The importance of t0 in forensic chromatography

On April 3, 2012, in Analysis, Chemistry, Equipment, Ethics, In The Courtroom, Laboratory Procedure, by Justin J. McShane

If you are bored and want to have a lit­tle fun, call up your local crime lab­o­ra­tory, ask for some­one who pre­forms their chro­matog­ra­phy (it doesn’t mat­ter if it is liq­uid chro­matog­ra­phy or gas chro­matog­ra­phy). When you get that per­son ask them a sim­ple question:

What’s the t0 (pro­nounced “t sub­zero”) for your method that you are running?

If you could see them (remem­ber you are over the phone), they would prob­a­bly look like this:

There will likely be a pause. Then per­haps some con­fus­ing bab­ble from the per­son that is sci­en­tific gib­ber­ish that means absolutely noth­ing as it was unre­spon­sive. Or per­haps the per­son might actu­ally con­fess that he/she doesn’t even know what t0 is or what you are talk­ing about or why it would be important.

The rea­son that I am gam­bling that this will be the likely response at your local crime lab­o­ra­tory is the way that crime lab­o­ra­to­ries hire and train these ana­lysts. It is a sad fact that there are very few under­grad­u­ate insti­tu­tions that actu­ally mean­ing­fully teach chro­matog­ra­phy prac­tice or the­ory. Even fewer under­grad­u­ate insti­tu­tions actu­ally have a GC or LC. The few that do usu­ally have 1. Usu­ally, it is bro­ken or abused. Even if it is run­ning, the use of it is reserved to grad­u­ate stu­dents or are ded­i­cated to the professor’s totally eso­teric research project, and there­fore is a hands-off expe­ri­ence for the poor under­grad­u­ate. As a byprod­uct of this sad real­ity, a gen­er­a­tion of state sci­en­tists have intel­lec­tu­ally grown up with­out any of the fun­da­men­tals or mean­ing­ful prac­ti­cal expe­ri­ence in chro­matog­ra­phy as we would hope that they would gain in under­grad­u­ate study. What fur­ther com­pounds the issue is the pay scale (which is appalling low) can­not pos­si­bly be designed to attract and main­tain in the crime lab­o­ra­tory ana­lyst posi­tion peo­ple who have any real expe­ri­ence and any real the­o­ret­i­cal train­ing in chro­matog­ra­phy. In fact, when I have looked at the CVs of crime lab­o­ra­tory ana­lysts from my col­leagues across the nation, it is highly unusual to find any one who has actu­ally used a GC or LC before they were hired by the crime lab­o­ra­tory. So when they show up for work, they are lit­er­ally tab­ula rasa. They know noth­ing more about the the­ory and the process than per­haps a 3 year old child. I have seen some truly weird back­grounds. I have seen a French Lit­er­a­ture major who had a minor in biol­ogy (con­sisted almost entirely of mam­malian biol­ogy classes) hired to con­duct LC-MS work which requires a lot of under­stand­ing in acid-base chem­istry to inter­pret the spec­tra cor­rectly. I have seen folks who are physics majors who never took a sin­gle chem­istry class hired to do this work. I have also fre­quently seen AA grad­u­ates with only 2 hard sci­ence classes (intro­duc­tion to biol­ogy and intro­duc­tion to chem­istry) with Foren­sic Science/Police Stud­ies AA degrees. Then there is the inter­net degree places where the per­son may be a chem­istry major, but never stepped foot in a lab­o­ra­tory at all and never seen a GC or LC in per­son let alone use it. In sum, very few have even an intro­duc­tory instru­men­ta­tion class. They cer­tainly have no expo­sure to metrol­ogy or in qual­ity con­trol or qual­ity assur­ance pro­ce­dures. It is a very sad state of affairs.

The crime lab­o­ra­tory, con­cerned with through­put, wants to get this per­son who is draw­ing a pay­check and tak­ing up space in the lab­o­ra­tory to test­ing sam­ples as soon as humanly pos­si­ble. The push to “get test­ing” leads to a totally incom­plete edu­ca­tion that amounts to lit­tle more than teach­ing some­one to be a but­ton pusher doing a mechan­i­cal process that requires no thought or inspection.

But back to our main topic—t0.

What is t0?

t0, T0, T sub zero, T naught, T zero, air peak, dead time, tM, hold time, or void time all refer to the same con­cept. It is the time that it takes for an unre­tained (mean­ing no inter­ac­tion with the sta­tion­ary phase) mol­e­cule to make it from injec­tion to detec­tor. The sig­nif­i­cance of it can­not be under­scored. Every lab­o­ra­tory should have in their val­i­da­tion study for the method the t0 time.

Why it is so impor­tant for us as exter­nal qual­ity assur­ance offi­cers for the crime laboratory?

If there is any peak before t0, then it must be from carry-over and not from the injec­tion that we are sup­posed to be analyzing.

Car­ry­over is the nice sci­en­tific name. The real­ity is that it is unplanned con­t­a­m­i­na­tion that calls into ques­tion the valid­ity of the reported results.

Here is data from a val­i­da­tion study from a foren­sic lab­o­ra­tory where they proved t0. Remem­ber that t0 will not be at the same time for every GC-FID in the world. It totally depends on the chro­mato­graphic col­umn and the chro­mato­graphic con­di­tions (pri­mar­ily tem­per­a­ture and pressure/flow).

There is actu­ally a the­o­ret­i­cal man­ner to mea­sure and accu­rately pre­dict t0 based upon your con­di­tions through phys­i­cal chem­istry. You then have to exper­i­men­tally con­firm the math on your instru­ment and test for robust­ness per the Huber/USP pro­to­col for method val­i­da­tion (more on method val­i­da­tion can be found here Method Val­i­da­tion for Lawyers: A Six Part Series).

Here is a sim­ple exem­plar chro­matogram using the same method (same col­umn and con­di­tions) where there is a peak detected before t0. The only way that hap­pens is if there is car­ry­over from the last sam­ples that pre­ceded it. There is an old say­ing in chro­matog­ra­phy which says “just because you are done with chro­matog­ra­phy doesn’t mean it is done with you” which means that just because you think that every­thing has eluted from the col­umn doesn’t make it so. This is a very basic rea­son why there must be blanks between unknowns.

Here is a par­tic­u­larly dan­ger­ous chro­matogram that demon­strates car­ry­over in the most impor­tant chro­matogram in the QC for the method which is the proof of separation/resolution:

Here is the other col­umn from the same batch. Note t0 is dif­fer­ent as the col­umn is dif­fer­ent, but same chro­mato­graphic con­di­tions. There is demon­strated car­ry­over and likely/possible car­ry­over demonstrated.

So ask your crime lab­o­ra­tory: “what’s your t0?”

Tagged with: • • • • • • • • •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Guest Blog Post From Dr. Frederic Whitehurst PhD JD: Issues in Explosive Residue Analysis

On March 29, 2012, in Analysis, Explosive Residue, Trace Evidence, by Justin J. McShane

Issues in Explo­sives Residue Analy­sis A Primer for the Bar Fred­eric White­hurst, Ph.D.[1]

[Edi­tors Note: This is a multi-part series deigned to edu­cate the defense bar on impor­tant issues con­cern­ing explo­sive and explo­sive residue investigations]

Part 1: Introduction

Part 2: Back to the Basics: Was it the result of an explo­sive device in the first place? How do we know that?

Part 3: Daubert pro­vides guid­ance and a means to expose lam­i­na­tions and eval­u­ate explo­sive inves­ti­ga­tions, meth­ods, and interpretation

Part 4: The Explo­sion Crime Scene: Sam­pling and Homo­gene­ity Issues

Part 5: Dis­po­si­tion Homo­gene­ity in explo­sive scene investigation

Part 6: Con­t­a­m­i­na­tion and Cross Con­t­a­m­i­na­tion in explo­sive scene investigation

Part 7: Con­t­a­m­i­na­tion by “Render-Safe” acts of explosives

Part 8: Trans­porta­tion and stor­age of evi­dence in explo­sive scene investigation

Part 9: Chem­i­cal analy­sis in explo­sive scene investigation

Part 10: Iden­ti­fy­ing Tech­niques in explo­sive scene investigation

Part 11: Inter­pre­ta­tion of data in explo­sive scene investigation

Part 12: Expe­ri­ence: What makes for a proper expert in explo­sive scene investigation?

Part 13: Conclusion]

______________________

On Feb­ru­ary 26, 1992, a mas­sive explo­sion ripped through the garage under­neath the World Trade Cen­ter in New York City. Ini­ti­ated by ter­ror­ist explo­sives, this bomb­ing left hun­dreds of mil­lions of dol­lars in dam­age, six dead and over 1000 injured cit­i­zens and this coun­try pro­foundly shaken in its belief in its own secu­rity. On April 19, 1995 another mas­sive explo­sion, the dead­liest ter­ror­ist attack ever on U.S. soil[2] ripped apart the Alfred P. Mur­rah Fed­eral Build­ing in Okla­homa City, Okla­homa, killing a large num­ber of small chil­dren among the vic­tims. Recent years have seen numer­ous inci­dents such as this. The down­ing of Pan Am Flight 103 over Locker­bie, Scot­land, is well known as is the bomb­ing of Avianca Flight 203 in Colom­bia and the bomb­ing of an Air India flight over the Atlantic Ocean. Crim­i­nal bomb­ings also attack pub­lic fig­ures such as occurred in the assas­si­na­tion of the Ital­ian judges Gio­vanni Fal­cone and Paulo Borcelino in Sicily and Fed­eral Judge Robert Vance in Atlanta, Georgia.

