ArticlePDF Available

The relationships of classic psychedelic use with criminal behavior in the United States adult population

October 2017
Journal of Psychopharmacology 32(1):026988111773568

October 2017
32(1):026988111773568

DOI:10.1177/0269881117735685

Authors:

Michael Scott Crawford

University of Alabama at Birmingham

Karen L Cropsey

University of Alabama at Birmingham

Heith Copes

University of Alabama at Birmingham

Show all 7 authorsHide

Criminal behavior exacts a large toll on society and is resistant to intervention. Some evidence suggests classic psychedelics may inhibit criminal behavior, but the extent of these effects has not been comprehensively explored. In this study, we tested the relationships of classic psychedelic use and psilocybin use per se with criminal behavior among over 480,000 United States adult respondents pooled from the last 13 available years of the National Survey on Drug Use and Health (2002 through 2014) while controlling for numerous covariates. Lifetime classic psychedelic use was associated with a reduced odds of past year larceny/theft (aOR = 0.73 (0.65–0.83)), past year assault (aOR = 0.88 (0.80–0.97)), past year arrest for a property crime (aOR = 0.78 (0.65–0.95)), and past year arrest for a violent crime (aOR = 0.82 (0.70–0.97)). In contrast, lifetime illicit use of other drugs was, by and large, associated with an increased odds of these outcomes. Lifetime classic psychedelic use, like lifetime illicit use of almost all other substances, was associated with an increased odds of past year drug distribution. Results were consistent with a protective effect of psilocybin for antisocial criminal behavior. These findings contribute to a compelling rationale for the initiation of clinical research with classic psychedelics, including psilocybin, in forensic settings.

. Characteristics of Psilocybin Only, Psilocybin & Other Psychedelics, Non-Psilocybin Psychedelics Only, and No Psychedelics groups.

…

. Results from generalized ordered logistic regression models predicting past year drug distribution, past year larceny/theft, and past year assault.

…

Figures - uploaded by Heith Copes

Content may be subject to copyright.

Content uploaded by Heith Copes

Content may be subject to copyright.

https://doi.org/10.1177/0269881117735685

Journal of Psychopharmacology

1 –12

Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0269881117735685

journals.sagepub.com/home/jop

Introduction

The United States places a greater proportion of its citizens under

criminal justice supervision—2.8% of all adults—than any other

country (Kaeble et al., 2016). Drug-related crimes (possession,

use, distribution, or manufacturing of illicit drugs) and property

crimes (burglary, larceny/theft, motor vehicle theft, or arson) are

the most common criminal offenses, comprising approximately

14% and 13.5% of all arrests, respectively (FBI Uniform Crime

Report, 2015). Though rates of violent crime (murder, man-

slaughter, rape/sexual assault, assault, or robbery) are at historic

lows and violent crime accounts for only 5% of all arrests (FBI

Uniform Crime Report, 2015; Truman and Langton, 2015), the

costs of violent crime remain high. Indeed, nearly all survivors of

violent crime experience mental health problems, with a majority

suffering from relationship difficulties or a decline in job/school

performance and about one-third evidencing severe distress

(Langton and Truman, 2014). Of all crimes, murder poses the

greatest financial cost to society, estimated at nearly $9,000,000

per murder (McCollister et al., 2010).

Recidivism rates are alarmingly high following release from

prison, with 67.8% and 76.6% of released offenders re-arrested

within three and five years for a new crime, respectively. Property

offenders have the highest five-year re-arrest rates (82.1%), fol-

lowed by drug offenders (76.9%), and violent offenders (71.3%).

Importantly, 33.1% of violent offenders go on to be re-arrested

for another violent crime (Durose et al., 2014). Several interven-

tions have been developed to decrease recidivism and include

employment programs, cognitive-behavioral approaches, Moral

Reconation Therapy, and specialty programs targeting juvenile

offenders. Overall, interventions that employ cognitive-behavio-

ral approaches and Moral Reconation Therapy demonstrate small

effects (Ferguson and Wormith, 2013; Pearson et al., 2002),

whereas those using employment strategies or targeting juveniles

demonstrate no effect (Schwalbe et al., 2012, Visher et al., 2005).

Notably, individuals who have committed a violent crime are

among the least likely to complete treatment and the most likely

to re-offend after treatment dropout (Oliver et al., 2011).

The development of innovative and effective interventions to

prevent criminal behavior is thus an obvious priority. Such inter-

ventions may involve classic psychedelics, non-selective

The relationships of classic psychedelic use

with criminal behavior in the United States

adult population

Peter S Hendricks1, Michael Scott Crawford1, Karen L Cropsey2,

Heith Copes3, N Wiles Sweat1, Zach Walsh4 and Gregory Pavela1

Abstract

Criminal behavior exacts a large toll on society and is resistant to intervention. Some evidence suggests classic psychedelics may inhibit criminal

behavior, but the extent of these effects has not been comprehensively explored. In this study, we tested the relationships of classic psychedelic

use and psilocybin use per se with criminal behavior among over 480,000 United States adult respondents pooled from the last 13 available years

of the National Survey on Drug Use and Health (2002 through 2014) while controlling for numerous covariates. Lifetime classic psychedelic use was

associated with a reduced odds of past year larceny/theft (aOR = 0.73 (0.65–0.83)), past year assault (aOR = 0.88 (0.80–0.97)), past year arrest for

a property crime (aOR = 0.78 (0.65–0.95)), and past year arrest for a violent crime (aOR = 0.82 (0.70–0.97)). In contrast, lifetime illicit use of other

drugs was, by and large, associated with an increased odds of these outcomes. Lifetime classic psychedelic use, like lifetime illicit use of almost all

other substances, was associated with an increased odds of past year drug distribution. Results were consistent with a protective effect of psilocybin

for antisocial criminal behavior. These findings contribute to a compelling rationale for the initiation of clinical research with classic psychedelics,

including psilocybin, in forensic settings.

Keywords

Hallucinogen, psilocybin, lysergic acid diethylamide, psychedelic, recidivism, criminal justice, forensic psychology

1 Department of Health Behavior, School of Public Health, University of

Alabama at Birmingham, Birmingham, USA

2 Department of Psychiatry and Behavioral Neurobiology, University of

Alabama at Birmingham, Birmingham, USA

3 Department of Criminal Justice, University of Alabama at Birmingham,

Birmingham, USA

4 Department of Psychology, University of British Columbia, Kelowna,

Canada

Corresponding author:

Peter S. Hendricks, Department of Health Behavior, School of Public

Health, University of Alabama at Birmingham, 227L Ryals Public Health

Building, 1665 University Blvd., Birmingham, AL 35294, USA.

Email: phendricks@uab.edu

735685JOP0010.1177/0269881117735685Journal of PsychopharmacologyHendricks et al.

research-article2017

Original Paper

2 Journal of Psychopharmacology 00(0)

5HT2AR agonists including lysergic acid diethylamide (LSD),

mescaline, and psilocybin that can occasion primary mystical

experiences (also known as primary religious experiences or

peak experiences) and that have been used for millennia across

cultures with therapeutic intention (Nichols, 2016). Indeed, dur-

ing the first wave of human classic psychedelic research from the

1950s through the early 1970s, at least three studies tested the

effect of classic psychedelic-assisted psychotherapy on criminal

offenders. Tenenbaum (1961) provided 10 treatment-resistant sex

offenders with multiple LSD-assisted group psychotherapy ses-

sions and noted demonstrable increases in empathy, insight, com-

munication, and treatment engagement in all but one participant.

Arendsen-Hein (1963) administered multiple LSD-assisted psy-

chotherapy sessions to 21 “criminal psychopaths” and reported

similar outcomes among 14 participants. In the largest and most

widely known study of classic psychedelics and recidivism,

Leary (1969) provided 32 prisoners with psilocybin-assisted psy-

chotherapy in his Concord Prison Experiment, reporting positive

findings. A later review, however, concluded that these findings

were overstated, inadequate support was provided outside of

psilocybin sessions, and that serious methodological flaws pre-

clude any conclusions (Doblin, 1998). Despite these mixed find-

ings, legal proscriptions and a dearth of funding ruled out further

study with classic psychedelics.

