Why a Quantifiable Amphetamine Result Is Not an Impairment Scale: The Swedish Driver Study

A blood laboratory’s ability to quantify amphetamine does not establish how impaired a driver was. In a Swedish observational study, amphetamine concentrations overlapped substantially across physician-rated impairment categories. The study is particularly useful for separating an analytical reporting limit from a biological performance threshold.

Two samples answered different questions

Alan Wayne Jones examined 300 apprehended drivers whose blood testing identified amphetamine as the only psychoactive drug. That sample described concentration patterns by age and sex. A separate, smaller retrospective sample of 70 drivers supplied the clinical impairment comparison. Treating all 300 as participants in the impairment analysis would overstate the evidence.

This was human primary research using forensic records, not a controlled administration experiment, an animal study or a review. Investigators did not assign doses, routes or times of use. Nor were these randomly selected members of the driving population: police suspicion and apprehension determined entry into the sample.

What the laboratory actually measured

Venous blood was collected with sodium fluoride and potassium oxalate. Screening was followed by more specific confirmation. Amphetamine in blood was quantified by gas chromatography–mass spectrometry after extraction and derivatization, using a deuterated internal standard. The routine limit of quantitation was 0.03 mg/L. Samples above the calibration range were reanalyzed with dilution.

The measured analyte was amphetamine. These data must not be relabeled methamphetamine concentrations, urine concentrations, or concentrations of an unspecified combined stimulant measure. A quantitation limit describes the analytical method’s reporting capability; it is not a concentration experimentally shown to impair every person.

The clinical categories had overlapping concentrations

In the 70-driver sample, physicians classified 11 drivers as not impaired, 18 as slightly impaired, 26 as moderately impaired and 15 as highly impaired. Their respective mean blood amphetamine concentrations were 0.60, 0.72, 0.69 and 0.73 mg/L. Ranges overlapped: the not-impaired category extended from 0.05 to 2.1 mg/L, whereas the highly impaired category extended from 0.15 to 2.3 mg/L.

After logarithmic transformation, the analysis did not establish concentration differences between the four categories. Figure 4 reports F = 0.61 and p = 0.608. This is a nonsignificant result for that particular comparison. It neither demonstrates that amphetamine has no behavioral effects nor proves that every concentration is compatible with safe driving.

Age and sex findings do not supply a driving threshold

In the larger 300-driver sample, the mean and median concentrations were 1.0 and 0.9 mg/L. Age had a weak positive association with concentration, r = 0.18, reported as statistically significant. Concentrations were numerically higher in women, but the sex comparison was not significant. These descriptive findings cannot establish that age causes a concentration difference, or that either age or sex determines impairment.

Why the clinical reference standard matters

The physicians knew they were examining suspected drug-impaired drivers. Their protocol included appearance, orientation, speech, coordination and balance, and had originated in alcohol assessment. Examinations occurred on average about an hour after driving or initial arrest. The paper itself questions potential expectancy bias and the relationship between these clinical observations and actual driving skills.

There was no drug-free comparison group for the clinical analysis and no controlled on-road performance outcome. Tolerance, fatigue and use history were discussed as possible explanations, not independently established causes of each driver’s results. The absence of another detected psychoactive drug also remains bounded by the testing panel and method.

How to apply the result without overstating it

A scientifically defensible opinion distinguishes three propositions: amphetamine was identified; a mass concentration was measured; a person had a particular functional limitation. The first two do not automatically decide the third. Equally, uncertainty about the third does not erase a valid analytical identification.

For comparison, our Norwegian registry analysis describes a different population, combined-analyte definition and statistical approach. Those findings should remain attached to their own methods rather than being merged into a universal cutoff. The Swedish paper’s historical policy discussion likewise is not a statement of current law in any jurisdiction.

Primary research

Jones AW. Age- and gender-related differences in blood amphetamine concentrations in apprehended drivers: lack of association with clinical evidence of impairment. Addiction. 2007;102:1085–1091. https://doi.org/10.1111/j.1360-0443.2007.01802.x.