Milwaukee Office
Oshkosh Office
West Bend Office
Waukesha Office
Free Consultation
CALL NOW - 24/7

Casting Doubt on the Reliability of Blood Alcohol Tests

The jagged hump phenomena raise serious doubts about the integrity of any test results.

Casting Doubt on the Reliability of Blood Alcohol Tests

Exculpatory Evidence in Blood Alcohol Tests Done at the Wisconsin State Laboratory of Hygiene

We have discovered a series of suspect or anomalous events in blood alcohol tests done at the Wisconsin State Laboratory of Hygiene (the Hygiene Lab).

These tests are a major component of evidence in cases involving citizens who are accused of alcohol-related offenses. We call this phenomena the “jagged hump,’’ for the manner that it appears on the chromatogram (or blood test result) graph.

These phenomena potentially affect hundreds and possibly thousands of blood test results and raise doubts about the reliability of tests done at the Hygiene Lab.

While the Hygiene Lab has not hidden the problem, neither has it affirmatively disclosed the issue to persons who may be affected. This article is to alert the Bar of the existence of this evidence, which we believe to be exculpatory, and subject to mandatory disclosure under Brady v. Maryland.

The Jagged Hump

Blood alcohol tests are done using a process called headspace gas chromatography. This process separates the component parts of a subject’s blood; once separated, these component parts are measured. The result is a graph called a chromatogram.

Ideally a chromatogram will have a flat baseline with discrete peaks, each representing a substance in the sample. Normally, a peak above the baseline of the graph indicates the presence of a substance (such as ethanol) in the blood sample. These peaks should be distinct and separate from other peaks, indicating good separation of substances, or resolution. Figure 1 shows chromatograms with two distinct peaks: one peak indicates ethanol in the sample; the other peak indicates 1-propanol, which was added to the sample by the Hygiene Lab. These are normal chromatograms (note that there are two chromatograms, corresponding to the two analysis “columns” or tubes), with good resolution, showing normal peaks.

Figure 2 is an example of an abnormal set of chromatograms. In Figure 2, there are anomalous, unexplained jagged humps at the lower left portion of each graph. The humps are not distinct or resolved from each other, indicating a malfunction of the instrument rather than the presence of an unknown substance. The position of the jagged hump at the left side of the chromatograms also indicates an instrument malfunction, as it appears to the left of the point where any peak at all should be visible.

A Hygiene Lab analyst recently testified that personnel from Perkin Elmer, the manufacturer of the test equipment, investigated the jagged hump phenomena. Nevertheless, neither the Hygiene Lab nor Perkin Elmer knows what causes these jagged humps; although they admit that they shouldn’t be there.

FIGURE 1: Normal chromatograms, showing distinct, resolved peaks, indicating the presence of ethanol and 1-propanol in the sample.

FIGURE 2, Jagged hump chromatograms, with jagged non-distinct, “non-resolved” peaks, near the left side of the graph, indicating an instrument malfunction. The peaks at .88 on top and .79 on the bottom were reported as ethanol.

Requirements of Accepted Laboratory Standards

The International Standards Organization (ISO) has promulgated the ISO 17025 laboratory standard, which is accepted as the consensus standard of the scientific community. ISO 17025 has been adopted by the American Society of Crime Laboratory Directors, the Wisconsin State Crime Laboratory, and the Wisconsin Occupational Health Laboratory (which is another department of the Hygiene Lab). This standard requires that any analytic laboratory instrument that repeatedly displays suspect results should be removed from service until a root-cause analysis has been completed and the problem is resolved. ISO 17025, sec. 5.5.7.

The ISO 17025 standard, although widely accepted, has not been adopted by the Forensic Toxicology Section of the Hygiene Lab. The Hygiene Lab is accredited by the American Board of Forensic Toxicology (ABFT), whose standards are less stringent. The ABFT, however, still requires a root cause analysis and corrective action for “repeated failures beyond that statistically expected.” ABFT Criteria E-9.

The Jagged Hump Calls Into Question the Reliability of Blood Tests

The Hygiene Lab maintains confidence in the integrity of their test results, despite the jagged hump anomaly. A Hygiene Lab analyst has testified that the abnormality “shouldn’t be there,” it is contrary to the manufacturer’s protocols, and it is unexplained. Nevertheless, because the jagged hump appears on the chromatograms in a location distinct from the ethanol peaks, and because the standard/control ethanol test results during the analysis batch run are within specification, the Hygiene Lab believes that ethanol tests are unaffected by this abnormality.

We believe, however, that the confidence of the Hygiene Lab in their test results is badly misplaced. As the frequency, scope and origin of the jagged hump phenomena are unknown, the effect on tests results must also be unknown. Unless the Hygiene Lab conducts a formal root-cause analysis and determines what is causing this repeated abnormality, it cannot conclude that its test results are reliable.

As the jagged hump phenomena are unexplained, they may be only the tip of the iceberg: the most visible symptoms of a larger, unknown problem affecting the Hygiene Lab’s test results. For example, it may be the result of a problem with the interface between the Hygiene Lab’s hardware and the software that runs the system. This could also cause other problems, such as transposed, mixed-up test results, or misreported test results (which has occurred in an Arizona laboratory using the same model instruments).

No analysis has been done as to whether the jagged hump appears randomly, in a pattern, or under specific conditions. These factors cannot be assumed without analysis; as, e.g., even the number Pi appears on its face to be random. The standard/control tests are placed in a specific pattern in the run. The jagged hump may simply be appearing in a different pattern from the standard/control samples.