Accord­ing to the 1993 FBI Bomb Data Cen­ter Sum­mary [3] two thou­sand nine hun­dred and eighty bomb­ing inci­dents were reported to the Bomb Data Cen­ter in 1993. Casu­al­ties from those bomb­ings num­bered 1,372 includ­ing the 1042 injured in the World Trade Cen­ter bomb­ing. Forty-nine of those injuries resulted in death. Dam­age to prop­er­ties in 1993 was cal­cu­lated at approx­i­mately $518,000,000 with $510,000,000 dam­age to the World Trade Cen­ter build­ing alone. [4]

The crim­i­nal inves­ti­ga­tions invari­ably lead to the iden­ti­fi­ca­tion of alleged per­pe­tra­tors and their being brought to trial. In the author’s expe­ri­ence, mem­bers of the bar have been ini­tially unpre­pared in their under­stand­ing of the issues that are asso­ci­ated with the part of those inves­ti­ga­tions, which involve the foren­sic analy­sis of the crime scenes. Berger[5] notes that the bar needed to begin to ask ques­tions before DNA analy­ses were questioned.

This paper will attempt to edu­cate the bar to also ask ques­tions in one of the areas of those inves­ti­ga­tions, the area of explo­sives and explo­sives residue analy­sis, in order that coun­sel will be bet­ter pre­pared to cor­rectly address those issues when they are met. The paper will present a “hard look” [6] at the issues pre­sented by this com­plex prob­lem and sug­gest meth­ods of han­dling the problems.

______________________

[1] Exec­u­tive Direc­tor, Foren­sic Jus­tice Project, Wash­ing­ton, D.C., B.S. Chem­istry, 1974, East Car­olina Uni­ver­sity, Ph.D. in Chem­istry, 1980, Duke Uni­ver­sity, J.D., 1996, George­town Uni­ver­sity School of Law. (202)342‑6980.

[2] Sue Anne Press­ley, Bomb­ing Kills Dozens in Fed­eral Build­ing, Chil­dren in Day Care Cen­ter Among Vic­tims of Appar­ent Ter­ror­ist Attack in Okla­homa City, Wash­ing­ton Post, April 20, 1995 at A1.

[3] Fed­eral Bureau of Inves­ti­ga­tion, US Deprat­ment of Jus­tice, 1993 Bomb Summary.

[4] Id, at 13,14.

[5] Mar­garet A. Berger, Pro­ce­dural Par­a­digms for Apply­ing the Daubert Test, 78 MINN. L. REV. 1345 1355 (1994) notes that pos­si­ble lim­i­ta­tions on find­ing matches in DNA sam­ples, as when DNA was graded or con­t­a­m­i­nated, were not explored until lawyers, aca­d­e­mics, and law reviews began ask­ing spe­cific questions.

[6] Turpin v. Mer­rell Dow Phar­ma­ceu­ti­cals, Inc. 959 F.2d 1349,1352 (6th Cir. 1992) sug­gests judges take a “hard look” at “the rea­son­able­ness of sci­en­tific the­o­ries and infer­ences before they decide whether there is enough to the case for it to go to the jury.”

 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Why per se Driving Under the Influence of Drug Based Prosecution is Not Scientific and How Probation Violations for Marijuana Use May Be Wrong

On March 13, 2012, in Analysis, Chemistry, Equipment, Ethics, Identification, In The Courtroom, Laboratory Procedure, Pharmacology, by Justin J. McShane

I can­not imag­ine that a sin­gle ratio­nal and sane per­son cares for drug impaired dri­vers to be on the road among us dri­ving what amounts to a guided mis­sile weigh­ing about 3,000 to 5,500 pounds or more. As has been shown time and again, a drug intox­i­cated dri­ver can be a weapon of mass destruc­tion. The right­ful pros­e­cu­tion and appro­pri­ate pun­ish­ment of drug impaired dri­vers is a good thing.

What I sug­gest is not “good,” “just,” or even “fair” is the pros­e­cu­tion and dra­con­ian pun­ish­ment of non-impaired dri­vers as if they were dan­ger­ous drugged drivers.

What ratio­nal per­son can disagree?

Is this what is hap­pen­ing in the United States today? Yes

This ques­tion is one of the most beau­ti­fully artic­u­late and hor­ri­bly com­pli­cated issues pos­si­ble when it comes to mar­i­juana in particular.

We have blogged here before about the gen­eral notion of the per­ils of inter­pret­ing an ana­lyt­i­cal chem­istry result stand alone: “Phar­ma­col­ogy For Lawyers Part 6: Elu­ci­dat­ing Phar­ma­co­dy­namic Effect from an Ana­lyt­i­cal Chem­istry Result”

What makes mar­i­juana also par­tic­u­larly inter­est­ing is the well-known and well-studied (pri­mar­ily by Mar­i­lyn A. Huestis, Ph.D., Senior Inves­ti­ga­tor Chief, Chem­istry and Drug Metab­o­lism Sec­tion in the National Insti­tutes on Drug Abuse) counter clock­wise hys­tere­sis effect in terms of  the user’s own sub­jec­tive reported phar­ma­co­dy­namic prop­er­ties and drug con­cen­tra­tion. This is shown on the graph below.

counterclockwise hysteresis of marijuana
The coun­ter­clock­wise hys­tere­sis of mar­i­juana sub­jec­tive report­ing over the course of time ver­sus blood con­cen­tra­tion of tetrahy­dro­cannabi­nol (THC)

As one can clearly see from the graph above, and what is impor­tant about these hys­tere­sis curves is that if you look at a par­tic­u­lar con­cen­tra­tion in plasma, such as 100 ng/mL of the par­ent drug Tetrahy­dro­cannabi­nol (THC), you have very lit­tle effect at one point in time, but at the same drug con­cen­tra­tion (100 ng/mL) at a later time one per­ceives very strong effects.

For this rea­son alone, it should be very very dif­fi­cult for the pros­e­cu­tion to con­vict a mar­i­juana user for an impair­ment based DUID. In fact this whole phe­nom­e­non is why per se drug offenses in terms of DUID are truly unscientific.

But there is more.

The state leg­is­la­tures around the United States have autho­rized the full pros­e­cu­tion and dra­con­ian pun­ish­ment of non-impaired dri­vers as if they were dan­ger­ous drugged dri­vers. The pros­e­cu­tion of mar­i­juana users at per se lev­els for mar­i­juana at or around the time of dri­ving can and is based upon pres­ence (some­times in trace amounts even) of phar­ma­co­dy­nam­i­cally inac­tive metabo­lite 11-nor-9-Carboxy-THC, also known as 11-nor-9-carboxy-delta-9-tetrahydrocannabinol, 11-COOH-THC, THC-COOH, and THC-11-oic acid. Pres­ence of only the phar­ma­co­log­i­cally inac­tive metabo­lite is evi­dence of NON-IMPAIRMENT at the time of the col­lec­tion of the spec­i­men. If any­one only has this inac­tive metabo­lite alone in their spec­i­men no mat­ter what the level, they are not impaired. It is just that sim­ple. This is even more pro­nounced if the spec­i­men is urine which is sim­ply an unpinned his­tory of use at some past time and not ever reflec­tive of impair­ment at the time of collection.

But there is more.

In addi­tion to the unsci­en­tific lia­bil­ity that our leg­is­la­tures have decided to attach to evi­dence of non-impaired dri­ving that we just dis­cussed, the evi­dence of past use of mar­i­juana and the pres­ence of the inac­tive metabo­lite THC-COOH could be far, far removed from last use.

From the abstract of “Uri­nary Elim­i­na­tion of 11-Nor-9-carboxy-9-tetrahydrocannnabinol in Cannabis Users Dur­ing Con­tin­u­ously Mon­i­tored Absti­nence” by Robert S. Good­win, William D. Dar­win, C. Nora Chi­ang, Ming Shih, Shou-Hua Li, and Mar­i­lyn A. Huestis:

The time course of 11-nor-9-carboxy-Δ9-tetrahydrocannnabinol (THCCOOH) elim­i­na­tion in urine was char­ac­ter­ized in 60 cannabis users dur­ing 24 h mon­i­tored absti­nence on a closed research unit for up to 30 days. 6158 indi­vid­ual urine spec­i­mens were screened by immunoas­say with val­ues ≥50 ng/mL clas­si­fied as pos­i­tive. Urine spec­i­mens were con­firmed for THCCOOH by gas chromatography/mass spec­trom­e­try fol­low­ing base hydrol­y­sis and liquid-liquid or solid phase extrac­tion. In 60%, the max­i­mum cre­a­ti­nine nor­mal­ized con­cen­tra­tion occurred in the first urine spec­i­men; in 40%, peaks occurred as long as 2.9 days after admis­sion. Data were divided into three groups, 0 – 50, 51 – 150, and >150 ng/mg, based on the cre­a­ti­nine cor­rected ini­tial THCCOOH con­cen­tra­tion. There were sta­tis­ti­cally sig­nif­i­cant cor­re­la­tions between groups and num­ber of days until first neg­a­tive and last pos­i­tive urine spec­i­mens; mean num­ber of days were 0.6 and 4.3, 3.2 and 9.7, and 4.7 and 15.4 days respec­tively, for the three groups. These data pro­vide guide­lines for inter­pret­ing urine cannabi­noid test results and sug­gest appro­pri­ate detec­tion win­dows for dif­fer­en­ti­at­ing new cannabis use from resid­ual drug excretion.

