Diffusion of Ecstasy in the Electronic Dance Music Scene

Abstract

Background:

Ecstasy (MDMA, Molly) is among the most prevalent drugs used by people who attend electronic dance music (EDM) events; however, little recent research has examined diffusion of ecstasy in this high-risk population.

Methods:

1,020 EDM event-attending adults (aged 18–40) were surveyed in NYC in 2018 using time-spacing sampling. Participants were asked about past-year ecstasy use, and those reporting use were asked where they initiated use and whether their first use was planned. They were also asked whether they have ever given someone their first dose and whether they were likely to use in the future. Prevalence and correlates of these outcomes were estimated among those reporting past-year use.

Results:

An estimated 31.0% of participants used ecstasy in the past year. Of these, 42.9% used ecstasy for the first time in an unplanned manner and initiation most commonly occurred at EDM festivals (33.4%), followed by nightclubs (24.3%). An estimated 39.4% reported having given someone their first dose of ecstasy and 60.2% reported being likely to use again. Hispanic and Asian participants were at higher risk for their first dose being unplanned, and those who used LSD in the past year were at higher risk for giving someone their first dose (aPR = 1.78, 95% CI = 1.20–2.65). Older participants (aPR = 1.03, 95% CI = 1.01–1.06), females (aPR = 1.46, 95% CI = 1.13–1.89), and those who used LSD in the past year (aPR = 1.42, 95% CI = 1.14–1.76) were more likely to report being likely to use again in the future.

Conclusions:

Results provide insight regarding diffusion and initiation of ecstasy in the EDM scene. Findings can inform prevention and harm reduction efforts.

Introduction

Ecstasy (3,4-methylenedioxymethamphetamine [MDMA]) has been among the most prevalent party drugs or “club drugs” over the past three decades (Schulenberg et al., 2019). In the United States (US), a tenth (10.5%) of young adults (ages 18–25) are estimated to have ever used ecstasy in their lifetime, and 3.1% of young adults are estimated to have used in the past year (Center for Behavioral Health Statistics & Quality, 2019). However, estimates are exponentially higher among electronic dance music (EDM) event attendees. For example, a national survey of dance festival-attending adults in Australia (which was not limited to EDM music) found that 78% of adults sampled reported having used an illegal drug at their last dance festival with 85% of these attendees using ecstasy (Hughes et al., 2017). Other recent studies estimate lifetime and past-year use among EDM event attendees in New York City (NYC) being 43.5% and 26.5%, respectively (Griffin et al., 2019; Palamar, 2017). Alarmingly, a fifth (20.3%) of such past-year users in the EDM scene in NYC have experienced an adverse effect after use in the past year (Palamar et al., 2019), and deaths related to ecstasy use in England and Wales reached an all-time high of 92 deaths in 2018 (Office for National Statistics, 2019). Since popularity of EDM is increasing (Watson, 2019) and use is currently associated with high prevalence of adverse outcomes including death (Office for National Statistics, 2019; Palamar et al., 2019), more research is needed on diffusion of ecstasy in this high-risk scene.

Ecstasy first boomed in popularity in dance music scenes in the early 1980s in the state of Texas in the US (Collin, 2009). Throughout the 1980s and 1990s, ecstasy diffused across multiple dance scenes and across continents (Collin, 2009). While indeed use appears to be most prevalent among EDM event attendees, use has also diffused beyond dance scenes into various youth networks, into urban neighborhoods, and into the general population (Collin, 2009; Golub et al., 2005; Schensul et al., 2005). The Diffusion of Innovation theory (Rogers, 2010) posits that ideas gain momentum and diffuse into a system depending on “innovators” ––in this case, those who use and/or promote use of ecstasy. Ecstasy is promoted both directly and indirectly via social influence by people in party scenes who might be viewed as role models (Schensul et al., 2005). Ecstasy use can also be viewed as a “contagion” in which those exposed to users can be seen as “susceptible” to use, and use can be “transmitted” through peers within and across social networks (Ferrence, 2001). Since use is not a transmittable disease, however, these susceptible individuals are not passive victims of drug initiation. People have choices regarding whether they initiate use, but such subcultures with normative social influence strongly affect individuals’ decisions regarding use (Golub et al., 2005; Vervaeke et al., 2008).

