Can Cocaine Remain Detectable in Newly Collected Hair Four Months After Reported Abstinence? A 15-Patient Study

The direct answer: the experiment tracked persistent hair findings, not fresh drug administration

In an original longitudinal investigation of 15 patients beginning cocaine treatment, compounds associated with prior cocaine consumption remained chemically measurable in the proximal portion of some subsequent scalp-hair samples. Four months into reported abstinence, six participants had quantifiable cocaine under the analytical method, but only three had cocaine above the study’s separate 0.5 ng/mg interpretive cutoff. Four still had quantifiable benzoylecgonine. By six months, no participant had a reported detectable result for the three analytes studied. A late-positive hair segment therefore did not automatically mean that the participant had secretly resumed use.

That answer has boundaries. The investigators monitored a selected rehabilitation cohort, excluded some cases inconsistent with continued abstinence, and collected hair about every two months. They did not create a controlled cocaine-dosing experiment or validate an exact individual last-use clock. The data are a useful demonstration of delayed disappearance from sampled hair, not a guarantee that every abstinent person will test negative at a particular calendar month.

What was the complete primary study?

A. Suárez-García, I. Álvarez-Freire, A. M. Bermejo-Barrera, P. Cabarcos-Fernández, and M. J. Tabernero-Duque published Duration of detection of cocaine and metabolites in hair after discontinuation of abuse in Microchemical Journal, volume 153, article 104335 (2020). The original DOI is 10.1016/j.microc.2019.104335. This was primary longitudinal empirical research combining repeated human hair sampling with laboratory method assessment, not a general review of hair toxicology.

The study’s narrow forensic question was whether quantifiable cocaine, benzoylecgonine, and cocaethylene would disappear promptly from the newest sampled scalp-hair portion once chronic users began an abstinence program. It is distinctive because the same cohort provided serial specimens at admission, month two, month four, and month six. Such paired longitudinal observations address an issue that a single cross-sectional positive or a hair-spiking study cannot resolve in the same manner.

How many biologically independent patients were followed?

The independent human unit was the patient, not the strand, one-centimeter segment, specimen vial, or injection into a gas chromatograph. The study reports fifteen rehabilitation participants. Each supplied an admission sample and then recurring scalp specimens at two-month intervals across a six-month follow-up. Those repeated observations remain linked to the same underlying individuals. Counting four sampling dates as sixty independent people would artificially enlarge the evidence and distort the apparent precision.

All fifteen entered a treatment program in Santiago de Compostela, Spain. The source describes urine monitoring during treatment to support reported abstinence, but it also explains that some initially considered cases showed rising later hair concentrations and did not remain in the analyzed abstinence series. Consequently the article describes the retained fifteen rather than everyone who might enter a drug-treatment service. It does not provide an unselected community cohort or validated population distribution of clearance times.

How were hair samples collected and separated?

Investigators sampled posterior vertex scalp hair as close to the scalp as feasible on enrollment and about every two months. They selected a proximal one-centimeter portion, treated the material with neutral soap and water to reduce external deposition, dried it, cut it into smaller fragments, and weighed fifty-milligram analytical aliquots. Their target was the new or recently available growth portion, but a bundle of hair is biologically heterogeneous.

Hair follicles alternate between growth, transitional, and resting phases; a lock taken close to the scalp is not made exclusively of fibers that began growing on the same day. Resting-phase hair can remain visible while newer hair grows. Sweat, sebum, and surface deposits can additionally contribute parent compound or metabolites. Thus the label “proximal one centimeter” identifies where analysts cut the sample. It does not certify a perfectly dated month of systemic exposure for every fiber in that bundle.

Why one centimeter of hair is not a one-month toxicology timestamp

Approximate scalp hair growth of one centimeter per month is a population-level planning convention. Real growth rates, follicular cycling, collection distance, hair length, and strand alignment vary. A segment assembled from many hairs is a composite; some strands may have stopped growing while others continued. The investigators themselves identify telogen retention and possible residual scalp deposits as explanations to consider when analytes persist after stopping drug use.