Mean detection rate each day after first negative <50 ng.mL cannabinoid urine specimen
Mean detec­tion rate each day after the first neg­a­tive (<50 ng/mL) cannabi­noid urine spec­i­men. Cannabis users were sep­a­rated into three groups (0–50, 51–150 and >150 ng/mg) accord­ing to creatinine-normalized THCCOOH con­cen­tra­tions in the first spec­i­men col­lected at admis­sion. Mean detec­tion rate (num­ber of pos­i­tive spec­i­mens in a day/total num­ber of spec­i­mens col­lected that day mul­ti­plied by 100) for the three groups was deter­mined each day after the first neg­a­tive specimen.

The under­ly­ing data that sup­ports the above fig­ure is pre­sented here. The data clearly shows that as long as 28 days days can elapse between the first recorded “neg­a­tive” (really under the limit of detec­tion) test and last pos­i­tive spec­i­men even after total absti­nence. In short due to lipid release a pos­i­tive can occur in chronic users that is arte­fac­tual and not pro­ba­tive (and truly prej­u­di­cial) to the crime sought to be pun­ished which is drug-impaired driving.

And later, the authors of this study conclude:

This study mon­i­tored cannabis users on a closed research unit under con­tin­u­ous med­ical sur­veil­lance dur­ing cannabis absti­nence for up to 30 days. The greater the cre­a­ti­nine cor­rected ini­tial THCCOOH con­cen­tra­tion, the greater the inter­val until the first neg­a­tive and last pos­i­tive spec­i­mens, the greater the win­dow of drug detec­tion and the higher the detec­tion rate of pos­i­tive spec­i­mens. Cannabis users who present with an ini­tial nor­mal­ized THCCOOH con­cen­tra­tion >150 ng/mg can be expected to have detec­tion rates between 60 and 100% for 28 days after the first neg­a­tive urine test. These data increase our under­stand­ing of THCCOOH uri­nary elim­i­na­tion and pro­vide guide­lines for the inter­pre­ta­tion of urine cannabi­noid test results. They also sug­gest appro­pri­ate detec­tion win­dows for dif­fer­en­ti­at­ing new cannabis use from resid­ual drug excre­tion based on cre­a­ti­nine nor­mal­ized THCCOOH urine data.

This phe­nom­e­non proves to be quite prob­lem­atic in pro­ba­tion or parole vio­la­tion hear­ings for alleged mar­i­juana use.  When one is sen­tenced to a term of pro­ba­tion and/or parole typ­i­cally there is a con­di­tion that there is to be no use of illicit drugs (mar­i­juana included). Typ­i­cally the spec­i­men col­lected is urine. Typ­i­cally the pro­ba­tion or parole offi­cer uses the 9 Panel ONESCREEN Multi-Drug Test Cup or some­thing sim­i­lar. Atyp­i­cally the results are sent off for con­fir­ma­tion by GC-MS. The screen­ing tests are pre­sented as truth and not as they should be which is as a screen­ing device with easy, cheap and rapid results with high rates of false pos­i­tive based upon cross-reactivity and lack of true speci­ficity. This same data above is typ­i­cally unknown to the courts or to the probation/parole offi­cers who when con­fronted with a neg­a­tive urine screen then days later are pre­sented with a pos­i­tive urine screen cry foul and accuse the probationer/parolee of use. This may be true in some cases for cer­tain, but the com­bi­na­tion of the lipid release phe­nom­e­non described above, the poor test­ing method, the under­trained and une­d­u­cated pro­ba­tion and parole offi­cer play­ing chemist, and the lack of knowl­edge of all of these lim­i­ta­tions may also result in wrongly accu­sa­tions and false find­ings of vio­la­tions. The issue is that we don’t know.

But there is more.

Plus, we can­not take for granted that the ana­lyt­i­cal chem­istry result is true and cor­rect. The detec­tion of THC and its metabo­lites by screen­ing tests from a bio­log­i­cal sam­ple can pro­vide for false pos­i­tives for any num­ber of rea­sons. Plus, the con­fir­ma­tion (if com­pleted at all) by GC-MS is a very tricky process involv­ing extrac­tion and deriva­ti­za­tion typ­i­cally in terms of the prepar­a­tive chro­matog­ra­phy steps to even make the pos­si­bil­ity of GC-MS detec­tion pos­si­ble. As recently dis­cov­ered and pub­lished in the Jour­nal of Ana­lyt­i­cal Tox­i­col­ogy (January/February 2012, Voume 36, No. 1, pages 61–65) enti­tled “Pro­duc­tion of Iden­ti­cal Reten­tion Times and Mass Spec­tra for Δ9–Tetrahy­dro­cannabi­nol and Cannabid­iol Fol­low­ing Deriva­ti­za­tion with Tri­flu­o­racetic Anhy­dride with 1,1,1,3,3,3-Hexafluoroisopropanol” by Rebecca Andrews and Sue Pater­son informs us of the pos­si­bil­ity of error in exam­in­ing bio­log­i­cal sam­ples (plasma or whole blood) in this cru­cial prepar­a­tive chro­matog­ra­phy step. The THC assays that use the TFAA (tri­flu­o­roacetic anhy­dride) deriva­ti­za­tion meth­ods have been found to result in con­vert­ing cannabid­iol (CBD) to the same deriv­a­tives as THC. The CBD deriv­a­tives have the same GC reten­tion times and mass spec­tra as the THC deriv­a­tives. The authors con­clude that when tar­get­ing and attempt­ing to quan­tify, this deriva­ti­za­tion method must not be used as it can pro­duce a false pos­i­tive in terms of the qual­i­ta­tive mea­sure­ment, and even if objec­tively present, will over­state the true blood concentration.

The inter­pre­ta­tion (elu­ci­da­tion) of the SIM and mass spec­tra often reveals match fac­tor scores and prob­a­bil­ity scores not at 100%. The selec­tion of diag­nos­tic ions and ratios that are thought to reflec­tive of THC is not absolutely set­tled in the ana­lyt­i­cal chemistry/forensic chem­istry world. The quan­ti­ta­tion of THC and its metabo­lites can be per­formed in a metro­log­i­cally unac­cept­able manner.

Yet despite know­ing this, we con­sciously choose to pros­e­cute and con­vict. Sad but true. The emperor has no clothes. False con­vic­tions based not on sci­ence and per­haps even not even ratio­nally related to the harm sought to be avoided abound.

Tagged with: • • • • • •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Score one for Science: The DRE program is found to be unreliable and not admissible

On March 8, 2012, in Analysis, Ethics, In The Courtroom, Pharmacology, by Justin J. McShane

As I have blogged before DUID cases should be very dif­fi­cult to pros­e­cute due to the phar­ma­co­log­i­cal dif­fer­ences among us. In this six part post we started to explore all of this:

Part 1. Intro­duc­tion
Part 2. Phar­ma­co­ki­net­ics
Part 3. Phar­ma­co­dy­nam­ics
Part 4. Bioavail­abilty
Part 5. “Free ver­sus Bound Drug“
Part 6. Elu­ci­dat­ing Phar­ma­co­dy­namic Effect from an Ana­lyt­i­cal Chem­istry Result

In a very well-litigated case where the pros­e­cu­tion brought in its very best national experts and the defense had fund­ing and knowl­edge into the DRE pro­gram, the trial court quite clearly found the Drug Recog­ni­tion Expert (DRE) pro­gram to be unsci­en­tific and unre­li­able. As a result the eval­u­a­tion and the con­clu­sion was inadmissible.

The case cita­tion is: State of Mary­land v. Charles David Bright­ful, et al, No. K-10–04-259, Cir­cuit Court for Car­roll County, MD March 5, 2012. The full opin­ion of the court is avail­able here.

What makes this vic­tory even more com­plete for sci­ence is that the court per Mary­land state case law used the much looser Frye stan­dard that gov­erns admis­si­bil­ity of the pros­e­cu­tion evi­dence. So under the Frye stan­dard, the pros­e­cu­tion should have had a much eas­ier time to have the evi­dence gain admissibility.

Here is the hold­ing of the court:

Apply­ing Md. R. 5–702 to the pro­posed DRE tes­ti­mony, the Court finds that a drug recog­ni­tion  expert is not suf­fi­ciently qual­i­fied to ren­der an  opin­ion, that the tes­ti­mony is not rel­e­vant, and the  pro­ba­tive value of the evi­dence is sub­stan­tially  out­weighed by its prej­u­di­cial effect.

Find­ings of Fact:

The DRE Pro­to­col fails to pro­duce an accu­rate and reli­able deter­mi­na­tion of whether a sus­pect is impaired by drugs and. by what spe­cific drug he is  impaired. The DRE train­ing police offi­cers receive does not enable DREs to accu­rately observe the signs and symp­toms of drug impair­ment, there­fore, police offi­cers are not able to reach accu­rate and reli­able  con­clu­sions regard­ing what drug may be caus­ing  impairment.

Six experts tes­ti­fied for the gov­ern­ment. Three tes­ti­fied for the defense: a clin­i­cal phar­ma­col­o­gist, an oph­thal­mol­o­gist, and a psy­chi­a­trist. The defense experts, in sum­mary, were all highly crit­i­cal of dif­fer­ent aspects of “drug recog­ni­tion experts” and the pro­to­col used by DRE’s. All three defense experts tes­ti­fied that the DRE pro­to­col and matrix are not gen­er­ally accepted in the fields of med­i­cine includ­ing specif­i­cally phar­ma­col­ogy, neu­rol­ogy, oph­thal­mol­ogy and psychiatry.