In theory, eventually everyone who frequents such scenes will have opportunity to use (Golub et al., 2005), and any attendance, especially more frequent attendance, is associated with increased risk for initiation/use (Abrahamsson & Hakansson, 2013; Golub et al., 2005; Palamar et al., 2017; Smirnov et al., 2013a, 2013b). Attendance of such high-risk venues can also link different types of individuals or individuals from different locations (e.g. urban and suburban) and lead to increased risk of exposure and initiation among these different groups (Schensul et al., 2005). Exposure to users is a strong risk factor for use (Palamar et al., 2017), and exposure is also associated with decreased stigma toward users, which is further associated with increased risk for use (Palamar et al., 2011, 2012). While entering a scene where ecstasy is available and/or promoted is a risk factor for use, individuals do still need to come into direct contact with users to be directly exposed to the drug (e.g. drug offers; although it should be noted that drugs such as ecstasy can now be purchased online from vendors). Most people who use report that most of their friends also use and ecstasy is most often first obtained from friends (Sherlock & Conner, 1999; Smirnov et al., 2013a, 2013b). People who use also typically have friends who approve of use (Martins et al., 2008a, 2008b; Vervaeke et al., 2008) and friends are often considered among the most accurate sources of information about ecstasy (Falck et al., 2004; Jacinto et al., 2008).

While most epidemiological research on ecstasy has examined risk factors for use, little research has focused on risk for initiating others into use. This is important to examine because peer influence is reciprocal as people who begin using often introduce others to the drug (Vervaeke et al., 2008). Ecstasy use can be promoted and/or offered to others by people who use. With regard to offers, sometimes ecstasy is offered to an individual for free or at reduced price (especially one’s first ever dose), and friends may persuade individuals into using it when they lack desire to initiate use or are reluctant to use (Schensul et al., 2005). Regardless of whether individuals were initially reluctant to use, ecstasy is typically first obtained from an individual one trusts, whether it be a friend or seller, and the individual providing this first dose (e.g. at an EDM event) often provides information on safe use and/or looks after the new initiate during his or her first experience (Hansen et al., 2001; Schensul et al., 2005; Southgate & Hopwood, 2001).

Friends, therefore, are in important part of diffusing ecstasy to both new initiates and to experienced users. In fact, unlike previous theories about dealers pushing ecstasy onto people, it appears that most ecstasy is purchased from friends and not unknown dealers. Of note, many people who sell ecstasy do not consider themselves dealers, particularly if they also use and only sell to friends (Jacinto et al., 2008). This occurs, in part, because such people often make an easy transition or drift into casual selling without making conscious decisions to do so (e.g. buying extra ecstasy to provide to their friends). This can be done to help friends obtain the drug, to make extra side money, or to achieve popularity (Jacinto et al., 2008; Remy et al., 2017), and many hope to share the ecstasy experience with others as they, personally, find it enjoyable (Sales & Murphy, 2007). Selling can also be easy in such scenes considering there is likely a preexisting customer base (Remy et al., 2017; Sales & Murphy, 2007). People who provide ecstasy to friends are often a different phenomenon from that of what we think of as traditional dealing. Indeed, people can form friendships with dealers and acquire a mutual level of trust, and many sellers look after their clientele, provide guidance regarding use, and some even refuse to sell to people they think may be abusing it (Jacinto et al., 2008). In addition, some dealers go as far as testing ecstasy they sell as a method of decreasing potential harm from use (Palamar et al., 2019). However, some dealers, particularly unknown dealers, may try to exploit users and increase the price or sell bunk product (Schensul et al., 2005). Therefore, purchasing from unknown or untrustworthy dealers, especially in a spontaneous or unplanned manner, may be particularly risky, especially if the ecstasy contains dangerous adulterants.

