Responses to a “typical” morning dose of kratom in people who use kratom regularly: A direct-observation study

Abstract

Introduction:

Use of kratom has outpaced systematic study of its effects, with most studies reliant on retrospective self-report.

Methods:

We aimed to assess acute effects following kratom use in adults who use regularly, and quantify alkaloids in the products, urine, and plasma. Between July-November 2022, ten adults came to our clinic and orally self-administered their typical kratom dose; blinding procedures were not used. Physiological measures included blood pressure, respiratory rate, heart rate, pulse oximetry, temperature, and pupil diameter. Subjective outcomes included Subjective Opioid Withdrawal Scale (SOWS), Addiction Research Center Inventory (ARCI), and Drug Effects Questionnaire (DEQ). Psychomotor performance was also assessed.

Results:

Participants were six men and four women, mean age 41.2 years. Nine were non-Hispanic white; one was biracial. They had used kratom for 6.6 (SD 3.8) years on average (2.0-14.1). Sessions were 190.89 minutes on average (SD 15.10). Mean session dose was 5.16 grams (median=4.38; range=1.1-10.9) leaf powder. Relative to baseline, physiological changes were minor. However, pupil diameter decreased (right, b=−.70, p<.01; left, b=−.73, p<.01) 40-80 minutes post-dose and remained below baseline >160 minutes. SOWS pre-dosing was mild (5.5±3.3) and decreased post-dose (b=[−4.0, −2.9], p<.01). DEQ “feeling effects” increased to 40/100 (SD 30.5) within 40 minutes and remained above baseline 80-120 minutes (b=19.0, p=.04), peaking at 72.7/100; six participants rated euphoria as mild on the ARCI Morphine-Benzedrine-scale. Psychomotor performance did not reliably improve or deteriorate post-dosing.

Conclusions:

Among regular consumers, we found few clinically significant differences pre- and post-kratom dosing. Alkaloidal contents in products were within expected ranges.

Introduction

Products derived from the botanical Mitragyna speciosa are marketed in the United States (US) as kratom.^1,2 Although the US Drug Enforcement Agency lists kratom among its “Drugs of Concern,” kratom and its alkaloids have not been scheduled under the Controlled Substances Act. The US Food and Drug Administration (FDA), citing “insufficient information to determine” that kratom products do not present significant or unreasonable risks, designates kratom an unapproved and unregulated “new dietary ingredient.” Preclinical studies suggest kratom alkaloids have opioidergic, adenosinergic, adrenergic, and serotonergic activity.^3-7 A concerning gap in knowledge accompanies a considerable increase in kratom use since 2010, with data (primarily from surveys or case reports) suggesting that kratom’s desired and adverse effects are consistent with its known pharmacology.^8-12 Among people who use kratom regularly, typical dosing frequency is 2-3 times/day, often within an hour of waking, with acute subjective effects often described as stimulatory or analgesic.^10,11,13,14 When doses are higher than intended, adverse effects may reflect adrenergic and/or serotonergic activation as much as mu opioid receptor (MOR) activation.^12,15

Commercial kratom products have not been studied systematically for their effects in humans. Existing knowledge about such effects, outside of self-report, is derived from three human-laboratory studies of kratom alkaloid pharmacokinetics after oral dosing; two of the studies used only one preparation type (a well-characterized kratom tea) and did not examine behavioral outcomes.^16,17 A third study, in Malaysia, used a placebo-controlled design to investigate kratom-induced analgesia.¹⁸ We know of no direct-observation studies of the acute effects of US commercial kratom products in humans. To begin addressing that gap, we conducted such a study as part of a larger, field-based national-history study (NCT05457803). Our aims with this substudy were to assess physiological, subjective, and psychomotor parameters before and after self-administration of participants’ typical morning dose of kratom and to quantify alkaloids in the products and in urine and plasma. The substudy and main study were approved by the National Institutes of Health Institutional Review Board.

Methods

Settings and Population

Between July-November 2022, ten adults who used kratom regularly were re-recruited from a 15-day remote study. To be eligible for this substudy, participants had to complete the remote study, live within 150 miles of our clinic, and report using kratom ≥3 times weekly for ≥4 weeks before enrollment. (A full description of remote-study methods and eligibility is available.¹⁹) Participants were told via email that they could be eligible and then phone calls with staff were scheduled. Those who met inclusion criteria could come to our clinic for informed consent and to give us two equally-sized doses of their typical kratom product, for self-administration on the session day and for analysis. These were weighed and stored by our pharmacy.

