Demographic and Behavioral Factors Associated with Kratom Use among U.S. College Students

Abstract

Kratom use represents a growing risk for public health. The present study examined demographic and behavioral factors linked with kratom use among a sample of college students in the United States. Kratom use was linked with being White, male or transgender/gender nonconforming, identifying as a sexual minority, use of alcohol or marijuana, and depressive symptoms. Kratom use was not uniquely linked to exercise or anxiety. The results of the present study can be used to inform initial targeting of efforts to reduce kratom use among college students.

Kratom (Mitragyna speciosa) is a tree native to Southeast Asia. Kratom has emerged as a substance of potential misuse and public health concern in the United States (U.S.).^1,2 The United States National Poison Data System indicated a 6-fold increase in kratom-related incidents from 2016 to 2017.^3,4 Given the relatively nascent spread of use of kratom, limited data are available on its lethality. One study of 27 deaths in Nevada in which mitragynine (the active compound in kratom) was listed as the cause of death indicated concentrations of kratom ranging from 8.7 to 1800 ng/mL. But, in the same study, a sample of 8 individuals who had taken consumed mitragynine but died of another cause, concentrations were also widely ranging (110 to 980 ng/mL).⁵ Animal studies have also been unclear with regard to lethal dose levels.⁶ Thus, it is unclear what dose of kratom is potentially dangerous, and lethality may depend on many factors including comorbid conditions and genetic risk factors.⁵ Given the increasing use, unknown risk factors, and consequences of use it is essential to discover demographic and behavioral corelates of kratom use to better define strategies to reduce risk for misuse.

Kratom is used as traditional medicine in Thailand and Malaysia to reduce fever, relieve pain, and increase work output.^7,8 Traditionally, kratom leaves are chewed or brewed and produce a simulant effect. Kratom has been used as an analgesic and an opioid substitute during opioid withdrawal in the U.S.,^2,7,9–11 but has not been found to have benefits over other treatments.¹² Kratom is not scheduled under the U.S. Controlled Substance Act and is legal in most of the U.S. (with the exception of Alabama, Arkansas, Indiana, Wisconsin, Rhode Island, Vermont, and DC).^11,13 Kratom can be commonly bought in the U.S. through online retailers and smoke shops,⁸ businesses that tend to geographically cluster around college campuses.¹⁴ Kratom users cite motivations for use including improving mood, increasing alertness, and managing pain,^7,15–17 suggesting that mental health concerns may motivate use. Some kratom users also indicate use of kratom to support fitness goals.^18,19

Case studies of individuals who presented for emergency care after kratom use and reports from people who used kratom indicate that effects can include, acute kidney injury, respiratory distress, loss of consciousness, coma, and death.^3,4,20–23 Withdrawal from kratom is more mild than opioid withdrawal.¹⁷

This rise of kratom use in the United States is potentially dangerous, particularly to groups that may have increased risk for use. Some work has investigated demographic correlates of kratom use, finding that kratom use in the U.S. is more prevalent among men, White individuals, adults between the ages of 18–50, and marijuana users.^24–27 However, several other possible correlates have not been examined as they pertain to kratom use. Sexual and gender minority groups are at higher risk for substance use and misuse than heterosexual, cisgender populations, as a result of systemic stressors facing those groups.^28–30 Further, links between kratom use and mental health variables such as depression and anxiety have not been explored in large national data sets despite indication that a potential motivation for kratom use mood enhancement, management of pain, or other desired psychological effects. Given links between marijuana use and kratom use, it is possible that use of alcohol also represents a risk factor for kratom use. Additionally, despite the marketing of kratom as a workout-enhancer, links between exercise and kratom use have not been explored in national data sets. Further work to identify risk factors for kratom use is important to help prevent further uptake of kratom in the U.S. population. The goal of this study was to explore kratom use using a national sample of individuals at risk for increased substance experimentation, use, and misuse: college students.³¹

Consistent with past research on kratom, we expected kratom use to be more common among White individuals, men, and those who reported recent marijuana use. Also consistent with past research suggesting links to kratom use, we expected use to be more common among gender and sexual minority individuals, those who reported higher depression or anxiety symptoms, those who report recent alcohol use, and those who exercise regularly.

Method

Participants and Procedures

Participants were U.S. college students who participated in the 2019–2020 Healthy Minds Study (HMS). Data for the study are available at https://healthymindsnetwork.org/hms/. The HMS is an online survey conducted annually using web-based sampling of graduate and undergraduate students from universities that participate in the data collection. The HMS is approved by the University of Michigan Health Sciences and Behavioral Sciences Institutional Review Board, and the HMS is also issued a certificate of confidentiality from the National Institutes of Health. Because students can skip items, a subsample of 81,513 students (out of the total 89,888 participants that completed the HMS) was analyzed using data from participants in the HMS who completed all the assessments that are included in the present analysis. Demographics for the analytic sample are presented in Table 1.

