My apologies for replying to old threads, but I'm just now catching up on
over a month's worth of posts.

My thanks to Brent and all those who have expressed their sympathy and
support; it is appreciated. Be assured, however, that I am in no danger of
leaving (or being "removed") from the profession. Like many news reports,
this one is basically factual but incomplete, and so is somewhat misleading
(unintentionally so).

In a nut shell, after over 24 years of professional work and the correct
completion of hundreds of proficiency tests and test samples, plus well over
ten thousand case samples, I made my first error in analysis (that I know
of). In fact I made two errors, among a series of nine drug analysis
proficiency test samples completed in a three-month period. The errors were
a result of overwork, overstress, chronic insufficient sleep, and rushing to
meet too many coincidental deadlines. I have identified the causes, and
taken appropriate measures to remedy the situation and ensure it will not
recur. As far as I'm concerned the problem is permanently resolved, but we
must still deal with the inevitable consequences (fair or not) of my two
mistakes, and we are working diligently to do that.

For those interested, I am willing to discuss at length what happened as an
educational exercise, in the hope that it may perhaps help someone else
avoid similar complications. So if you'd like to learn how a well-educated
and well-trained person with decades of experience could still manage to
make a couple of "bonehead" errors in proficiency testing, read on. If not,
you can stop here.

---------------------------------------------------------

For the record, everything I write here is of my own volition and at my own
impetus as an individual professional and private citizen. No one has
requested this commentary and none of it has been reviewed or endorsed in
any way by my laboratory's administration. The viewpoint expressed is
entirely my own.

First, I'd like to say that, contrary to the impression some have gotten
from the newspaper article, the administration of my laboratory has been
nothing but supportive of me; and that all of my supervisors (at every
level) have expressed their continued confidence in my professional
capabilities. I am extremely honored and humbled, in fact, by their
steadfast fidelity. The challenge for me lies not in reestablishing my
administration's confidence (since I never lost it), but rather in
reestablishing the confidence of the public (who don't know me or have any
reason to have confidence in me, beyond what they hear and read). As has
been noted, many people unreasonably expect the profession of forensic
science to have a zero error rate, a standard that no other profession I
know of is held to. Even surgeons and hospitals have recorded and tracked
error rates, but so long as there is no gross negligence or ongoing patterns
of repeated errors, they continue to practice in spite of occasional
missteps because the medical profession recognizes that in any human
endeavor error is not only possible, but inevitable, no matter how high the
level of competence. Medicine and forensic science share an especially
demanding burden of excellence because in both professions the stakes are
about as high as they come. We can not (and should not) tolerate anything
less than excellence, but even the highest degree of excellence does not
(and can not) guarantee a zero error rate. The key is to have a quality
system in place to promote the highest level of performance achievable to
begin with, and then to detect, identify, and correct errors when they do
occur, remedy their causes, and prevent recurrences. All reputable
professions have such mechanisms, and forensic science is no exception.
Accreditation organizations (including ASCLD-LAB and ISO) universally
recognize and support such a process; none I know of require the mythical
and unattainable "zero error" standard.

As for me, my first error was in the Collaborative Testing Service (CTS)
drug analysis proficiency test (PT) I completed in June of last year, in
which I correctly identified alprazolam in one of the two test samples, but
mistakenly identified it again in the other test sample. The second sample
in fact contained no controlled substances. This was a sampling error, not
an analytical error, but it still constitutes what is referred to as a
"false positive," surely the most serious of reporting errors in this kind
of work. My initial analysis of the second sample revealed no controlled
substances, and had I left it at that all would have been well. However,
following our protocol for apparently "negative" samples, I repeated the
analysis of the "negative" sample at a higher concentration (this is in
order to detect low concentrations of drugs and avoid a "false negative").
Apparently, when I attempted to re-sample the second test item to repeat its
analysis, I accidentally re-sampled the wrong bag, such that I actually
reanalyzed not the second item but rather the first one (thinking it was the
second one). Since, unbeknownst to me, I was in fact reanalyzing the
previously identified alprazolam sample, I naturally again identified
alprazolam, not realizing I had mixed up the two items. When the PT results
were published by CTS in August, we became aware of my error and made the
only responsible decision we could: I stepped down (at least temporarily)
from drug analysis casework until such time as we could determine the source
of my error and permanently correct it. We launched a thorough review and
investigation to determine what had gone wrong in my analysis of this PT
sample. My data was seen to completely support both identifications
(alprazolam was clearly indicated in both sets of the data I produced), yet
when I and another analyst analyzed the second sample again, we both
confirmed there was no alprazolam present; so some kind of contamination or
sample mix-up had to have taken place during my previous analyses of this
particular sample. All instruments, supplies, and processes were checked
and verified, and all possible sources of contamination ruled out, so the
only possible cause of the error left was sample confusion.