From the opin­ion itself:

Although the DRE pro­gram is uti­lized in 45 states, the pres­ence of the DRE pro­gram does not  equate to wide­spread judi­cial accep­tance by appel­late  courts nor accep­tance in the med­ical community.

Mr. Tower (he is a liai­son from NHTSA and gov­ern­ment expert) fur­ther tes­ti­fied that in addi­tion to  the wide dis­cre­tion in what weight to give the  indi­ca­tors on the matrix, the DRE is not even required  to com­plete the 12-step pro­to­col to reach an opin­ion  as those steps are merely “pre­ferred.” (Tr. 2/14/11  at 95–96). Mr. Tower tes­ti­fied that even if no drugs  at are found in the subject’s blood, the DRE is  “not going to change [their] opin­ion after you get the  blood.” (Id. at 103–04) Mr. Tower stated that the  rea­son there would be no change in the officer’s  opin­ion is that “you are lim­ited on what the lab can  test for.” (Id. at 104) (Empha­sis supplied.)

Part of the defense expert psychiatrist’s testimony:

I have got to tell you, your Honor, DRE is some­thing that’s not fore­most in the mind of those of us who take care of sub­stance  abusers, clin­i­cally or foren­si­cally. Peo­ple  are aware of it. But it’s– no one I  know of, no physi­cian I know of would even con­sider using this matrix or the — even pieces of it in deter­min­ing either whether some­one was impaired on drugs or even more  ridicu­lously to tell which spe­cific drug  cat­e­gory. It’s ridiculous-I can’t empha­size  that enough.

The DRE matrix

The DRE matrix

For both sides it was the most com­pre­hen­sive hear­ing to date look­ing directly at the sci­en­tific reli­a­bil­ity and admis­si­bil­ity of the DRE process.

To validly inter­pret the total­ity of the cir­cum­stances, these cases take experts–real cre­den­tialed phar­ma­col­o­gists with clin­i­cal expe­ri­ence who have dosed folks and real ana­lyt­i­cal chemists. To defend these DUID cases, defense attor­neys need to hire experts. Sadly and incor­rectly most of my col­leagues think they can go with­out. They infre­quently win when they do. If the accused has a knowl­edge­able attor­ney who is schooled and truly under­stands the sci­ence com­bined with real cre­den­tialed experts these cases are quite easy to defend because the sci­ence is wholly on the side of the motorist accused in terms that the gov­ern­ment very rarely has all of the infor­ma­tion nec­es­sary to form a valid conclusion.

The bot­tom line is that claim­ing some­thing is valid and prov­ing it is valid are two sep­a­rate things.

Sci­ence was well-served by this brave and hon­est jurist. Jus­tice was well served by the lit­i­gat­ing attor­neys: Brian DeLeonardo and Alex Cruickshank.

So far, in past hear­ings where it was very lop­sided with the defense not pre­sent­ing experts, the gov­ern­ment could only claim that the DRE process was valid.

Other than per­sonal belief and per­haps anec­do­tal experience/observation, what is there? A bro­ken clock is infi­nitely pre­cise and totally accu­rate twice a day. That coin­ci­dence of obser­va­tion at that exact moment when it is objec­tively cor­rect doesn’t make that clock as a mea­sur­ing device valid, does it?

The source of the empir­i­cal data that was used in the case to inval­i­date the DRE tool came exclu­sively from the stud­ies NHTSA funded or indi­rectly funded. So it is their own research that objec­tively inval­i­dates the tool. It was not defense friendly research that was cited in the hearings.

In short, the cru­cial ques­tion was: “Where is the sci­en­tific val­i­da­tion of this method?” or “Is the DRE pro­to­col proven to be valid for its intended pur­pose which is to opine impair­ment due to source?”

Not only doesn’t the val­i­da­tion stud­ies exist, but the research funded from osten­si­bly pro-prosecution quar­ters fal­si­fies the hypoth­e­sis that it could be a use­ful tool.

I am quite sure that there are a lot of police offi­cers who use the DRE pro­to­col out there who hon­estly believe that it works and is valid. That’s quite fine. It is still the United States of Amer­ica, they are per­son­ally free to believe what­ever they like. I’m all for per­sonal opin­ions. A per­sonal opin­ion doesn’t have to be grounded in sta­tis­tics or val­i­da­tion stud­ies. It is your own.

How­ever, what isn’t just or right is to espouse in court an opin­ion (expert or not) that a tool is valid when it hasn’t been proven as such, and much worse when the evi­dence proves that the method has no valid­ity.  Anyone’s (not the true believ­ers) BELIEF in the use­ful­ness and valid­ity of a tool and/or someone’s anec­do­tal expe­ri­ence is clearly not reli­able sci­en­tific proof of anything.

In essence, the basic ques­tion can be stated thusly: “Is it legit­i­mate to sup­pose that an eval­u­a­tion of a total stranger at one time and one time only with­out know­ing any rel­e­vant phar­ma­co­log­i­cal or phys­i­cal his­tory of the eval­uee by a very min­i­mally trained, non cre­den­tialed per­son, who has never done clin­i­cal rounds, dosed any­one and has no clin­i­cal expe­ri­ence, could be valid?”

A con­vic­tion based upon DRE even if per­fectly exe­cuted is sim­ply a sin against true sci­ence and a sin against justice.

Tagged with: •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Guest Blog Post From Dr. Frederic Whitehurst PhD JD: Of moos and clucks-the myth of FTIR identification

On March 8, 2012, in Analysis, Chemistry, Contextual Bias, Equipment, Ethics, Identification, In The Courtroom, Laboratory Procedure, by Justin J. McShane

Of moos and clucks-the myth of FTIR identification

By:  Fred­eric White­hurst, J.D., Ph.D. [1]

OK, so we will now dis­cuss moos and clucks. Which is the basis for the prob­lem with Fourier Trans­form Infrared Spec­troscopy (FTIR) in foren­sic sci­ence. [2] If you close your eyes in a stock yard you can’t tell what moo came from what cow. If you close your eyes in a chicken house you can’t tell what cluck came from what chicken.
FTIR

FTIR

Infrared spec­tra can be seen to be “fin­ger­prints” of organic com­pounds. How­ever, we sel­dom if ever see organic com­pounds by them­selves in foren­sic sam­ples. (We have what can be described as stock yards or chicken houses full of organic cows/chickens.) What we see are com­plex sam­ples the con­tent of which is unknown. The infrared spec­tra are based upon absorp­tion of infrared radi­a­tion by par­tic­u­lar moi­eties (func­tional groups) (parts of the) mol­e­cule. But dif­fer­ent mol­e­cules can have the same func­tional groups. So when we have a mix­ture of mate­ri­als we can’t tell what mol­e­cule con­tain­ing what func­tional group gave rise to the peak on the spectrum.
The foren­sic ana­lyst in order to cut short his work load sim­ply decides that because he sees, for instance, car­bonyl groups together in a par­tic­u­lar area of the spec­trum, must have cocaine when in fact there are tens of thou­sands of mol­e­cules which have car­bonyl groups. We don’t know what is in the alleged cocaine plus dilu­ent mix­ture so we don’t know what mol­e­cules are giv­ing rise to those car­bonyl peaks. The foren­sic ana­lyst ignores the prob­lem and with a magic wave of his or her hand sim­ply assigns the peaks based upon a request from the field to “check for the pres­ence of con­trolled sub­stances.” The pro­to­col he is using has not been val­i­dated nor deter­mined to be reli­able. He or she sel­dom if ever sep­a­rates the com­po­nents with chro­matog­ra­phy, leaves the data in the file, and upon ques­tion­ing says he or she knows that he or she is look­ing at cocaine peaks based upon his or her vast knowl­edge and expe­ri­ence. Then he or she may run a GC/MS with its sep­a­ra­tion pow­ers and detect the pres­ence of cocaine.
How­ever, the FTIR and GC/MS must stand alone. The FTIR gives an answer and the GC/MS tests the hypoth­e­sis. But with the FTIR data in the file the FTIR data is dif­fer­ent from cocaine and proves the null-hypothesis pre­sented by the GC/MS and the gov­ern­ment data proves that cocaine is not present.
The proper way to use FTIR in a foren­sic set­ting is GC/FTIR or sol­vent or mechan­i­cal separation.
Another trick that the gov­ern­ment lab is using every­where is to present the spec­tra with­out the peaks labeled accord­ing to their absorp­tion energy. The x-axis on the spec­tra goes from 400 to 4000 cm-1.  The res­o­lu­tion on that axis is of such poor qual­ity that one can not dis­cern with the naked eye what the actual absorp­tion ener­gies are so two spec­tra may look alike but can not be compared.
an analysis of an unknown by FTIR
an analy­sis of an unknown by FTIR (note that the peaks are not labelled)
standard by FTIR of cocaine (note it the peaks are not labelled)

stan­dard by FTIR of cocaine (note that the peaks are not labelled)