Outcomes and Measurements

Before substudy enrollment, participants completed a questionnaire on demographics, drug use, and health. For the session day, we asked participants to refrain from using kratom from 11:59 p.m. the night before until after arrival at our clinic, and to not use psychoactive substances the morning of their session except those they routinely used (e.g., caffeine, nicotine, vitamins, daily medications). Although we permitted participants to eat breakfast the day of their session, participants fasted during the session until blood and urine samples were collected. Our goal was to minimize life disruptions while modeling “real-world” kratom use. Kratom is rarely used in isolation from other substances, so there is public-health relevance in this approach. Before taking their kratom, participants underwent a breathalyzer and a urine immunoassay for: alcohol, amphetamines, barbiturates, buprenorphine, benzodiazepines, cocaine, fentanyl, 3,4-methylenedioxy-methamphetamine, methadone, methamphetamine, morphine, oxycodone, phenylcyclohexyl piperidine, and tetrahydrocannabinol. Baseline assessments for all outcomes were then completed. The study nurse then provided the participant their kratom dose to self-administer in their usual fashion under observation.

The following measures were taken at baseline and post-dose at 5 timepoints, except for the Subjective Opioid Withdrawal Scale (SOWS) and psychomotor tasks, which were assessed at baseline and 3 post-dose timepoints. The Drug Effects Questionnaire (DEQ) was not assessed at baseline (pre-dosing).

Physiological outcomes.

We assessed pupil diameter and vital signs: systolic and diastolic blood pressure (SBP, DBP), respiratory rate (RR), heart rate (HR), pulse oximetry (SpO₂), and temperature.

Subjective outcomes.

Participants rated acute drug effects on the following scales.

SOWS²⁰(16 items, total score 0-64). Scores of 1-10 and 11-20 indicate mild and moderate withdrawal in people withdrawing from opioids. A modified Drug-Effects Questionnaire (DEQ),²¹ consisting of five 100-point visual-analog scales (VAS scales): “feel drug,” “like drug,” “high,” “intoxicated,” and “want more.” DEQ VAS items were unipolar except “liking,” which was bipolar. The 49-item short-form of the Addiction Research Center Inventory (ARCI),²² whose empirically derived subscales reflect experiences consistent with effects of drugs that produce euphoria (MBG [Morphine-Benzedrine Group]), dysphoria (LSD), stimulation (Amphetamine), and/or sedation (PCAG [Pentobarbital-Chlorpromazine-Alcohol Group]).

Psychomotor outcomes.

We administered two computerized tasks: the Four-Choice Reaction-Time Test (FCRTT)²³ and the Number-Vigilance Task (NVT),²⁴ both validated for detecting changes from baseline in cognitive processing speed (reaction time), and accuracy (errors of omission/commission) in response to alphanumeric symbols. Participants were oriented to tasks via brief standardized practice.

Acute adverse events.

As with all NIH clinical studies, participants were monitored for adverse events (AEs), and we report these as an outcome due to this study’s novelty. An exhaustive list of all possible AEs was impractical, so AEs were broadly defined by the study team as any unanticipated medical event reported by participants, or measured/observed by the study team, that would require contacting the study physician. Participants were monitored continuously during the study by nursing and research staff. No follow-up contact was made after the session. Vertigo and dizziness were prespecified foreseeable events for the driving-simulation task (results reported elsewhere). Had participants experienced vertigo during a simulation, the simulation would have been stopped.

Quantitative assays for kratom-product alkaloids, and for alkaloid metabolites in urine and blood.

MassLynx 4.2 was used for acquisition and processing of assay data. For kratom products, ten alkaloids were quantified simultaneously as previously described.²⁵ For urine and plasma samples, the major alkaloids were simultaneously quantified as previously described.²⁶ Both processes used a Waters Acquity Class-I ultra-performance liquid chromatography (UPLC) coupled with a Xevo TQ-S Micro triple quadrupole mass spectrometer (Milford, MA, USA). These methods were partially modified to include quantification of mitraciliatine and isopaynantheine, along with metabolites (see Supplement).

Outcomes reported elsewhere included simulated-driving performance and qualitative interviews.¹² This report includes some interview statements to contextualize main results.