Classic psychedelic research has experienced a modest renais-

sance over the past two and a half decades, with recent investiga-

tions demonstrating that classic psychedelics and psilocybin in

particular may confer a number of long-lasting psychological

benefits. For instance, in two separate randomized clinical trials,

psilocybin produced substantial and sustained decreases in anxi-

ety and depression among patients with life-threatening cancer

(Griffiths et al., 2016; Ross et al., 2016). Similarly, in a single-

arm, open-label feasibility study of treatment-resistant depres-

sion, psilocybin produced large and enduring improvements in

mood (Carhart-Harris et al., 2016). These findings comport with

population-based studies indicating that having ever used a clas-

sic psychedelic (Hendricks et al., 2015a) and having ever used

psilocybin per se (Hendricks et al., 2015b) are both associated

with a decreased likelihood of psychological distress and suici-

dality. In addition, single-arm, open-label feasibility studies sug-

gest psilocybin promotes abstinence from alcohol (Bogenschutz

et al., 2015) and tobacco (Johnson et al., 2014; 2017), and a pop-

ulation-based study indicates that having ever used a classic psy-

chedelic is associated with a decreased risk of opioid abuse and

dependence (Pisano et al., 2017).

Considering that many individuals in the criminal justice sys-

tem suffer from numerous comorbid psychopathologies that exac-

erbate criminality (Chandler et al., 2009; Perry et al., 2006), these

individuals may benefit from classic psychedelic-assisted inter-

ventions. Unfortunately, there is scant contemporary data to

inform this issue. To our knowledge, only two recent studies have

yielded data relevant to the potential impact of classic psyche-

delics on criminal behavior. Hendricks et al. (2014) found that

naturalistic hallucinogen use predicted a reduced likelihood of

supervision failure (adjusted odds ratio (aOR) = 0.60) among over

25,000 individuals under community corrections supervision with

a history of substance involvement. Walsh et al. (2016) found that

naturalistic hallucinogen use predicted reduced arrest for intimate

partner violence (aOR = 0.62) among 302 jail inmates. Though

suggestive of protective effects, because hallucinogens are a

broader class of substances that includes classic psychedelics in

addition to other substances, neither Hendricks et al. (2014) nor

Walsh et al. (2016) were able to test the unique relationships of

classic psychedelics with criminal behavior outcomes.

Replication and extension of prior results is merited given the

high rates of non-replication of research findings (Ioannidis,

2014; Open Science Collaboration, 2015), and also because the

magnified political lens surrounding classic psychedelics suggests

that rigorous and persuasive results will be required for applica-

tion in forensic settings. Accordingly, the purpose of the current

study was to examine the relationships of lifetime classic psyche-

delic use with past year criminal behavior using data drawn from

the last 13 available years of the National Survey on Drug Use and

Health (NSDUH) at the time of analysis (2002–2014). Consistent

with prior research (Hendricks et al., 2015b), the specific relation-

ships of lifetime psilocybin use with past year criminal behavior

also was examined to help inform decisions by the United States

Food and Drug Administration and regulatory bodies of other

nations regarding future approved use. Based on findings from

prior research, we hypothesized that lifetime classic psychedelic

use and lifetime psilocybin use per se would be associated with a

decreased likelihood of past year criminal behavior.

Data and methods

The NSDUH is an annual cross-sectional survey conducted by

the Substance Abuse and Mental Health Services Administration

of the United States Department of Health and Human Services.

It is designed to estimate the prevalence of substance use and

mental illness in the general United States civilian non-institu-

tionalized population (United States Department of Health and

Human Services, 2003, 2004, 2005, 2006, 2007, 2008, 2009,

2010, 2011, 2012, 2013, 2014, 2015). Participants in the present

research were adult (≥18 years old) respondents of the NSDUH

pooled across years 2002 through 2014 (the maximum number of

survey years providing the same variables for analysis;

unweighted Ns for all analyses > 480,000). Similar approaches

have been employed by five prior investigations evaluating pop-

ulation-level associations of classic psychedelic use with indices

of mental health (Hendricks et al., 2015a, b; Johansen and Krebs,

2015; Krebs and Johansen, 2013; Pisano et al., 2017). Detailed

information on the NSDUH can be found elsewhere (https://

nsduhweb.rti.org/respweb/homepage.cfm). The NSDUH survey

was approved by the institutional review board of the Research

Triangle Institute and the present research was approved by the

institutional review board of the University of Alabama at

Birmingham.

Measures

Classic psychedelic use

The measurement of classic psychedelic use followed those of

two prior studies (Hendricks et al., 2015a, b). Respondents report-

ing that they had ever, even once, used ayahuasca, dimethyl-

tryptamine (DMT), LSD, mescaline, peyote or San Pedro, or

psilocybin mushrooms (referred to as “psilocybin” hereafter)

were coded as positive for lifetime classic psychedelic use,

whereas those indicating that they had never used any of these

Hendricks et al. 3

substances were coded as negative (the NSDUH specifically que-

ried lifetime LSD, mescaline, peyote, and psilocybin use, whereas

respondents had the opportunity to volunteer lifetime use of aya-

huasca, DMT, and San Pedro in open-ended questions). To better

identify outcomes associated specifically with psilocybin use,

four mutually exclusive groups of respondents were examined:

(1) Psilocybin Only (those reporting lifetime use of psilocybin but

no other classic psychedelic); (2) Psilocybin & Other Psychedelics

(those reporting lifetime use of psilocybin in addition to other

classic psychedelics); (3) Non-Psilocybin Psychedelics Only

(those reporting lifetime use of any classic psychedelic with the

exception of psilocybin); and (4) No Psychedelics (those report-

ing no lifetime use of any classic psychedelic substance).

Criminal behavior

We examined six outcome variables pertaining to criminal behav-

ior, including (1) past year drug distribution (“During the past 12

months, how many times have you sold illegal drugs?”; variable

SNYSELL), (2) past year larceny/theft (“During the past 12

months, how many times have you stolen or tried to steal anything

worth more than $50?”; variable SNYSTOLE), and (3) past year

assault (“During the past 12 months, how many times have you

attacked someone with the intent to seriously hurt them?”; variable

SNYATTAK). Each of these variables used the following response

set: 1 = 0 times, 2 = 1 or 2 times, 3 = 3 to 5 times, 4 = 6 to 9 times,

5 = 10 or more times. Responses were collapsed into three catego-

ries (coded 0 = 0 times, 1 = 1 or 2 times, 3 = > 2 times) to maintain

the ordinal nature of the measure while improving data scarcity

(e.g. even in a sample size as large as ours, very few individuals

have assaulted others more than five times in the past year).

The NSDUH incorporates a number of yes/no questions

regarding past year arrest history (“In the past 12 months, were

you arrested and booked for…”), including: “…possession, man-

ufacture, or sale of drugs?” (variable BKDRUG); “…burglary or

breaking and entering?” (variable BKBURGL); “…larceny or

theft? Do not include motor vehicle theft.” (variable

BKLARCNY); “…motor vehicle theft?” (variable BKMVTHFT);

“…arson?” (BKARSON); “…aggravated assault…forcible

rape…murder, homicide, or nonnegligent manslaughter?” (vari-

able BKSRVIOL); “…other assault, such as simple assault or

battery?” (variable BKSMASLT); and “…robbery?” (variable

BKROB). We condensed these questions into one of three con-

ceptually similar categories, estimating each as a dichotomous

outcome. These categories composed our final three outcome

variables: (4) past year arrest for a drug-related crime (coded 0 if

BKDRUG = no, 1 if response to this variable = yes), (5) past year

arrest for a property crime (coded as 0 if variables BKBURGL,

BKLARCNY, BKMVTHFT, or BKARSON each = no, 1 if

responses to any of these variables = yes), and (6) past year arrest

for a violent crime (coded as 0 if variables BKSRVIOL,

BKSMASLT, or BKROB each = no, 1 if responses to any of

these variables = yes).