While various studies have examined diffusion of ecstasy, these studies have largely been qualitive in nature and are over a decade old (Hansen et al., 2001; Jacinto et al., 2008; Sales & Murphy, 2007; Schensul et al., 2005; Vervaeke et al., 2008). Most previous studies have also not focused solely on this high-risk scene which limits their generalizability to the current cohort of EDM event attendees. In addition, previous research has not specifically asked about Molly use when asking participants about ecstasy. Molly is a nickname for powder or crystalline ecstasy in the US since the early 2000s (Palamar, 2018) and not including Molly in the definition of ecstasy appears to lead to underreporting of ecstasy use (Palamar, Keyes, & Cleland, 2016). This paper uses a quantitative approach to examine the diffusion of ecstasy among EDM event attendees in NYC and attempts to provide estimates of relevant behaviors and contextual variables.

Methods

Procedure

This study used a multi-stage sampling design. Through time-space sampling (MacKellar et al., 2007), each week, EDM events in NYC were randomly selected (at the design stage) and then adult attendees entering each randomly selected event were surveyed. Survey administration typically occurred on 1–2 nights per week on Thursday through Sunday and recruitment was conducted from June through September of 2018. While most events were held at nightclubs, participants were also surveyed entering two large daytime EDM dance festivals. Dance festivals were not randomly selected because they are so infrequent in NYC and we had to ensure their inclusion as part of the parent study. Events at nightclubs and dance festivals must have been EDM events (featuring EDM DJs). Events with mixed music (e.g. hop hop) were not considered for selection. Recruiters approached individuals, and if eligible, were asked if they would be willing to take a survey about drug use. Individuals were eligible if they were 1) between 18–40 years old, and 2) about to enter the randomly selected event. Individuals were ineligible if they appeared inebriated. Study staff tried to ensure that potential participants were not visibly inebriated and ensured that those approached did not display impaired attention, impaired gait, or exhibit slurred speech. Surveys were taken on electronic devices after informed consent was provided. Those completing the survey, which typically took 10–15 min to complete, were compensated $10 USD. The survey response rate was 73% and 1,020 participants completed the survey and provided complete data on ecstasy use. This study only focused on data from those reporting past-year ecstasy use. The study was approved by the New York University Langone Medical Center institutional review board.

Measures

Participants were first asked their age, sex, race/ethnicity, and sexual orientation. They were then asked about past-year use of various drugs including LSD, ketamine, and methamphetamine. They were also asked if they have used ecstasy/MDMA/Molly in the past year. Molly was included in the definition of ecstasy as it is a nickname for powder or crystalline ecstasy in the US (Palamar, 2018). Those reporting past-year ecstasy use were asked a series of additional questions. They were asked, “Where did you first use Ecstasy/Molly/MDMA?” and answer options were “a nightclub,” “a festival,” “a concert,” “at another type of party or venue,” “at a home”, and “other.” This variable was also dichotomized into “nightclub” versus other venues because ecstasy has traditionally been considered a “club drug.” Participants were also asked, “Were you planning to use ecstasy before you used it the first time?” and answer options were “no,” “yes, I was willing to use it, but wasn’t planning to use it,” and “yes, I was planning to use it.” This was dichotomized into “unplanned use” vs. “planned use or was willing to use.” Participants reporting past-year ecstasy use were also asked, “Have you ever given someone their FIRST dose ever of Ecstasy/Molly/MDMA?” and answer options were “no,” “yes, at an EDM party,” “yes, outside of an EDM party,” “yes, inside and outside of EDM parties,” and “not sure.” This variable was dichotomized into “yes” vs. “no” and the 46 participants reporting that they were unsure were removed from analyses. Finally, participants were asked “How likely are you to use Ecstasy/Molly in the future?” with response options being “very likely,” “likely,” “neutral,” “unlikely,” and “very unlikely,” and responses were dichotomized into “likely/very likely” versus all other responses.

