Psychopharmacology (2019) 236:2677–2685 https://doi.org/10.1007/s00213-019-05238-8 ORIGINAL INVESTIGATION Psychopathological symptoms associated with synthetic cannabinoid use: a comparison with natural cannabis Vincent T. Mensen 1 & Annabel Vreeker 2 & Johan Nordgren 3 & Amanda Atkinson 4 & Rafael de la Torre 5 & Magi Farré 5,6 & Johannes G. Ramaekers 7 & Tibor M. Brunt 8,9 Received: 6 November 2018 / Accepted: 22 March 2019 / Published online: 9 April 2019 # The Author(s) 2019 Abstract Background Synthetic cannabinoids (SCs) are a class of new psychoactive substances that have been rapidly evolving around the world throughout recent years. Many different synthetic cannabinoid analogues are on the consumer market and sold under misleading names, like Bspice^ or Bincense.^ A limited number of studies have reported serious health effects associated with SC use. In this study, we compared clinical and subclinical psychopathological symptoms associated with SC use and natural cannabis (NC) use. Methods A convenience sample of 367 NC and SC users was recruited online, including four validated psychometric question- naires: The Drug Use Disorders Identification Test (DUDIT), Insomnia Severity Index (ISI), Altman Mania Scale (Altman), and Brief Symptom Inventory (BSI). The two groups were compared with analysis of variance (ANOVA) and covariance (ANCOVA), chi2 tests, and logistic regression when appropriate. Results The SC user group did not differ in age from the NC user group (27.7 years), but contained less females (21% and 30%, respectively). SC users scored higher than NC users on all used psychometric measures, indicating a higher likelihood of drug abuse, sleep problems, (hypo)manic symptoms, and the nine dimensions comprising the BSI, somatization, obsessive- compulsive behavior, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism. Odds ratios (95% CI) for the SC user group vs NC user group were, respectively, drug dependence 3.56 (1.77–7.16), (severe) insomnia 5.01 (2.10–11.92), (hypo-)mania 5.18 (2.04–13.14), and BSI psychopathology 5.21 (2.96–9.17). Discussion This study shows that SC use is associated with increased mental health symptomatology compared to NC use. Keywords Cannabis . Synthetic cannabis . Spice . Mental health . Psychology . Psychiatry . Questionnaire . BSI . ISI . Altman . DUDIT Introduction recreational drugs on the consumer market, mostly through the internet (European Monitoring Centre for Drugs and There has been a dramatic increase in the diversity of new Drug Addiction 2015, 2017). The number of different NPS psychoactive substances (NPS) that are being sold as that has been reported to the early warning system of the * Tibor M. Brunt 6 Clinical Pharmacology, Hospital Universitari Germans Trias i Pujol, orbit.biomed@gmail.com Badalona, Universitat Autònoma de Barcelona, Bellaterra, Spain 7 1 Department of Neuropsychology and Psychopharmacology, Faculty Department of Drug Monitoring, Utrecht, The Netherlands of Psychology and Neuroscience, Maastricht University, 2 University Medical Center Utrecht, Department of Psychiatry, Maastricht, The Netherlands University of Utrecht, Utrecht, The Netherlands 8 Department of Psychiatry, Academic Medical Centre, University of 3 Department of Social Work, Malmö University, Malmö, Sweden Amsterdam, Meibergdreef 9, Amsterdam, The Netherlands 9 4 Public Health Institute, Liverpool John Moores University, Department of Developmental Psychopathology, Radboud Liverpool, UK University, Nijmegen, The Netherlands 5 Integrative Pharmacology and Systems Neurosciences, Hospital del Mar Medical Research Institute (IMIM), Barcelona, Spain 2678 Psychopharmacology (2019) 236:2677–2685 European Union is already up to 600 since 2005 and is still toxicological screens in saliva or urine. In the USA, the num- increasing. Whereas the diffusion of NPS on the consumer ber of emergency department visits related to SCs increased market initially started out with stimulants, such as synthetic from 11,406 to 28,531 between 2010 and 2011 (Bush and cathinones (Bossong et al. 2005; Brunt et al. 2011), these were Woodwell 2013). In Europe, SCs were reported in 0.3% of followed by a wide array of other chemicals belonging to a all drug toxicity cases in 16 emergency rooms (Dines et al. range of drug classes. These include hallucinogens such as n- 2015). Most patients present with relatively mild physical benzyl-oxy-methyl derivatives (NBOMes) and the 2C-class (nausea, shortness of breath, hypertension) and mental (rest- drugs (Burns et al. 2014; Caudevilla-Gálligo et al. 2012; lessness, agitation, and anxiety) symptoms and recover within Hondebrink et al. 2015), entactogens such as benzofurans 24 h with supportive care. In some cases, however, patients (Hondebrink et al. 2015; Soh and Elliott 2014) and synthetic are hospitalized for serious physical (myocardial infarction, opioids and the cannabis analogues, synthetic cannabinoids stroke, seizures, kidney injury) and mental (psychosis, para- (SCs) (Martinotti et al. 2017). noia) complications associated with SC use (Castaneto et al. SCs refer to herbal or powder preparations containing syn- 2014; Musshoff et al. 2014; Tait et al. 2016). While causality thetic cannabinoid receptor agonists. SCs were originally devel- is difficult to establish, and many SC users also use other oped in order to investigate the selective therapeutic effects of drugs (Winstock and Barratt 2013), case studies suggested Δ9-tetrahydrocannabinol (THC) and the role of the that SC use is able to trigger long-term psychotic symptoms endocannabinoid system by selectively targeting cannabinoid in otherwise healthy individuals (Fattore 2016). Furthermore, receptors (Pertwee 2006; Huffman et al. 1996). In fact, some among psychiatric patients, SC users presented with more potential for therapeutic purposes was found for certain synthet- severe psychotic symptoms and agitation compared to natural ic cannabinoid agonists and antagonists (De Luca and Fattore cannabis (NC) users (Bassir Nia et al. 2016; Shalit et al. 2016). 