Covariates

The relationships between classic psychedelic use and each of our

dependent variables may be confounded by sociodemographic

variables, illicit use of other substances, secular changes in both

classic psychedelic use and our dependent variables over time, as

well as other unmeasured variables. As such, we statistically

adjusted for a range of measured variables. All analyses were

adjusted for survey year, age in years (dummy coded as 18–25,

26–24, 35–49, and 50 or older), sex (male = 0, female = 1), ethno-

racial identity (dummy coded as non-Hispanic White, non-His-

panic African American, non-Hispanic Native American/Alaska

Native, non-Hispanic Native Hawaiian/Pacific Islander, non-His-

panic Asian, non-Hispanic more than one race, or Hispanic), edu-

cational attainment (dummy coded as 5th grade or less through

senior college year or more across 11 categories), annual house-

hold income (dummy coded as less than $20,000, $20,000–

$49,999, $50,000–$74,999, or $75,000 or more), marital status

(dummy coded as married, divorced/separated, widowed, or never

married), employment status (dummy coded as full time, part

time, unemployed, or other (including those not in the labor

force)), self-reported engagement in risky behavior (dummy

coded as never, seldom, sometimes, or always), degree to which

religious beliefs influence decisions (“Your religious beliefs influ-

ence how you make decisions in your life.”; dummy coded as

strongly disagree, disagree, agree, or strongly agree; one of three

interrelated questions on the NSDUH pertaining to religiosity/

spirituality and deemed most relevant to the current analyses), and

lifetime illicit use of cocaine, other stimulants, sedatives, tran-

quilizers, heroin, pain relievers, marijuana, 3, 4-methylenediox-

ymethamphetamine (MDMA)/ecstasy, phencyclidine (PCP), and

inhalants (each substance category coded as separate covariates;

no lifetime use = 0, lifetime use = 1). These covariates mirror

those of prior investigations (Hendricks et al., 2015a, b) with the

exception of survey year, employment status, and degree to which

religious beliefs influence decisions, which were added to further

address sources of possible confounding.

Statistical analyses

We used generalized ordered logit regression to test the relation-

ships of classic psychedelic use and group membership with past

year drug distribution, past year larceny/theft, and past year

assault controlling for the covariates listed above (Williams,

2006). Generalized ordered logit regression relaxes the assump-

tion that the effects of independent variables are proportional

across levels of the dependent variable, improving model fit

when the proportional odds assumption is violated (Williams,

2016). For each of these three ordinal outcomes, we first esti-

mated models in which the estimated effects of drug use varia-

bles were relaxed while holding the effects of covariates constant

across levels of the dependent variable. Based on these results,

we tested whether estimated coefficients differed across levels of

the dependent variable using Wald tests. We then re-estimated the

models, allowing any coefficient found to significantly differ

across levels of the dependent variable to vary across said levels,

while holding all other coefficients constant. An ordered logit

model in which the proportional odds assumption is relaxed for

some but not all variables is also known as a partial proportional

odds model (Peterson and Harrell, 1990). Because all variables

pertaining to arrest histories were dichotomously coded, we used

binomial logistic regression to test the relationships of classic

psychedelic use and group membership with past year arrest for

a drug-related crime, past year arrest for a property crime, and

past year arrest for a violent crime while controlling for the afore-

mentioned covariates.

4 Journal of Psychopharmacology 00(0)

All data management was conducted in SAS 9.4 and all anal-

yses were conducted in STATA 14. Generalized ordered logit

regression was performed using the ‘gologit2’ user written

STATA package (Williams, 2006). All analyses accounted for the

complex study design variables, sampling weights, and pooling

of data from multiple survey years as recommended by the

NSDUH.

Results

Descriptive statistics

Supplemental Table 1 presents frequencies of classic psychedelic

use and criminal behavior variables. Consistent with prior reports

(Hendricks et al., 2015a, b), lifetime classic psychedelic use was

reported among approximately 1/7th of respondents; most

respondents fell into the No Psychedelics group, followed by the

Psilocybin & Other Psychedelics group, the Non-Psilocybin

Psychedelics Only group, and finally, the Psilocybin Only group.

Past year drug distribution, past year larceny/theft, past year

assault, past year arrest for a drug-related crime, past year arrest

for a property crime, and past year arrest for a violent crime were

each infrequent outcomes, with weighted percentages ranging

from 0.2 to 1.2.

Table 1 reports weighted descriptive statistics of lifetime clas-

sic psychedelic users versus non-lifetime classic psychedelic

users. Similar to prior analyses (Hendricks et al., 2015a), lifetime

classic psychedelic use was concentrated among 26 to 34 year

olds and 35 to 49 year olds, but less common among those 50

years of age or older. Furthermore, lifetime classic psychedelic

use was more common among men, non-Hispanic Whites and

Native Americans/Alaska Natives, those with greater educational

attainment and income, those who were divorced/separated or

who had never been married, individuals with greater self-

reported engagement in risky behavior, and those who reported

lifetime illicit use of each of the other substances. Lifetime clas-

sic psychedelic use was also more common among employed

individuals and those who reported less influence of religious

beliefs on their decisions. Though lifetime classic psychedelic

use varied by survey year, findings revealed no readily interpret-

able pattern (results not presented in this report).

Table 2 reports weighted descriptive statistics of the Psilocybin

Only, Psilocybin & Other Psychedelics, Non-Psilocybin

Psychedelics Only, and No Psychedelics groups. As shown in the

table, the groups differed significantly on all covariates. The

Psilocybin Only group tended to be younger, more educated,

higher income, and more likely to have never been married than

the other three groups. The Psilocybin Only group comprised

more men and non-Hispanic Whites than the No Psychedelics

and Non-Psilocybin Psychedelics Only group, but fewer men and

non-Hispanic Whites than the Psilocybin & Other Psychedelics

group. In addition, the Psilocybin Only group was more likely to

be employed than the No Psychedelics group and reported greater

engagement in risky behavior and less influence of religious

beliefs on their decisions than the No Psychedelics and Non-

Psilocybin Psychedelics Only groups. Finally, the Psilocybin

Only group was more likely to report lifetime illicit use of each

of the other substances than the No Psychedelics group, but less

likely to report lifetime illicit use of each of the other substances

than the Psilocybin & Other Psychedelics group. The Psilocybin

Only group was more likely to report lifetime illicit use of tran-

quilizers, pain relievers, marijuana, MDMA/Ecstasy, and inhal-

ants but less likely to report lifetime illicit use of cocaine, other

stimulants, sedatives, heroin, and PCP as compared with the

Non-Psilocybin Psychedelics Only group. Group membership

also varied by year, but with no obvious pattern (results not pre-

sented in this report).

Outcomes

Table 3 reports results from generalized ordered logistic regres-

sion models predicting past year drug distribution, past year lar-

ceny/theft, and past year assault. As shown in this table, lifetime

classic psychedelic use was associated with an increased odds of

past year drug distribution one or more times. Results of a Wald

test indicated the estimated association between lifetime classic

psychedelic use and past year drug distribution differed across

levels of the dependent variable and that the association was

stronger among individuals who distributed drugs more fre-

quently. Lifetime illicit use of each of the other substances was

associated with an increased odds of past year drug distribution,

with the exception of lifetime sedative and PCP use, which were

not significantly associated with this outcome (both aORs ≥ 1.0).

Lifetime classic psychedelic use was also associated with a

decreased odds of past year larceny/theft and retained the same

association in both models because the effect was assumed to be

proportional across levels of the dependent variable. Conversely,

lifetime illicit use of each of the other substances was associated

with an increased odds of past year larceny/theft, with the excep-

tion of lifetime sedative and PCP use, which were not signifi-

cantly associated with this outcome (both aORs > 1.0). Finally,

lifetime classic psychedelic use was associated with a decreased

odds of past year assault and retained the same association in

both models because the effect was assumed to be proportional

across levels of the dependent variable. Lifetime illicit use of

each of the other substances was associated with an increased

odds of past year assault, with the exception of lifetime cocaine,

other stimulant, and heroin use, which were not significantly

associated with this outcome (all aORs > 1.0).