Statistical Analyses

This study was the first of its kind and exploratory; we made no a priori hypotheses. For physiological, subjective, and psychomotor outcomes, we used R (v 4.2.2) to fit generalized linear mixed-effect (GLMER) models examining changes from baseline in vital signs, pupil size (mm), SOWS, ARCI subscales, DEQ scores, and psychomotor tasks. For practical reasons (scheduling the driving simulator), collection timepoints could not be uniform across participants. We binned time into 6 levels: Baseline before kratom dose (n=10), 0-40min post-dose (n=10, mean=21.7±13.8 min), 40-80min post-dose (n=10, mean=59.3±16.6 min), 80-120min post-dose (n=10, mean=110.2±7.9 min), 120-160min post-dose (n=10, mean=145.9±10.9 min), >160min post-dose (n=9-10, mean=186.7±15.0 min).

Response data were initially examined using unconditional growth models, estimating only change over time for each response variable, allowing levels of response variables to vary between participants. Change-from-baseline comparisons with timepoints after kratom dose (0-40min, 41-80min, etc.) were modeled as fixed effects, such that their coefficients reflect change from baseline to a given timepoint. Sex, age, and urine drug results were tested as fixed-effect covariates and retained if they improved fit (likelihood-ratio tests). For each variable found to change over time, the number of hours between most recent pre-session dose and session-day dose was examined as a fixed-effect covariate. Model overdispersion was tested by comparing Pearson residuals with χ² distributions of the same degrees of freedom. Intraclass Correlation Coefficients were used to estimate variance accounted for by random effects.

Associations of responses with kratom doses and alkaloid exposure

For outcomes whose slopes deviated from zero in GLMER models of change over time, we examined marginal associations with participants’ kratom dose weight (g/kg), product alkaloid content (mg/kg), change in urine alkaloid concentration pre- to post-session, and post-dose plasma alkaloid concentration. Alkaloids were examined as weight relative to body mass. Where applicable, we examined log10, log-sin, and root transformations. This was repeated for the hours between participants’ last kratom dose and their self-administered session dose. Given the small number of observations for examining marginal effects, response slopes were examined qualitatively.

Results

Participants were six men and four women, average age 41.2 years (SD 10.3; range 26-60; Table 1). Nine were non-Hispanic white; one was biracial. Participants had used kratom for a mean of 6.6 years (SD 3.8; range = 2-14), though some reported taking periodic tolerance breaks since having initiated use.¹² Mean session dose, reflective of participants’ typical dose, was 5.16 grams (median = 4.38; range =1.14-10.9). Seven participants reported using kratom 7 days/week; 2 reported using 5 days/week; 1 reported using 3 days/week. Typical doses/day ranged from 1-5 (Supplemental Table 1). Substances used in the 24 hours before and during the study session appear in Supplemental Table 4. Average time between most recent kratom dose and session dose was 16.5 +/− 4.6 hours (range, 10.3 - 25.8). Inclusion of this covariate in models examining responses over time did not indicate significant associations (p range = .10 - .70) and did not influence the estimates described below.

Table 1.

Participant demographic characteristics, substance use, and kratom-product dosing

Participant	Past-24-hour substance use1	Last kratom dose, pre- session	Last dose amount	Session dosing time	Session dose weight	Route of administration
1: White, 60-year-old female	kratom, duloxetine, fludrocortisone, trazodone, aspirin, lorazepam, clobetasol cream, loratadine, turmeric, vitamin B12, vitamin D, ibuprofen, caffeine	02:00 on session day	1 g	12:17	1.14 g	“toss-n-wash” (loose powder followed by water)
2: White, 26-year-old female	kratom, bupropion, escitalopram, atomoxetine, ibuprofen, naproxen, caffeine	23:00 day before session	5 g	12:01	4.03 g	powder mixed with sports drink
3: White 45-year-old male	kratom, multivitamin, caffeine	18:30 day before session	4 g	11:32	4.33 g	capsules
4: White 49-year-old male	kratom, lorazepam, dexlansoprazole, lamotrigine, multivitamin, cannabis, caffeine	19:30 day before session	2.5 g	11:22	3.07 g	powder mixed with warm water
5: White 35-year-old female	kratom, caffeine	17:30 day before session	5g	11:13	4.43 g	capsules
6: White 34-year-old male	kratom, amlodipine, cannabis, caffeine	16:00 day before session	3 g	11:37	3.22 g	powder mixed with water
7: White 52-year-old male	kratom, aspirin, multivitamin, caffeine	16:30 day before session	2 tbsp	11:00	10.9 g	powder mixed with orange juice
8: White/Asian 32-year-old male	kratom, cannabis, modafinil, caffeine	10:00 day before session	8 g	11:50	7.93 g	powder mixed with orange juice
9: White 41-year-old female	kratom, bupropion, clonidine, cannabidiol, multivitamin, acetoglutathione, colostrum, vitamins B and C, caffeine	00:30 on session day	5.5 g	11:13	5.82 g	powder mixed with water and 1 sachet of H2 fizzy drink
10: White 38-year-old male	multivitamin, fruit/vegetable supplement powder, kava, doxylamine	Not recorded	Unrecorded	11:40	6.57 g	powder mixed with liquid