2018; Muller et al. 2019). The SCs currently used recreationally While this data from emergency rooms and clinics highlights are potent cannabinoid receptor agonists often synthesized in the risks associated with SCs, little is known about the majority clandestine laboratories in China with a much higher receptor of recreational users, not admitted in any health care system. A affinity and more intense psychoactive effects than THC, which couple of studies have appeared recently, indicating a higher is a partial agonist of these receptors (Castaneto et al. 2014; proneness to abuse or risk-taking behavior in adolescent users ElSohly et al. 2014; European Monitoring Centre for Drugs of SC compared to users of natural cannabis (Clayton et al. and Drug Addiction 2017; Hoffman et al. 2017; Schifano 2017; Blevins et al. 2016). From online discussions we know et al. 2015; van Amsterdam et al. 2015). that recreational users report acute symptoms very similar to SCs form the largest category of NPS: they represent 32% those observed in emergency rooms, both physically (tachycar- of all NPS and a total of over 240 different SCs have been dia, respiratory issues, nausea) and mentally (fear, memory im- reported to the United Nations Office on Drugs and Crime by pairment), in addition to sub-acute (hangovers, lethargy) and 65 member states (United Nations Office on Drugs and Crime longer-term issues (dependence, memory impairment, mood 2017). In addition, the quantity of SCs seized between 2010 swings) (Soussan and Kjellgren 2014). This study aims to and 2015 showed a sharp increase. SCs are usually sold in quantify a wide range of psychopathological symptoms in a colorful packaging marked with labels to suggest that it is not group of recreational SC users in a non-clinical setting and for human consumption, such as Bincense^ or Bspice^ in order compare them to natural cannabis (NC) users. to mislead and circumvent drug laws. Different countries have different laws concerning SCs, sometimes via compound- specific bans or via generic cannabinoid analogue legislation Methods by a blanket ban. SCs are still easily obtained however, espe- cially via the internet (European Monitoring Centre for Drugs Sample and Drug Addiction 2015, 2017). SCs are mainly used out of curiosity, to avoid detection in routine drug tests, and because This convenience sample included SC and NC users, re- of availability and non-illegality (Vandrey et al. 2012; cruited in three European countries (United Kingdom, Winstock and Barratt 2013). Sweden, and the Netherlands). The subjects have been The relatively high potency compared to natural cannabis contacted through social media (Facebook, Twitter) and and increasing prevalence of use of SCs together with an by drug-related forums that are the most used sales/ unfamiliarity with their effects have led to an increase in the diffusion channels, as reported by a review (Miliano et al. number of cases with acute SC intoxication around the world. 2018). The online drug-related forums included www. In Russia, Poland, and the USA, SCs have been associated flashback.org and www.bluelight.org, from January to with poisoning and deaths (Adamowicz 2016; Law et al. June 2017. As prevalence of SC use is rather uncertain, we 2015; Shevyrin et al. 2015). Systematic data on prevalence directed most effort into recruiting users of SC through the is scarce, partly because SCs do not show up in most standard various nightlife and drug-related forums and social media. Psychopharmacology (2019) 236:2677–2685 2679 Inclusion criteria for the participants were that they had the NC users. We compared the SC and NC user groups on age used either natural cannabis and/or synthetic cannabinoids in using ANOVA, and gender, frequency of use, and use of other the past 12 months and were 18 years of age or older. drugs using chi2 tests. We then compared both groups on the Participants were obligated to give online informed consent primary outcome measures (DUDIT total score, ISI total score, before filling out the survey and time to complete the survey Altman total score, and the BSI global severity index, or GSI) took approximately 25 min. The study was approved by the using ANCOVA with the primary outcome measure as depen- research ethical committee of Maastricht University. dent variable, group (SC vs NC) as independent variable, and age and gender as covariates. The nine categories in the BSI Survey and psychometric measures were then analyzed separately using this same method. First of all, the survey started with some demographic items, Clinical outcome analysis like age, gender, education, work status, and housing status. The next part dealt with substance use history, in which last To