Table 4 reports results from generalized ordered logistic

regression models predicting past year drug distribution, past

year larceny/theft, and past year assault as a function of group

membership. Relative to the Psilocybin Only group, both the No

Psychedelics and Non-Psilocybin Psychedelics Only groups

were less likely to report past year drug distribution. The

Psilocybin Only and Psilocybin & Other Psychedelics groups did

not differ on this outcome. These effects were assumed to be pro-

portional across levels of the dependent variable. The No

Psychedelics group was more likely to report past year larceny/

theft as compared with the Psilocybin Only group, whereas the

Psilocybin & Other Psychedelics group was less likely to report

past year larceny/theft as compared with the Psilocybin Only

group. However, a Wald test indicated that the association of the

Psilocybin & Other Psychedelics group with past year larceny/

theft differed across levels of the dependent variable and was sig-

nificantly stronger among individuals who reported past year lar-

ceny/theft more than two times. Consequently, the estimated

odds of past year larceny/theft more than two times did not differ

between the Psilocybin Only and Psilocybin & Other Psychedelics

group. The Psilocybin Only and Non-Psilocybin Psychedelics

Hendricks et al. 5

Only groups did not differ with regard to past year larceny/theft.

Finally, the No Psychedelics group was more likely to report past

year assault as compared with the Psilocybin Only group. Neither

the Psilocybin & Other Psychedelics nor the Non-Psilocybin

Psychedelics Only group differed from the Psilocybin Only

group on this outcome. These effects were assumed to be propor-

tional across levels of the dependent variable.

Table 5 reports results from binomial logistic regression mod-

els predicting past year arrest for a drug-related crime, past year

arrest for a property crime, and past year arrest for a violent

crime. Lifetime classic psychedelic use was not associated with

past year arrest for a drug-related crime. Lifetime illicit use of

each of the other substances was associated with an increased

odds of past year arrest for a drug-related crime, with the excep-

tion of lifetime other stimulant and sedative use, which were not

associated with this outcome (both aORs > 1.0), and lifetime

inhalant use, which was associated with a decreased odds of this

outcome. Lifetime classic psychedelic use was associated with a

reduced odds of past year arrest for a property crime. All other

lifetime illicit substance use variables were either associated with

an increased odds of past year arrest for a property crime (cocaine,

tranquilizers, heroin, and pain relievers) or not associated with

this outcome (other stimulants, sedatives, MDMA/Ecstasy, PCP,

Inhalants; all aORs > 1.0). Finally, lifetime classic psychedelic

Table 1. Characteristics of lifetime classic psychedelic users versus

non-lifetime classic psychedelic users.

Variable Lifetime

classic

psychedelic

users

Non-lifetime

classic

psychedelic

users

Weighted % Weighted %

Age, years <0.0001

18 to 25 15.4 14.7

26 to 34 21.2 15.1

35 to 49 36.1 27.0

50 and older 27.2 43.3

Sex <0.0001

Male 62.4 46.0

Female 37.6 54.1

Ethnoracial identity <0.0001

Non-Hispanic White 84.0 65.9

Non-Hispanic African American 3.9 12.6

Non-Hispanic Native Ameri-

can/Alaska Native

1.0 0.4

Non-Hispanic Native Hawai-

ian/Pacific Islander

0.2 0.3

Non-Hispanic Asian 1.2 5.0

Non-Hispanic more than one

race

1.9 1.1

Hispanic 7.8 14.5

Educational attainment <0.0001

5th grade or less 0.3 1.7

6th grade 0.2 1.5

7th grade 0.2 0.6

8th grade 0.8 1.9

9th grade 1.9 2.5

10th grade 3.2 3.1

11th grade 5.1 4.7

12th grade 28.6 31.1

Freshman college year 10.1 8.5

Sophomore or junior

college year

20.3 16.6

Senior college year or more 29.3 27.8

Annual household income <0.0001

Less than $20,000 17.0 19.0

$20,000 to $49,999 31.4 34.3

$50,000 to $74,999 18.2 17.4

$75,000 or more 33.3 29.2

Marital status <0.0001

Married 46.7 55.6

Widowed 1.6 6.9

Divorced/Separated 18.3 12.8

Never married 33.4 24.7

Employment status <0.0001

Full time 63.5 51.0

Part time 14.1 13.5

Unemployed 6.1 4.3

Other 16.3 31.2

Self-reported engagement in

risky behavior

<0.0001

Never 26.7 55.0

Variable Lifetime

classic

psychedelic

users

Non-lifetime

classic

psychedelic

users

Weighted % Weighted %

Seldom 44.9 32.8

Sometimes 25.5 11.1

Always 2.9 1.1

Religious beliefs influence

decisions

<0.0001

Strongly disagree 21.0 11.4

Disagree 21.6 13.6

Agree 37.3 38.7

Strongly agree 20.1 36.2

Lifetime illicit substance use

Lifetime cocaine use 71.1 7.3 <0.0001

Lifetime other stimulant use 38.5 3.8 <0.0001

Lifetime sedative use 19.4 1.2 <0.0001

Lifetime tranquilizer use 37.6 4.9 <0.0001

Lifetime heroin use 10.6 0.4 <0.0001

Lifetime pain reliever use 45.6 9.1 <0.0001

Lifetime marijuana use 98.1 35.7 <0.0001

Lifetime MDMA/ecstasy use 31.2 1.9 <0.0001

Lifetime PCP use 18.5 0.4 <0.0001

Lifetime inhalant use 40.6 3.8 <0.0001

Weighted N (%) 30,711,342

(13.6)

194,394,381

(86.4)

Note. All percentages rounded to the nearest tenth of a percent; cumulative

percentages may not add to 100.0. Rao–Scott chi-square tests were used to

examine the characteristics of lifetime classic psychedelic users versus non-

lifetime classic psychedelic users. MDMA: 3,4-methylenedioxymethamphetamine;

PCP: phencyclidine.

(continued)

Table 1. (Continued)

6 Journal of Psychopharmacology 00(0)

Table 2. Characteristics of Psilocybin Only, Psilocybin & Other Psychedelics, Non-Psilocybin Psychedelics Only, and No Psychedelics groups.

Variable Psilocybin Only Psilocybin & Other

Psychedelics

Non-Psilocybin

Psychedelics Only

No Psychedelics p

Weighted % Weighted % Weighted % Weighted %

Age, years <0.0001

18 to 25 31.8 14.6 8.7 14.7

26 to 34 25.8 22.7 17.0 15.1

35 to 49 29.4 36.9 38.3 26.9

50 and older 12.9 25.8 36.0 43.3

Sex

Male 60.2 69.1 54.0 45.9

Female 39.8 30.9 46.0 54.1

Ethnoracial identity <0.0001

Non-Hispanic White 83.3 87.7 79.2 65.9

Non-Hispanic African American 2.8 1.8 7.3 12.6

Non-Hispanic Native American/Alaska Native 0.5 0.7 1.7 0.4

Non-Hispanic Native Hawaiian/Pacific Islander 0.3 0.2 0.2 0.3

Non-Hispanic Asian 2.1 0.9 1.2 5.0

Non-Hispanic more than one race 1.9 1.9 1.8 1.1

Hispanic 9.1 6.7 8.6 14.5

Educational attainment <0.0001

5th grade or less 0.4 0.2 0.3 1.7

6th grade 0.1 0.1 0.3 1.5

7th grade 0.2 0.2 0.3 0.6

8th grade 0.5 0.8 1.0 1.9

9th grade 0.9 1.8 2.5 2.5

10th grade 2.1 3.2 3.7 3.1

11th grade 4.1 5.0 5.7 4.7

12th grade 22.4 28.1 32.3 31.1

Freshman college year 9.9 10.2 9.9 8.5

Sophomore or junior college year 20.3 20.6 19.8 16.6

Senior college year or more 38.9 29.8 24.0 27.8

Annual household income <0.0001

Less than $20,000 16.6 17.1 17.2 19.0

$20,000 to $49,999 28.9 31.8 32.1 34.3

$50,000 to $74,999 16.8 18.0 19.3 17.4

$75,000 or more 37.7 33.1 31.4 29.2

Marital status <0.0001

Married 39.7 44.6 52.9 55.6

Widowed 0.8 1.3 2.3 6.8

Divorced/Separated 11.0 18.6 21.4 12.8

Never married 48.5 35.4 23.3 24.7

Employment status <0.0001

Full time 63.2 64.8 61.8 51.0

Part time 17.5 13.8 12.9 13.5

Unemployed 6.5 6.2 5.7 4.3

Other 12.8 15.2 19.5 31.2

Self-reported engagement in risky behavior <0.0001

Never 20.8 22.7 35.2 55.0

Seldom 47.4 45.1 43.5 32.8

Sometimes 28.3 28.7 19.8 11.1

Always 3.5 3.5 1.6 1.1

Religious beliefs influence decisions <0.0001

Strongly disagree 24.8 22.9 16.5 11.4

Disagree 23.1 22.2 20.1 13.6

Agree 36.3 36.8 38.4 38.7

(continued)

Hendricks et al. 7

use was associated with a decreased odds of past year arrest for a

violent crime. All other lifetime illicit substance use variables

were either associated with an increased odds of past year arrest

for a violent crime (cocaine, heroin, pain relievers, marijuana,

MDMA/Ecstasy) or were not associated with this outcome (other

stimulants, sedatives, PCP, inhalants; all but PCP aOR > 1.0).