The way to bring that out is to ask the ana­lyst on the stand to label each peak in the spec­trum. If he or she can’t do that, which he or she can’t ever, then you ask how you com­pare one spec­trum on paper to another.  When the spec­tro­scopist uses a res­o­lu­tion of 4 cm-1 on the energy axis and a pro­to­col that demands that peaks be within +/-4cm-1 to match then with­out the labeled peaks he or she can’t say he or she is within the vari­ance required by the pro­to­col in order to make a match.
Another issue with FTIR is that even very small dif­fer­ences in spec­tra can mean very large dif­fer­ences in struc­ture of mol­e­cules. The gov­ern­ment ana­lyst nor­mally ignores small dif­fer­ences. One can see that clearly in the size of the peaks. Two peaks together have not only loca­tion but ampli­tude (size). If you com­pare the FTIR spec­trum to another the ratios of the two peaks or any num­ber of peaks should remain the same or else the ana­lyst must be able to explain why the ratios are the same. Peaks on the edge of larger peaks are often ignored also. These issues can be detected by care­fully putting two spec­tra in front of you and sim­ply com­par­ing them for the slight­est dif­fer­ences. You will find those dif­fer­ences in almost all gov­ern­ment IR spec­tra comparisons.
To teach about how this is per­formed in the real world, one needs to present what the gov­ern­ment presents: the spec­tra with­out the peaks labeled, and then what the gov­ern­ment knows: the same spec­tra with the peaks labeled. The infor­ma­tion con­cern­ing the ener­gies absorbed is infor­ma­tion the gov­ern­ment has and yet does not present, in vio­la­tion of dis­cov­ery requests and Brady–like oblig­a­tions. If ener­gies are not within accept­able range then the infor­ma­tion is Brady infor­ma­tion when the gov­ern­ment opines that two spec­tra are the same. Attor­neys should be able to see how this infor­ma­tion is hidden.
Cocaine base with no peaks labeled

Cocaine base with no peaks labeled

The way to label peaks is to sim­ply push the but­ton that says “Label Peaks” either on the face of the screen or in a set of tools in the FTIR han­dling algo­rithm. So  what about peak res­o­lu­tion? Do you (the resent­ing ana­lyst in court) under­stand the dif­fer­ence between dig­i­tal and ana­log data col­lec­tion? It can make a huge dif­fer­ence  in spec­tral inter­pre­ta­tion. It can mean match or no match.
Cocaine base with peaks labeled

Cocaine base with peaks labeled

A way to han­dle the mix­ture prob­lem is to do a peak sub­trac­tion using the com­puter stored spec­tra. Sup­pose you think you have cocaine and caf­feine and lido­caine. You can’t say you have cocaine but you think you have cocaine. So you ask that the spec­tra of caf­feine and lido­caine be sub­tracted from the total spec­trum to see if you get cocaine. But no one knows how much caf­feine and lido­caine are present or even if they are present so when you sub­tract those spec­tra you will never get a good spec­trum for cocaine. Just won’t.
Unknown possibly with cocaine

Unknown pos­si­bly with cocaine

The dif­fer­ence between an aca­d­e­mic course and the prac­ti­cal appli­ca­tion is that you learn the­ory in the aca­d­e­mic courses but acad­e­mia has not yet really entered into the extremely com­plex world of foren­sic sam­ples. We are at the front lines of apply­ing sci­ence to real world sit­u­a­tions, find­ing the gov­ern­ment under­funded, under­e­d­u­cated, under­manned, while try­ing to com­plete impos­si­ble tasks.  So they ignore the small stuff, which in sci­ence is always very important.

_______________

[1] Exec­u­tive Direc­tor, Foren­sic Jus­tice Project, Wash­ing­ton, D.C., B.S. Chem­istry, 1974, East Car­olina Uni­ver­sity, Ph.D. in Chem­istry, 1980, Duke Uni­ver­sity, J.D., 1996, George­town Uni­ver­sity School of Law. (202)342‑6980.

[2] FTIR is an instru­ment that is used pri­mar­ily in solid drug dose (per-consumption) form iden­ti­fi­ca­tion of an unknown. The result of the analy­sis is a spec­trum. The spec­trum of the unknown is com­pared against a known stan­dard. In inter­pre­ta­tion of the spec­tra is made by the ana­lyst with some sort of con­clu­sion offered typ­i­cally as an absolute iden­ti­fi­ca­tion as if with per­fect specificity.

Tagged with: • •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Zero is not a valid data point for calibration purposes

On March 6, 2012, in Analysis, Chemistry, Equipment, Ethics, Identification, In The Courtroom, Laboratory Procedure, Metrology, by Justin J. McShane
Zero-A simple concept to understand, an impossible point to measure

Zero-A sim­ple con­cept to under­stand, an impos­si­ble point to measure

When the quan­ti­ta­tive mea­sure­ment really mat­ters to deter­mine guilt, we have to under­stand how that mea­sure­ment is achieved. In the case of the analy­sis of bio­log­i­cal sam­ples that are tested, quan­ti­ta­tion of an unknown is derived from a cal­i­bra­tion curve. The proper con­struc­tion of the cal­i­bra­tion curve is a cru­cial com­po­nent of qual­ity con­trol. It serves as the very foun­da­tion of the valid­ity of the quan­ti­ta­tive mea­sure­ment. If the cal­i­bra­tion curve is not prop­erly formed then there will be bias. The mea­sure­ment will be wrong. The reported result will not be true.

A par­tic­u­lar­ized prob­lem in ana­lyt­i­cal chem­istry as pre­formed in a mod­ern crime lab­o­ra­tory is that some crime lab­o­ra­to­ries use zero as a data point on the con­structed cal­i­bra­tion curve or force the cal­i­bra­tion curve through the ori­gin. For two basic rea­sons, this is wrong.

First, no machine is capa­ble of mea­sur­ing true zero. True zero is the total and com­plete absence of every­thing. As all ana­lyt­i­cal instru­ments have an inputted sig­nal that is reg­is­tered on some sort of detec­tor, there will always be noise. Noise is detected sig­nal. It is not true zero. The clos­est that the machine can mea­sure zero is by run­ning a true blank, mean­ing test­ing a sam­ple with no ana­lyte of inter­est in it and with what is thought to have noth­ing detectable by the detec­tor. Even with a true blank, there will always be an off­set on the y-axis that is the ran­dom noise sig­nal detec­tion. It will never be able to detect true zero. There­fore, the ori­gin (the inter­cept of 0 on the x-axis and 0 on the y-axis) can­not ever be truly achieved as a true mea­sured data point. So the very notion that an ana­lyt­i­cal instru­ment can mea­sure true absence is a myth. We need to be care­ful with declar­ing a true con­cen­tra­tion of an ana­lyte is “zero,” as Avogadro’s num­ber is extremely large, (N = 6.0223 x 1023 par­ti­cles mol-1), it could be argued that it is impos­si­ble to ever say that a sub­stance is not present. For the pur­poses of this dis­cus­sion, “zero” means that for all prac­ti­cal pur­poses, the ana­lyte of inter­est is not present. But this is like call­ing the reflec­tion of the sun on the moon, moon­light. Nei­ther moon­light or mea­sured zero truly exist.

Sec­ond, the resul­tant cal­i­bra­tion curve will be biased. The results from the zero-data-point-inclusive-calibration-curve will be biased high in the begin­ning of the lin­ear dynamic range and will be be equiv­a­lent at one point with it being biased low at the high­est end of the lin­ear dynamic range at some point.

The above video is a demon­stra­tion of the bias that can occur if zero is included as a data point. The same exact response from the unknown can be dan­ger­ously miscalculated.

An even worse form of inap­pro­pri­ate con­struc­tion of a cal­i­bra­tion curve is “force through ori­gin” which is an option with some sys­tems. What this does is force lin­ear­ity through the ori­gin. This will result in even greater bias than using zero as a data point in con­struct­ing the cal­i­bra­tion curve.

From EM 200–1-1. Title: Envi­ron­men­tal Qual­ity — Val­i­da­tion of Ana­lyt­i­cal Chem­istry Laboratories

A cal­i­bra­tion curve (whether lin­ear or non­lin­ear) must not be forced through the ori­gin unless it is demon­strated (e.g., dur­ing method devel­op­ment) that the inter­cept (i.e., y[x = 0]) is not sta­tis­ti­cally dif­fer­ent from zero (e.g., by per­form­ing a t-test for the y-intercept or com­par­ing it to the MDL.) Arbi­trar­ily forc­ing a cal­i­bra­tion curve through the ori­gin may adversely impact low-level quan­ti­ta­tive results. Sim­i­larly, when cal­i­bra­tion curves are gen­er­ated using regres­sion analy­sis, the curves must not be arti­fi­cially weighted toward the ori­gin by includ­ing the point (0,0) one or more times.

In short, lin­ear­ity through zero is a sta­tis­ti­cal assump­tion and not a ratio­nale for report­ing results below the cal­i­bra­tion range demon­strated by the analy­sis of the standards.

Addi­tional sources include:

  • Envi­ron­men­tal Pro­tec­tion Agency’s (EPA) MDL procedure1 pro­mul­gated at 40 CFR (Code of Fed­eral Reg­u­la­tions) Part 136, Appen­dix B, rev. 1.11
  • Ana­lyt­i­cal Detec­tion Limit Guid­ance & Lab­o­ra­tory Guide for Deter­min­ing Method Detec­tion Lim­its (“It is not accept­able to force any cal­i­bra­tion curve through zero.”
  • Amer­i­can Soci­ety for Test­ing and Mate­ri­als, “Stan­dard Prac­tice for Intral­ab­o­ra­tory Qual­ity Con­trol Pro­ce­dures and a Dis­cus­sion on Report­ing Low-Level Data”, Annual Book of ASTM Stan­dards, Vol. 11.01, des­ig­na­tion D 4210–89, pp 14 — 20.

Do you know what your crime lab­o­ra­tory is doing?