Physiological measures

Mean baseline SBP was 125.4±11.5mmHg (Figure 1). We found a nonsignificant (b=5.3, p=.07) increase in SBP at 40-80 minutes post-dose, followed by a decrease, such that SBP fell significantly below baseline by 160 minutes post-dose (b=−8.0, p=.01). Mean baseline DBP was 76.0±7.8mmHg and did not significantly increase post-dose; it decreased below baseline at >160 minutes (b=−5.3, p=.02). Mean baseline SpO₂, RR, and core temperature values did not change significantly post-dose and were 97.7±1.7%, 16.4±2.0 breaths per minute, and 36.7±0.3 degrees Celsius.

Figure 1.

Figure panels display mean vital sign measurements (represented by thick blue dots and lines) during each time bin, and raw data points for individual participants are shown in transparent blues and golds. Measurement units are displayed on y -axes, and all vital sign measurements were collected during each time bin.

Mean baseline pupil diameters were 4.8±1.1 mm (right and 4.7±0.9 mm left). There was a significant decrease 40-80 minutes post-dose (b=−.70 and b=−.73, p<.001 for both); this remained lower than baseline through >160 minutes (b=[−.86,−.75] and b=[−.80,−.65], p<.001 for both). Participants with higher doses (g/kg) had greater left-pupil constriction >160 minutes post-dose (b=−11.22, p=.03) (see Supplement).

Subjective measures

Mean baseline SOWS score was 5.5±3.3; the highest score was 10 (Figure 2). In people using opioids, those scores indicate mild withdrawal. SOWS scores were significantly lower at all timepoints post-dose (b=[−4.0,−2.9], p<.001), lowest at 80-120 minutes (b=−4.0, p<.001), but remaining >0 in most participants.

Figure 2.

Figure panels display mean scores on self-report instruments, which are represented by thick blue dots and lines, and raw data points for individual participants are shown in transparent blues and golds. The number of data points collected varied by instrument, which is reflected in the x-axes of each. SOWS clinical severity cut-offs are indicated by yellow, orange, and red lines.

Note: SOWS = Subjective Opioid Withdrawal Scale; ARCI = Addiction Research Center Inventory; MBG = Morphine-Benzedrine Group; LSD = Lysergic Acid Diethylamide; DEQ = Drug Effects Questionnaire.

For the DEQ, in response to the question “Do you feel any drug effects?” mean VAS scores were 40±30.5 at 40 minutes post-dose. Felt effects became higher at 40-80 minutes (b=32.7, p<.001) and 80-120 minutes (b=19.0, p=.04); peak was 72.7±11.1. Felt effects at subsequent points were not significantly different from the 40-minute point. For the DEQ question “Do you like the effects you are feeling now?” mean VAS scores were 63.4±13.0 at 40 minutes post-dose. Liking became higher at 40-80 minutes post-dose (b=14.4, p=.02). Liking at subsequent points was not significantly different from the 40-minute point. For the DEQ question “Would you like more of what you took, right now?” mean VAS score was 19.3±30.0 at 40 minutes post-dose. At 120-160 minutes, scores displayed nonsignificant increases (b=13.7, p=.06); by >160 minutes, they had significantly increased (b=26.6, p<.001). For the DEQ question “Do you feel high?” mean VAS score was 15.2±23.6 at 40 minutes post-dose. This remained constant at the group level, although one participant gave higher ratings over time. For the question “Do you feel intoxicated?” mean VAS score was 4.8±9.8 at 40 minutes post-dose. This remained constant at all time points for all participants.