Analyses

First, characteristics of the past-year ecstasy-using sample were described and prevalence estimates were calculated for all dependent variables—where ecstasy was first used, whether one’s first use was planned, whether the participant ever gave someone their first dose, and whether the participant reported being likely to use again in the future. Next, Rao-Scott chi-square was used to determine whether there were bivariable differences regarding each of these variables according to participant characteristics, although differences in age, which is a ratio variable, were determined using linear regression. Finally, all participant characteristic variables were fit as covariates into separate multivariable generalized linear models using Poisson and log link to determine associations with all else being equal. These models were used to generate adjusted prevalence ratios (aPRs) because odds ratios can lead to inflation of estimates when outcomes are prevalent (>10%) (Thompson et al., 1998). It should be noted that past-year use of LSD, ketamine, and methamphetamine were only included as covariates when examining correlates of giving someone his or her first dose and likelihood of future use. This was done because the other dependent variables focus on the participants’ initiation and literature suggests that initiation of these drugs tends to come after ecstasy initiation (Gagnon et al., 2011; Halkitis & Palamar, 2008; Reid et al., 2007). A Bonferroni statistical correction was applied to bivariable models to correct for any inflation of family-wise error introduced by multiple testing.

Since time-space sampling was used, selection probabilities were computed for each participant to provide estimates. These probabilities were based on frequency of past-year EDM event attendance and number of attendees (tracked via a clicker) who passed a predetermined line near the event entrance (as an indicator of response rates) (MacKellar et al., 2007). Weights for frequency of event attendance were inversely proportional to frequency of attendance. With regard to number of individuals tracked entering each event, weights were inversely proportional to this event-level response rate. The two weight components were combined via multiplication and normalized. Thus, those believed to have a lower probability of selection were up-weighted and those believed to have a higher probability of selection were down-weighted (Jenness et al., 2011; MacKellar et al., 2007). This helped ensure that frequent attendees (who were more likely to be surveyed) were not over-represented. These probability weights were used in all analyses to account for differential selection probability and clustering of individuals entering each event. Data were analyzed using Stata 13 SE and Taylor series estimation was used to obtain accurate standard errors (Heeringa et al., 2010). Probability weights were used for all analyses and randomly selected event was specified as the primary sampling unit.

Results

31.0% of participants surveyed (n = 422) reported past-year ecstasy use. Table 1 presents descriptive statistics for the full sample. Most of the sample identified as male (66.1%) and heterosexual (80.9%). A plurality identified as white (48.8%). Past-year use of various other drugs was common with 28.7% reporting LSD use and 19.1% reporting ketamine use. With respect to ecstasy initiation, an estimated 42.9% used their first dose in an unplanned manner, with 22.5% willing, but not planning use, and 34.6% planning use. A third (33.4%) initiated a dance festival, 24.3% initiated at a nightclub, 19.7% initiated at a home, 13.8% initiated at a concert, and 8.8% initiated at a party of different type of venue. An estimated 39.4% of participants have given someone else their first ever dose of ecstasy and 60.2% reported likelihood of future use.

Table 1.

Sample characteristics (n = 422).

	n	Weighted %
Age	Mean = 25.8	SE = 0.4
Sex
Male	257	66.1
Female	165	33.9
Race/Ethnicity
White	217	48.8
Black	25	4.0
Hispanic	77	22.9
Asian	65	18.1
Other/Mixed	38	6.2
Sexual Orientation
Heterosexual	317	80.9
Gay/Lesbian	50	12.3
Bisexual	40	5.9
Other Sexuality	15	0.9
Other Past-Year Drug Use
LSD	166	28.7
Ketamine	130	19.1
Methamphetamine	27	3.2
First Dose of Ecstasy Unplanned
No	250	57.1
Yes	139	42.9
Where Took First Dose
Non-Nightclub Venue	285	75.7
Nightclub	118	24.3
Gave Someone Their First Dose
No	204	60.6
Yes	180	39.4
Likely to Use Ecstasy in the Future
No	146	39.8
Yes	272	60.2

Note. Percentages are valid percentages and exclude missing data.

SE = standard error.