Tagged with: • • •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Guest Blog Post From Dr. Frederic Whitehurst PhD JD: Circuitous thought-the myth of reliable and valid marijuana identification in the courtroom

On March 1, 2012, in Analysis, Chemistry, Contextual Bias, Equipment, Ethics, Identification, In The Courtroom, by Justin J. McShane

Cir­cuitous thought-the myth of reli­able and valid mar­i­juana iden­ti­fi­ca­tion in the courtroom

By:  Fred­eric White­hurst, J.D., Ph.D. [1]

As sci­en­tists we all agree that the hall­marks of sci­ence are valid­ity and reli­a­bil­ity. What do we mean by that? Well, the gov­ern­ment sci­en­tist has to agree that his method of answer­ing a ques­tion put to him has to be valid and reli­able. So no mat­ter how he iden­ti­fies mar­i­juana that method must be valid and reli­able. Sci­en­tists deter­mine valid­ity and reli­a­bil­ity with data. What does valid­ity mean? It means that the method we use to answer a ques­tion must appro­pri­ate for the use to which we put it. Bet­ter said, just because we have a ham­mer does not mean every­thing is a nail.

The ques­tion we are ask­ing about mar­i­juana is “Is this mate­r­ial mar­i­juana?” That means is it mar­i­juana to the exclu­sion of all other plants? So we have to have a pro­to­col that will show us that the mate­r­ial we have is mar­i­juana to the exclu­sion of all other plants. In order to show that our pro­to­col is valid we must either have tested all other plants or have a the­o­ret­i­cal argu­ment as to why no other plant can be mis­taken for mar­i­juana using our pro­to­col. The only attempt at val­i­da­tion that has been con­ducted in the past is the work of George Naka­mura and John Thorn­ton back in the early 1970’s. That did not show the pro­to­col iden­ti­fied mar­i­juana to the exclu­sion of all other plants but only tested the pro­to­col on 31874 plants. So no one has val­i­dated the pro­to­col. The gov­ern­ment sci­en­tist can not show that. In a Frye state we accept the opin­ion if the method used to arrive at it is accepted by the com­mu­nity of peers. The com­mu­nity of peers do not accept that the mar­i­juana analy­sis pro­to­col is valid because no one has val­i­dated the pro­to­col. They just use the method. No one has val­i­dated it. As for reli­a­bil­ity, that means if the tests is per­formed by one per­son many times or by many peo­ple will one get the same answer. How­ever, there are no reli­a­bil­ity stud­ies known. Gov­ern­ment sci­en­tists sim­ply say that they always get the right answer and never have got­ten the wrong answer but they are not being tested for reli­a­bil­ity with known stan­dards. They are sim­ply say­ing that the unknowns must be mar­i­juana because the police say they are mar­i­juana so they must be marijuana.

Cir­cuitous thought. Wrong think­ing. Com­pound that with the paucity of sci­en­tific data that exists to test the reli­a­bil­ity hypoth­e­sis and you have noth­ing to refer to to indi­cate that the pro­to­col is actu­ally reli­able. So the peer group would agree that in order for the pro­to­col to be used it must be shown to be valid and reli­able and it has not been shown to be so it can’t be used…but they use it anyhow.

_______________

[1] Exec­u­tive Direc­tor, Foren­sic Jus­tice Project, Wash­ing­ton, D.C., B.S. Chem­istry, 1974, East Car­olina Uni­ver­sity, Ph.D. in Chem­istry, 1980, Duke Uni­ver­sity, J.D., 1996, George­town Uni­ver­sity School of Law. (202)342‑6980.

Tagged with: •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

The myth of the neutral scientist doing forensic science

On February 28, 2012, in Analysis, Contextual Bias, by Justin J. McShane

Oné of the most ridicu­lous claims that I hear in court is a claim by a foren­sic tech­ni­cian or expert wit­ness (regard­less of what par­tic­u­lar dis­ci­pline he or she is from) that some­how he or she is a neu­tral and detached sci­en­tist doing sci­ence, and there­fore he or she has no bias what­so­ever. This total lack of bias is argued in that the per­son gets the same pay­check every pay period if they turn in results that are con­sis­tent with con­vic­tion or incon­sis­tent with con­vic­tion. There­fore, they are not any dif­fer­ent than an automa­ton. This is more than Spock, greater than a robot (that could have been pro­grammed by some­one who may be biased), but a true autonomous machine that is imper­vi­ous to exter­nal stim­uli or inter­nal emo­tions or stim­uli to do any­thing other than the task at hand. As it is sim­ply put “They don’t have a dog in the fight” or so the say­ing goes.

The forensic scientist as an automaton

The foren­sic sci­en­tist as an automaton

Hog­wash. Sim­ply put, every­one is biased about nearly every­thing. In the com­mon ver­nac­u­lar the word “bias” has taken on a very neg­a­tive con­no­ta­tion. In sci­ence, bias also has a neg­a­tive con­no­ta­tion mean­ing the degree of dif­fer­ence from the true value. How­ever, in real­ity bias is not a bad thing nec­es­sar­ily. It is a very impor­tant func­tion of our evo­lu­tion as a species. Bias is what keeps us from get­ting into cars with strangers. It is part of what keeps us safe and sound.

In the court­room and the lab­o­ra­tory, the peo­ple who scare me the most are those who ascribe to this claim that they are totally free of bias. What is worse per­haps are the folks who work in pri­vate lab­o­ra­to­ries who claim that because they work in a for-profit lab­o­ra­tory set­ting that they are immune from all forms of bias.

If you are biased, I am totally fine with that. I don’t see that as a flaw. Embrace it. Declare it. How­ever, you have to insti­tute reme­di­a­tion mea­sures designed to mit­i­gate it. You are a flawed per­son and a flawed scientist/technician/expert if you believe that you have no bias.

As no one is free from bias, foren­sic lab­o­ra­to­ries need to start teach­ing every­one who works there that they are biased and stop the fool­ish claim. If we are inter­ested in pro­mot­ing error free work that is the high­est qual­ity, then we need to acknowl­edge it and take steps to mit­i­gate it.

Tagged with:
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

The Confrontation Clause: Who has to show up in court and why

On February 24, 2012, in Analysis, Ethics, In The Courtroom, Lectures, by Justin J. McShane

The below is a full lec­ture on the Con­fronta­tion Clause and the expla­na­tion of the “Pri­mary Pur­pose Rule,” the “Par­tic­u­lar Wit­ness Rule,” and an exam­i­na­tion of all of the cur­rent cases as decided or pend­ing before the United States Supreme Court. This lec­ture is pre­sented from a prac­ti­cal point of view for the ben­e­fit of all parties.

This video is specif­i­cally ded­i­cated to my col­leagues, friends, and fel­low mem­bers of the National Col­lege for DUI Defense, Inc.

Tagged with: • • • • • • •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

Guest Blog Post From Dr. Frederic Whitehurst PhD JD: On sampling when testing multiple packets

On February 23, 2012, in Analysis, Chemistry, Identification, Laboratory Procedure, by Justin J. McShane

Sam­pling Crime: On sam­pling when test­ing  mul­ti­ple packets

By:  Fred­eric White­hurst, J.D., Ph.D. [1]

If we accept that foren­sic crime lab­o­ra­to­ries are com­pletely over­whelmed, under­staffed, under­equipped, and exhaust­ing envi­ron­ments in which to func­tion, we see real­ity.  Despite polit­i­cal rhetoric to the con­trary, there has been lit­tle, if any, relief for those folks in crime labs who are toil­ing every­day under impos­si­ble work­loads.  The US War on Drugs is brought to the doors of crime labs hun­dreds of times a day because, finally, pow­ders and veg­etable mate­ri­als must be chem­i­cally and phys­i­cally ana­lyzed in order to charge offend­ers prop­erly. If we believe the Chem­i­cal Abstracts Reg­istry, which tells us that there are over 56,000,000 known chem­i­cal com­pounds, and if we real­ize that a large por­tion of those mate­ri­als can occur as pow­ders, then we see the daunt­ing task faced by foren­sic chemists. White pow­ders seized by law enforce­ment per­son­nel are not proven to be con­trolled sub­stances sim­ply because they are white pow­ders which give pos­i­tive results for pre­sump­tive field tests. Those pow­ders must be sub­jected to rig­or­ously val­i­dated sci­en­tific ana­lyt­i­cal pro­to­cols in order to test the hypothe­ses of seiz­ing offi­cers that the mate­ri­als are actu­ally con­trolled sub­stances. These tests take an enor­mous amount of time, time that crime labs do not have.  In order to address the issue of lim­ited resources, the North Car­olina State Bureau of Inves­ti­ga­tion (NC SBI) Crime Lab­o­ra­tory in the past has turned to case law, State v. Brenda Hard­ing, and sam­pling pro­to­cols which allow one to deter­mine the whole from an analy­sis of a part of that whole.  This arti­cle will dis­cuss the sci­en­tific and legal impli­ca­tions of the sam­pling pro­to­cols which the SBI Lab­o­ra­tory has uti­lized to achieve deter­mi­na­tions of seized multi­u­nit sam­ples of alleged con­trolled substances.

When law enforce­ment offi­cers seize sus­pected con­trolled sub­stances, often those seizures are in many small units of pow­ders. An inves­ti­ga­tion may end in the seizure of hun­dreds of small bags or bindles. The chem­i­cal ana­lyst can­not real­is­ti­cally ana­lyze every small bag; there­fore, for many years the NC SBI Lab uti­lized a sam­pling pro­to­col referred to as the “square root of n plus 1” pro­to­col. That pro­to­col in its entirety is pro­duced here:

Name of Procedure:

Ran­dom Sam­pling

Ran­dom Sam­pling of Mul­ti­ple Pack­ages or Units

Sug­gested Uses:

Ran­dom sam­pling is a pro­ce­dure that is used when ana­lyz­ing an item of evi­dence that con­sists of mul­ti­ple pack­ages or units.  This pro­ce­dure allows a chemist to deter­mine the com­po­si­tion of the evi­dence by ana­lyz­ing some ran­domly selected pack­ages or units and extrap­o­lat­ing the results.  Ran­dom sam­pling is an accepted pro­ce­dure used in foren­sic sci­ence and has been upheld by the Appel­late Courts of North Car­olina (see lit­er­a­ture references).

