Can Immersed Hair Reference Materials Make Forensic Drug Results More Consistent? An Eight-Laboratory Study

Which original empirical experiment answers this question?

Jing Hu, Jiaojiao Ji, Xinyu Cao, Xiaodan He, Zheng Qiao, Xianyu Fan, Wei Liu, and Hang Chen published Development and application of hair matrix reference materials for the determination of morphine, cocaine, 3,4-methylenedioxymethamphetamine, methamphetamine, and their metabolites in Forensic Science International, volume 378 (2026), article 112705. The underlying original study is identifiable at its primary research DOI. The investigation combines preparation of artificially impregnated human-hair controls, studies of washing and storage, bottle-to-bottle quality assessment, and a comparison involving eight actual analytical laboratories.

Its central question was methodological: would drug-containing human-hair matrix reference materials, made through a controlled immersion procedure, behave more consistently during quantitative laboratory analysis than material prepared by adding standard solutions to hair? This is different from asking whether a given person’s tested hair proves voluntary ingestion. A control designed to challenge an instrument can be highly useful without recreating the biological history of the person involved in a forensic case.

What is a hair matrix reference material, and why does the matrix matter?

A matrix reference material is a material with a defined analyte content intended to help a laboratory evaluate or control a measurement process. Hair is unusually difficult because the test does not simply introduce a freely dissolved compound into a vial. Drug molecules may be bound to or associated with structural proteins, pigments, cuticle layers, superficial residues, or complex sample fragments. A laboratory must wash, cut, homogenize, extract, separate, detect, and quantify the analytes in the presence of this matrix.

Adding a reference drug to a blank hair sample immediately before analysis produces one kind of control. Immersing hair in a drug-containing solvent for many days and washing it afterward produces another. Neither procedure reproduces natural incorporation from blood during hair growth in a verified dosed person. The researchers investigated whether the latter manufactured matrix would produce more reproducible measured results after the full analytical workflow. A better reference material may strengthen quality control while still leaving interpretation of external exposure and internal administration to separate evidence.

How many genuinely independent people and specimens were involved?

For making laboratory reference materials, the authors collected drug-free hair from 10 consenting healthy volunteers, aged 20 to 45, and confirmed the target substances were not detected at initial screening. These donor samples were used to create intentionally drug-fortified experimental hair. In a separate application experiment, the investigators evaluated 18 previously positive authentic hair specimens whose legal retention periods had expired: six in the morphine/6-acetylmorphine category, six in the MDMA/MDA category, and six in the cocaine/benzoylecgonine category. The manuscript does not establish that all these source individuals can be added into a single independent recruitment denominator.

Eight laboratories participated in measuring prepared portions of authentic hair. Eight laboratories are eight measurement settings, not eight new exposed patients. The 180 vials generated for some prepared materials, 90-vial sets made at each methamphetamine soaking concentration, repeated time-point withdrawals, replicate extractions, and instrument injections were technical or within-material observations. Treating a hundred prepared bottles as a hundred drug-using subjects would fundamentally misstate the experiment and exaggerate its external validity.

Which chemicals were deliberately incorporated into drug-free hair?

Seven target compounds were analyzed: morphine (MOR), 6-acetylmorphine (6-AM), 3,4-methylenedioxymethamphetamine (MDMA), 3,4-methylenedioxyamphetamine (MDA), cocaine (COC), benzoylecgonine (BZE), and methamphetamine (MAMP). Their chemical identities matter. Benzoylecgonine is a metabolite associated with cocaine exposure; 6-AM is associated with heroin metabolism; MDA and MDMA are chemically distinct. The laboratory’s ability to identify each in an artificially fortified hair reference preparation does not prove the biological route by which the same compound reached another person’s hair.

For six targets the team prepared a mixed drug-containing solution, immersed cleaned hair, took samples over time, and compared content before and after washing. Methamphetamine was separately studied at three different soaking-solution concentrations. None of these quantities represented an administered oral, smoked, injected, or transdermal dose in a participant. They were known external laboratory preparations, and the resulting hair values were reported as nanograms of analyte per milligram of hair, not as drug concentrations in an individual’s circulation.

How did immersion time affect measured drug content?

The investigators monitored the uptake of six target analytes during roughly 45 days of exposure to the soaking solution. The traces rose unevenly, sometimes with abrupt jumps, and generally approached a plateau around day 30. The methamphetamine series was followed through 39 days and taken out of the bath on day 40, with a reported plateau around day 33. The study therefore supports a practical observation: short immersion may not create the same measured matrix concentration or steadiness as a prolonged preparation under these specific laboratory conditions.

These data should not be read as human pharmacokinetic curves. The horizontal axis measures the number of days a dead hair specimen was soaked in a chemical bath. It does not measure time since human ingestion, distribution in living tissues, or excretion. The temporary increases and dips also caution against assuming that a single time-point result reflects simple linear uptake. The authors discussed several possible explanations, but the study did not isolate a complete molecular mechanism for every fluctuation.