One cannot use these results to subtract four months from a positive hair test and identify the exact last-use weekend. At most, the serial design shows that detectable material remained in the selected proximal collection despite contemporaneous rehabilitation observation. To turn a hair measurement into event timing would require separate, validated subject-specific growth and deposition information that this experiment did not measure. Even a technically sound segmentation protocol cannot transform the specimen into a synchronized calendar.

Which cocaine-related molecules were chemically quantified?

The targeted analytes were unchanged cocaine, benzoylecgonine (called BZE in the source), and cocaethylene (CE). Benzoylecgonine is an important cocaine metabolite, while cocaethylene can arise when cocaine and ethanol are present in the body. Detecting these molecules provides chemical information about the extracted hair, subject to assay specificity and external-contamination assessment. It does not directly reveal dose, moment of exposure, subjective intoxication, consent, or how a particular encounter occurred.

Results were expressed in nanograms of analyte per milligram of hair, not nanograms per milliliter of blood. No legal venous whole-blood specimen, separated plasma or serum parent-drug concentration, oral-fluid sample, sweat patch, exhaled-breath assay, or numerical brain concentration was paired with these monthly hair results. Periodic urine monitoring addressed compliance with treatment, but it was not a simultaneous individual quantitative blood pharmacokinetic study. The physical matrix should remain explicit in any expert summary.

What analytical method identified the substances?

The laboratory used an established solid-phase extraction procedure followed by gas chromatography–mass spectrometry, or GC–MS. Fifty-milligram hair aliquots were prepared with enzymatic and extraction steps, processed through polymeric reversed-phase cartridges, derivatized, and analyzed by selected-ion monitoring with deuterated reference compounds. The published instrumentation included an HP 6890 gas chromatograph and a mass-selective detector operated with electron ionization.

This workflow differs from an unconfirmed immunoassay screen: chromatographic separation and selected ions help distinguish target compounds under the stated laboratory method. Nonetheless, analytical confirmation of a molecule and interpretation of its route into hair are separate questions. The study’s chemical specificity does not automatically eliminate all residual external contamination, nor does it show whether any person had acute behavioral effects when the hair was collected.

Evidence table 1: longitudinal detection counts, not independent new users

The following reorganizes the source’s reported detection counts for the same fifteen participants. A reported count means the compound was quantifiable in that study’s hair analysis at that visit. These counts are descriptive observations. They are not results of a statistical test showing a significant or nonsignificant treatment effect.

There is an important difference between being quantifiable at the method’s analytical limit and meeting an interpretive decision cutoff. At month four, the six quantifiable cocaine measurements did not all exceed 0.5 ng/mg. That distinction reconciles the six-person analytical count with the separate three-person statement about cases exceeding the selected cocaine cutoff; neither should be silently replaced by the other.

How do measurement limits differ from reporting cutoffs?

The analytical limit of quantification concerns whether a numerical amount can be measured within the performance characteristics of the method. An interpretive cutoff is a decision boundary used to classify a result for a defined program or question. Neither is a universally valid dividing line between a truthful and an untruthful person. In this study, the laboratory could report cocaine and benzoylecgonine at 0.2 ng/mg, whereas the discussion identified a higher 0.5 ng/mg cocaine decision threshold.

One consequence is that a concentration can be quantifiable yet fall below the chosen interpretive cutoff. Such a result should not be mislabeled both “chemically absent” and “proof of recent ingestion.” Laboratories must specify which meaning of positive they intend, along with uncertainty, sample preparation, target ions, quality controls, and decontamination method. The result below a cutoff may still be scientifically relevant while failing to satisfy a particular legal or administrative reporting criterion.

Evidence table 2: what the original laboratory actually validated

The following information is drawn from the primary method-validation results but reorganized around the questions a forensic reviewer would ask. These are analytical attributes of the method, not a biological elimination curve or a validation of precise last-use dates.

The limit of detection is lower than the validated quantitative reporting level for each substance. A trace at the limit of detection is not equivalent to a trustworthy, precise quantified result. Recovery describes extraction behavior for analytical materials; it does not mean the laboratory can recover a known percentage of someone’s ingested dose from hair or distinguish every source of deposition.

What did the two-month findings establish?