Table 6 reports results from binomial logistic regression mod-

els predicting past year arrest for a drug-related crime, past year

arrest for a property crime, and past year arrest for a violent crime

as a function of group membership. No significant associations

were found.

Discussion

The aim of the current research was to evaluate the associa-

tions of classic psychedelic use, and psilocybin use in particu-

lar, with criminal behavior in a large sample generalizable to

the United States adult population. In support of hypotheses,

lifetime classic psychedelic use was associated with a 27%

decreased odds of past year larceny/theft, a 12% decreased

odds of past year assault, a 22% decreased odds of past year

arrest for a property crime, and an 18% decreased odds of past

year arrest for a violent crime. Lifetime illicit use of other sub-

stances, in contrast, was largely associated with an increased

odds of criminal behavior at or above the trend level. These

findings are consistent with a growing body of research sug-

gesting classic psychedelics confer enduring psychological

and prosocial benefits (Bogenschutz et al., 2015; Carhart-

Harris et al., 2016; Griffiths et al., 2016; Hendricks et al.,

2015a; Johnson et al., 2014, 2017; Pisano et al., 2017; Ross

et al., 2016) and an older (Arendsen-Hein, 1963; Tenenbaum,

1961) and newer (Hendricks et al., 2014; Walsh et al., 2016)

literature indicating that classic psychedelics may be effective

in preventing criminal behavior. The replication and extension

of prior results is notable considering that scientific findings

often fail to be replicated (Ioannidis, 2014; Open Science

Collaboration, 2015). It is also noteworthy that whereas other

recent studies have examined criminal recidivism, the present

findings are based on a non-forensic sample with a majority of

individuals who did not report any involvement with the crimi-

nal justice system. As such, the current results speak favorably

to the generalizability of prior findings with correctional sam-

ples. Moreover, results were consistent across the complemen-

tary criteria of self-reported criminal behavior and arrest,

which suggests that the apparent protective effects of classic

psychedelic use are attributable to genuine reductions in anti-

social behavior rather than reflecting improved evasion of

arrest. Simply put, the positive effects associated with classic

psychedelic use appear to be reliable. Acknowledging political

hurdles, it is hoped that the current findings will contribute to

a compelling rationale for the initiation of clinical research

with classic psychedelics among groups at increased risk of

engaging in criminal behavior, including released inmates and

those engaged in problematic substance use. Given the cost to

society and recalcitrance of criminal behavior, the potential

represented by this treatment paradigm is significant. However,

we feel it is essential to note that, given the intense, sensitive,

and personal nature of primary mystical experiences, individ-

ual autonomy in the decision to engage in classic psychedelic-

assisted treatment must be foregrounded, particularly with

regard to vulnerable populations. The potential for coercion

and undue incentivization in the administration of classic

psychedelics to individuals in the criminal justice system is an

issue that requires further scrutiny to ensure conformity with

the highest possible ethical standards. It is our opinion that

classic psychedelic-assisted treatments never be applied in a

context that might be perceived as institutionally mandated.

Variable Psilocybin Only Psilocybin & Other

Psychedelics

Non-Psilocybin

Psychedelics Only

No Psychedelics p

Weighted % Weighted % Weighted % Weighted %

Strongly agree 15.7 18.1 25.0 36.2

Lifetime illicit substance use

Lifetime cocaine use 55.4 83.4 61.4 7.3 <0.0001

Lifetime other stimulant use 27.2 48.9 29.6 3.8 <0.0001

Lifetime sedative use 7.5 26.0 16.0 1.2 <0.0001

Lifetime tranquilizer use 30.7 47.2 27.6 4.9 <0.0001

Lifetime heroin use 3.6 15.0 7.7 0.4 <0.0001

Lifetime pain reliever use 43.7 55.1 33.3 9.1 <0.0001

Lifetime marijuana use 97.3 99.3 96.7 35.7 <0.0001

Lifetime MDMA/ecstasy use 32.3 41.5 16.3 1.9 <0.0001

Lifetime PCP use 2.7 25.9 15.9 0.4 <0.0001

Lifetime inhalant use 30.6 53.8 27.2 3.8 <0.0001

Weighted N (%) 5,143,062 (2.3) 14,872,415 (6.6) 10,695,865 (4.8) 194,394,381 (86.4)

Note. Psilocybin Only: respondents reporting lifetime use of psilocybin but no other classic psychedelic; Psilocybin & Other Psychedelics: respondents reporting lifetime

use of psilocybin in addition to other classic psychedelics; Non-Psilocybin Psychedelics Only: respondents reporting lifetime use of any classic psychedelic with the excep-

tion of psilocybin; No Psychedelics: respondents reporting no lifetime use of any classic psychedelic substance. All percentages rounded to the nearest tenth of a percent;

cumulative percentages may not add to 100.0. Rao–Scott chi-square tests were used to examine the characteristics of the four groups. MDMA: 3,4-methylenedioxymeth-

amphetamine; PCP: phencyclidine.

Table 2. (Continued)

8 Journal of Psychopharmacology 00(0)

Table 3. Results from generalized ordered logistic regression models predicting past year drug distribution, past year larceny/theft, and past year

assault.

Variable Past year drug distribution pPast year larceny/theft pPast year assault p

0 vs. 1 or more aOR (95% CI) aOR (95% CI) aOR (95% CI)

Lifetime classic psychedelic use 1.47(1.35–1.59) <0.0001 0.73(0.65–0.83) <0.0001 0.88(0.80–0.97) 0.009

Lifetime cocaine use 1.52(1.40–1.65) <0.0001 1.23(1.09–1.39) 0.001 1.10(0.98–1.22) 0.097

Lifetime other stimulant use 1.32(1.23–1.43) <0.0001 1.20(1.07–1.34) 0.001 1.06(0.97–1.17) 0.197

Lifetime sedative use 1.05(0.94–1.18) 0.347 1.05(0.88–1.25) 0.603 1.11(0.92–1.35) 0.267

Lifetime tranquilizer use 1.46(1.34–1.59) <0.0001 1.19(1.07–1.32) 0.002 1.17(1.05–1.30) 0.005

Lifetime heroin use 1.26(1.13–1.40) <0.0001 1.73(1.48–2.03) <0.0001 1.04(0.87–1.23) 0.679

Lifetime pain reliever use 1.91(1.79–2.05) <0.0001 1.94(1.77–2.12) <0.0001 1.66(1.53–1.80) <0.0001

Lifetime marijuana use 1.72(1.56–1.90) <0.0001 1.32(1.20–1.45) <0.0001 1.37(1.26–1.50) <0.0001

Lifetime MDMA/ecstasy use 1.70(1.56–1.85) <0.0001 1.33(1.17–1.51) <0.0001 1.38(1.26–1.50) <0.0001

Lifetime PCP use 1.00(0.89–1.13) 0.915 1.13(0.95–1.34) 0.175 1.29(1.13–1.48) <0.0001

Lifetime inhalant use 1.25(1.17–1.33) <0.0001 1.41(1.27–1.57) <0.0001 1.32(1.19–1.47) <0.0001

Variable Past year drug distribution pPast year larceny/theft pPast year assault P

1 to 2 times vs. > 2 aOR (95% CI) aOR (95% CI) aOR (95% CI)

Lifetime classic psychedelic use 1.68(1.54–1.85) <0.0001 0.73(0.65–0.83) <0.0001 0.88(0.80–0.97) 0.009

Lifetime cocaine use 1.52(1.40–1.65) <0.0001 1.23(1.09–1.39) 0.001 1.10(0.98–1.22) 0.097