Why were unwashed and washed artificial hair compared?

Forensic laboratories frequently wash hair to address surface residues before analyzing material from within the shaft. The researchers compared three washing programs on artificially impregnated hair: water alone, acetone alone, and a combination of water followed by acetone. The experimentally unwashed material offered a baseline. A removal percentage was calculated from the difference between concentrations measured in washed and unwashed prepared hair.

Across the source’s six-compound comparison, removal varied substantially with the analyte and procedure. It is incorrect to present a mixed solvent protocol as universally superior for every drug: cocaine was one notable exception in the reported results. Equally, a high removal percentage cannot by itself prove which molecular compartment held the lost material. A wash can remove accessible surface deposits and some compound associated with deeper layers. It can also change the specimen that a downstream extraction must recover. For a specific method, laboratories need documented decontamination behavior and validated quality-control criteria.

Evidence table 1: How did the washing protocol change artificially impregnated hair?

Amounts are reported hair-content measurements and removal percentages from the original washing comparison. They are not fractions of an administered human dose and cannot be converted to blood concentrations.

What do the different wash responses actually show?

Morphine and 6-acetylmorphine showed relatively large reductions after combined washing: 68% and 74% in the original table. By contrast, the same protocol removed approximately one fifth to one quarter of the four other reported targets. Acetone alone removed only 2% of MDMA and 3% of benzoylecgonine under the tested conditions; it removed 26% of cocaine compared with 23% using the combined protocol. This heterogeneity matters more than a single universal claim that a particular wash is always best.

The outcome represents what these agents did to this specific artificially loaded hair after a specified immersion and drying procedure. Native biological deposition in growing hair, cosmetic bleaching, pigment content, storage histories, powder or smoke deposition, sweat transfer, and use of different solvents may change the result. Successful washing of an externally immersed control does not by itself distinguish a real user’s ingestion from environmental contamination. This limitation is particularly important when testimony turns on whether the molecule was taken internally rather than how well a control passed a laboratory extraction check.

What exactly did the LC–MS/MS instrument quantify?

The investigators weighed small portions of prepared or authentic hair, subjected them to cleaning, cutting and mechanical homogenization, then used methanol-containing internal standard, ultrasound, centrifugation, and liquid chromatography–tandem mass spectrometry. They operated in positive-ion electrospray multiple-reaction monitoring and tracked selected transitions for each target. In the stated analytical procedure, results were reported as ng of specific drug per mg of hair.

That unit is fundamental. A value such as 6.80 ng/mg refers to a defined analytical analyte-to-hair-mass ratio under the processing conditions. It is not 6.80 ng/mL in venous whole blood, not 6.80 ng/mL of centrifuged serum or plasma, and not a measured quantity in urine, saliva, breath, sweat, or vitreous humor. Likewise, a methamphetamine concentration of 120 micrograms per milliliter in a soaking bath describes the manufactured reference liquid, not the blood of the volunteer who donated blank hair. The study provides no mathematical or clinically validated conversion across those compartments.

Was the methamphetamine impregnation concentration controlled?

For methamphetamine, the team prepared three nominal soaking solutions: 1.2, 12, and 120 micrograms per milliliter. The associated homogeneity-assessment measurements in finished hair were about 0.58, 1.93, and 29.09 ng/mg, respectively. Increasing the externally prepared bath concentration corresponded to greater measured methamphetamine retained by the hair. The source also reported a closely fitted relationship across its studied preparation settings, with a coefficient of determination near 0.998.

These were three laboratory preparation levels, not three patient dose groups. A high coefficient of determination over three nominal concentrations does not validate interpolation across all hair colors, damaged hair, bathing conditions, drug formulations, or living people. The result is useful for planning manufacturing trials and checking the final reference-material concentration. It is not a methamphetamine use-history or toxicokinetic regression, and it must not be used to reverse engineer an individual’s administered dose or time of last use from a positive strand.

Did the researchers conduct actual statistical tests of homogeneity?

Yes. For prepared reference-material batches the investigators used one-way analysis of variance to evaluate between-bottle differences. All nine reported values were above their nominal 0.05 criterion. Examples include morphine p = 0.43, 6-acetylmorphine p = 0.36, MDMA p = 0.34, cocaine p = 0.35, benzoylecgonine p = 0.42, and the medium-concentration methamphetamine preparation p = 0.08. The last number is particularly relevant because it is closer to 0.05 than the other listed comparisons while still being formally nonsignificant at the stated threshold.

A non-significant ANOVA result means the test did not identify sufficient evidence of a between-bottle mean difference under this design and threshold. It does not prove an absolute absence of variation, establish a legal equivalence margin, or guarantee the performance of every future manufacturing lot. The individual bottle values, analytical repeatability, selection plan and intended acceptance specifications still matter when laboratories decide whether a reference material is suitable for a particular casework method.