analyze the relative risk of clinically relevant mental health year use and past use (last 2 weeks) was asked for several problems of SC use compared to NC use we investigated the substances, including alcohol and tobacco. categorical outcome measures of the questionnaires. To get a broad overview of any potential psychological prob- The DUDIT has three outcome categories related to the lems, four standardized psychometric instruments were used, total score: no problems (< 6), drug abuse (6–19), and drug based on symptoms that were frequently mentioned in relationship dependence (20+). We compared the no problem group to the to (synthetic) cannabinoids in scientific literature (e.g., depression, other two groups. psychosis, anxiety, sleep disorders, abuse, and dependence): The ISI has four outcome categories related to the total To assess the level of substance abuse problems in both score: no significant insomnia (< 8), subthreshold insomnia groups, we used the 11-item Drug Use Disorders (8–14), clinical insomnia (15–21), and severe clinical insom- Identification Test (DUDIT) (Berman et al. 2005; Hildebrand nia (22+). We compared the participants with clinical and 2015). The DUDIT consists of nine questions on a 5-point severe clinical insomnia groups to the participants with no or scale, and two questions on a 3-point scale, all of which con- subthreshold insomnia. tribute to a total score which can fall in three categories: no The Altman mania scale has two outcome categories: no problems, substance abuse (6+), or substance dependence manic symptoms (< 6) or (hypo)manic symptoms (6+). These (20+). To assess any problems surrounding sleep disturbances outcome categories were compared. and insomnia, the Insomnia Severity Index (ISI) was used The BSI judges an individual to be a potential psychiatric (Morin et al. 2011). This is a 7-item questionnaire where the case with a GSI score equal to or greater than a T-score of 63. participants rate sleep-related problems on a 5-point scale For adult non-patients this translates to a GSI score of 0.58 for which all contribute to a total score which can be categorized men and 0.78 for women according to Appendix A of the BSI as no problems, subthreshold insomnia (8+), clinical insomnia manual (Derogatis 1993). Participants were assigned to two (15+), and severe clinical insomnia (22+). The Altman Self- groups (psychiatric problems or not) accordingly. Rating Mania Scale (Altman) contains 5 questions which are These categories were analyzed using logistic regression, answered on a 5-point scale and is used to assess the presence with user group (SC or NC) as independent variable, clinical and severity of (hypo)manic symptoms (Altman et al. 2001; category as dependent variable, and age and gender as covar- Altman et al. 1997). The total score on these 5 questions can iates. Bias-corrected accelerated bootstrapping was used to be categorized as no symptoms, or a high probability of a produce confidence intervals. (hypo)manic condition (6+). To gain insight into any other mental problems, the Brief Symptom Inventory (BSI) was used (Derogatis 1993). It contains 53 questions which are answered Results on a 5-point Likert scale. Each question contributes to a global severity index as well as one of nine psychopathological dimen- Participants sions: somatization, obsessive-compulsive, interpersonal sensi- tivity, depression, anxiety, hostility, phobic anxiety, paranoid The total number of responses was 589 SC users and 417 NC ideation, and psychoticism. users. From this total, cases were removed if they did not give explicit consent, if they did not provide an age over 18 or Statistics gender, if they did not complete at least one of the question- naires, and if they did not explicitly state they had used NC First, we compared the SC and NC user groups to each other on (for the NC users) or SC (for the SC users) in the past year. age using analysis of variance (ANOVA) and gender using a After removing these cases, we included 367 participants from chi2 test. For our main analyses, we compared the SC users to three different countries: 238 SC users and 129 NC users (see 2680 Psychopharmacology (2019) 236:2677–2685 Table 1). For all analyses, cases with missing data on any 1.14), NC 0.38 (95% CI 0.31–0.44). This was the case for all measure were pairwise excluded. nine BSI categories underlying the GSI (Table 5). Age and gender Clinical outcome analysis There was no difference in age between the SC and NC user Using logistic regression to examine the clinical outcome mea- groups (F(1,365) = 0.575, p = 0.449). There were fewer wom- sures of the psychometric instruments, we found that partici- en in the SC user group (21%) than in the NC user group pants in the SC user group were between 3.56 and 5.21 times (30%) (chi2 = 8.308, p = 0.016). more likely than participants in the NC user group to score in the clinical range of the psychometric instruments (Table 6). Substance use There was a difference in frequency of use (chi2 = 100.029, Discussion p < 0.001), with significantly fewer SC users (30%) reporting monthly or more frequent SC use than the NC use among the In this study we found that compared to NC use, SC use is NC users group (84%). There was no difference in duration of more strongly associated with a broad range of self-reported use (as calculated by the intervals for age of onset and