Table 1.

Demographics and variable data.

Count variables	# (%)
Kratom use	310 (0.4%)
Gender
Male	24588 (30.2%)
Female	55595 (68.2%)
Transgender/NB	1330 (1.6%)
Sexual orientation
Heterosexual	66927 (82.1%)
Sexual minority	14586 (17.9%)
Race/ethnicity
White	52126 (63.9%)
Black/African American	4803 (5.9%)
Asian/Asian American	8579 (10.5%)
Hispanic	6205 (7.6%)
Pacific Islander	105 (0.1%)
Middle Eastern	1164 (1.4%)
Other	852 (1.0%)
Multi	7679 (9.4%)
Past 2-week Alcohol Use	44628 (54.7%)
Past 30-days Marijuana Use	15917 (19.5%)
Interval variables	M (SD)
Age	23.19 (6.62)
Exercise	2.33 (1.15)
Depression	8.45 (6.17)
Anxiety	7.33 (5.68)

Measures

Kratom use was assessed with the item, “Over the past 30 days, have you used any of the following drugs,” for which one response option was kratom. Use was coded dichotomously (0 = no use in past 30 days, 1 = use in past 30 days).

Race/ethnicity was assessed using the item, “What is your race/ethnicity?” Participants could select more than one option; any participants who selected more than one option were coded into a new multiracial category. Because of very low counts for the group American Indian or Alaskan Native, participants who selected only membership in this group were merged into the self-identify group. White was used as the reference category, given prior work suggesting that White people are more likely to use kratom.

Gender was assessed using an item about gender identity that included the options male [sic], female [sic], trans male/trans man, trans female/trans woman, genderqueer/gender non-conforming, and an option for self-identification; participants could select more than one option. For the purpose of this study, participants were coded as cisgender men and cisgender women (if only the male or female option were selected, respectively), or transgender/gender nonconforming (TGNC) if they selected any other option. If participants selected both one of the TGNC options as well as “male” or “female” they were coded as TGNC. Cisgender men were selected as the reference category.

Marijuana use was assessed using the item, “Over the past 30 days, have you used any of the following drugs?” if the response option for marijuana was selected. Endorsement of marijuana use was coded dichotomously (0 = no, 1 = yes).

Sexual orientation was assessed using the item “How would you describe your sexual orientation?” with the response options heterosexual, lesbian, gay, bisexual, queer, questioning, and an option for self-identification. Participants were coded as heterosexual if they selected only heterosexual, and as sexual minority if they selected any single other option or any combination of options. Heterosexual was used as the reference category.

Depression was assessed using the Patient Health Questionnaire-9 (PHQ-9), a well-established measure of depressive symptoms.³² Anxiety was assessed using the General Anxiety Disorder-7 (GAD-7), a well-established measure of anxiety symptoms.³³ For both the PHQ-9 and GAD-7 the continuous total scores were used consistent with other work that uses these measures in non-clinical samples.³⁴

Alcohol use was assessed using the item, “Over the past 2 weeks, did you drink any alcohol?” Responses were dichotomous (0 = no, 1 = yes).

Exercise was assessed using the item, “In the past 30 days, about how many hours per week on average did you spend exercising? (Include any exercise of moderate or higher intensity, where “moderate intensity” would be roughly equivalent to brisk walking or bicycling);” response options were 1 = less than 1 hour, 2 = 2–3 hours, 3 = 3–4 hours, and 4 = 5 or more hours.

Age was assessed using a single item asking participants their age.

Data Analysis

The data were analyzed in SPSS v27³⁵ using a logistic regression model to account for the binary nature of the dependent variable, kratom use. All other variables were entered as independent variables, except for age, which was entered as a covariate.

Results

The overall model was significant, χ² (16) = 519.102, p < .001, Nagelkerke R² = 0.13. The Homer and Lemeshow test was not significant, χ² (8) = 7.601, p = 0.47, indicating that the model was not a poor fit to the data. Results of the logistic regression are presented in Table 2.

Table 2.

Logistic regression of factors linked with kratom use.

