D.C. DNA Crime lab fails accreditation

[Boyd Baumgartner]

WaPo Article

The problems with the District lab have centered on the analysis of evidence that includes DNA from more than one person — and their conclusions regarding the likelihood that a certain person’s genetic material is included in the sample.

In their initial findings earlier this year, experts contracted by the U.S. attorney’s office said the problem came to light when an outside expert reviewed the DNA analysis conducted in a rape case. The biggest mistake involved the analysis of DNA found on a stolen car’s gearshift, prosecutors said. D.C. analysts looking at the evidence found that the car owner’s DNA could have been on the gearshift and said the chance that a randomly selected person had the same genetic traits was 1 in 3,290. The outside experts said the more accurate finding was 1 in 9.

I'd say the interesting point of discussion comes from the fact that the NAS called DNA the Gold Standard for forensic science. As I mentioned in the FBI Exaggerates Thread, it seems as if the 'grass is greener' recommendations that were proposed and that our industry wants to adopt (statistical foundations, accreditation, etc) is not exactly the panacea we are lead to believe.

The other big point of discussion in the article is this:

The D.C. lab was audited last fall by the same accreditation organization that the city used to perform the new review. Houck said his lab passed the previous audit by the group. The FBI also audited the lab last year, he said.

In an interview, Terry Mills, an accreditation manager for the board that issued the report, said the audits were only a “snapshot in time” of a lab’s operations and that board officials would have to be “camped out for about six months” to thoroughly investigate a lab’s procedures.

The age old adage of who polices the police rears its head. This could be said about any of the auditing organizations from ANAB to Ron Smith and Associates to Matthew Schwartz. It would be interesting to see if errors made by auditing bodies have been found or reported and what the outcomes were. It seems that accreditation/audit is really just a compounded problem. It's just an appeal to authority or a perceived independent body which may be overstating it's findings, which is exactly the problem audits/accreditation is supposed to prevent.

While I don't agree with all the findings, I would have to commend the FBI's method of handling of Mayfield with an independent contractor reviewing, Robert Stacey's Report and the OIG report working comprehensively.

[Kasey Wertheim]

Director of DC's embattled DNA lab resigns after suspension of testing

http://www.washingtonpost.com/local/dir ... story.html

The embattled director of the District’s first independent DNA lab resigned Thursday, a week after two audits found that the lab’s procedures were inadequate, thereby forcing a national accreditation body to suspend all of the lab’s DNA testing.

Max M. Houck has been the director of the District’s Department of Forensic Sciences since the lab opened its $220 million facility in Southwest Washington three years ago.

In addition to Houck’s resignation, two other senior officials, the chief scientist for the lab and the senior manager for DNA testing, were let go.

The interim director of the lab will be the city’s medical examiner, Roger A. Mitchell Jr.

Last week, the ANSI-ASQ National Accreditation Board determined in its audit that analysts at the lab were “not competent and were using inadequate procedures.” The authors of the audit, which was ordered by the District, gave the lab 30 days to address the concerns.

Despite the sweeping executive personnel changes, people familiar with the operations of the lab said its accreditation still remained at risk as a result of the board’s findings.

The Washington Post first reported last month that District prosecutors within the U.S. attorney’s office alerted city officials to what they described as numerous errors with the DNA analysis performed at the lab. Houck, whose resignation is effective this week, will not be paid for the rest of his contract.

Prosecutors stopped sending DNA evidence to the lab earlier this year.

Houck had a year and a half left in his four-year contract with the city as director of the lab.

Local prosecutors have ordered the review of 182 cases as a result of the errors they said they discovered in the lab’s DNA results. Prosecutors, as well as the accreditation board, had problems with the interpretation in DNA mixtures cases, those in which more than one person’s DNA is present in the evidence.

It remains unclear how many cases have been affected by the lab’s problems. The prosecutors said the errors have not resulted in the dismissal of any cases or in any exonerations.