For the ARCI subscales (Figure 2), mean baseline MBG (euphoria) score was 9.1±4.0. This remained constant, except for a nonsignificant increase at 40-80 minutes (b=1.9, p=.07), driven largely by three participants. Mean baseline ARCI Amphetamine (stimulation) baseline score was 5.20±0.77. This increased at 40-80min post-dose (b=1.80, p=.02). Mean baseline ARCI LSD score (dysphoria, somatic symptoms) was 4.7±2.4, remaining constant. Mean baseline ARCI PCAG score (sedation) was 5.3±2.45, remaining constant (see Supplement).

Psychomotor measures

For baseline FCRTT (reaction time), participants made an average of 95.9±2.7 correct responses out of 100 (Figure 3). This high score increased significantly 40-80 minutes post-dose (b=2.0, p=.03) but was not significantly different from baseline at other points. To assess changes, we controlled for variability within trials; mean reaction time was significantly higher than baseline 80-120 minutes post-dose (b=23.1, p=.01) but not at any other point. On the NVT baseline, participants made an average of 98.5±1.8 correct responses out of 100. This remained constant post-dose, as did numbers of target hits (baseline: 24.3±1.1 out of 25) and false alarms (baseline: 0.7±0.7 out of 75 target-absent trials).

Figure 3.

Figure panels display mean psychomotor task performance indices (represented by thick blue dots and lines) during each time bin, and raw data points for individual participants are shown in transparent blues and golds. Psychomotor tasks were administered during four of the six time bins.

Note: FCRTT = Forced-choice Reaction Time Task; NVT = Number Vigilance Task

Acute Adverse Events

No AEs were reported by participants or observed by staff, including events that have occurred in other studies where participants ingested kratom material while fasting (e.g., nausea).¹⁶

Quantification of alkaloids in kratom products, urine, and plasma

Alkaloid concentrations in the products and biosamples are in Table 2. Compound parameters for the alkaloids and internal standards (for the kratom product, plasma, and urine samples) are in Supplemental Tables 2 and 3.

Table 2.

Alkaloid concentrations in kratom products and participant plasma and urine samples.

	Mean (SD)	Median (IQR)	Range
Session Dose (g/kg)	0.061 (0.037)	0.048 (0.044)	0.015 - 0.129
Alkaloids in Session Day Product Dose (mg/kg)
mitragynine	0.77 (0.535)	0.594 (0.477)	0.190 - 1.698
7-HMG	0.006 (0.005)	0.003 (0.004)	0.001 - 0.017
corynantheidine	0.015 (0.012)	0.011 (0.008)	0.003 - 0.037
speciogynine	0.097 (0.071)	0.079 (0.023)	0.021 - 0.235
speciociliatine	0.154 (0.115)	0.114 (0.081)	0.034 - 0.369
mitraciliatine	0.028 (0.023)	0.021 (0.006)	0.006 - 0.071
paynantheine	0.109 (0.077)	0.09 (0.037)	0.023 - 0.268
isopaynatheine	0.018 (0.014)	0.013 (0.007)	0.004 - 0.045
Alkaloids in Plasma (ng/ml)
mitragynine	175.573 (114.822)	174.507 (154.042)	36.575 - 396.494
7-HMG	34.021 (25.778)	27.416 (31.191)	2.944 - 77.655
corynantheidine	3.867 (3.455)	3.536 (3.038)	0.690 - 12.300
speciogynine	28.671 (14.638)	31.183 (22.538)	6.230 - 49.509
speciociliatine	187.586 (93.658)	176.202 (159.267)	75.141 - 327.703
mitraciliatine	88.964 (38.836)	89.763 (63.32)	36.247 - 145.423
paynantheine	33.807 (24.325)	31.386 (24.555)	7.505 - 91.339
isopaynatheine	44.105 (21.075)	46.009 (36.977)	14.668 - 71.780
Alkaloids in Urine (ng/ml) – Pre-dose
mitragynine	12069.23 (17617.432)	6756.433 (8283.38)	770.940 - 56912.895
7-HMG	25467.47 (31761.077)	13332.755 (12793.606)	2587.220 - 94640.910
corynantheidine	117.349 (140.298)	57.56 (83.588)	9.420 - 406.380
speciogynine	5871.224 (7571.886)	3139.948 (4589.875)	331.840 - 20049.120
speciociliatine	41535.978 (37817.962)	25711.342 (40689.838)	11267.260 - 107559.440
mitraciliatine	18331.603 (21742.93)	11385.895 (11480.424)	3044.220 - 68520.335
paynantheine	2073.623 (2917.661)	713.117 (1757.416)	155.220 - 8192.000
isopaynatheine	8530.667 (11371.026)	3702.892 (5974.604)	1000.080 - 32535.555
Alkaloids in Urine (ng/ml) – post-dose
mitragynine	10675.759 (9537.519)	9205.885 (8593.404)	1052.100 - 27497.890
7-HMG	25306.96 (25943.424)	16940.775 (21785.22)	2418.240 - 88063.470
corynantheidine	138.212 (103.795)	126.145 (115.58)	24.600 - 312.345
speciogynine	3881.323 (2853.086)	4433.9 (4231.966)	416.775 - 8756.325
speciociliatine	27939.445 (20562.91)	30447.067 (29257.727)	6216.525 - 70766.055
mitraciliatine	8632.584 (6959.564)	7972.275 (7816.871)	1095.975 - 21289.940
paynantheine	1693.86 (1075.724)	2060.72 (1757.384)	211.620 - 3209.360
isopaynatheine	4055.032 (3437.492)	3519.745 (4565.494)	443.475 - 10127.320