I believe I made this first error because I rushed through the proficiency
test in order to meet a variety of deadlines that could not be extended. I
had been heavily involved with our very recently (and successfully)
completed ASCLD-LAB accreditation inspections, so I was far behind in my
casework to begin with. We were at the time short one chemist, who was away
for an extended period at a training course, and I was hard pressed to keep
up with the increased drug caseload necessarily assigned to me. In the same
two-week period that I conducted the June CTS drug analysis PT, I also had
two other PTs with immediate deadlines in my other specialty (blood alcohol
analysis), a large number of new "rush" drug analysis requests for cases
scheduled for trial in the immediate future, and a dozen blood alcohol cases
that were rapidly approaching a regulatory deadline for speedy analysis.
None of these deadlines could be missed or extended. Rushing through case
analysis was of course unthinkable because other peoples' lives and liberty
were at stake, so I made the only choice open to me in order to meet all
deadlines - I rushed through the proficiency testing, reasoning that the
only risk of doing so was to me personally, and to my personal reputation
(in hindsight, I should have realized the reputation of the laboratory might
also be affected). I successfully completed both blood alcohol proficiency
tests (six separate test items) without difficulty or error, but as we've
seen, I made a mistake in the drug proficiency test by mixing up the two
samples in that test.

We next tackled the tasks of making sure no such error would happen again,
verifying that I had made no errors in any of my recent casework, and
rehabilitating public confidence. A poll was taken of other forensic
laboratories to see what their procedures called for in similar
circumstances, and a decision was made to go so far beyond what any of them
required as to leave no possibility of any lingering doubt regarding my
current drug analysis capabilities. Since there was no analytical error,
there really was no remedial analytical training applicable; but we decided
I would nevertheless review our training program reference materials. We
also decided that not six months, but more than a full year of my previous
drug casework (everything done since my immediately previous, successful,
CTS proficiency test) would be reexamined by other chemists. We also
decided I would immediately perform not one, not three, but a series of five
additional drug analysis external proficiency tests (each involving multiple
samples). When all this was completed without any errors found, it would be
followed by a period of supervised casework, wherein my work would be
watched closely by another staff member to verify that I made no sample
mix-ups or mismarkings, had only one sample open at a time (which is our
standard procedure), etc. This seemed a reasonable plan that would not
only confirm I had made no other errors, but also return me to drug analysis
service as soon as prudence would allow.

What we could not predict, however, is that before this
reexamination/reevaluation period could be completed, my physical and
cognitive condition would deteriorate due to a variety of converging
stressors. Not the least of these was my natural concern for the possible
repercussions of my first error -- for my lab, for my career, and for my
family; and my even greater concern that, despite my firm belief to the
contrary, it was conceivable that I could have made an undetected error in
casework during that same period of high deadline pressure. Frankly, I can
live with complicating my own life, and even my family's lives, through an
unintentional fault of my own (I can forgive myself for being human), but I
could not countenance making an error that adversely affected the life of an
innocent under investigation. Sources of considerable additional stress
included the death of a very close family member in July, and a very serious
car accident in August in which my car was broadsided by pickup truck and
totaled, sending me via air evacuation to a trauma center for tests and
observation (fortunately, the initial suspicion of serious injuries
ultimately proved to be unfounded - my injuries were relatively minor), and
the aftermath of that accident (insurance claims, transportation problems,
replacement car shopping, etc.). As a result of all this, my personal
condition had begun rapidly deteriorating, to a degree far more than anyone
realized (including me).