Ran­dom Sam­pling Procedures:

1. Visu­ally exam­ine all of the pack­ages or units in the item of evi­dence, as well as the con­tents, for dif­fer­ences in size, weight, color, pack­ag­ing, mark­ings, signs of tam­per­ing, label­ing or other char­ac­ter­is­tics.  If there are no appre­cia­ble dif­fer­ences, all of the pack­ages or units should be con­sid­ered together for the selec­tion of ran­dom sam­ples.  If there are appre­cia­ble dif­fer­ences, seg­re­gate the pack­ages or units into indi­vid­ual groups, based upon such observed differences.

2. To deter­mine the num­ber of ran­dom sam­ples to be selected from a total num­ber of pack­ages or units, where n equals total num­ber of pack­ages or units:

a. If n is less than or equal to 4, then ran­dom sam­pling is not done.

b. If n is greater than or equal to 5, then the num­ber of ran­dom sam­ples selected is equal to the square root of n plus 1, expressed as:

ran­dom sam­ples = √n + 1

c. Weight deter­mi­na­tion — the total weight of all pack­ages or units may be extrap­o­lated from the weight of a ran­dom sam­ple of the pack­ages or units.

d. Weight count — the total num­ber of all pack­ages or units may be extrap­o­lated from the weight of a ran­dom sam­ple of the pack­ages or units.

Clas­si­fi­ca­tion of Evidence:

There are three main forms of con­trolled substances:

1. Plant material.

2. Con­trolled sub­stances con­sist­ing of marked dosage units from legit­i­mate phar­ma­ceu­ti­cal manufacturers.

3. Con­trolled sub­stances derived from clan­des­tine manufacturers.

a. Pack­ages or units con­tain­ing pow­der or solids.

b. Pack­ages or units con­tain­ing liquid.

c. Pack­ages or units con­sist­ing of any sub­stance which is used as a median to absorb or con­tain a con­trolled sub­stance (plas­tic bags, glas­sine bindles, paper bindles, blot­ter paper, gelatin, sugar cubes, tea leaves, pars­ley, etc.)

Appli­ca­tion of Pro­ce­dure on Evidence:

1. Ran­dom sam­pling of plant material:

a. Visual exam­i­na­tion of all pack­ages or units and a com­plete analy­sis of one pack­age or unit is required to con­firm iden­ti­fi­ca­tion (min­i­mum requirements).

2. Ran­dom sam­pling of marked dosage units from legit­i­mate phar­ma­ceu­ti­cal manufacturers:

a. The visual exam­i­na­tion and the mark­ings on the dosage units pro­vide iden­ti­fi­ca­tion of the con­trolled sub­stance and a com­plete analy­sis of one dosage unit is required to con­firm iden­ti­fi­ca­tion (min­i­mum requirement).

3. Ran­dom sam­pling of con­trolled sub­stances derived from clan­des­tine manufacturers:

a. Ran­dom sam­ples of pack­ages or units must be selected and sub­jected to at least one screen­ing test.  A com­plete analy­sis of a por­tion of the ran­dom sam­ples is required to con­firm iden­ti­fi­ca­tion (min­i­mum requirement).

Safety Con­cerns:

Not applic­a­ble.

Lit­er­a­ture References:

Colon, Rodriguez, and Diaz, Rep­re­sen­ta­tive Sam­pling of Street Drug Exhibits, Jour­nal of Foren­sic Sci­ences, Vol. 38, No. 3, May 1993, pp. 641–648.

Siegel, and Safer­stein, Foren­sic Iden­ti­fi­ca­tion of Con­trolled Sub­stances, Foren­sic Sci­ence Hand­book, Vol. 2, Pren­tice Hall, 1988.

Tzi­dony and Ravreby, A Sta­tis­ti­cal Approach to Drug Sam­pling: A Case Study, Jour­nal of Foren­sic Sci­ences, Vol. 37 Novem­ber, 1992, pp. 1541–1549.

Frank, Hink­ley, and Hoff­man, Rep­re­sen­ta­tive Sam­pling of Drug Seizures in Mul­ti­ple Con­tain­ers, Jour­nal of Foren­sic Sci­ences, Vol. 36, March 1991, pp. 350–357.

Wag­goner, R.W., t Dis­tri­b­u­tion and Pre­dic­tion Inter­vals, North Car­olina State Bureau of Inves­ti­ga­tion, 1996.

State v. Myers, 301 S.E. 2d 401, 402 (N.C. App. 1983)

State v. Wil­helm, 296 S.E. 2d 664, 667 (N.C. App. 1982)

State v. Absher, 237 S.E. 2d 1325, 1328 (N.C. App. 1977)

State v. Clark, 197 S.E. 2d 81, 82 (N.C. App. 1973)

Lit­er­a­ture Ref­er­ences (continued):

State v. Riera, 172 S.E. 2d 535, 539 (N.C. App. 1970)

State v. Hard­ing, 429 S.E. 2d 416 (N.C. App. 1993)

Com­plete with sci­en­tific lit­er­a­ture ref­er­ences and case law, the pro­to­col might at first seem to be an unad­dress­able issue when one is ques­tion­ing its valid­ity. Indeed State v. Hard­ing seems at first glance to stand solidly behind the pro­to­col. Recall that, Hard­ing stands for the premise that one can deter­mine the whole from an analy­sis of a part. How­ever, Hard­ing does not address which sam­pling pro­to­col one must uti­lize. Fur­ther­more, Hard­ing does not sup­port the use of the pro­to­col listed above. In order to demon­strate this lack of sup­port, let us look at an exam­ple of a bindle of heroin seized as evidence.

This bindle is a cel­lo­phane con­tainer which has been folded twice over and then sealed with a piece of tape. Pow­der within the bindle can­not be seen. One can­not, there­fore, fol­low the pro­to­col that is repeated in part below by visu­ally exam­in­ing the con­tents for dif­fer­ences in size, weight, or color…:

1. Visu­ally exam­ine all of the pack­ages or units in the item of evi­dence, as well as the con­tents, for dif­fer­ences in size, weight, color, pack­ag­ing, mark­ings, signs of tam­per­ing, label­ing or other char­ac­ter­is­tics. If there are no appre­cia­ble dif­fer­ences, all of the pack­ages or units should be con­sid­ered together for the selec­tion of ran­dom sam­ples. If there are appre­cia­ble dif­fer­ences, seg­re­gate the pack­ages or units into indi­vid­ual groups, based upon such observed differences.

The opaque nature of the bindle stops an exam­iner from deter­min­ing the color and the mor­phol­ogy (size/shape) of the pow­der crys­tals inside the indi­vid­ual pack­ages. Fur­ther­more, street sam­ples of heroin found in these bindles are usu­ally in the hun­dredths of a gram size.

To give us some idea of the size of those sam­ples, notice that a pack­age of sugar sub­sti­tute that one finds on the table in a restau­rant con­tains 1 gram of material.

Now imag­ine two to four hun­dredths of a gram is all the pow­der within the bindle. Imag­ine then clos­ing your eyes while some­one removes 10% of the pow­der in that pack­age with­out your knowl­edge and being asked to look back at the mate­r­ial. One can­not really detect such an appre­cia­ble change in such a small amount of mate­r­ial. There­fore, the pro­to­col set forth above does not accu­rately func­tion for such sam­ples. Addi­tion­ally color is not a very pro­ba­tive char­ac­ter­is­tic of mate­ri­als when numer­ous chem­i­cals and com­bi­na­tions of chem­i­cals in the world might very well have the same color.

Because the visual exam­i­na­tion through a mul­ti­lay­ered semi-opaque mate­r­ial is not pos­si­ble, the exam­iner must open all the bindles in order to visu­ally ana­lyze the con­tents of those bindles. In the expe­ri­ence of the author, Dr. White­hurst, at the point of open­ing these pack­ages, very often the ana­lyst will sim­ply mix the con­tents of the pack­ages, thus destroy­ing infor­ma­tion con­cern­ing those con­tents. The undis­turbed con­tents of these pack­ages is evi­dence which a defen­dant has a right to review and an act, such as mix­ing these con­tents, is a direct vio­la­tion of N.C. GS 14–221.1 and 15–11.1 which pro­hibit the destruc­tion of evi­dence with­out court autho­riza­tion.  Mix­ing con­tents often accom­plishes the aim of any­one wish­ing to estab­lish guilt of traf­fick­ing. How­ever, assum­ing that the prosecution’s duty is to uphold jus­tice, this mix­ing cir­cum­vents both the role of the pros­e­cu­tion as well as defense counsel.

The ran­dom sam­pling pro­to­col above also neglects to instruct an ana­lyst regard­ing how to actu­ally con­duct a ran­dom sam­pling. Ana­lysts should be instructed that in order to prove a lack of bias in choos­ing which sam­ples to weigh and ana­lyze one must “ran­domly” select samples.