Table 2 presents bivariable and multivariable correlates of unplanned ecstasy initiation. The mean age for those who planned their first use (26.6 [SE = 0.5]) was significantly higher than the mean age for those reporting unplanned first use (24.6 [SE = 0.7]; p = .016). In the multivariable model, with all else being equal, age was associated with reduced risk for unplanned ecstasy initiation (aPR = 0.96, 95% CI: 0.91–1.00, p = .044) and compared to white participants, those identifying as Hispanic (aPR = 1.83, 95% CI: 1.12–2.98; p = .015) or Asian (aPR = 1.78, 95% CI: 1.13–2.81; p = .014) were at higher risk for using in an unplanned manner. With regard to where participants’ first dose was taken (Table 3), the mean age for those who first used at a nightclub (27.8 [SE = 0.8]) was significantly higher than the mean age for those who first used in a non-nightclub setting (25.2 [SE = 0.5]; p = .007) and there was a significant difference with regard to sexual orientation with 61.3% of those initiating at a nightclub identifying as heterosexual compared to 87.0% of those not initiating at a nightclub identifying as heterosexual (p < .001). In the multivariable model, compared to white participants, those identifying as black were at higher risk for initiating at a nightclub (aPR = 2.39, 95% CI: 1.05–5.48, p = .039). Further, those identifying as gay or lesbian were at significantly higher likelihood of initiating at a nightclub (aPR = 2.81, 95% CI: 1.54–5.12, p = .001) as were those identifying as other sexuality (aPR = 2.56, 95% CI: 1.02–6.45, p = .046).

Table 2.

Correlates of unplanned initiation of ecstasy.

	Bivariable tests			Multivariable model
	Planned weighted %	Unplanned weighted %	P	aPR	(95% CI)	P
Age – Mean (SE)	26.6 (0.5)	24.6 (0.7)	.016	0.96	(0.91, 1.00)	.044
Sex			.131
Male	71.2	58.8		1.00
Female	28.8	41.2		1.26	(0.88, 1.82)	.206
Race/Ethnicity			.041
White	60.8	32.4		1.00
Black	2.8	3.9		1.80	(0.97, 3.33)	.062
Hispanic	18.1	29.7		1.83	(1.12, 2.98)	.015
Asian	13.1	25.8		1.78	(1.13, 2.81)	.014
Other/Mixed	5.2	8.2		1.60	(0.72, 3.56)	.248
Sexual Orientation			.515
Heterosexual	77.7	86.4		1.00
Gay/Lesbian	14.4	9.8		1.03	(0.45, 2.33)	.949
Bisexual	6.8	3.3		0.64	(0.18, 2.29)	.490
Other Sexuality	1.2	0.6		0.65	(0.24, 1.74)	.390

Note. Rao-Scott chi-square was used to determine whether there were bivariable differences, although differences in age, which is a ratio measurement, were determined using linear regression. All variables were then fit as covariates into separate multivariable generalized linear models using Poisson and log link to determine associations with all else being equal. This generated adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs). A Bonferroni statistical correction was applied to bivariable tests, so results are only deemed statistically significant when p < .0125 (alpha = .05/4 outcomes).

Table 3.

Correlates of first use occurring at a nightclub.

	Bivariable tests			Multivariable model
	Non-nightclub venue weighted %	Nightclub weighted %	P	aPR	(95% CI)	P
Age – Mean (SE)	25.2 (0.5)	27.8 (0.8)	.007	1.04	(0.99, 1.09)	.102
Sex			.230
Male	62.3	72.2		1.00
Female	37.7	27.8		1.04	(0.54, 1.99)	.906
Race/Ethnicity			.391
White	48.0	47.9		1.00
Black	3.0	7.5		2.39	(1.05, 5.48)	.039
Hispanic	21.2	28.3		1.21	(0.65, 2.27)	.543
Asian	20.4	12.7		0.95	(0.40, 2.23)	.903
Other/Mixed	7.4	3.6		0.72	(0.24, 2.09)	.540
Sexual Orientation			<.001
Heterosexual	87.0	61.3		1.00
Gay/Lesbian	6.4	31.3		2.81	(1.54, 5.12)	.001
Bisexual	5.9	5.8		1.38	(0.49, 3.86)	.540
Other Sexuality	0.7	1.6		2.56	(1.02, 6.45)	.046

Note. Rao-Scott chi-square was used to determine whether there were bivariable differences, although differences in age, which is a ratio measurement, were determined using linear regression. All variables were then fit as covariates into separate multivariable generalized linear models using Poisson and log link to determine associations with all else being equal. This generated adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs). A Bonferroni statistical correction was applied to bivariable tests, so results are only deemed statistically significant when p < .0125 (alpha = .05/4 outcomes).