[GrayMatter]

http://www.washingtonpost.com/opinions/ ... story.html

The article primarily deals with the suspicion that the personnel changes were related to lab management's proactive approach to scientific neutrality, there are several other illuminating comments regarding the interpretation of DNA mixtures:

To draw conclusions from such evidence, analysts must make subjective judgments about matters that cannot be determined with certainty, such as the exact number of contributors and the probability that the test failed to detect certain genetic characteristics of certain contributors. Reasonable experts have long been divided over the best way to analyze such evidence and how to report the results. While the method used by the D.C. lab is open to criticism, a 2013 study by the National Institute of Standards and Technology suggests that a majority of DNA laboratories in the United States follow the same method. The audit team pointed to several cases in which it deemed the lab’s interpretations to be problematic, but one could find similar examples in the casework of accredited crime laboratories nationwide.

[Boyd Baumgartner]

And the plot thickens....

http://dailycaller.com/2015/05/27/dc-ad ... t-firings/

A member of the Scientific Advisory Board for D.C.’s Department of Forensic Science sent in his resignation letter Wednesday claiming that political pressure, not bad science, led to the shuttering of a DNA lab last month.

This all sounds vaguely familiar with regards to the independence of ASCLD from ASCLD-LAB.

A quick look through the wayback machine shows us that there are a couple changes to the line up including Christine Funk, who has also had her hands in the very same topic as it pertains to our discipline. Anyone know why she left?

DCLab.jpg

LPC-Stats.jpg

I found this quote from Mr. Siegel to be interesting, and adds a new aspect to the 'emerging paradigm shift':

The only scientific issue of merit in this process was the protocols used in the interpretation of the level of significance of contributors to certain DNA mixtures. It is well known in the forensic science community that there are no standards for the statistical interpretation of such mixtures.

Someone correct me if I'm wrong, but it looks like Max Houck sought a culture shift towards scientific neutrality, and some people didn't like that. As a result they used the accreditation boards to re-audit and 'find' an issue that according to GrayMatter's link isn't really an issue at all and didn't raise any red flags on any previous audit. This then was used to railroad Houck and cast aspersions on the practitioners. This ultimately flies in the face of the NAS Recommendations that certification, accreditation and oversight be the trinity of quality assurance.

[Bill Schade]

Someone correct me if I'm wrong, but it looks like Max Houck sought a culture shift towards scientific neutrality, and some people didn't like that. As a result they used the accreditation boards to re-audit and 'find' an issue that according to GrayMatter's link isn't really an issue at all and didn't raise any red flags on any previous audit.

I see where you are going, but it is hard to believe that anyone has the power to have an accreditation board find a problem that isn't a problem. Thats not how it's supposed to work.

I don't understand the issues but there must be some objective support to take the action that was taken.

I prefer to believe that we just don't have all the information since the press is notorious for presenting facts in a way that "makes for a good story"

[Tazman]

I prefer to believe that we just don't have all the information since the press is notorious for presenting facts in a way that "makes for a good story"

An old friend who works in another discipline close geographically to the storm in DC tells me the mess is political, too. The press is in it for the controversy. If you had six protesters picketing a situation in which they were the only dissenters in a company of thousands of employees, the press would make it look like a 50-50 split in the organization. I think Bill has hit the nail on the head -- it makes for a good story.

I suspect the action was about politics and the publicity is about selling newspapers. Can anyone give me a reason to think that conclusion is wrong?

[GrayMatter]

To continue Boyd's play on words: Someone added cornstarch! The plot continues to thicken!

http://dailycaller.com/2015/06/14/us-pr ... irlfriend/

Here is what I find interesting. Some critics (Roger Koppl) advocate for silos and 3rd party laboratories to each independently analyze the evidence. The DC lab is a case where at some points two "independent" laboratories did actually analyze the evidence, one publically funded and one privately funded. Whose results do we trust? Whose are accurate? Interesting considering we have Jay Siegel (NAS 2009 Committee Member) saying that there is no community standard on how to interpret complex mixtures like this. Is the problem crime lab independence or lack of standards? I will bet they say it is both. Some one is always moving the goal post.