Does six-month chemical stability mean reference hair lasts indefinitely?

No. The authors studied storage behavior in selected environmental conditions, including refrigerated storage near 4°C and room temperature near 25°C for six months, plus a short elevated-temperature and high-humidity challenge. They assessed stability with a regression-based acceptance rule rather than asserting that no individual molecule could ever degrade. Within those tested windows the paper reported that the prepared analytes met its stability criterion.

Stability depends on the chosen analyte, initial material, container, light, humidity, temperature, handling, and assay. A six-month outcome does not automatically extend to two years in an uncontrolled warehouse or to a different manufacturer, hair pigment, or extraction method. The investigators also reported expanded uncertainty estimates. Those uncertainty values require careful interpretation rather than treating the apparently successful stability signs as proof of zero error. For casework, the validity of a control should be established by its actual certificate or characterization evidence, expiry, storage history, and the laboratory’s quality management practices.

What was the real eight-laboratory comparison?

For the application phase the investigators obtained 18 authentic historical hair specimens from earlier drug-positive cases. They prepared and distributed analyzed portions among eight laboratories, which measured target groups using either conventional spiked hair reference materials or the newly immersed matrix reference material for quantitative adjustment. The table in the paper reports results across the laboratories for six target analytes: MOR, 6-AM, MDMA, MDA, COC and BZE. Methamphetamine belonged to the preparation and characterization work, but it is not one of the six columns in that eight-laboratory authentic-sample comparison.

The principal reported difference was consistency of quantitative outputs among participating laboratories. The authors listed lower coefficients of variation and narrower ranges with immersed controls. Those observed differences are informative for interlaboratory reproducibility. They do not directly compare two independent groups of patients, and the paper does not provide an inferential p-value establishing the statistical significance of the difference in CV for each analyte. The reference-control type is only one possible source of laboratory-to-laboratory disagreement; extraction recovery, analysts, instrument calibration, internal standards and local practices may also contribute.

Evidence table 2: Which interlaboratory coefficients of variation became smaller?

These are the source’s across-laboratory CV figures reorganized without reproducing its original table layout. A lower CV describes less observed relative dispersion; it is not itself a p-value or a guarantee of lower systematic bias.

Why does a lower CV not necessarily prove higher accuracy?

Precision and accuracy are different analytical properties. A set of laboratories may report closely clustered numbers that are collectively biased high or low. The coefficient of variation captures dispersion relative to the mean, not the closeness of that mean to an independently certified true concentration. The original comparison showed that immersed hair reference controls were associated with less cross-laboratory scatter; it did not independently establish an absolute reference value for every authentic specimen that makes all participating measurements demonstrably unbiased.

The authors noted a practical constraint: for the authentic hair evaluation only one concentration of immersed reference material was available, so each laboratory used single-point correction with an internal standard. A single adjustment point does not establish calibration linearity, matrix response, or measurement uncertainty across the entire concentration range. An effective commercial laboratory reference program would need several relevant levels, appropriate blank material, metrological characterization, control of lot-to-lot variation, and evidence that method performance transfers to real test specimens. The eight-lab result is promising but not an all-purpose validation certificate.

Could immersion-made reference hair mistake environmental contamination for ingestion?

There is a crucial conceptual difference. In this experiment the researchers themselves deliberately immersed previously negative hair in a solution containing named drugs. That process demonstrates that drug molecules can be placed in and later recovered from hair without those donors having ingested them. The study was not designed to determine what fraction of a positive clinical or criminal hair result arose from true internal exposure versus external deposition. Some molecules could remain after washing, but persistence alone is not proof of a biological origin.

In an actual investigation, surface contamination, decontamination-wash findings, drug/metabolite patterns, hair treatment, sampling site, segment length, growth variation, and independent clinical or investigative records may be relevant. None should be treated as a universal stand-alone discriminator. A laboratory should be able to explain how controls were manufactured and how those controls relate to the case specimen, particularly if the report’s interpretation will address ingestion rather than the narrower fact that a chemical was identified in prepared material.

What does the study leave unmeasured?

The study did not administer cocaine, methamphetamine, MDMA, heroin or any other target drug to the ten drug-free hair donors. It did not obtain paired quantitatively assayed human forensic venous whole blood for the identified analytes, nor did it report accompanying serum, plasma, urine, oral-fluid, sweat, breath, or vitreous concentrations. The artificial liquid in which hair was immersed was not a human specimen. The authentic 18 hair specimens had a prior positive testing history, but no verified dose, exact route, precise last-use time, contemporaneous blood measure, or behavioral impairment measure was established by this interlaboratory method experiment.