Table 4 presents bivariable and multivariable correlates of participants ever giving someone their first dose of ecstasy. In bivariable models, only past-year LSD use was a risk factor, given the Bonferroni correction, with 42.5% of those giving someone their first dose reporting use compared to 20.6% of those not giving someone their first dose reporting use (p = .002). With all else being equal, LSD remained the only significant correlate of offering someone their first dose (aPR = 1.78, 95% CI: 1.20–2.65, p = .004). Finally, regarding correlates of reporting likelihood of using ecstasy again in the future (Table 5), those reporting past-year LSD use (38.2%) and/or past-year ketamine use (26.4%) were more likely to report intention to use than those not reporting past-year use of LSD or ketamine (14.7% and 8.3%, respectively, ps < .001). In the multivariable model, each additional year of age was associated with increased risk of self-reported likelihood of future use by 3% (aPR 1.03, 95% CI: 1.01–1.06, p = .015). Females were also at higher risk (aPR = 1.46, 95% CI: 1.13–1.89, p = .004) as were those reporting past-year LSD use (aPR = 1.42, 95% CI: 1.14–1.76, p = .001).

Table 4.

Correlates of giving someone their first dose of ecstasy.

	Bivariable tests			Multivariable model
	No weighted %	Yes weighted %	P	aPR	(95% CI)	P
Age – Mean (SE)	25.5 (0.7)	26.1 (0.5)	.462	1.03	(0.99, 1.07)	.141
Sex			.339
Male	63.6	70.7		1.00
Female	36.4	29.3		0.91	(0.61, 1.35)	.633
Race/Ethnicity			.154
White	45.1	57.2		1.00
Black	2.6	4.9		1.36	(0.75, 2.49)	.313
Hispanic	30.9	14.3		0.60	(0.34, 1.07)	.086
Asian	13.9	18.5		1.32	(0.79, 2.22)	.293
Other/Mixed	7.5	5.0		0.83	(0.33, 2.05)	.684
Sexual Orientation			.535
Heterosexual	77.6	82.1		1.00
Gay/Lesbian	14.2	11.7		0.83	(0.42, 1.65)	.599
Bisexual	7.7	4.4		0.64	(0.25, 1.67)	.358
Other Sexuality	0.5	1.8		1.47	(0.82, 2.61)	.193
Other Drug Use
LSD	20.6	42.5	.002	1.78	(1.20, 2.65)	.004
Ketamine	15.1	27.2	.042	1.10	(0.74, 1.65)	.642
Methamphetamine	2.3	2.9	.742	1.23	(0.47, 3.24)	.672

Note. Rao-Scott chi-square was used to determine whether there were bivariable differences, although differences in age, which is a ratio measurement, were determined using linear regression. All variables were then fit as covariates into separate multivariable generalized linear models using Poisson and log link to determine associations with all else being equal. This generated adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs). A Bonferroni statistical correction was applied to bivariable tests, so results are only deemed statistically significant when p < .0125 (alpha = .05/4 outcomes).

Table 5.

Correlates of self-reported likelihood of using again in the future.

	Bivariable tests			Multivariable model
	No weighted %	Yes weighted %	P	aPR	(95% CI)	P
Age – Mean (SE)	25.0 (0.6)	26.3 (0.6)	.130	1.03	(1.01, 1.06)	.015
Sex			.058
Male	74.8*	60.4		1.00
Female	25.2	39.6		1.46	(1.13, 1.89)	.004
Race/Ethnicity			.029
White	37.8	55.6		1.00
Black	4.8	3.5		0.96	(0.52, 1.77)	.896
Hispanic	32.8	16.5		0.68	(0.46, 1.02)	.063
Asian	22.2	15.5		0.91	(0.63, 1.32)	.627
Other/Mixed	2.4	8.8		1.61	(1.00, 2.59)	.048
Sexual Orientation			.200
Heterosexual	88.3	76.2		1.00
Gay/Lesbian	9.1	14.2		1.31	(0.90, 1.91)	.151
Bisexual	2.4	8.2		1.11	(0.86, 1.43)	.429
Other Sexuality	0.2	1.3		0.89	(0.62, 1.27)	.526
Other Drug Use
LSD	14.7	38.2	<.001	1.42	(1.14, 1.76)	.001
Ketamine	8.3	26.4	<.001	1.21	(1.00, 1.47)	.055
Methamphetamine	2.5	3.6	.575	0.90	(0.64, 1.28)	.563