At the second scheduled collection, thirteen patients still had quantifiable cocaine in the proximal hair portion, ten had benzoylecgonine, and three had cocaethylene. Those were repeat observations of the same participants, collected during a treatment period in which urine testing was used to corroborate reported cessation. They demonstrate a mismatch between a simple elapsed-time assumption and the chemically measurable contents of a newly collected hair sample.

The investigators did not introduce a known dose at admission or experimentally control every past use pattern, hair treatment, skin exposure, or follicular growth rate. Therefore, the persistent findings do not identify a single chemical storage mechanism, and they cannot define a detection probability for an unrelated person. Some of the remaining hair burden may reflect the biology of strands already present before treatment and other deposition processes; the laboratory observations alone cannot apportion those contributions.

What did the four-month findings establish?

Month four is especially important when an authority alleges that any detectable cocaine in scalp hair must represent renewed use during treatment. At this visit, six patients had numerically quantifiable cocaine, four had quantifiable benzoylecgonine, and none had reported quantifiable cocaethylene. Yet only three participants’ cocaine results exceeded the study’s 0.5 ng/mg interpretive cutoff. The six-person count and the three-person count ask different questions.

The original serial observations challenge a categorical inference from one late hair positive to new cocaine use. They do not authorize the opposite categorical conclusion that a positive result must always come from historic exposure. If urine, admissions, observed behavior, hair washing conditions, and test methodology in a specific case contradict abstinence, those facts must be evaluated independently. The point is that hair persistence makes the timing inference underdetermined, not that all apparent relapse findings are artifacts.

Why was everything reported negative at month six?

All fifteen retained participants had results reported as no longer detectable for cocaine, benzoylecgonine, and cocaethylene at the six-month collection. The source authors regarded this as evidence that prolonged persistence after reported cessation can resolve over several months in their selected cohort. In a descriptive five-page report, however, zero detections in fifteen people is not a population-wide guarantee of clearance by that date.

The observation does not distinguish every possible cause of a negative result, such as new growth, changing follicle composition, washing efficiency, dilution of residual burden across sampled fibers, or analyte concentration falling below the method. Nor does it mean every person who refrains from cocaine for six months will have a negative result under every laboratory protocol or cutoff. The scientifically supported conclusion is narrower: all fifteen selected participants in this particular serial design were negative in the reported tests by month six.

Were significant and nonsignificant group differences statistically tested?

The article presents validation statistics such as calibration fit and analytical precision, plus descriptive patient-level trajectories and counts. It does not report a randomized treatment comparison of cocaine persistence, a formal paired significance test of month-two versus month-four hair results, or a hypothesis test comparing one person’s hair clearance with another person’s. A downward sequence in the graphs should not be described as “statistically significant” unless the original study actually reported that test. It did not.

Likewise, “nonsignificant” is not a synonym for a measurement that was missing or for a result below a chemical cutoff. A genuine nonsignificant result requires a defined statistical comparison and an actual analysis that failed the stated criterion. Here the relevant absence is not tested, not an invented p-value. This is a central distinction for scientific reporting: descriptive evidence can be valuable without being treated as proof of statistical generalization.

What do benzoylecgonine and cocaethylene add?

Benzoylecgonine is routinely considered when identifying cocaine exposure. In this cohort it was reported in thirteen patients’ admission samples, ten at month two, and four at month four. Cocaethylene was less frequent: six at admission, three at month two, and none at months four or six. That pattern indicates different persistence of the measured chemicals under this protocol; it is not a validated clock separating simultaneous cocaine-alcohol exposure from later isolated cocaine use.

Cocaethylene’s biochemical association with cocaine and ethanol does not tell an investigator which drink or ingestion event produced an individual hair concentration. It also cannot independently establish intoxication during an allegation months earlier. For a case-specific conclusion, the relevant questions include authentic specimen identity, tested analytes, observed concentration and cutoff, laboratory recovery, possible external contribution, and corroborating history. Each analyte adds context, but none converts retrospective hair analysis into an eyewitness of what happened.

What role did the urine checks have?

Urine testing formed part of the rehabilitation program’s monitoring and supported participants’ accounts of discontinuing cocaine use. This is stronger contextual evidence than a hair-only study with no contemporaneous abstinence checks. Nevertheless, the authors did not provide a complete independently quantified every-day urine time series with case-specific chain-of-custody findings for all participants across the six months. They also describe excluding cases that failed to maintain the intended abstinence pattern.