It also did not prospectively determine the true clinical sensitivity or specificity of hair testing for drug ingestion, the false-positive rate under all contamination scenarios, the probability of impaired driving at a given hair concentration, or causation of a particular assault, crash, medical condition or death. Those questions demand different designs and evidence. Successful instrument quantification does not transform the reference material into an epidemiological exposure study or an individualized medico-legal opinion.

How should a forensic laboratory or expert apply these findings?

A useful application is to ask whether a laboratory’s chosen quality-control material genuinely challenges the extraction and mass-spectrometric workflow used for the questioned hair. Review the certificate or characterization evidence, homogeneity and stability records, target matrix and pigment range, blank controls, acceptance windows, chromatographic identification criteria, extraction recovery, calibration levels, replicate performance, and documented corrective actions. Compare lot-specific results with the uncertainty of the actual measurement rather than simply citing the label “reference material.”

For opposing-expert review, distinguish three independent propositions: first, that a laboratory method detected a named substance in a particular hair specimen; second, that the quantitative value was measured with acceptable precision and accuracy for the task; and third, that the measured material demonstrates an individual’s ingestion, impairment or participation in an event. Hu and colleagues primarily inform the second proposition, with some preparation-method implications for the first. They do not independently decide the third. An expert should avoid using improved reference-material consistency to resolve exposure origin or legal causation without other validated evidence.

Frequently asked forensic-science questions

1. Did the ten healthy volunteers receive methamphetamine, MDMA, cocaine or opioids?

No. They provided initially drug-free hair for artificial preparation of reference materials. Named chemicals were placed into soaking solutions under laboratory control. A hair strand containing a standard after that procedure is not evidence that its donor took the substance, and none of the bath masses should be described as an administered human drug dose.

2. Were the 180 bottles and multiple laboratory aliquots independent people?

No. Bottles, sample segments, repeated withdrawals and injections are technical subunits or repeated assessments. The study describes hair donated by ten initially drug-free volunteers, separate authentic historical hair specimens, and eight participating measurement laboratories. Inflating those counts into hundreds of independent exposed subjects would misrepresent the design.

3. Did the experiment find a statistically significant difference between immersed and spiked control CVs?

The table reports sizeable descriptive reductions in between-laboratory coefficients of variation for all six reported analytes. The authors did not report analyte-by-analyte hypothesis-test p-values establishing statistical superiority of those CV reductions. In contrast, they did report genuine nonsignificant bottle-homogeneity ANOVA p-values for the prepared material. These two statistical questions should not be conflated.

4. Can ng of methamphetamine per mg of hair be converted to ng/mL in whole blood?

No conversion was validated by this study. Drug content in extracted artificially immersed hair and concentration in circulating venous whole blood are different quantities from different compartments. The methamphetamine solution concentrations were laboratory fortification levels, not measured drug concentrations in any human donor. A numerical blood estimate would require additional validated pharmacokinetic and sampling data that were not collected.

5. Does water-and-acetone washing always remove more drug than acetone alone?

No. Most compounds in the reported combined-washing comparison showed greater removal than with acetone alone, but cocaine did not: the source lists 23% removal with the combination and 26% with acetone alone. Each method must be evaluated for the analyte and hair preparation actually at issue, and washing efficiency is not itself evidence of an individual’s drug-ingestion history.

6. Do six months of acceptable stability results guarantee that any laboratory hair sample is stable for six months?

No. The reported storage tests concern particular manufactured batches under specified conditions and a defined statistical acceptance rule. Authentic case hair may have a different pigment composition, treatment history, container, environmental exposure, moisture and analyte profile. Storage conclusions should not be extrapolated without validated evidence for the actual sample and conditions.

7. Does lower cross-laboratory variation prove a positive hair result establishes intoxication or a criminal incident?

No. Lower CV can help characterize reproducibility, but the study did not test intoxication, psychomotor performance, impairment thresholds, a person’s last drug use, or a specific event. Source-of-exposure and causation questions require separate evidence even when a laboratory has accurately identified and quantified a chemical in hair.

Bottom line: a better reference material is not a better drug-use clock

The original experiment supports a restrained conclusion: extended immersion can produce drug-containing hair matrices with measurable behavior under washing, homogeneity and stability tests, and eight laboratories showed markedly less relative scatter when applying immersed reference material to selected authentic hair analyses. It also exposed limitations in single-level correction, chemical-specific wash effects, and the distinction between a highly reproducible value and a demonstrably accurate biological interpretation. Those are useful forensic laboratory lessons, not an inference that all hair positives equal ingestion or clinical impairment.

For complementary discussion of evidence types, see the independently available same-site guides on biological specimen matrices, confirmatory drug testing, and specimen collection and handling. The underlying original numbers and limitations should be checked against Hu and colleagues’ original primary empirical paper before using them in a case-specific opinion. No original experimental figure or publisher table is reproduced here.