If, in fact, the jagged hump is appearing randomly, it may simply have not affected the ethanol reading of a standard/control sample, by chance. The Hygiene Lab’s contrary conclusion is unsupported by data; no statistical analysis has been done regarding the frequency or timing of the phenomena as it relates to the placement of standard/control samples.

As the abnormality appears in results from all three of the Hygiene Lab’s instruments, it may be a function of some other part of the process, such as sample preparation, that systematically excludes the manner in which standard/control samples are handled.

Most importantly, to ignore these abnormal phenomena violates the scientific method. It introduces an unacceptable, unknown variable into the process. It cannot be assumed that this unknown variable is unimportant. Scientific due diligence requires a proper investigation of the jagged hump phenomena. In analytic chemistry, methods of analysis must be validated; validity may not, simply, be assumed.

In response to our open records request, the Hygiene Lab has stated that no root cause analysis or corrective action have occurred, nor any documented communications made to any governmental or non-governmental entity regarding the abnormality (despite the testimony that Perkin Elmer, itself, has investigated).

We have retained two experts — a laboratory auditor and a pharmacologist — who have advised us that the jagged hump phenomena may reasonably be considered “suspect results” under the ISO 17025 standard, which requires the instruments to be removed from service. Further, these phenomena are “repeated failures” under the ABFT Criteria E-9. Thus, the Hygiene Lab’s ABFT accreditation standard requires a formal, properly documented, root-cause analysis and corrective action. This has not been done; nor has the matter been reported to the ABFT.

Who Is Affected by the Jagged Hump?

The Hygiene Lab has three separate instruments, each with two columns, or channels. The jagged hump has appeared on both columns of all three instruments. Nevertheless, all three instruments remain in service. Tens of thousands of test results have been reported, with an unknown but substantial number displaying a jagged hump. There has been no investigation of when, or the number of times the phenomena has appeared. The Hygiene Lab tests about 20,000 samples per year, and we have seen many jagged humps in our cases. Therefore, we believe it to have affected at least hundreds, probably thousands of samples over the last year.

The jagged hump phenomena raise serious doubts about the integrity of any test results that display the phenomena; and any other test results in the same batch. As such, it is exculpatory evidence that should be disclosed under the ABFT laboratory accreditation standards, as well as Brady v. Maryland. At least one jury has agreed with us; as, attorney Jeffrey Oswald obtained an acquittal in a jagged hump case.

We recommend that defense counsel demand the disclosure of the chromatograms of clients’ tests, as well as all other chromatograms in the batch, as well as other appropriate laboratory records. The phenomenon may be of evidentiary significance, simply if it appears in the same batch as that of a client.

We have formed an ad hoc committee for the purpose of investigating this and other issues related to blood alcohol testing in Wisconsin. Please contact us if you have any questions or comments.

Andrew Mishlove
Lauren Stuckert
Emily Jane Bell
Mishlove and Stuckert, LLC

Aaron Nelson
Doar, Drill and Skow, S.C.

Todd Schroeder
Devanie, Belzer and Schroeder, S.C.

Jeffrey Oswald
Hammet, Bellin and Oswald, LLC

Michael Cohen
Cohen Law Offices, LLC

Karyn T. Missimer, Esq.
Missimer Law, S.C.

Wisconsin OWI Lawyer Andrew Mishlove
Attorney Andrew Mishlove has over 30 years' experience defending clients accused of DUI. He is considered a leading authority on drunk driving defense, often teaching other attorneys the latest defense techniques.
To speak with Andrew, call 414-332-3499 or click here for a free consultation.

4 Responses

  1. Roderick F. Mollson

    One of the major problems with gas chromatography is its extreme sensibility. There used to be a joke that if you walked a person through a room where people were smoking marijuana, the gas chromatograph would show cannabinol in that person’s blood.
    Given that the Wisconsin lab appears to be picking up something else along with the propanol and the ethanol fairly accurately, why should there be any concern about anomalies that creep in? Maybe people in Wisconsin have something odd in their drinking water, but how exactly does that impeach the test result?

    1. Thank you for your comment. This is a interesting hypothesis, but it does not seem to fit all the facts. The jagged hump does not seem to be a substance in the sample at all. Its also appearing in the laboratory prepared control samples. This is much more likely to be an abnormality in the process.

      First, the peaks are entirely unresolved, hence the phrase “jagged hump” indicating something other than a substance in the matrix.

      Second, this type of gas chromatography uses flame ionization detection, which would not show anything other than volatile substances. It would never show a cannabinol; as that would require a different type of detector. It is unlikely that the jagged hump reflects the presence of some unknown volatile compound from the water (unless, perhaps, the water from West Virginia). If it was compound, it looks to be a strange and deadly one.

      Third, the jagged hump precedes T0, meaning that it appears on the chromatogram as eluting faster than the carrier gas itself, or possibly from the preceding sample. Either way, that’s quite abnormal.

      Most of all, it is invalid to simply assume that everything is okay. The jagged hump is a cause for concern because it is unexplained. The burden is on the government to disclose this to all persons who may be affected, which they have not done. Also, the burden is on the government to prove that the machines are operating properly, which they have not done. The burden is not and should not be on the accused to prove anything. The jagged hump may or may not be causing false ethanol test results; no one really knows. The government should not make any assumptions without doing a root-cause analysis.

      Thank you, again,
      Andrew Mishlove

  2. Roderick F. Mollison

    So basically a subpoena to the lab for the bench notes and the pretest calibration check should give the answer.