current mental health problems. The SC user group scored significant- age of SC use and NC use) between the NC user group and the ly higher on all measured indices of general and specific psy- SC user group. chopathology, and odds of scoring in the clinical range on any Use of 12 psychoactive drugs in the previous year and past of our measurements were 3.5 to over 5 times higher in the SC 2 weeks was assessed. NC use was also highly prevalent in the user group compared to the NC user group. These results SC user group; SC use was non-prevalent in the NC user warrant serious concern about SC use. group however. NC use was lower in the SC user group only Few studies have attempted to investigate mental health in the last 2 weeks of reported substance use. In addition, problems associated with SC use in a non-clinical population. salvia was used by more SC users in the past 2 weeks and past In a clinical population differences between NC and SC users 12 months (8.5%) than NC users (0%). There was no differ- were observed, indicating an association of SC use with psy- ence between groups for all other substances at a Bonferroni chotic problems, agitation, and longer hospitalizations (Bassir corrected p value of 0.004 (see Tables 2 and 3). Nia et al. 2016; Shalit et al. 2016). While the studied popula- tions are very different, this supports our findings of more Psychometric instruments severe problems related to SC use compared to NC use. On our main index of psychiatric problems, the BSI, SC users The SC user group showed a higher DUDIT total score than have higher scores than NC users on the GSI and all underly- the NC user group (F(1,357) = 21.336, p = 0.000, eta2 = ing categories. The probability (odds ratio) of the GSI to ex- 0.056). Adjusted means: SC user group 13.73 (95% CI ceed the threshold for a psychiatric diagnosis, as is formulated 12.49–14.96), NC user group 9.36 (95% CI 8.30–10.40). in the BSI manual (Derogatis 1993), is five times higher for Likewise, the SC user group showed a higher ISI total score participants in the SC group than the NC group. This is espe- than the NC user group (F(1,222) = 20.052, p < 0.001, eta2 = cially worrying considering that many studies have found that 0.083). Adjusted means: SC 9.66 (95% CI 8.50–10.83), NC NC use itself is already more often associated with psychiatric 5.60 (95% CI 4.61–6.70) (Table 4). problems, including depression, anxiety, and in particular psy- The Altman total score was higher in the SC user group than chotic symptoms and schizophrenia (Moore et al. 2007; the NC user group (F(1,229) = 23.982, p < 0.001, eta2 = 0.095). Volkow et al. 2014). Adjusted means: SC 3.84 (95% CI 3.23–4.43), NC 1.72 (95% Aside from the general psychiatric problems indexed by CI 1.27–2.14). The SC user group showed a higher GSI total the BSI, the results from the other psychometric measures score compared to the NC user group (F(1,289) = 49.936, p = were also in line with previous studies. Cannabis use has been 0.000, eta2 = 0.147). Adjusted means: SC 1.00 (95% CI 0.87– associated with insomnia symptoms and impairment of sleep Table 1 Participant information Substance Age range Mean age Women Men Other (years old) (years old ± SD) SC users (N = 238) 18–66 27.6 (± 8.5) 51 (21%) 177 (74%) 10 (4%) NC users (N = 129) 18–71 27.7 (± 10.2) 39 (30%) 90 (70%) 0 (0%) SD standard deviation, SC synthetic cannabis, NC natural cannabis Psychopharmacology (2019) 236:2677–2685 2681 Table 2 Additional psychoactive drug use in past 12 months Drug NC user group SC user group Chi2 Natural cannabis 100.0% 81.5% 2.561, p = 0.126 Amphetamines 18.7% 31.3% 3.871, p = 0.051 Cocaine/crack 9.0% 15.4% 2.888, p = 0.115 Tranquilizers 9.3% 19.2% 4.925, p = 0.024 Ecstasy/MDMA 5.4% 9.3% 2.056, p = 0.189 LSD/other hallucinogens 4.8% 7.5% 3.974, p = 0.067 Magic mushrooms 0.7% 4.2% 4.041, p = 0.087 GHB/GBL 0.0% 1.9% 4.122, p = 0.065 Ketamine 2.0% 8.1% 6.877, p = 0.106 Amyl nitrate/poppers 4.1% 1.4% 0.711, p = 0.612 Salvia divinorum 0.0% 8.8% 9.129, p = 0.000* Nitrous oxide/laughing gas 4.1% 5.6% 0.874, p = 0.471 Methylone 4.6% 3.8% 2.125, p = 0.231 SC synthetic cannabis, NC natural cannabis *Significant at Bonferroni-corrected p = 0.004 quality in various studies (Babson et al. 2017; P. Gates et al. likely to report a lifetime history of cannabis use (Agrawal 2016; P. J. Gates et al. 2014). Mainly, long-term use of can- et al. 2011). Cannabis use may also cause incidental episodes nabis (mainly the chief component THC) seems to have a of mania or even long-term bipolar disorder, with cannabis use negative impact on sleep quality, whereas some cannabinoids often preceding onset of the first mania (Bally et al. 2014; show some promise for treatment of insomnia (Babson et al. Henquet et al. 2006). Furthermore, NC use was also associat- 2017). However, the causative relationship of cannabis and ed with a younger age of onset of the first mania and an insomnia remains unclear and insomnia might as well be a exacerbation of depressive and manic symptomatology in pa- risk factor for adolescent cannabis use (Roane and Taylor tients diagnosed with bipolar disorder (Bally et al. 2014; 2008). Notwithstanding cause and effect, our study confirms Gibbs et al. 2015). The results of our study underline this a relationship between insomnia symptoms and the use of association between the cannabinoid system and mania, while cannabinoid