There were no statistically significant associations of changes in physiological, subjective, and psychomotor outcomes with kratom dose (g/kg), product alkaloid content (mg/kg), change in urine alkaloids pre- to post-session, or plasma alkaloids post-dose. However, slopes differing from zero were qualitatively observed for changes in SBP and pupil diameters (Supplemental Figure 1): greater product alkaloid content and plasma alkaloid content were associated with smaller decreases in SBP at >160 minutes and greater decreases in pupil diameter 40-80 minutes post-dose.

Discussion

This was the first laboratory investigation of the acute effects from a typical morning dose of kratom among US adults who use kratom regularly, a group that may number in the millions.²⁷ Given kratom’s constellation of acute effects, and given that our participants used regularly, several outcomes were possible. The morning dose might have been energizing/focusing, or it might have relieved overnight withdrawal so participants felt essentially normal after taking it, or it might have been impairing/intoxicating—or it might have had some combination of those effects. We found more evidence for the first two effect types (energizing or normalizing) than for impairment or intoxication, with several findings meriting comment.

Physiological measures reflected acute effects of the dose, including decreases in pupil diameter that presumably reflect MOR activation. However, these changes in physiological measures were usually not statistically significant and were never clinically significant (Figure 1).

Before dosing, all participants had SOWS scores that would reflect mild withdrawal in people who use opioids (1-10, where >10/64 is the cutoff between mild and moderate). Kratom dosing decreased SOWS scores, although only three participants reached 0. Nonzero scores may partly reflect opioidlike withdrawal (none of our participants reported current opioid use, as verified by drug screen), but may also reflect drug-nonspecific symptoms, especially when items endorsed were anxiety, sweating, yawning, and restlessness (all mild or moderate). The item “felt like using” was usually endorsed only at baseline and at the last point in the session. Among those who use kratom daily, it may be that continued daily use is partly driven by withdrawal avoidance, however, the persistence of low SOWS scores >0 after use does not necessarily indicate that withdrawal symptoms detected here went unrelieved or that they can directly be attributable to kratom use. We suggest this, in part, because some SOWS items (e.g., goosebumps; anxiety) are nonspecific for opioid withdrawal. We have argued elsewhere that, because the kratom’s pharmacology is not merely opioidergic, future kratom withdrawal assessment should include additional scales.⁹ We chose to administer the SOWS, in part, because some of kratom’s major alkaloids are MOR agonists. However, given kratom’s activity on multiple other systems, the SOWS may not be sufficient to assess kratom withdrawal. For example, caffeine-withdrawal scales might be useful. Ultimately, supervised withdrawal studies are needed to determine the extent to which kratom withdrawal resembles classic opioid withdrawal.