Lifetime other stimulant use 1.32(1.23–1.43) <0.0001 1.20(1.07–1.34) 0.001 1.06(0.97–1.17) 0.197

Lifetime sedative use 1.05(0.94–1.18) 0.347 1.05(0.88–1.25) 0.603 1.46(1.08–1.97) 0.015

Lifetime tranquilizer use 1.46(1.34–1.59) <0.0001 1.19(1.07–1.32) 0.002 1.17(1.05–1.30) 0.005

Lifetime heroin use 1.26(1.13–1.40) <0.0001 2.88(2.37–3.49) <0.0001 1.04(0.87–1.23) 0.679

Lifetime pain reliever use 1.91(1.79–2.05) <0.0001 1.94(1.77–2.12) <0.0001 1.66(1.53–1.80) <0.0001

Lifetime marijuana use 5.47(4.35–6.89) <0.0001 3.00(2.50–3.61) <0.0001 1.86(1.56–2.22) <0.0001

Lifetime MDMA/ecstasy use 1.70(1.56–1.85) <0.0001 1.33(1.17–1.51) <0.0001 1.67(1.43–1.95) <0.0001

Lifetime PCP use 1.12(0.99–1.27) 0.066 1.13(0.95–1.34) 0.175 1.99(1.53–2.59) <0.0001

Lifetime inhalant use 1.25(1.17–1.33) <0.0001 1.41(1.27–1.57) <0.0001 1.09(0.91–1.31) 0.335

N (unweighted) 484,616 485,071 485,185

N (weighted) 221,687,999 221,791,920 221,824,983

Note. For past year drug distribution, parallel slopes not assumed for lifetime classic psychedelic, marijuana, and PCP use. For past year larceny/theft, parallel slopes not

assumed for lifetime marijuana and heroin use. For past year assault, parallel slopes not assumed for lifetime sedative, marijuana, MDMA/ecstasy, PCP, and inhalant use.

Associations of demographic variables, study year, self-reported engagement in risky behavior, and degree to which religious beliefs influence decisions are not presented.

MDMA: 3, 4-methylenedioxymethamphetamine; PCP: phencyclidine.

Mandated treatments are common in correctional settings, yet

are generally ineffective (Parhar et al., 2008). Moreover, clas-

sic psychedelics were administered coercively and/or surrepti-

tiously to unwitting human subjects in 1950s through the 1970s

(Lee and Shlain, 1992), including to incarcerated samples

without therapeutic intent (e.g. Isbell, 1959; Isbell et al., 1956,

1959; Isbell & Logan, 1957; Rosenberg et al., 1964). This

unfortunate history should serve a strong cautionary tale for

future investigations.

Contrary to hypotheses, lifetime classic psychedelic use was

associated with a 47% to 68% increased odds of past year drug

distribution. Lifetime illicit use of almost all other substances

was also associated with an increased odds of this outcome. An

obvious conclusion, therefore, is that one drug-related crime

(use) predicts another (distribution), a contention supported by

the research literature (e.g. Pedersen and Skardhamar, 2009).

Considering that experienced drug users believe that classic

psychedelics hold therapeutic potential (Carhart-Harris and

Nutt, 2013), it may be that classic psychedelic use is associated

with the distribution of classic psychedelics motivated in part

by prosocial intention. There is historical precedent for this per-

spective, for instance, in organizations such as the Brotherhood

of Eternal Love, which distributed LSD with the goal of improv-

ing the human condition (Lee and Shlain, 1992). It could not be

determined which specific substances were distributed in the

present analyses. However, lifetime classic psychedelic use was

not associated with past year arrest for a drug-related crime,

whereas lifetime illicit use of almost all other substances was

associated with an increased odds of this outcome. Considering

that approximately 70% of all drug-related arrests in the United

States relate to heroin or cocaine and their derivatives, or mari-

juana, with arrests relating to classic psychedelics presumably

so low that they are not specified (FBI Uniform Crime Report,

2015), it may be that classic psychedelic use is not associated

with past year arrest for a drug-related crime because it largely

aligns with possession, use, distribution, or manufacturing of

classic psychedelics. In any event, these findings highlight that

classic psychedelic use is not associated with a uniform reduc-

tion in all behavior deemed criminal. In support of this interpre-

tation, recent survey results suggest that classic psychedelic use

is associated with liberal and anti-authoritarian political views

(Nour et al., 2017). Drug policy is contentious and drug-related

crimes are generally not considered antisocial (Global

Commission on Drug Policy, 2011; Room and Reuter, 2012). It

Hendricks et al. 9

Table 4. Results from generalized ordered logistic regression models predicting past year drug distribution, past year larceny/theft, and past year

assault as a function of group membership.

Variable Past year drug distribution pPast year larceny/theft pPast Year assault p

0 vs. 1 or more aOR (95% CI) aOR (95% CI) aOR (95% CI)

Psilocybin Only (reference group) 1.00 – 1.00 – 1.00 -

Psilocybin & Other Psychedelics 0.90(0.80–1.02) 0.089 0.85(0.72–0.99) 0.042 0.94(0.81–1.09) 0.416

Non-Psilocybin Psychedelics Only 0.56(0.48–0.65) <0.0001 1.02(0.83–1.25) 0.850 1.12(0.95–1.32) 0.181

No Psychedelics 0.55(0.49–0.61) <0.0001 1.29(1.13–1.48) <0.0001 1.15(1.03–1.30) 0.016

Variable Past year drug distribution pPast Year larceny/theft pPast Year assault p

1 to 2 times vs. > 2 aOR (95% CI) aOR (95% CI) aOR (95% CI)

Psilocybin Only (reference group) 1.00 – 1.00 – 1.00 -

Psilocybin & Other Psychedelics 0.90(0.80–1.02) 0.089 1.05(0.83–1.32) 0.695 0.94(0.81–1.09) 0.416

Non-Psilocybin Psychedelics Only 0.56(0.48–0.65) <0.0001 1.02(0.83–1.25) 0.850 1.12(0.95–1.32) 0.181

No Psychedelics 0.50(0.44–0.56) <0.0001 1.29(1.13–1.48) <0.0001 1.15(1.03–1.30) 0.016

N (unweighted) 484,616 485,071 485,185

N (weighted) 221,687,999 221,791,920 221,824,983

Note. Psilocybin Only: respondents reporting lifetime use of psilocybin but no other classic psychedelic; Psilocybin & Other Psychedelics: respondents reporting lifetime

use of psilocybin in addition to other classic psychedelics; Non-Psilocybin Psychedelics Only: respondents reporting lifetime use of any classic psychedelic with the excep-

tion of psilocybin; No Psychedelics: respondents reporting no lifetime use of any classic psychedelic substance. For past year larceny/theft, parallel slopes not assumed

for Psilocybin & Other Psychedelics. Associations of demographic variables, study year, self-reported engagement in risky behavior, degree to which religious beliefs influ-

ence decisions, and lifetime illicit substance use are not presented.

Table 5. Results from binomial logistic regression models predicting past year arrest for a drug-related crime, past year arrest for a property crime,

and past year arrest for a violent crime.