receptor agonists, whereby use of SCs is associ- demonstrating that SCs show a greater association with mania ated with poorer sleep quality than use of NC. or bipolar disorder. Cannabis use has also been associated with increased risk NC use may also lead to substance use disorders or depen- of mania and subsequent bipolar disorder (Strakowski et al. dence (Hall and Degenhardt 2009). The results of our study 2007). Individuals with bipolar disorder were 6.8 times more suggest a much higher dependence liability of SCs than NC. Table 3 Additional psychoactive drug use in the past 2 weeks Drug NC user group SC user group Chi2 Natural cannabis 74.7% 21.3% 7.862, p = 0.002* Amphetamines 11.0% 21.5% 3.594, p = 0.071 Cocaine/crack 6.1% 12.3% 2.419, p = 0.155 Tranquilizers 7.1% 17.1% 4.634, p = 0.044 Ecstasy/MDMA 3.4% 7.3% 1.735, p = 0.212 LSD/other hallucinogens 2.1% 6.9% 4.865, p = 0.033 Magic mushrooms 0.0% 3.8% 4.512, p = 0.046 GHB/GBL 0.0% 1.4% 3.855, p = 0.085 Ketamine 1.2% 7.3% 6.138, p = 0.120 Amyl nitrate/poppers 4.1% 1.4% 0.560, p = 0.517 Salvia divinorum 0.0% 8.4% 9.929, p = 0.000* Nitrous oxide/laughing gas 3.4% 4.9% 0.982, p = 0.381 Methylone 3.6% 1.6% 6.950, p = 0.013 SC synthetic cannabis, NC natural cannabis *Significant at Bonferroni-corrected p = 0.004 2682 Psychopharmacology (2019) 236:2677–2685 Table 4 Means adjusted for gender and age, between-subject Questionnaire NC adjusted mean SC adjusted mean p Partial eta2 significance, and partial eta squared for the effect of group on DUDIT 9.36 13.73 < 0.001 0.056 the questionnaires ISI 5.60 9.66 < 0.001 0.083 Altman 1.72 3.84 < 0.001 0.095 BSI (GSI) 0.38 1.00 < 0.001 0.147 DUDIT Drug Use Disorders Identification Test, ISI Insomnia Severity Index, BSI Brief Symptom Inventory, GSI General Symptom Inventory, SC synthetic cannabis, NC natural cannabis There is evidence to suggest that SCs, with a much greater SC mixtures or SCs alone do not contain cannabidiol, a com- potency and affinity for the endocannabinoid system, are ca- pound found in NC that is associated with reducing psychotic pable of inducing dependence. For instance, recent neuroim- experiences and anxiety, further increasing the potential risk aging research showed a modest, but significant increase of of mental health problems emerging after the use of SCs (Di striatal dopamine in humans after cannabis use and also im- Forti et al. 2015; Murray et al. 2016; van Amsterdam et al. paired striatal release in dependent cannabis users (Bossong 2015). Even though there is much debate on the causality of et al. 2009; van de Giessen et al. 2017). Our findings of a NC use and psychiatric problems, potent activation of CB1 higher DUDIT score for SC users are supported by a study receptors by increasingly potent agonists is a likely mecha- executed among high school students, showing higher sub- nism to cause long-term brain changes (Haney and Evins stance use disorder psychopathology (Blevins et al. 2016). 2016). Combining our findings of much higher scores on all The higher DUDIT score in the SC user group cannot be indices of psychiatric problems with the knowledge from attributed to frequency of use, which was higher for NC in in vitro and animal studies about SC potency, again strong the NC user group. However, it has to be stipulated that the caution is warranted in the case of SC use. DUDIT is not able to differentiate between which specific While our data clearly reveal that SC use is associated with substance is most likely to be underlying increased DUDIT increased mental health problems, there are some limitations scores and most users are poly-drug users. to the study. First of all, it is a study which relies on effects and Our results on SCs are also supported by in vitro and ani- symptoms that are self-reported on recall basis. And, because mal studies at the moment. The pharmacological effects are it is a cross-sectional study, no statements can be made regard- similar to the main active ingredient of NC: Δ9-THC, but with ing causality as we do not know if SC leads to increased a much greater binding affinity for the cannabinoid receptors, mental health problems, or if mental health problems instigate in particular the cannabinoid 1 receptor (CB1R) (Castaneto SC use. To investigate causality, future studies should employ et al. 2014; Martinotti et al. 2017). This role as a potent a longitudinal design. In addition, the fact that the majority of CB1R agonist has been observed to cause disturbed synaptic SC users also used NC in the past year might substantially functioning, potentially much stronger in SCs than Δ9-THC confound current findings, since effects of NC could be un- which is a partial agonist (Hoffman et al. 2017). Furthermore, derlying effects associated with the SC using group. Still, it does not confound the magnitude in difference in symptoms Table 5 The 95% confidence intervals adjusted for gender and age, found between the two groups. There were no general statis- between-subject significance, and partial eta squared for the effect of tical differences found in most other substances used between group on BSI category score these two groups. However, we cannot discount the role other BSI category NC 95% CI SC 95% CI p Partial eta2 drugs in the development of mental health status of our par- ticipants. Furthermore, recruitment methods might underlie Somatization 0.23–0.36 0.61–0.87 < 0.001 