Subjective ratings on the ARCI and the DEQ may speak to the question of kratom as a mild euphoriant. The ARCI MBG scale seems well suited to test this because it consists of responses elicited by both opioids and psychostimulants, with euphoria being their main shared effect.²² We found only a small group-level increase in MBG scores post-dose, and the increase was driven by three of our ten participants, one of whom also showed increases in his ratings of “feel high” on the DEQ (without accompanying increases in “intoxicated” or “want more.”) MBG items endorsed by those three participants, post-dose, included “I would be happy all the time if I felt as I feel now” and “I feel as if something pleasant had just happened to me.”

We can contextualize these findings in two ways. First, prior research has shown small increases in ARCI MBG responses elicited by caffeine (300mg orally) in four out of ten daily coffee drinkers,²⁸ a finding not unlike ours. Second, we have insight into our participants’ responses because, at the end of the session, we interviewed them about their daily kratom use.¹² The participant with post-dose increases in DEQ “high” and MBG scores stated: “It gives me, like, a euphoria, that feels, like, typical of an opiate….When I use it, it more enhances my, like, functioning, rather than impair it….cognitively I can focus better…I can get more into whatever it is that I’m doing, and then, you know, the highness and euphoria also helps drive that feeling to keep doing whatever it is that I’m into.” The other two participants compared their euphoria not to that of an opioid, but that of a cup of coffee or bar of chocolate; one described enhancement of academic productivity.¹² During the interviews, none described impairment from kratom at their typical doses, nor did they report intoxication on the DEQ. Direct staff observations during sessions found no signs of impairment (e.g., slurred or illogical speech) and the psychomotor tasks did not reliably indicate declines in performance after kratom. On both tasks, response accuracy was high pre- and post-dosing; only at 80-120 minutes post-dosing was there a modest increase in reaction time. Evidence of mild euphoria, without slow or inaccurate psychomotor performance, merits follow-up using more sensitive tasks.

Alkaloidal contents of the products were similar to those previously found in leaf-based kratom material available in native settings and other commercial products,^29-31 with no elevated concentrations to suggest (for example) “spiking” with 7-hydroxymitragynine.³² As many kratom products do not provide recommended serving sizes or list levels of mitragynine, we cannot compare detected concentrations to those reported on labels. Labels of products (some of which were purchased by participants in bulk with minimal labeling) were not collected; only the product themselves were collected and analyzed.

Limitations

By design, our findings generalize only to a single morning dose in regular consumers. Our sample size of 10 participants was not powered to detect dos-response effects for specific alkaloids; the few such effects that seemed qualitatively observable (Figure S1-S4) will require replication in larger studies. We did not directly compare kratom with other substances, or use scales for non-opioidergic withdrawal. We enrolled only people who answered our initial advertisements, completed the remote study from which we re-recruited, and resided within one geographic region. Although the kratom consumed was assayed, we did not test post-exposure urine for the presence of other substances because the kratom consumed during the session was the same product batch that participants had been consuming prior to their drug screen; self-reported use aligned with urine test results, with no participant testing positive for opioids or stimulants; and intoxication was not evinced during the session. Finally, our findings cannot speak to long-term AEs from chronic use. Indeed, one participant planned to discontinue kratom use after the study, feeling that the benefits were not proportionate to perceived unknowns (e.g., risk of liver injury) or side effects (e.g., tolerance, decreased libido).¹² Others described minor side effects (e.g., constipation, irritability) or physical dependence.¹² Even with respect to acute AEs, it is important to remember that participants were under direct observation for only four hours. The acute and chronic effects of kratom need to be investigated in controlled studies with larger samples. Studies examining the effects of well-characterized kratom formulations in healthy populations are desperately needed.

Conclusions

In a direct-observation laboratory assessment of acute kratom effects in ten adults who regularly use kratom, we found that a low-to-moderate single dose of whole-leaf kratom products did not produce consistent improvement or impairment in psychomotor tasks. However, there were acute physiological and subjective effects that included mild euphoria, at least for three participants. As most participants reported tolerance to kratom,¹² the detection of acute stimulatory effects without intoxication is noteworthy; it suggests that kratom remains acutely activating at low-to-moderate doses after tolerance develops. It also suggests that consumers may titrate their doses to produce desired effects without impairment.¹² Motivations to use kratom often center around productivity more than recreation.^10,12 These direct-observation data support such accounts, and our novel methods provide a template for conducting future studies.

Data sharing and availability:

Requests for data sharing may be made to Drs. David H. Epstein and Kirsten E. Smith.