Nevertheless, in September I retuned from sick leave and immediately began
the series of remedial proficiency testing, while continuing my routine
casework in alcohol toxicology and routine administrative functions. I
successfully completed the first two new drug PTs in the planned series of
five, analyzing four separate samples and drawing correct conclusions in
each. At the end of the month I conducted the third PT, and therein made my
second mistake. There were three analytical samples in this particular PT,
and I correctly identified the controlled substance present in two of them.
In the third sample, I failed to detect the presence of a low concentration
of (ironically enough) the same drug I made my first error with:
alprazolam. Despite analyzing the sample twice, including using a much
higher concentration than is standard, I found no identifiable controlled
substances. There were instrumental indications of something being present,
but not enough data to make an identification. Since anything less than an
identification is to be reported as "no controlled substances demonstrated,"
that is the result I reported. Both the test provider and my supervisor
reanalyzed the source material, and both confirmed that alprazolam was
present at a low, but detectible, level. My results for the third sample of
this third proficiency test therefore constituted a "false negative."

This second error came as a great shock to everyone, and it was only then,
with the help of the observations of family, friends and colleagues, that I
realized how debilitated I had become over the course of the preceding two
months. After much introspection and consultation with these very
supportive people, I decided to make some drastic changes in my work and
home habits, all designed to reduce my stress level, ensure I obtained
sufficient sleep, and improve my physical conditioning. I gave up some
voluntary activities and commitments that were worthwhile but which
unnecessarily contributed to my stress level and inadequate sleep time; I
stopped working until 7, 8, or 9 PM (and occasionally later) at the lab,
setting a firm goal of leaving work between 5 and 5:30 pm, and absolutely no
later than 6 pm; and I eliminated late-night/early-morning activities at
home (no matter what deadlines I faced), setting for myself a minimum
nightly sleep period - all measures designed to ensure I obtain sufficient
rest (at 50 years of age, I have learned the hard way to accept that I no
longer have the stamina I routinely had in my younger days). I have also
embarked on a physical conditioning program to reverse the weight gain and
loss of physical fitness I have experienced in the few years since retiring
from reserve military service, as the result of a too-sedentary daily
routine.

In the months since, I have come to feel like a new (or restored) man, and I
am now confidently ready to resume all my old duties. Before I can do that,
however, we must meet the remaining challenge: to find adequate ways to
clearly demonstrate my restored capabilities to the communities we serve
(including LEOs, judges, attorneys, and of course the public, who make up
our juries). We are still working that part out, but we're making progress.
The reanalysis of my old drug casework continues, and with almost half
complete, all my analytical conclusions have thus far been confirmed. If
the results in all of the remaining old cases are likewise confirmed by
reanalysis, that should go a long way toward restoring public confidence.

In the mean time, I continue to perform my other professional analytical
specialty (alcohol toxicology, including five annual BAC determination
proficiency tests, each with multiple samples), continue my review of drug
analysis reference and training materials, and perform a variety of
administrative tasks (some I've long had, plus some new ones to lessen the
burden that this unfortunate turn of events has caused my coworkers - which
I greatly regret). I have also decided to go back to school in pursuit of a
master's degree. I never had the time to do this when I was a military
reservist, and I've always strived to continue my professional education
though journals, seminars, short courses, association meetings, and the
like; but it's never too late to continue formal academic education as well,
and I can now spare the time to do so (my first class began this week). I
expect to return to drug analysis casework in the not too distant future,
assuming all progresses as expected.


Some of the lessons I hope others may gain from my unfortunate recent
experiences are these:

1. No one is infallible -- you're never too intelligent, too well educated,
too well trained, or too experienced to make a mistake; it can happen to
anyone at any time, given the wrong stressors and/or the wrong
circumstances.

2. We all have human frailties and human breaking points, of which we must
strive to be constantly aware in order to guard against exceeding them.