Consequently, one should neither ignore the urine monitoring nor exaggerate it into an infallible independent proof of no possible exposure at every hour. Urine and hair differ in collection timing, incorporation, concentration units, and detection windows. A negative contemporaneous urine specimen is not the same as a negative hair segment, and a positive hair segment is not automatically evidence that urine screening missed fresh consumption. Both matrices require their own appropriate interpretations.

How could scalp deposits and resting hair complicate a timeline?

The original authors discuss at least two plausible contributors to unexpectedly persistent hair findings: strands in the resting phase that retain an earlier drug history, and residual material associated with the scalp or skin surface. Neither process was isolated or quantitatively apportioned by a controlled experiment in this cohort. Hair growth and environmental deposition can be heterogeneous, so the disappearance of one compound need not mirror the disappearance of another.

A laboratory may attempt to minimize external material through washing, but no single washing protocol demonstrates that every remaining molecule entered exclusively through blood during follicular growth. Conversely, the existence of possible contamination does not make every hair positive meaningless. The scientifically sound approach identifies which mechanisms the study tested, which it only discussed, and which remain unresolved for a given sample.

Does this experiment measure venous whole-blood cocaine?

No. The chemical results come from cut scalp hair measured in ng/mg by GC–MS. They are not parent cocaine concentrations in human venous whole blood, and they do not establish a validated hair-to-whole-blood conversion. The study did not use longitudinal serum or plasma parent concentrations to reconstruct dose-time kinetics, and it did not measure real-time impairment, attention, vehicle control, or eyewitness memory.

Even a seemingly precise hair value—for example, 0.55 ng/mg—does not carry an associated legal venous whole-blood result simply because both are called toxicology. These are physically different compartments with different growth, accumulation, clearance, and sampling properties. A court or investigator should not mistake mass per hair mass for mass per blood volume or infer an on-road physiological effect months before specimen collection.

Could the study prove recent relapse or concealment?

No single hair result from this work establishes that proposition. Because selected abstinent participants sometimes remained quantifiably positive into month four, a late-positive result alone cannot be a definitive relapse timestamp. The study did not evaluate admissions, credibility assessments, secret dosing, intercepted drugs, actual transport events, or prospective clinical relapse adjudication across a representative population. It did not establish a test for intentional deception.

At the same time, the observation is not a license to dismiss every unexpected positive. A real review should examine sequence and trend of results, specimen length and collection site, decontamination records, absolute analyte measurements, independent urine or other testing, permitted medications, prior exposure pattern, and potential contamination. A conclusion about a particular person must come from the complete record, not one generic assertion that a four-month positive is always historic.

How should a forensic reviewer use the patient-by-patient pattern?

The individual trajectories in the original table are more informative than a single average because not everyone began with the same burden and several people had measurable low-level material at follow-up. The correct analytical unit remains the participant. Repeated measures from one person show a trend for that person; they should not be treated as independently sampled people when describing apparent precision or estimating a general cutoff.

Reviewers should distinguish admission level from follow-up level, take care with units and reporting thresholds, and ask whether a finding was above the interpretive cutoff or merely above the laboratory LOQ. The same calendar month can contain different reported numerical levels across patients. That heterogeneity limits simplistic statements about universal clearance and supports a measured evidentiary description tied to the collected scalp portion and laboratory protocol.

What is the defensible implication for licensing, employment, or litigation?

A disciplinary, licensing, or legal inference often asks a binary question—did this person use cocaine during a specified prohibited period? This primary study cautions that a proximal hair positive during rehabilitation does not, by itself, resolve that binary question. The analyte might still be measurable after cessation, and different threshold definitions can change how a specimen is classified. Causation and precise timing require independent foundation.

The reviewer should obtain the complete chain-of-custody and laboratory record, validated method and limits, raw target-ion confirmation, contamination-control and washing procedures, exact scalp collection and cut dimensions, and a longitudinal account of any comparable specimens. These are not formal recommendations derived from an experimental intervention; they are questions that help align a legal proposition with what the measurements actually answer. Drug detection, last-use timing, present impairment, and witness credibility are not interchangeable claims.

