A significant methamphetamine result and a nonsignificant d-amphetamine result do not, by themselves, establish that the two drugs differ. A historical controlled experiment illustrates this distinction unusually well. Methamphetamine significantly increased measured cardiovascular endpoints relative to placebo, while the other drugs did not. Yet several direct comparisons between the stimulants were nonsignificant, and two significant comparisons reflected greater responses to a comparator than to low methamphetamine doses.
That is a question about the comparison actually tested. It is different from asking whether participants could recognize a stimulant’s subjective effects, whether a drug was present in a specimen, or whether someone was impaired.
What experiment produced these findings?
Sevak, Stoops, Hays and Rush published the study in the March 2009 issue of the Journal of Pharmacology and Experimental Therapeutics. Ten healthy adults enrolled; seven completed and entered the analyses. Two did not acquire the required drug discrimination and one was lost to follow-up. The seven completers were 20–24 years old, included four men and three women, and reported recreational stimulant experience without meeting the study’s criteria for stimulant abuse. This selected sample is not representative of all people using methamphetamine.
Participants first learned to distinguish 10 mg of oral methamphetamine from placebo. The subsequent test phase used double-blind, medically supervised oral administration, with randomized order subject to scheduling restrictions. The active dose ranges were:
- Methamphetamine: 2.5, 5, 10 and 15 mg.
- D-amphetamine: 2.5, 5, 10 and 15 mg.
- Methylphenidate: 5, 10, 20 and 30 mg.
- Triazolam: 0.0625, 0.125, 0.25 and 0.375 mg, as a pharmacologically different comparison.
These are administered doses, not blood concentrations or dose recommendations. Heart rate and blood pressure were measured before dosing and hourly for five hours afterward. Sessions were separated by at least 24 hours. The test-phase cardiovascular analysis used each participant’s maximal observed value during the one-to-five-hour period. It did not estimate an all-day average or continuously record every transient change.
Which placebo comparisons were significant?
Across the tested dose ranges, methamphetamine was the only drug with significant increases above placebo in heart rate and systolic and diastolic pressure. The discussion identifies the 10 and 15 mg methamphetamine conditions as producing the significant cardiovascular increases. The reported peak values after methamphetamine were 91.1 ± 5.1 beats/min, 147.1 ± 1.7 mm Hg systolic pressure and 81.7 ± 3.7 mm Hg diastolic pressure; the uncertainty terms are standard errors of the mean.
These are reported peak levels, not increases of those amounts above baseline. Nor are they thresholds for toxicity, impairment or a particular dose in an unknown person. A standard error describes uncertainty in a sample mean; it does not describe the full range expected among individual patients.
What happened when the researchers compared the drugs directly?
The investigators separately tested between-drug effects using drug and dose factors. Those comparisons omitted placebo. Reading them changes the meaning of a simple claim that “only methamphetamine was significant.”
| Endpoint and comparison | Reported result | Interpretive limit |
|---|---|---|
| Heart rate: methamphetamine versus d-amphetamine | No significant difference in dose-related effects; F(4,24) = 1.9, p > 0.05. | The unequal placebo-test outcomes did not establish a heart-rate difference between these drugs. |
| Systolic pressure: methamphetamine, d-amphetamine and methylphenidate comparisons | No significant differences among these stimulants; reported F(4,24) values ≤ 1.6, p > 0.05. | The study did not demonstrate greater systolic effects for methamphetamine across these comparisons. |
| Diastolic pressure: methamphetamine versus methylphenidate | No significant difference; F(4,24) = 1.9, p > 0.05. | A significant placebo comparison for only one drug did not establish their difference. |
| Diastolic pressure: methamphetamine versus d-amphetamine | A significant drug effect; F(1,6) = 5.9, p < 0.05. | The authors attributed it partly to greater d-amphetamine responses than responses after 2.5 and 5 mg methamphetamine. |
The heart-rate comparison with methylphenidate also differed significantly, partly because methylphenidate produced greater increases than low methamphetamine doses. A significant difference therefore cannot automatically be described as evidence that methamphetamine produced the larger response.
Why can the two kinds of analysis disagree?
Each drug-versus-placebo analysis asks whether the evidence supports a change from its reference condition. A direct drug comparison asks whether the responses differ from one another. Crossing a significance threshold in one analysis but not another does not itself test that second question.
The appropriate response is to inspect the actual comparison, endpoint, dose structure and direction. Conversely, a nonsignificant between-drug result does not establish equivalence. With seven analyzed participants, the experiment has substantial limits on the differences it can resolve. No formal equivalence margin was established for these cardiovascular comparisons.
Does the study rank cardiovascular danger or driving impairment?
No. The authors described the tested methamphetamine doses as tolerable in these screened volunteers and characterized the observed cardiovascular changes as not clinically significant. That historical assessment applies to this protocol. The study did not measure long-term cardiovascular events, recruit a cardiovascular-disease cohort, compare smoked or intravenous exposure, or reproduce high-dose repeated use.
It also found no significant drug effect on the percentage of correct responses in its digit-symbol task. That particular null result is not a general finding of unimpaired driving. Neither the cardiovascular values nor the task score provides a blood-concentration impairment scale. A case-specific opinion requires the relevant behavioral observations, specimen information, exposure context and alternative explanations.
What should a careful report say?
State that only methamphetamine significantly increased the three cardiovascular endpoints versus placebo in this small oral-dose experiment. Then preserve the direct comparisons: several were nonsignificant, and some significant contrasts partly reflected larger comparator responses than low-dose methamphetamine responses. This wording retains the finding without converting a pattern of p-values into an unsupported ranking of drugs.
Scientific source
Sevak RJ, Stoops WW, Hays LR, Rush CR. Discriminative Stimulus and Subject-Rated Effects of Methamphetamine, d-Amphetamine, Methylphenidate, and Triazolam in Methamphetamine-Trained Humans. Journal of Pharmacology and Experimental Therapeutics. 2009;328(3):1007–1018. doi:10.1124/jpet.108.147124. Full text. Historical controlled human experiment; this commentary focuses on cardiovascular comparisons.
Leave a Reply