3. Our normal capabilities can be adversely affected by external stressors
both professional and personal, so we must take actions and make changes as
needed to neutralize these stressors and avoid exceeding our limits.

4. Some of our human limits will naturally change as we age, and so it will
eventually be necessary for us to surrender some of the things we took for
granted in our youth, and to make prudent adjustments in others.

And finally,

5. There is no substitute or compensation for adequate rest and physical
fitness, especially as we enter middle age and our later years.

I've learned my lessons well -- I hope others may benefit from this
discussion and do the same. Thanks for reading, and a belated happy
holidays to all.

Bob Parsons, B.S., F-ABC
Forensic Chemist

"The forensic scientist's goal is the evenhanded use of all available
information to determine the facts and, subsequently, the truth."
American Academy of Forensic Sciences web site, Choosing a Career page

"If the law has made you a witness, remain a man of science. You have no
victim to avenge, no guilty or innocent person to convict or save -- you
must bear testimony within the limits of science."
Dr. P.C.H. Brouardel, 19th Century French Medico-legalist

> -----Original Message-----
> From: forensic-science@yahoogroups.com [mailto:forensic-
> science@yahoogroups.com] On Behalf Of Brent Turvey
> Sent: Thursday, November 23, 2006 3:11 PM
> To: Forensic Science
> Cc: FORENS-L
> Subject: [forensic-science] Indian River Crime Lab Analyst Fails
Proficiency Testing
>
> Members;
>
> I think we have all come to appreciate the contributions of Bob Parsons to
> this list over the years.
>
> The revelation that he failed a proficiency this summer, and the legal
> fall-out (the retesting is mandatory I would think, but the alleged
> statements from the IR Lab Director are of interest), hit me from perhaps
a
> different direction than most might expect.
>
> First, proficiencies are an important barometer of ability and competency,
> but only one. As such, mistakes are not only allowed, but they are
actually
> expected to some degree. This is how learning occurs. The notion that
> learning ends once someone gets accreditation or a degree or a certain
level
> of experience is an ignorant one - but one that some in the forensic
> sciences have encouraged to buttress certain kinds of experience based
> testimony.
>
> As such, quote below from Bob Parson's boss, Mr. Nippes ("In our
profession,
> there's no room for error,") is not just tragic for Bob Parson's future
> testimony, but misinformed in general about the nature, purpose and
ultimate
> value of proficiency testing. Unfortunately, in a climate where experience
> is so heavily relied upon in court, this extreme position on error is to
be
> expected, if not totally anticipated. Too many forensic scientists like to
> infer or state their infallibility as a matter of course, and as such when
> some forensic scientists fail proficiencies, it is seen as a violation of
> their covenant.
>
> Second, and equally disturbing, is the statement by Mr. Nippes that "this
is
> the first time in his 32 years at the crime lab that anyone has failed a
> proficiency exam." This is a red flag to me. Is this possibly true? I
> suppose it might be if analysts knew they were being tested and knew what
> the expected result was (this is fairly common). But then it wouldn't be a
> very rigorous proficiency - and the result would have little internal
value
> (but tremendous PR value).
>
> I hope we will learn more about this instance, and that the "once wrong -
> always wrong" adage doesn't take Bob Parsons out of the profession. That
> would be a loss.
>
> As long as human beings are performing tests, and as long as human beings
> are interpreting results, there has to be room for error. And there must
> also be an admission during testimony that this is a reality. The method
may
> be flawless (unlikely, but hypothtically), but we humans aren't.
>
> Article below.
>
> Brent
>
> Evidence from 189 drug cases will be reanalyzed
>
http://www1.tcpalm.com/tcp/local_news/article/0,2545,TCP_16736_5132523,00.ht
> ml
> By DAN GARCIA
> dan.garcia@...
> November 10, 2006
>
> INDIAN RIVER COUNTY - Evidence from 189 drug cases in the 19th Judicial
> Circuit will be reanalyzed after a chemist with the Indian River Crime
> Laboratory failed two proficiency exams, according to the lab's director.
>
> The State Attorney's Office asked the Crime Laboratory, located in Fort