One way to achieve this ran­dom sam­ple is to assign a num­ber to each sam­ple. Then, uti­liz­ing a ran­dom num­ber gen­er­a­tor com­puter pro­gram, select those sam­ples directed by that pro­gram. An inex­pen­sive ver­sion of the ran­dom sam­pling con­sists of putting all the sam­ples into a box or bag and pick­ing out sam­ples with­out look­ing at the sam­ples cho­sen for analy­sis. Thus, despite being named “Ran­dom Sam­pling” pro­ce­dure, the ran­dom­ness of the sam­pling is not defined to ensure ran­dom selection.

Let’s review how this pro­to­col directs us to look at the num­ber of sam­ples to be analyzed:

2. To deter­mine the num­ber of ran­dom sam­ples to be selected from a total num­ber of pack­ages or units, where n equals total num­ber of pack­ages or units:

a. If n is less than or equal to 4, then ran­dom sam­pling is not done.

b. If n is greater than or equal to 5, then the num­ber of ran­dom sam­ples selected is equal to the square root of n plus 1, expressed as:

ran­dom sam­ples = √n + 1

c. Weight deter­mi­na­tion — the total weight of all pack­ages or units may be extrap­o­lated from the weight of a ran­dom sam­ple of the pack­ages or units.

d. Weight count — the total num­ber of all pack­ages or units may be extrap­o­lated from the weight of a ran­dom sam­ple of the pack­ages or units.

For the reader who suc­cess­fully may have for­got­ten any lessons from math classes in high school the “square root” of a num­ber is another num­ber which when mul­ti­plied by itself will equal the num­ber. For exam­ple the square root of one hun­dred is ten. The square root of six­teen is four. There­fore, if there are one hun­dred bindles in a seizure, one will then ran­domly select ten plus one and ana­lyze eleven sam­ples. In Dr. Whitehurst’s expe­ri­ence, exam­in­ers in the NC SBI lab will some­times choose the square root of n plus one sam­ples and then imme­di­ately mix them together. Once mixed, the court will never know which sam­ples con­tained what mate­r­ial. Using our com­mon sense we know that if we mix unknown pow­ders together (in order to prove a traf­fick­ing weight), this mix­ture is a newly cre­ated mix­ture. Because NC statutes crim­i­nal­ize the man­u­fac­ture of pow­der con­tain­ing con­trolled sub­stance x, the state’s ana­lyst is pos­si­bly in vio­la­tion of these man­u­fac­tur­ing statutes by mix­ing together sam­ples of this pow­der con­tain­ing x.

Fur­ther­more, sup­pose, for exam­ple, that the defen­dant has 100 bags of pow­der, pro­duced by an aver­age street phar­ma­cist (con­sist­ing of unre­li­able con­tents), and some of those bindles con­tain a con­trolled sub­stance while oth­ers con­sist of a non-controlled sub­stance. There is no qual­ity con­trol in those street phar­ma­cies. This prob­lem is evi­denced by statutes in NC which pro­hibit the sale of false con­trolled sub­stances. We will not assume naively that the buyer, upon find­ing he has been cheated by pur­chas­ing a non-controlled sub­stance, will go to the bet­ter busi­ness bureau. In fact the level of law­less­ness in the illicit drug busi­ness is legendary.

Com­pli­cat­ing this pic­ture fur­ther, the US Drug Enforce­ment Sci­en­tific Work­ing Group on Drugs has pub­lished its 2008 “Sci­en­tific Work­ing Group for the Analy­sis of Seized Drugs (SWGDRUG) Rec­om­men­da­tions” which can be found at www.swgdrug.org. Page 8 of that doc­u­ment describes sam­pling strat­egy and notes that the sam­pling pro­ce­dures are divided into sta­tis­ti­cal and non-statistical pro­ce­dures.  Among the sta­tis­ti­cal pro­ce­dures the hyper­ge­o­met­ric, Bayesian, and other probability-based approaches are listed and among the non-statistical pro­ce­dures are the square root n, man­age­ment direc­tive, and judi­cial require­ments stan­dards. One can infer from this list that the NC SBI lab’s ran­dom sam­pling pro­to­col is a non-statistical sam­pling pro­ce­dure because page 9 explains that: “If an infer­ence about the whole pop­u­la­tion is to be drawn from a sam­ple, then the plan shall be sta­tis­ti­cally based and lim­its of the infer­ence shall be documented.”

Page 11 goes on to say: “Depend­ing upon the infer­ence to be drawn from the analy­sis for a mul­ti­ple unit pop­u­la­tion, the sam­pling plan may be sta­tis­ti­cal or non-statistical…Statistical approaches are applic­a­ble when infer­ences are made about the whole pop­u­la­tion. For exam­ple:  b) The total net weight of the pop­u­la­tion is to be extrap­o­lated from the weight of a sample.”

Page 12 also indi­cates that: “Non-statistical approaches are appro­pri­ate if no infer­ence is to be made about the whole population.”

Based on these pas­sages, it seems as though the NC SBI lab has been uti­liz­ing a non-statistical sam­pling pro­to­col from which one can­not deter­mine the total weight of a sam­ple of mul­ti­ple units. Fur­ther­more, these pas­sages indi­cate that the total net weight for multi­u­nit sam­ples has been deter­mined with­out valid foun­da­tion. If one reviews the pro­to­col as pro­vided under dis­cov­ery by the NC SBI lab dur­ing past cases, one will see that this pro­to­col has been in use since 1996. The use of this non-statistical pro­to­col for nearly 15 years is trou­bling and leads us to A.J. Izenman’s “Sta­tis­ti­cal and Legal Aspects of the Foren­sic Study of Illicit Drugs in Sta­tis­ti­cal Sci­ence, 2001, Vol. 16, No. 1, 35–57.  On page 47 Izen­man writes: “The square-root and other pop­u­lar rules. A world­wide sur­vey of sam­pling prac­tices and choices of sam­ple sizes for foren­sic drug analy­sis …found that the most pop­u­lar rule for decid­ing how many con­tain­ers or items, whether homo­ge­neous or not, to sam­ple for drug test­ing was not a sta­tis­ti­cally moti­vated one. Instead, the most pop­u­lar rule was the square-root rule, … used by lab­o­ra­to­ries in Aus­tralia, Aus­tria, Canada, Eng­land, New Zealand, Hong Kong, North­ern Ire­land and the United States and U.S.A. Army-Europe. One would assume, there­fore, that the square-root rule would be an accepted part of sam­pling prac­tice. Yet, in an infor­mal, but exten­sive, sur­vey of sam­pling prac­ti­tion­ers, we found that most sam­pling experts had never encoun­tered the square-root rule and no text­book on sam­pling the­ory or prac­tice nor review of the field…even refers to it…

The square-root rule appar­ently orig­i­nated in the 1920’s from a need to pro­vide agri­cul­tural reg­u­la­tory inspec­tors (specif­i­cally, those who knew how to extract a square root) with a con­ve­nient, mem­o­riz­able rule for sam­ple size deter­mi­na­tion.”

The his­toric con­text of the use of this pro­to­col ver­sus its present appli­ca­tion is trou­bling. Since 1996 has the North Car­olina SBI lab­o­ra­tory been decid­ing weight of total pop­u­la­tion in drug traf­fick­ing cases with a sam­pling pro­to­col which the com­mu­nity of sta­tis­ti­cians opines is not valid for such a use? Have law enforce­ment, pros­e­cu­tors and juries decided a defen­dant pos­sessed a cer­tain amount of mate­r­ial when in fact the NC SBI Lab­o­ra­tory could not accu­rately make that deter­mi­na­tion with the pro­to­col that it utilized?

Recently a col­league requested a review of the lab dis­cov­ery mate­r­ial in a crim­i­nal mat­ter. Under Item 1 on the SBI Lab­o­ra­tory report dated Sep­tem­ber 2010 one can read that a sta­tis­ti­cal sam­pling plan that demon­strates 95% con­fi­dence was uti­lized.  That report goes on to call that plan the “hyper­ge­o­met­ric” sam­pling plan. Does this pro­to­col change indi­cate  that the NC SBI lab is now rec­og­niz­ing that the pre­vi­ous non-statistical sam­pling plan was indeed fun­da­men­tally flawed? If there truly is a new sam­pling plan—a sta­tis­ti­cal sam­pling plan—what con­se­quences will this changed pro­to­col have not only for post con­vic­tion relief mat­ters, but also for the many traf­fick­ing cases which were ana­lyzed using the square root of n plus one sam­pling plan, a plan which the NC SBI lab may now admit to itself, if not to defen­dants in courts of law under Brady, pro­lif­er­ated mis­takes and pos­si­bly caused mis­car­riages of justice?

__________

[1] Exec­u­tive Direc­tor, Foren­sic Jus­tice Project, Wash­ing­ton, D.C., B.S. Chem­istry, 1974, East Car­olina Uni­ver­sity, Ph.D. in Chem­istry, 1980, Duke Uni­ver­sity, J.D., 1996, George­town Uni­ver­sity School of Law. (202)342‑6980.

Tagged with: • • • •
 
About the author

Justin J. McShane

Harrisburg DUI attorney Justin J. McShane is the President/CEO of The McShane Firm, LLC - Pennsylvania's top criminal law and DUI law firm. He is the highest rated DUI attorney in PA as rated by Avvo.com. Justin McShane is a double Board certified attorney. He is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc. He is also a Board Certified Criminal Trial Advocate by the National Board of Trial Advocacy, a Pennsylvania Supreme Court Approved Agency. Justin McShane is the first and so far the only Pennsylvania attorney to achieve American Bar Association recognized board certification in DUI defense from the National College for DUI Defense, Inc.

Visit Authors Website

0 Comments
Leave A Response

« Previous Entries Next Entries »
Go To Top »