Variable	B	SE	Wald (df)	p	OR	95% CI LB	95% CI UB
Constant	−7.338	0.312	553.125 (1)	< .001
Gender1			82.203 (2)	< .001
Women	−1.111	0.123	82.203 (1)	< .001	0.329	0.259	0.419
Trans/NB	0.022	0.263	.007 (1)	0.933	1.022	0.611	1.711
Age	0.033	0.008	164.107 (1)	< .001	1.033	1.017	1.050
Sexual minority	0.291	0.100	8.458 (1)	0.004	1.338	1.100	1.627
Race/ethnicity2			22.961 (7)	0.002
Black/African American	−1.082	0.415	6.786 (1)	0.009	0.339	0.150	0.765
Asian/Asian American	−0.843	0.311	7.336 (1)	0.007	0.430	0.234	0.792
Hispanic	−0.891	0.310	8.262 (1)	0.004	0.410	0.224	0.753
Pacific Islander	0.878	1.020	0.742 (1)	0.389	2.407	0.326	17.763
Middle Eastern	−0.320	0.585	0.299 (1)	0.585	0.726	0.231	2.287
Other	−0.516	0.586	0.777 (1)	0.378	0.597	0.189	1.881
Multi	−0.245	0.193	1.615 (1)	0.204	0.783	0.537	1.142
Exercise	−0.008	0.051	0.027 (1)	0.870	0.992	0.897	1.097
Alcohol	0.600	0.152	15.574 (1)	< .001	1.821	1.352	2.453
Marijuana	1.682	0.131	164.107 (1)	< .001	5.376	4.156	6.953
Depression	0.053	0.013	18.073 (1)	< .001	1.055	1.029	1.081
Anxiety	0.009	0.014	.418 (1)	0.518	1.009	0.981	1.038

Note. CI = confidence interval, LB = lower bound, UB = upper bound.

¹Reference group = men.

²Reference group = White.

Kratom use was reported by 310 of 81,513 participants included in the analyses. Consistent with past research, being Black/African American, Asian, or Hispanic was related to lower kratom use as compared to White participants; all other race/ethnicity groups were not different from White individuals. Gender was linked to kratom use in that women were less likely to use kratom (OR = 0.33, 95% CI = .26, .42) than men; transgender and non-binary participants were not different from men in kratom use. Sexual orientation was linked with kratom use, such that sexual minority identity was linked with increased kratom use (OR = 1.34, 95% CI = 1.10, 1.63).

Depression, as measured by the PHQ-9, was linked with kratom use (OR = 1.06, 95% CI = 1.02, 1.05). Past-2-week alcohol use was linked to more risk for kratom use (OR = 1.82, 95% CI = 1.35, 2.45). Marijuana use was also linked with more risk for kratom use (OR = 5.38, 95% CI = 4.16, 6.95). Exercise and anxiety were not significantly linked to kratom use.

Discussion

Kratom represents an emerging substance with potential for misuse among college students. The aim of the present study was to better understand factors associated with kratom use among a national sample of U.S. college students. Kratom use was linked to several demographic correlates. Consistent with prior research, kratom use was more common among White students relative to Black/African American, Asian, and Hispanic students, men relative to women, and those who reported recent alcohol and marijuana use. TNGC identity was linked to kratom use levels equivalent to cisgender men. Sexual minority identity was linked with greater kratom use, similar to other work that links sexual minority identity to increased risk for substance use due to systematic social marginalization.³⁰ Contrary to our expectations, kratom use was not linked with exercise behaviors despite kratom being openly marketed as an exercise supplement.^19,36 It is possible that kratom remains relatively less well-known as a workout supplement. Use of kratom in the context of sports supplements should continue to be monitored. Motivations to use kratom for mood enhancement or for other desired psychological effects should also be explored; depression, but not anxiety, was linked with kratom use in the present study.

The present study adds to the growing literature on kratom use by examining demographic and behavioral correlates of kratom use. The present findings support prior literature indicating that kratom use is more common among White individuals, men, sexual minorities, those with more depression symptoms, those who used alcohol, and those who recently used marijuana. Marijuana use was a particularly substantial risk factor for kratom use; participants who used marijuana within the past 30 days were more than 5 times more likely to also use kratom. Given the links between alcohol and marijuana use, both of which are elevated among college students,³⁷ and kratom use, further examination of kratom use among college students is important. To better inform interventions, more research is needed on kratom so that accurate information can be presented on perceived benefits and risks of its use. Further work to identify risk factors for kratom use will be useful to understand patterns of use and potential for misuse in the U.S. given that there is much to be learned about the effects of kratom. Public health would be enhanced by precautionary efforts to reduce kratom use, particularly among those at risk for other health concerns. These results can help to inform interventions to increase awareness of the potential effects of kratom and reduce use of kratom on campuses. Given the results, intervention materials could be distributed among men’s groups such as fraternities, spaces for sexual and gender minorities, and college-adjacent bars, and integrated into existing interventions for students managing depression. Crucially, such interventions should be informed by further research on students’ knowledge of, attitudes toward, and behaviors related to kratom use.

The present study should be interpreted in light of its limitations. This study used data from a national sample of college students and thus does not represent the U.S. population as a whole. The present study used cross-sectional data and thus we cannot infer causality or temporal ordering of kratom use and many related variables, such as depression or alcohol use. Also, kratom may have been used by some students outside of the 30-day time-frame of our variable measurement or in a different month in which this data was collected. Future studies should consider seasonal aspects of kratom use and how similar factors may impact measurement of substance misuse (e.g., use during the academic year versus summer break). Finally, as we used a national data set, our analyses were limited to variables assessed in the Health Minds Study. Many other factors may potentially be related to kratom use and can help to inform interventions that were not included in the Healthy Minds Study.