction. Alabama, Arizona, California, Missouri, Wisconsin, Tennessee, New Mexico, Kentucky, New Jersey, and Virginia are among the states with similar provisions."

    How fucked-up is that?

  • phenryinohio||

    Wait! This is a real thing? Like in the law and all that official stuff?

    I call shenanigans!

  • Half-Virtue, Half-Vice||

    Batman will always have the best rogues' gallery.

  • Citizen X - #6||

    ...though you wouldn't know it from watching Suicide Squad.

  • Fist of Etiquette||

    The Austin Police Department's DNA lab closed last June after state officials discovered it had been using outdated statistical methods for years, which may have led it to overstate its confidence in DNA matches.

    Well, prosecutors aren't going to use labs who keep exonerating defendants. Doy.

  • Bubba Jones||

    You don't expect me to admit that I couldn't get a full DNA sequence out of that bloody sock, do you?

  • Radioactive||

    CSI: Dogpatch...no dental records and the DNA is all the same

  • Radioactive||

    and they only have thumb prints, cause their fingers was all burnt from the meth cookin...

  • Fuck You - Cut Spending||

    I consider DNA-matching and fingerprint-matching to be junk science. They are based on probabilities and by definition are not accurate; they are the climate science of forensics - relying on consensus rather than proof. If I were on a jury, DNA and fingerprint evidence would not be enough to get me to convict, although if junk science can't even provide a match, then I'm happy to have a convicted person go free based on DNA lack-of-evidence.

  • phenryinohio||

    The statistical analysis of DNA is pretty good today in private labs. They can call somewhere between 12 and 27 markers in a $100 paternity test(test also includes paternity calculation on the 3 parties).

    Meaning for $50 you can get let's say 12/17 markers for one human. There can be contamination. But a knowledgeable Ph.D. can sort that out. Hell, I could using software. And rerun test. I think CODIS which is the US database, uses 8 or maybe 12. I forget what I learned but a match was like 6 to 8. I don't know the stats as the number of markers increase. At some point, it's diminishing returns. But most Paternity labs require 12 to 20 today for "chain of custody" testing to prove you are the Daddy.

    QC is everything in a lab. So I'd go for external testing in 2 accredited labs. Or at least a Gov lab and one external. Much less likely an external lab that does more than CODIS testing is going to FU QC. CODIS testing even in private is a very fast and cheap, down and dirty process.

  • Hank Phillips||

    Freezer failures is one way of putting it. Opening a police lab freezer in Austin Texas to find it is 80ºF inside is... is... too much information! Sumptuary laws are turning These States into a nation of predatory snitches and fraudulent vampires. But as long as illiterate rubes can be frightened out of "wasting" a vote by letting the LP have another 21 votes' worth of law-changing clout, the soft machine's got the sanction of the victim on its side.