0.088 some differences we found, as different drug-related forums Obsessive-compulsive 0.49–0.72 1.20–1.53 < 0.001 0.131 and social media were used to suit each country, which might Interpersonal 0.30–0.49 0.77–1.10 < 0.001 0.078 have resulted in bias. However, they were selected to optimize sensitivity sample size, and using identical recruitment platforms, if even Depression 0.43–0.70 1.05–1.39 < 0.001 0.088 possible, would have likely led to small sample sizes with its Anxiety 0.24–0.42 1.00–1.32 < 0.001 0.173 own limitations. Since this study is based on data acquired Hostility 0.15–0.28 0.58–0.86 < 0.001 0.089 using an online survey, the veracity of the answers cannot be Phobic anxiety 0.19–0.36 0.77–1.10 < 0.001 0.115 determined. Notwithstanding this argument, online surveys do Paranoid ideation 0.17–0.28 0.75–1.05 < 0.001 0.149 not appear to differ from traditional surveys in this regard Psychoticism 0.27–0.45 0.67–0.98 < 0.001 0.068 (Miller et al. 2002). Finally, we have no information on the BSI Brief Symptom Inventory, CI confidence interval, SC synthetic can- subtype or amount of SC used by the participants. nabis, NC natural cannabis Unfortunately, this mimics the real-world situation and Psychopharmacology (2019) 236:2677–2685 2683 Table 6 The relationship between clinical symptoms in the Clinical measure Questionnaire OR SC vs NC 95% CI p SC group compared to the NC group Drug dependence DUDIT 3.56 1.77–7.16 < 0.001 (Severe) insomnia ISI 5.01 2.10–11.92 < 0.001 (Hypo-)mania Altman 5.18 2.04–13.14 0.001 Psychiatric problems BSI 5.21 2.96–9.17 < 0.001 Logistic regression with age and gender as covariates OR odds ratio, CI confidence interval, DUDIT Drug Use Disorders Identification Test, ISI Insomnia Severity Index, BSI Brief Symptom Inventory highlights an additional danger of SC use. There is significant Altman EG, Hedeker D, Peterson JL, Davis JM (1997) The Altman Self- Rating Mania Scale. Biol Psychiatry 42(10):948–955. https://doi. variance of subtype and quantity of SCs, even between pack- org/10.1016/S0006-3223(96)00548-3 ages with the same name (Fattore 2016). Altman EG, Hedeker D, Peterson JL, Davis JM, McAllister T, Alahi P Our study demonstrates that there are significant mental et al (2001) A comparative evaluation of three self-rating scales for health problems in this population of SC users, which should acute mania. Biol Psychiatry 50(6):468–471. https://doi.org/10. be cause for concern for treatment and prevention profes- 1016/S0006-3223(01)01065-4 Babson KA, Sottile J, Morabito D (2017) Cannabis, cannabinoids, and sionals. Use of SCs is associated with a probability for mental sleep: a review of the literature. Curr Psychiatry Rep 19(4):23. health risks up to over five times greater than NC use, which is https://doi.org/10.1007/s11920-017-0775-9 in line with previous results assembled in clinical and preclin- Bally N, Zullino D, Aubry J-M (2014) Cannabis use and first manic ical (experimental animal) studies with SCs which showed a episode. J Affect Disord 165(August):103–108. https://doi.org/10. much more potent mechanism of action than NC. This study 1016/j.jad.2014.04.038 Bassir Nia A, Medrano B, Perkel C, Galynker I, Hurd YL (2016) should be considered as a strong caution about the potential Psychiatric comorbidity associated with synthetic cannabinoid use danger of SC use to mental health. compared to cannabis. J Psychopharmacol 30(12):1321–1330. https://doi.org/10.1177/0269881116658990 Acknowledgments We want to thank all participants in this study. Berman AH, Bergman H, Palmstierna T, Schlyter F (2005) Evaluation of the Drug Use Disorders Identification Test (DUDIT) in criminal jus- Funding This study was funded by the European Commission (Drugs tice and detoxification settings and in a Swedish population sample. Policy Initiatives, Justice Programme 2014–2020, contract no. Eur Addict Res 11(1):22–31. https://doi.org/10.1159/000081413 HOME/2014/JDRU/AG/DRUG/7082, Predicting Risk of Emerging Blevins CE, Banes KE, Stephens RS, Walker DD, Roffman RA (2016) A Drugs With In Silico and Clinical Toxicology, PREDICT project). M. preliminary evaluation of synthetic cannabinoid use among adoles- Farré was granted by Instituto de Salud Carlos III (ISCIII-FIS-FEDER, cent cannabis users: characteristics and treatment outcomes. Addict PI14/00715). Behav 63:114–119 Bossong MG, Van Dijk JP, Niesink RJM (2005) Methylone and mCPP, two new drugs of abuse? Addict Biol 10(4):321–323. https://doi. Compliance with ethical standards org/10.1080/13556210500350794 Bossong MG, van Berckel BNM, Boellaard R, Zuurman L, Schuit RC, Conflicts of interest The authors declare that they have no conflict of Windhorst AD, van Gerven JMA, Ramsey NF, Lammertsma AA, interest. Kahn RS (2009) Delta 9-tetrahydrocannabinol induces dopamine release in the human striatum. Neuropsychopharmacology 34(3): Open Access This article is distributed under the terms of the Creative 759–766. https://doi.org/10.1038/npp.2008.138 Commons Attribution 4.0 International License (http:// Brunt TM, Poortman A, Niesink RJ, van den Brink W (2011) Instability creativecommons.org/licenses/by/4.0/), which permits unrestricted use, of the ecstasy market and a new kid on the block: mephedrone. J distribution, and reproduction in any medium, provided you give appro- Psychopharmacol 25(11):1543–1547. https://doi.org/10.1177/ priate credit to the original author(s) and the source, provide a link to the 0269881110378370 Creative Commons license, and indicate if changes were made. Burns L, Roxburgh A, Matthews A, Bruno R, Lenton S, Van Buskirk J (2014) The rise of new psychoactive substance use in Australia. Drug Test Anal 6(7–8):846–849. https://doi.org/10.1002/dta.1626 Bush DM, Woodwell DA (2013) Update: drug-related emergency depart- ment visits involving synthetic cannabinoids. The CBHSQ Report References Castaneto MS, Gorelick DA, Desrosiers NA, Hartman RL, Pirard S, Huestis MA (2014) Synthetic cannabinoids: epidemiology, pharma- Adamowicz P (2016) Fatal intoxication with synthetic cannabinoid codynamics, and clinical implications. Drug Alcohol Depend 144: MDMB-CHMICA. Forensic Sci Int, 261(July 2015), e5–10. 