Variable Past year arrest for a

drug-related crime

pPast year arrest for a

property crime

pPast year arrest

for a violent crime

aOR (95% CI) aOR (95% CI) aOR (95% CI)

Lifetime classic psychedelic use 0.99(0.82–1.18) 0.871 0.78(0.65–0.95) 0.015 0.82(0.70–0.97) 0.023

Lifetime cocaine use 1.71(1.43–2.05) <0.0001 1.59(1.32–1.90) <0.0001 1.58(1.34–1.87) <0.0001

Lifetime other stimulant use 1.19(0.99–1.43) 0.058 1.05(0.88–1.25) 0.580 1.08(0.92–1.27) 0.364

Lifetime sedative use 1.08(0.85–1.37) 0.545 1.18(0.89–1.56) 0.246 1.19(0.93–1.53) 0.166

Lifetime tranquilizer use 1.39(1.17–1.66) <0.0001 1.42(1.21–1.67) <0.0001 1.17(0.99–1.38) 0.058

Lifetime heroin use 1.80(1.50–2.17) <0.0001 1.97(1.57–2.48) <0.0001 1.42(1.12–1.81) 0.005

Lifetime pain reliever use 1.44(1.21–1.71) <0.0001 1.51(1.28–1.77) <0.0001 1.57(1.36–1.81) <0.0001

Lifetime marijuana use 6.51(5.02–8.45) <0.0001 1.92(1.62–2.27) <0.0001 1.96(1.69–2.28) <0.0001

Lifetime MDMA/ecstasy use 1.79(1.54–2.08) <0.0001 1.18(1.00–1.40) 0.053 1.23(1.06–1.43) 0.006

Lifetime PCP use 1.25(1.02–1.55) 0.035 1.19(0.91–1.56) 0.192 1.19(0.95–1.49) 0.135

Lifetime inhalant use 0.80(0.70–0.92) 0.002 1.06(0.88–1.28) 0.565 0.92(0.77–1.10) 0.360

N (unweighted) 480,756 480,746 480,751

N (weighted) 220,416,915 220,409,740 220,412,915

Note. Associations of demographic variables, study year, self-reported engagement in risky behavior, and degree to which religious beliefs influence decisions are not

presented. MDMA: 3,4-methylenedioxymethamphetamine; PCP: phencyclidine.

would not be surprising, then, if classic psychedelic use were

associated with more liberal and anti-authoritarian attitudes

toward drug use, as opposed to a more broadly antisocial pro-

pensity for transgressive behavior. This also leaves open the

possibility that, consistent with contemporary critique of drug

policy (Global Commission on Drug Policy, 2011; Room and

Reuter, 2012), the prohibition of classic psychedelics causes

harm insofar that it requires users to engage with the black mar-

ket and risk adverse legal consequences, without in fact reduc-

ing antisocial crime.

We also hypothesized that lifetime psilocybin use, per se,

would be associated with a decreased likelihood of past year

criminal behavior. In partial support of hypotheses, those who

reported lifetime use of psilocybin but no other classic psyche-

delic were less likely to report past year larceny/theft and past

year assault than those reporting no lifetime use of any classic

psychedelic. Contrary to hypotheses, those who reported life-

time use of psilocybin but no other classic psychedelic were

more likely to report past year drug distribution than those

reporting no lifetime use of any classic psychedelic and those

10 Journal of Psychopharmacology 00(0)

reporting lifetime use of any classic psychedelic with the excep-

tion of psilocybin, and more likely to report larceny/theft than

those reporting lifetime use of psilocybin in addition to other

classic psychedelics. No other differences were found. It is noted

that with the exception of past year drug distribution (the impli-

cations of which are discussed above), those who reported life-

time use of psilocybin but no other classic psychedelic were less

likely to report criminal behavior at or above the trend level as

compared with those reporting no lifetime use of any classic psy-

chedelic and those reporting lifetime use of any classic psyche-

delic with the exception of psilocybin (i.e. all those reporting no

lifetime use of psilocybin). Results are therefore consistent with

a protective effect of psilocybin for antisocial criminal behavior

and suggest clinical investigations making use of this particular

agent may hold promise.

As described elsewhere (Hendricks et al., 2015a), there are a

number of limitations of the current methodological design.

Response biases inherent in self-report may have obscured the

true relationships between classic psychedelic use and criminal

behavior. Analyses were restricted to the available data, which

preclude evaluating more precise dose-response relationships or

associations between classic psychedelic use and specific types

of drug distribution, among others. In addition, analyses relied on

cross-sectional naturalistic data, which limit causal inferences.

Though we attempted to control for multiple sources of potential

confounding, a number of shared underlying or “third” variables

may be responsible for the associations reported here (e.g. per-

sonality openness, spirituality, political orientation; Lerner and

Lyvers, 2006; Lyvers and Meester, 2012; Móró et al., 2011; Nour

et al., 2017). We also cannot rule out the possibility that classic

psychedelic use may have caused harm at the individual level.

Classic psychedelic use can aggravate certain mental health con-

ditions and occasion challenging experiences characterized by

anxiety, fear, panic, and paranoia. If such harms occurred, they

failed to obfuscate population-level associations suggesting pro-

tective effects. This is consistent with recent survey data showing

that almost 85% of those who have had challenging experiences

after ingesting psilocybin mushrooms report benefiting from the

experience (Carbonaro et al., 2016). Finally, we could not evalu-

ate potential mechanisms of action underlying the associations of

classic psychedelic use with criminal behavior. Though specula-

tive, it is possible that mystical-type experiences are typified by

awe, an emotion that may promote prosocial behavior via dimin-

ishment of the individual self (e.g. Piff et al., 2015).

Conclusion

The current study demonstrates that having ever used a classic

psychedelic, and to some degree, having ever used psilocybin per

se is associated with a decreased likelihood of larceny/theft and

other property crimes as well as a decreased likelihood of assault

and other violent crimes. These findings, coupled with both older

and emerging bodies of evidence suggesting that classic psyche-

delics may provide enduring benefits for criminal justice popula-

tions, compel much-needed clinical research in forensic settings.

Acknowledgements

The authors thank Emma E. Green and James D. Sexton for their work on

the project.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship,

and/or publication of this article.

References

Arendsen-Hein GW (1963) LSD in the treatment of criminal psycho-

paths. In: Crocket RW, Sandison RA and Walk A (eds) Hallucino-

genic Drugs and Their Psychotherapeutic Use. London: H. K. Lewis

& Co. Ltd., pp. 101–106.

Bogenschutz MP, Forcehimes AA, Pommy JA, et al. (2015) Psilocybin-

assisted treatment for alcohol dependence: A proof-of-concept study.

J Psychopharmacol 29: 289–299.

Carbonaro TM, Bradstreet MP, Barrett FS, et al. (2016) Survey study

of challenging experiences after ingesting psilocybin mushrooms:

Acute and enduring positive and negative consequences. J Psycho-

pharmacol 30: 1268–1278.

Carhart-Harris RL, Bolstridge M, Rucker J, et al. (2016) Psilocybin with

psychological support for treatment-resistant depression: An open-

label feasibility study. Lancet Psychiatry 3: 619–627.

Table 6. Results from binomial logistic regression models predicting past year arrest for a drug-related crime, past year arrest for a property crime,

and past year arrest for a violent crime as a function of group membership.

Variable Past year arrest

for a drug-related crime

pPast year arrest

for a property crime

pPast year arrest

for a violent crime

aOR (95% CI) aOR (95% CI) aOR (95% CI)

Psilocybin Only (reference group) 1.00 - 1.00 - 1.00 -

Psilocybin & Other Psychedelics 1.03(0.84–1.25) 0.806 0.80(0.61–1.05) 0.111 0.82(0.63–1.06) 0.134

Non-Psilocybin Psychedelics Only 1.07(0.88–1.29) 0.515 1.13(0.84–1.52) 0.407 1.17(0.85–1.60) 0.338

No Psychedelics 1.05(0.86–1.27) 0.638 1.23(0.95–1.60) 0.122 1.20(0.93–1.56) 0.162

N (unweighted) 480,756 480,736 480,751

N (weighted) 220,416,915 220,409,740 220,412,915

Note. Psilocybin Only: respondents reporting lifetime use of psilocybin but no other classic psychedelic; Psilocybin & Other Psychedelics: respondents reporting lifetime

use of psilocybin in addition to other classic psychedelics; Non-Psilocybin Psychedelics Only: respondents reporting lifetime use of any classic psychedelic with the

exception of psilocybin; No Psychedelics: respondents reporting no lifetime use of any classic psychedelic substance. Associations of demographic variables, study year,

self-reported engagement in risky behavior, degree to which religious beliefs influence decisions, and lifetime illicit substance use are not presented.

Hendricks et al. 11

Carhart-Harris RL and Nutt DJ (2013) Experienced drug users assess the

relative harms and benefits of drugs: A web-based survey. J Psycho-

active Drugs 45: 322–328.

Chandler RK, Fletcher BW and Volkow ND (2009) Treating drug abuse

and addiction in the criminal justice system: Improving public health

and safety. JAMA 301: 183–190.

Doblin R (1998) Dr. Leary’s concord prison experiment: A 34-year fol-

low-up study. J Psychoactive Drugs 30: 419–426.