How does this cohort differ from contamination-only hair experiments?

Some hair toxicology studies deliberately expose hair samples to external cocaine, wash them, and measure removal or persistence of chemical residues. Others compare hair and skin swabs at a single moment. The present research instead enrolled people in a treatment setting, collected new clinical specimens repeatedly, and observed measured hair concentrations through reported abstinence. These distinct designs should not be merged into a single pooled estimate of false positives, last-use clocks, or hair-test accuracy.

Laboratory hair-spiking research can address removal of artificially deposited drug; population screening can address prevalence of analytical positives in those sampled; longitudinal treatment observations can address within-person changes under monitored conditions. None, alone, validates the same ultimate forensic inference. Here the specific evidence is the disappearance sequence across fifteen retained people, not a controlled external-contamination challenge or a randomized exposure experiment.

Which questions remain unmeasured?

The original investigators did not quantify every person’s individual follicular growth rate, percentage of resting strands, scalp deposition burden, hair-cosmetic history, exact residual cocaine dose, venous whole-blood kinetics, or probability of relapse given a result in a licensed-driver population. They did not record a statistically powered head-to-head validation of alternative cutoffs across hundreds of proven abstinent and current-use subjects. These limitations cannot be converted into supposed negative statistical tests.

Equally, no measured on-road driving task, genuine incident, neuropsychological memory examination, or objective assessment of truthful testimony was paired with the hair results. The study can inform appropriate caution, not adjudicate a driver, survivor, employee, or witness. Precisely defining its unanswered questions makes it more useful than claiming it provides a universal six-month legal rule.

Seven frequently asked scientific questions

Does a cocaine-positive scalp segment four months into abstinence prove recent cocaine use?

No. Six of fifteen selected monitored participants still had quantifiable cocaine at month four, and three were above the separate 0.5 ng/mg decision cutoff. The study cannot date consumption from that finding alone; examine the individual’s complete test sequence and independent evidence.

Were the fifteen people given cocaine by the researchers?

No. They were patients entering a rehabilitation program after previous cocaine use. Hair samples were collected at scheduled visits, and urine checks supported reported cessation. The investigators did not assign a controlled cocaine dose or build a dose-to-hair calibration curve.

Why do the study counts say six quantifiable cocaine samples but only three above cutoff at month four?

The laboratory could quantify some concentrations below the separate 0.5 ng/mg interpretive threshold. Analytical quantifiability and a decision cutoff are different claims. Report which criterion is used rather than declaring the figures contradictory or silently dropping measurements.

What are benzoylecgonine and cocaethylene measuring here?

They are cocaine-related compounds identified in extracted hair, reported as ng/mg. Benzoylecgonine is a cocaine metabolite; cocaethylene can form with cocaine and ethanol exposure. Neither chemical hair result alone proves blood concentration, impairment, or the exact time of drug use.

Were month-to-month changes statistically significant?

The primary article reports descriptive serial concentration data and validated analytical method characteristics, but not a formal month-by-month paired significance test. Calling the decline statistically significant, or reporting a nonsignificant p-value, would fabricate an analysis absent from the source.

Does a negative hair result at month six establish a universal elimination deadline?

No. All fifteen retained participants were negative in this specific protocol by six months. The cohort was small and selected, and hair growth, sample preparation, previous use patterns, and cutoff decisions vary. A population-wide six-month guarantee was not tested.

Can these results determine whether a person was intoxicated, driving poorly, or telling the truth?

No. The test matrix was scalp hair, with no matched quantitative venous whole-blood drug level, on-road performance, verified incident, or credibility endpoint. Such claims would require different measurements and independent case evidence.

Read the original study and related specimen guidance

The primary empirical source is Suárez-García and colleagues, Microchemical Journal (2020), original DOI and complete-study record. For related matrix distinctions, see the independently published biological-matrices guide and confirmatory-testing overview. These contextual pages do not change the source-specific numbers or supply a missing individual blood concentration.

Bottom line: serial testing in fifteen selected patients documented persisting cocaine-related analytes in proximal hair after reported cessation, with decreasing detection over months and zero reported detections by month six. The result argues against a simple last-use clock; it does not create a universal innocence, relapse, or impairment rule.