12–41. https://doi.org/10.1016/j.drugalcdep.2014.08.005 https://doi.org/10.1016/j.forsciint.2016.02.024 Caudevilla-Gálligo F, Riba J, Ventura M, González D, Farré M, Barbanoj Agrawal A, Nurnberger JI, Lynskey MT, Bipolar Genome Study (2011) MJ, Bouso JC (2012) 4-Bromo-2,5-dimethoxyphenethylamine (2C- Cannabis involvement in individuals with bipolar disorder. B): presence in the recreational drug market in Spain, pattern of use Psychiatry Res 185(3):459–461. https://doi.org/10.1016/j.psychres. and subjective effects. J Psychopharmacol 26(7):1026–1035. https:// 2010.07.007 doi.org/10.1177/0269881111431752 2684 Psychopharmacology (2019) 236:2677–2685 Clayton HB, Lowry R, Ashley C, Wolkin A, Grant AM (2017) Health Huffman JW, Yu S, Showalter V, Abood ME, Wiley JL, Compton DR, risk behaviors with synthetic cannabinoids versus marijuana. Martin BR, Bramblett RD, Reggio PH (1996) Synthesis and phar- Pediatrics 139(4):e20162675 macology of a very potent cannabinoid lacking a phenolic hydroxyl De Luca MA, Fattore L (2018) Therapeutic use of synthetic cannabi- with high affinity for the CB2 receptor. J Med Chem 39(20):3875– noids: still an open issue? Clin Ther 40(9):1457–1466. https://doi. 3877 org/10.1016/j.clinthera.2018.08.002 Law R, Schier J, Martin C, Chang A, Wolkin A, Centers for Disease Derogatis LR (1993) Brief Symptom Inventory: BSI; administration, Control (CDC) (2015) Notes from the field: increase in reported scoring, and procedures manual. Pearson adverse health effects related to synthetic cannabinoid use— Di Forti M, Marconi A, Carra E, Fraietta S, Trotta A, Bonomo M et al United States, January–May 2015. Morb Mortal Wkly Rep (2015) Proportion of patients in South London with first-episode 64(22):618–619 psychosis attributable to use of high potency cannabis: a case- Martinotti G, Santacroce R, Papanti D, Elgharably Y, Prilutskaya M, control study. Lancet Psychiatry 2(3):233–238. https://doi.org/10. Corazza O (2017) Synthetic cannabinoids: psychopharmacology, 1016/S2215-0366(14)00117-5 clinical aspects, psychotic onset. CNS Neurol Disord Drug Targets Dines AM, Wood DM, Yates C, Heyerdahl F, Hovda KE, Giraudon I et al 16(5):567–575. https://doi.org/10.2174/ (2015) Acute recreational drug and new psychoactive substance 1871527316666170413101839 toxicity in Europe: 12 months data collection from the European Miliano C, Margiani G, Fattore L, De Luca MA (2018) Sales and adver- Drug Emergencies Network (Euro-DEN). Clin Toxicol (Phila) tising channels of new psychoactive substances (NPS): internet, so- 53(9):893–900. https://doi.org/10.3109/15563650.2015.1088157 cial networks, and smartphone apps. Brain Sci 8(7). https://doi.org/ ElSohly MA, Gul W, Wanas AS, Radwan MM (2014) Synthetic canna- 10.3390/brainsci8070123 binoids: analysis and metabolites. Life Sci 97(1):78–90. https://doi. Miller ET, Neal DJ, Roberts LJ, Baer JS, Cressler SO, Metrik J, Marlatt org/10.1016/j.lfs.2013.12.212 GA (2002) Test-retest reliability of alcohol measures: is there a European Monitoring Centre for Drugs and Drug Addiction (2015) New difference between internet-based assessment and traditional psychoactive substances in Europe. Lisbon. https://doi.org/10.2810/ methods? Psychol Addict Behav 16(1):56–63. https://doi.org/10. 372415 1037/0893-164X.16.1.56 European Monitoring Centre for Drugs and Drug Addiction (2017) Moore THM, Zammit S, Lingford-Hughes A, Barnes TRE, Jones PB, Perspectives on drugs: synthetic cannabinoids in Europe Burke M, Lewis G (2007) Cannabis use and risk of psychotic or Fattore L (2016) Synthetic cannabinoids—further evidence supporting affective mental health outcomes: a systematic review. Lancet the relationship between cannabinoids and psychosis. Biol (London, England) 370(9584):319–328. https://doi.org/10.1016/ Psychiatry 79(7):539–548. https://doi.org/10.1016/j.biopsych. S0140-6736(07)61162-3 2016.02.001 Morin CM, Belleville G, Bélanger L, Ivers H (2011) The insomnia se- Gates PJ, Albertella L, Copeland J (2014) The effects of cannabinoid verity index: psychometric indicators to detect insomnia cases and administration on sleep: a systematic review of human studies. evaluate treatment response. Sleep 34(5):601–608. https://doi.org/ Sleep Med Rev 18(6):477–487. https://doi.org/10.1016/j.smrv. 10.1093/sleep/34.5.601 2014.02.005 Muller C, Morales P, Reggio PH (2019) Cannabinoid ligands targeting Gates P, Albertella L, Copeland J (2016) Cannabis withdrawal and sleep: TRP channels. Front Mol Neurosci, 11(487). https://doi.org/10. a systematic review of human studies. Subst Abus 37(1):255–269. 