Durose MR, Cooper AD and Snyder HN (2014) Recidivism of prison-

ers released in 30 states in 2005: Patterns from 2005 to 2010. Spe-

cial Report. Washington, DC: United States Department of Justice,

Federal Bureau of Investigation, Criminal Justice Information Ser-

vices Division. Available at: https://www.bjs.gov/content/pub/pdf/

rprts05p0510.pdf

FBI Uniform Crime Report (2015) Crime in the United States 2015.

Washington, DC: United States Department of Justice, Federal

Bureau of Investigation, Criminal Justice Information Services Divi-

sion. Available at: https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-

in-the-u.s.-2015

Ferguson LM and Wormith JS (2013) A meta-analysis of moral recona-

tion therapy. Int J Offender Ther Comp Criminol 57: 1076–1106.

Global Commission on Drug Policy (2011) War on Drugs: Report on the

Global Commission on Drug Policy. Global Commission on Drug

Policy. Available at: https://www.globalcommissionondrugs.org/

reports/war-on-drugs/

Griffiths RR, Johnson MW, Carducci MA, et al. (2016) Psilocybin pro-

duces substantial and sustained decreases in depression and anxiety

in patients with life-threatening cancer: A randomized double-blind

trial. J Psychopharmacol 30: 1181–1197.

Hendricks PS, Clark CB, Johnson MW, et al. (2014) Hallucinogen use

predicts reduced recidivism among substance-involved offenders

under community corrections supervision. J Psychopharmacol 28:

62–66.

Hendricks PS, Thorne CB, Clark CB, et al. (2015a) Classic psychedelic

use is associated with reduced psychological distress and suicidal-

ity in the United States adult population. J Psychopharmacol 29:

280–288.

Hendricks PS, Johnson MW and Griffiths RR (2015b) Psilocybin, psycho-

logical distress, and suicidality. J Psychopharmacol 29: 1041–1043.

Ioannidis JP (2014) How to make more published research true. PLoS

Med: 11: e1001747.

Isbell H (1959) Comparison of the reactions induced by psilocybin and

LSD-25 in man. Psychopharmacology 1: 29–38.

Isbell H, Belleville RE, Fraser HF, et al. (1956) Studies on lysergic acid

diethylamide (LSD-25). I. Effects in former morphine addicts and

development of tolerance during chronic intoxication. AMA Arch

Neurol Psychiatry 76: 468–478.

Isbell H and Logan CR (1957) Studies on the diethylamide of lysergic

acid (LSD-25). II. Effects of chlorpromazine, azacyclonol, and res-

perine on the intensity of the LSD-reaction. AMA Arch Neurol Psy-

chiatry 77: 350–358.

Isbell H, Logan CR and Miner EJ (1959) Studies on lysergic acid diethyl-

amide (LSD-25). III. Attempts to attenuate the LSD-reaction in man

by pretreatment with neurohumoral blocking agents. AMA Arch Neu-

rol Psychiatry 81: 20–27.

Johansen PØ and Krebs TS (2015) Psychedelics not linked to mental

health problems or suicidal behavior: A population study. J Psycho-

pharmacol 29: 270–279.

Johnson MW, Garcia-Romeu A, Cosimano MP, et al. (2014) Pilot study

of the 5-HT2AR agonist psilocybin in the treatment of tobacco addic-

tion. J Psychopharmacol 28: 983–992.

Johnson MW, Garcia-Romeu A and Griffiths RR (2017) Long-term

follow-up of psilocybin-facilitated smoking cessation. Am J Drug

Alcohol Abuse 43: 55–60.

Kaeble D, Glaze L, Tsoutis A, et al. (2016) Correctional populations in

the United States, 2014. Washington, DC: United States Department

of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Available at: https://www.bjs.gov/content/pub/pdf/cpus14.pdf

Krebs TS and Johansen PØ (2013) Psychedelics and mental health: A

population study. PLoS One 8: e63972.

Langton L and Truman J (2014) Socio-emotional impact of violent crime.

Special Report. Washington, DC: United States Department of

Justice, Office of Justice Programs, Bureau of Justice Statistics.

Available at: https://www.bjs.gov/content/pub/pdf/sivc.pdf

Leary T (1969) The effects of consciousness-expanding drugs on pris-

oner rehabilitation. Psychedelic Rev 10: 29–45.

Lee MA and Shlain B (1992) Acid Dreams: The Complete Social His-

tory of LSD: The CIA, the Sixties, and Beyond, Revised Edition. New

York: Grove Press.

Lerner M and Lyvers M (2006) Values and beliefs of psychedelic drug

users: A cross-cultural study. J Psychoactive Drugs 38: 143–147.

Lyvers M and Meester M (2012) Illicit use of LSD or psilocybin, but

not MDMA or nonpsychedelic drugs, is associated with mystical

experiences in a dose-dependent manner. J Psychoactive Drugs 44:

410–417.

McCollister KE, French MT and Fang (2010) The cost of crime to soci-

ety: New crime-specific estimates for policy and program evalua-

tion. Drug Alcohol Depend 108: 98–109.

Móró L, Simon K, Bárd I, et al. (2011) Voice of the psychonauts: Coping,

life purpose, and spirituality in psychedelic drug users. J Psychoac-

tive Drugs 43: 188–198.

Nichols DE (2016) Psychedelics. Pharmacol Rev 68: 264–355.

Nour MM, Evans L and Carhart-Harris RL (2017) Psychedelics, person-

ality and political perspectives. J Psychoactive Drugs 49: 182–191.

Oliver ME, Stockdale KC and Wormith JS (2011) A meta-analysis of

predictors of offender treatment attrition and its relationship to recid-

ivism. J Consult Clin Psychol 79: 6–21.

Open Science Collaboration (2015) Estimating the reproducibility of

psychological science. Science 349: aac4716.

Parhar KK, Wormith JS, Derkzen DM, et al. (2008) Offender coercion

in treatment: A meta-analysis of effectiveness. Criminal Justice and

Behavior 35: 1109–1135.

Pearson FS, Lipton DS, Cleland CM, et al. (2002) The effects of behav-

ioral/cognitive-behavioral programs on recidivism. Crime & Delin-

quency 48: 476–496.

Pedersen W and Skardhamar T (2009) Cannabis and crime: Findings

from a longitudinal study. Addiction 105: 109–118.

Perry A, Coulton S, Glanville J, et al. (2006) Interventions for drug-using

offenders in the courts, secure establishments and the community.

Cochrane Database Syst Rev CD005193. doi: 10.1002/14651858.

CD005193.pub3.

Peterson B and Harrell FE (1990) Partial proportional odds models for

ordinal response variables. Applied Statistics 39: 205–217.

Piff PK, Dietze P, Feinberg M, et al. (2015) Awe, the small self, and

prosocial behavior. J Per Soc Psychol 108: 883–899.

Pisano VD, Putnam NP, Kramer HM, et al. (2017) The association of

psychedelic use and opioid use disorders among illicit users in the

United States. J Psychopharmacol 31: 606–613.

Room R and Reuter P (2012) How well do international drug conventions

protect public health? Lancet 379: 84–91.

Rosenberg DE, Isbell H, Miner EJ, et al. (1964) The effect of N,

N-dimethyltryptamine in human subjects tolerant to lysergic acid

diethylamide. Psychopharmacology 5: 217–277.

Ross S, Bossis A, Guss J, et al. (2016) Rapid and sustained symptom

reduction following psilocybin treatment for anxiety and depression

in patients with life-threatening cancer: A randomized controlled

trial. J Psychopharmacol 30: 1165–1180.

Schwalbe CS, Gearing RE, MacKenzie MJ, et al. (2012) A meta-analysis

of experimental studies of diversion programs for juvenile offenders.

Clin Psychol Rev 32: 26–33.

Tenenbaum B (1961) Group therapy with LSD-25. (A preliminary

report). Dis Nerv Syst 22: 459–462.

12 Journal of Psychopharmacology 00(0)

Truman JL and Langton L (2015) Criminal Victimization, 2014. Wash-

ington, DC: United States Department of Justice, Office of Justice

Programs, Bureau of Justice Statistics. Available at: https://www.bjs.

gov/content/pub/pdf/cv14.pdf

United States Department of Health and Human Services. Substance

Abuse and Mental Health Services Administration. Center for

Behavioral Health Statistics and Quality (2003) National Survey on

Drug Use and Health, 2002. Ann Arbor, MI: Inter-university Con-