> Pierce, to reanalyze the drugs to ensure defendants who were convicted and
> sentenced to confinement were prosecuted on sound evidence.
>
> Daniel C. Nippes, director of the Indian River Crime Laboratory, said he
> alerted the State Attorney's Office on Aug. 24 after Criminalist Robert W.
> Parsons Jr. failed a routine proficiency exam, then failed a follow-up
test.
> The State Attorney's Office recommended retesting all drugs chemically
> analyzed by Parsons in the past year. Parsons was removed from his role in
> analyzing controlled substances and reassigned within the laboratory.
>
> Parsons could not be reached Thursday for comment.
>
> Nippes described Parsons as a highly qualified chemist who made mental
> mistakes in the proficiency exams. He said Parsons has worked with the
crime
> lab for 25 years and previously passed 47 proficiency exams.
>
> "This really hurts me an awful lot because he's a very qualified,
> experienced analyst," Nippes said. "As bad as I feel about the analyst,
the
> most important thing is the credibility of the lab. We have to be fair to
> the citizens, we don't want anyone to be unfairly punished."
>
> The Indian River Crime Laboratory employs three chemists who analyze
> suspected drugs for law enforcement agencies to determine whether they are
> controlled substances. The State Attorney's Office relies on the lab's
> findings to prosecute defendants.
>
> In some cases, chemists testify in court about the chemical makeup of the
> evidence.
>
> Nippes informed Chief Assistant State Attorney Thomas Bakkedahl that in a
> proficiency exam Aug. 23, Parsons "reported a controlled substance was
> present in an item, when in fact no drug was actually present in this
item."
>
> In a follow-up exam, a sample tested by Parsons "contained a low level of
> controlled substance that was determined to be negative by the analyst."
>
> Nippes said the failed exams were "human error."
>
> "People make mistakes," Nippes said. "But it's a shame our protocols don't
> allow for that. There's just no room for error in our profession."
>
> By law, the State Attorney's Office is required to notify defendants and
> their attorneys of any information that may be favorable to defendants.
>
> The State Attorney's Office on Thursday notified 19th Circuit Public
> Defender Diamond Litty and private attorneys for the 189 defendants, whose
> cases date back a year to when Parsons last passed a proficiency exam.
>
> "Obviously we're concerned about this," Litty said. "If there are legal
> challenges to be made, we are certainly going to make them."
>
> Assistant Public Defender Edward J. Abare of Indian River County said he
> would consider asking the State Attorney's Office to recheck drug evidence
> in some cases handled by Parsons that date back more than one year.
>
> "It may create a lot of work, but so be it, because somebody's due process
> may have been compromised," Abare said.
>
> Asked if he believed someone may have been convicted on faulty evidence,
> Abare replied: "It's possible, and possible is enough for me."
>
> Nippes said retesting in older cases likely would prove futile because the
> laboratory has retested about 20 percent of the evidence in the 189 cases,
> and tests have confirmed Parsons' findings. He said it will take several
> months to complete the retesting.
>
> Bakkedahl said prosecutors "understand the importance of accurate tests."
>
> "We want to ensure there is no miscarriage of justice," Bakkedahl said.
"If
> some evidence was found not to be a drug and the test was a false
positive,
> we're not going to dispute that case.
>
> "But that is unlikely to happen because all the retesting so far has
> confirmed Bob Parsons' original findings," Bakkedahl said.
>
> Attorney Rusty Akins of Fort Pierce said he was notified his office
> represented five clients convicted of drug offenses in cases handled by
> Parsons.
>
> "I'm kind of shocked because usually those guys are straight and narrow,"
> Akins said. "Maybe because they reported it themselves, that proves they
are
> straight and narrow. I have to give them credit for that."
>
> Akins said he had not looked at the five cases yet to determine what
action
> he would take.
>
> Number of cases affected: 189
>
> Location of cases: St. Lucie, Martin, Indian River and Okeechobee
counties.
>
>
> Brent E. Turvey, MS - Forensic Science
> Forensic Solutions, LLC
> bturvey@...
> http://www.forensic-science.com