Note. Rao-Scott chi-square was used to determine whether there were bivariable differences, although differences in age, which is a ratio measurement, were determined using linear regression. All variables were then fit as covariates into separate multivariable generalized linear models using Poisson and log link to determine associations with all else being equal. This generated adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs). A Bonferroni statistical correction was applied to bivariable tests, so results are only deemed statistically significant when p < .0125 (alpha = .05/4 outcomes).

Discussion

This study, focusing on past-year ecstasy-using EDM event attendees in NYC, adds to a wealth of older literature, which was largely qualitative in nature, and provides insight into ecstasy diffusion in the current EDM scene. An estimated 31% of EDM party attendees in NYC have used ecstasy in the past year and these individuals were the focus of this report. The main aim of this report was to estimate prevalence and correlates of unplanned initiation, place of initiation, providing others their first dose of ecstasy, and self-reported likelihood of future use.

With regard to planning of first use, an estimated 42.9% of people using ecstasy in the past year used their first ever dose in an unplanned manner. However, this estimate can actually be viewed as a conservative estimate as the 22.5% who reported willingness to use (but did not report planned use) were treated as planned users in this study. Results suggest that younger participants were more likely to use in an unplanned manner. It is unknown when these individuals initiated; thus, it is not clear whether this is more of an age, period, or cohort effect, but at face value, it appears that currently, younger attendees are at higher risk for using in an unplanned manner. Hispanic and Asian participants were also at higher risk for using in an unplanned manner compared to white participants, and further research is needed to help determine reasons for increased risk in these groups. Unplanned initiation can be hazardous if the individual does not know what effects to expect or if he or she engages in behaviors that may contradict a potential safer experience— such as drinking alcohol or not having enough rest. Unplanned use can also have an adverse effect regarding responsibilities such as having to attend work the next morning. Event-level intercept studies with follow-up have found that 7.3%−11.9% of individuals attending such events used ecstasy the night of the survey in an unplanned manner (Palamar et al., 2019; Ramchand et al., 2013), but this is among the first studies to examine unplanned initiation, albeit in a retrospective manner.

Initiation occurred in a variety of different venues or contexts, but the plurality of initiation occurred at dance festivals (33.4%), followed by nightclubs (24.3%). Although more than half of ecstasy initiation occurred at dance festivals or nightclubs, which is unsurprising, we must not ignore the fact that a fifth (19.7%) initiated at a home, 13.8% initiated at a concert, and 8.8% initiated at a party or different type of venue. Although all participants in this study were surveyed entering EDM events at nightclubs or at EDM dance festivals, it is unknown whether these individuals who initiated in other contexts were already attending such EDM events. Regardless, ecstasy initiation is by no means limited to nightlife scenes. Although no longer significant in the multivariable model, bivariable results suggest that older participants were more likely to initiate in a nightclub compared to younger participants. However, it should be noted that these current data do not allow us to determine whether this is more of an age effect, or a cohort or period effect.

Black participants were also more likely than white participants to initiate ecstasy at a nightclub as compared to contexts of dance festivals and private homes. There is currently a dearth of literature focusing on ecstasy use among black individuals, but recent research suggests a link between hip hop music and Molly use among black individuals who attend nightclubs (Rigg & Estreet, 2019). Gay and lesbian participants as well as those identifying as other sexuality were also more likely to initiate use at a nightclub. More research is needed, but such sexual minorities may be more likely to be introduced to the drug in nightclub scenes, which also often have gay parties, while currently, EDM dance festivals appear to have more of a heterosexual attendance.