3389/fnmol.2018.00487 https://doi.org/10.1080/08897077.2015.1023484 Murray RM, Quigley H, Quattrone D, Englund A, Di Forti M (2016) Gibbs M, Winsper C, Marwaha S, Gilbert E, Broome M, Singh SP (2015) Traditional marijuana, high-potency cannabis and synthetic canna- Cannabis use and mania symptoms: a systematic review and meta- binoids: increasing risk for psychosis. World Psychiatry 15(3):195– analysis. J Affect Disord 171:39–47. https://doi.org/10.1016/j.jad. 204. https://doi.org/10.1002/wps.20341 2014.09.016 Musshoff F, Madea B, Kernbach-Wighton G, Bicker W, Kneisel S, Hutter Hall W, Degenhardt L (2009) Adverse health effects of non-medical M, Auwärter V (2014) Driving under the influence of synthetic cannabis use. Lancet (London, England) 374(9698):1383–1391. cannabinoids (Bspice^): a case series. Int J Legal Med 128(1):59– https://doi.org/10.1016/S0140-6736(09)61037-0 64. https://doi.org/10.1007/s00414-013-0864-1 Haney M, Evins AE (2016) Does cannabis cause, exacerbate or amelio- Pertwee RG (2006) The pharmacology of cannabinoid receptors and their rate psychiatric disorders? An oversimplified debate discussed. ligands: an overview. Int J Obes 30(Suppl 1):S13–S18. https://doi. Neuropsychopharmacology 41(2):393–401. https://doi.org/10. org/10.1038/sj.ijo.0803272 1038/npp.2015.251 Roane BM, Taylor DJ (2008) Adolescent insomnia as a risk factor for Henquet C, Krabbendam L, de Graaf R, ten Have M, van Os J (2006) early adult depression and substance abuse. Sleep 31(10):1351– Cannabis use and expression of mania in the general population. J 1356 Affect Disord 95(1–3):103–110. https://doi.org/10.1016/j.jad.2006. Schifano F, Orsolini L, Duccio Papanti G, Corkery JM (2015) Novel 05.002 psychoactive substances of interest for psychiatry. World Hildebrand M (2015) The psychometric properties of the drug use disor- Psychiatry 14(1):15–26. https://doi.org/10.1002/wps.20174 ders identification test (DUDIT): a review of recent research. J Subst Shalit N, Barzilay R, Shoval G, Shlosberg D, Mor N, Zweigenhaft N, Abus Treat 53(January 2005):52–59. https://doi.org/10.1016/j.jsat. Weizman A, Krivoy A (2016) Characteristics of synthetic cannabi- 2015.01.008 noid and cannabis users admitted to a psychiatric hospital: a com- Hoffman AF, Lycas MD, Kaczmarzyk JR, Spivak CE, Baumann MH, parative study. J Clin Psychiatry 77(8):e989–e995. https://doi.org/ Lupica CR (2017) Disruption of hippocampal synaptic transmission 10.4088/JCP.15m09938 and long-term potentiation by psychoactive synthetic cannabinoid Shevyrin V, Melkozerov V, Nevero A, Eltsov O, Shafran Y, Morzherin Y, Bspice^ compounds: comparison with Δ(9)-tetrahydrocannabinol. Lebedev AT (2015) Identification and analytical characteristics of Addict Biol 22(2):390–399. https://doi.org/10.1111/adb.12334 synthetic cannabinoids with an indazole-3-carboxamide structure Hondebrink L, Nugteren-van Lonkhuyzen JJ, van der Gouwe D, Brunt bearing a N-1-methoxycarbonylalkyl group. Anal Bioanal Chem TM (2015) Monitoring new psychoactive substances (NPS) in the 407(21):6301–6315. https://doi.org/10.1007/s00216-015-8612-7 Netherlands: data from the drug market and the Poisons Information Soh YNA, Elliott S (2014) An investigation of the stability of emerging Centre. Drug Alcohol Depend 147(January):109–115. https://doi. new psychoactive substances. Drug Test Anal 6(7–8):696–704. org/10.1016/j.drugalcdep.2014.11.033 https://doi.org/10.1002/dta.1576 Psychopharmacology (2019) 236:2677–2685 2685 Soussan C, Kjellgren A (2014) The flip side of Bspice^: the adverse van de Giessen E, Weinstein JJ, Cassidy CM, Haney M, Dong Z, effects of synthetic cannabinoids as discussed on a Swedish Ghazzaoui R, Abi-Dargham A (2017) Deficits in striatal dopamine Internet forum. Nordic Stud Alcohol Drugs 31(2):207–220. https:// release in cannabis dependence. Mol Psychiatry 22(1):68–75. doi.org/10.2478/nsad-2014-0016 https://doi.org/10.1038/mp.2016.21 Strakowski SM, DelBello MP, Fleck DE, Adler CM, Anthenelli RM, Vandrey R, Dunn KE, Fry JA, Girling ER (2012) A survey study to Keck PE, Arnold LM, Amicone J (2007) Effects of co-occurring characterize use of spice products (synthetic cannabinoids). Drug cannabis use disorders on the course of bipolar disorder after a first Alcohol Depend 120(1–3):238–241. https://doi.org/10.1016/j. hospitalization for mania. Arch Gen Psychiatry 64(1):57–64. https:// drugalcdep.2011.07.011 doi.org/10.1001/archpsyc.64.1.57 Volkow ND, Baler RD, Compton WM, Weiss SRB (2014) Adverse Tait RJ, Caldicott D, Mountain D, Hill SL, Lenton S (2016) A systematic health effects of marijuana use. N Engl J Med 370(23):2219– review of adverse events arising from the use of synthetic cannabi- 2227. https://doi.org/10.1056/NEJMra1402309 noids and their associated treatment. Clin Toxicol 54(1):1–13. Winstock AR, Barratt MJ (2013) Synthetic cannabis: a comparison of https://doi.org/10.3109/15563650.2015.1110590 patterns of use and effect profile with natural cannabis in a large United Nations Office on Drugs and Crime. (2017) World Drug Report global sample. Drug Alcohol Depend 131(1–3):106–111. https:// 2017 doi.org/10.1016/j.drugalcdep.2012.12.011 van Amsterdam J, Brunt T, van den Brink W (2015) The adverse health effects of synthetic cannabinoids with emphasis on psychosis-like Publisher’s note Springer Nature remains neutral with regard to effects. J Psychopharmacol 29(3):254–263. https://doi.org/10.1177/ jurisdictional claims in published maps and institutional affiliations. 0269881114565142