With regard to “spreading” ecstasy use, 39.4% of past-year ecstasy users reported knowingly giving someone their first dose of ecstasy. Past-year LSD use was the only significant risk factor for having given someone his or her first dose, which may suggest that people with more extensive recent drug experience (or perhaps more extensive repertoires) are more likely to introduce people to the drug. According to the Diffusion of Innovation theory (Rogers, 2010) diffusion of ecstasy is driven, in part, by these “innovators” who promote or actually provide others with their first ever dose (inside and/or outside the EDM scene). Such sharing of the drug in this scene further demonstrates ecstasy as a form of social “contagion” in which those exposed to users can be seen as “susceptible” to use, and the drug is “transmitted” through peers (Ferrence, 2001). Thus, attendance, and especially frequent attendance of such events with high prevalence of use can be seen as increasing susceptibility for initiation—especially when one’s friends use. Whether or not these first doses were given free of charge, such high prevalence suggests that peer influence is reciprocal as people who use often introduce others to the drug after initiation (Vervaeke et al., 2008). However, more research is needed into situations in which people receive their first dose. It is important that people who use or are at risk for using are educated about potential effects and dangers, and again, it may be important for those introducing drugs like ecstasy to others to at least provide education and harm reduction information and perhaps even look after the new initiate during his or her first experience (Hansen et al., 2001; Schensul et al., 2005; Southgate & Hopwood, 2001).

Finally, 60% of the sample reported being likely to use again in the future. Older participants, and surprisingly, females, were at higher risk for reporting likelihood of using again in the future. Females are typically at lower risk for use of various drugs compared to their male counterparts (Johnston et al., 2018), although some studies focusing on EDM event attendees have found that females are not at higher risk for ecstasy use compared to males (Palamar et al., 2017). Those reporting past-year LSD use were also more likely to report intention to use ecstasy again. This might suggest that those with more extensive repertoires of recent use of psychedelics are more likely to use again in the future. Intention to use is an important risk factor as intention to engage in a behavior is often viewed as the most proximal predictor of a behavior (Ajzen, 1985). Corroborating previous research (Sherlock & Conner, 1999; Umeh & Patel, 2004), current or more recent users are at high risk for reporting intention to use again. This suggests that many individuals may indeed use again, or use many more times, so targeting current users in particular may be most ideal for disseminating relevant harm reduction information.

Although this study examined various contexts of self-reported ecstasy diffusion and initiation in the NYC EDM scene, more current research is needed to examine such associations within the general population and in other scenes or subcultures. Likewise, more information is needed to examine how use diffuses from EDM scenes to the general population. While direct exposure to users is an important factor regarding ecstasy initiation/use, more research is also needed to focus on the influence of subgenres of music, and availability through the Internet. Regardless, these findings are based on the first in-depth analyses examining diffusion of ecstasy in current times, and findings were based on quantitative data rather than qualitative interviews.

Limitations

This study was cross-sectional; therefore, temporality of associations could not be determined. This was a particular limitation regarding outcomes related to ecstasy initiation, which is why variables indicating use of other drugs were omitted from analyses examining initiation. Initiation could have occurred many years ago, especially in older participants, and a limitation is that participants were not asked how long ago they initiated, so it is unknown to what extent these findings describe recent initiation. Limited recall is of concern, especially regarding initiation questions, and dependent variables were not validated. However, self-reported drug use in an earlier version of this survey was found to be highly reliable (Palamar et al., 2019). Finally, while results are generalizable to EDM attendees, they may not be as generalizable outside of EDM scenes. Likewise, all surveys were conducted among people entering EDM events in NYC so results may not be generalizable to individuals outside of NYC.

Conclusions

This was the first study in recent times to examine diffusion of ecstasy use in the EDM scene in a quantitative manner. Results suggest that a substantial portion of EDM event-attending people who use ecstasy initiated in an unplanned manner and have introduced others to their first doses. Likewise, most use appears to now first occur at dance festivals and more than half of people who have used ecstasy in the past year believe they will use again. Results suggest that people who are currently or have recently used ecstasy in particular should be targeted to disseminate prevention, treatment, and/or harm reduction information. Although many individuals already look after their friends in such scenes, future research can help determine the efficacy of ecstasy users educating and looking out for one another in this high-risk scene.