User perceptions of long-term costs and benefits of MDMA use: findings from a large online sample

Abstract

Background

Previous research has focused on risks associated with non-clinical MDMA use, contrasting short-sighted hedonistic motivations with negative long-term consequences. User perceptions may be more nuanced, incorporating numerous underexplored positive long-term effects. We aimed to: (1) Assess MDMA user perceptions regarding both positive and negative long-term effects, and (2) Generate a large, open dataset with correlates to explore for future research.

Methods

886 non-clinical MDMA users completed an online, opt-in survey, with self-reported measures of acute, subacute, and long-term effects, motivations, use context, and polydrug use.

Results

User perceptions of MDMA’s long-term effects were far more positive than negative. Respondents endorsed long-term increases in aesthetic appreciation, deeper social connections, and positive changes in life perspectives, among other positive outcomes. A minority of users reported negative effects (e.g. on concentration/memory and mood), which remain important. Long-term effects showed several correlations with acute effects and motivations, with use due to peer influence being repeatedly linked to lower positive and higher negative long-term effects.

Conclusion

MDMA users report getting much more than a momentary high from taking the drug. A nuanced appreciation of user perceptions can inform drug safety communication, and advance our understanding of drug effects by highlighting targets for further investigation.

Keywords:

• MDMA
• ecstasy
• user perceptions
• costs and benefits
• deficit model

Introduction

MDMA (3,4-methylenedioxy-methamphetamine, also known as Ecstasy) is one of several largely prohibited psychoactive drugs receiving renewed interest among researchers and clinicians as a potential adjunct to psychotherapy (Carhart-Harris et al., 2018a; Gasser et al., 2015; Mithoefer et al., 2018). Early clinical findings suggest that MDMA-assisted psychotherapy appears to be both effective and safe, with applications in depression-, trauma-, and anxiety-related mental health problems (Feduccia et al., 2019; Mithoefer et al., 2019). Beyond symptom reduction, participants in such clinical trials reportedly show long-term changes in aspects of their personality (e.g. increased openness to experience), as well as benefits in daily quality of life, such as deeper self-awareness and more positive relationships with family and friends (Barone et al., 2019; Wagner et al., 2017). Although such benefits can be expected from successful typical psychotherapy, it has been suggested that specific qualities of the acute MDMA experience – such as increased trust, positivity, and openness – may help catalyze these changes (Bershad et al., 2016; Mithoefer, 2013).

The therapeutic context is integral to safe and effective MDMA-assisted psychotherapy (Carhart-Harris et al., 2018b). Nevertheless, it is possible that non-clinical (‘recreational’) use of MDMA may also produce long-term changes. Indeed, it has been argued that the possible long-term effects of classical psychedelics such as psilocybin and lysergic acid among healthy individuals has also been under-appreciated (Elsey, 2017). For example, users of classical psychedelics have reported a long-lasting positive impact on their appreciation of life and nature (Doblin, 1991), as well as reporting a facilitation of creative pursuits (Sessa, 2008). Understanding whether and how MDMA impacts personally meaningful areas such as self-insight, worldview, and relationships over the long-term among recreational users may inform both a fuller picture of the effects of MDMA, as well as more targeted and effective public health communication aiming to educate users or reduce harms associated with use. Research covering acute, subacute, and long-term effects of MDMA among recreational users is outlined in the following paragraphs.

Considerable research has investigated acute effects of MDMA among recreational users via self-report, or among healthy volunteers in laboratory settings (Bershad et al., 2016; Green et al., 2019; Kamilar-Britt & Bedi, 2015). These studies found that MDMA typically produces euphoria (hence ‘ecstasy’), heightened sociality and feelings of interpersonal closeness, a rush of energy, as well as enhancement of aesthetic sensations and sometimes visual distortions or hallucinations. On the negative side, users may also experience discomfort and anxiety. Researchers have also investigated sub-acute effects of MDMA (effects of drug use that can occur in the days following use, when the acute effects have worn off). Such studies note a possible dip in mood occurring in the days following MDMA use (a ‘comedown’), though this may be attributable to sleep disturbance rather than MDMA per se (McKetin et al., 2014; Pirona & Morgan, 2010; Scott et al., 2013).

Investigations of long-term effects of MDMA have predominantly focused on risks associated with MDMA use, including neurocognitive deficits in areas such as concentration and memory, as well as negative mental health consequences such as depression and anxiety (Parrott, 2013). Understanding the potential negative long-term impact of MDMA is vital for understanding its risk profile, but some research (e.g. Parrott, 2015) seems to approach the user experience of drug effects from what has been referred to in health psychology as a ‘deficit model’ (Joffe, 2003). In short, because of some deficit in either appreciation of risks or capacity to resist short-term gains, users of potentially risky drugs are seen to trade-off the benefit of a short-term high for the cost of medium to long-term effects, which are seen to be predominantly or wholly negative. From this perspective, users are essentially either making an irrational trade-off between short- vs. long-term effects (a deficit in willpower or decision making), or are unaware of or under appreciate the risks (a deficit in knowledge). We believe that user experiences and perceptions of drug use and effects are likely considerably more nuanced than this and, even if one’s key concern is long-term risk reduction, more fully understanding user perceptions across a wider range of long-term effects than has been investigated previously – including possible positive effects – is important. Furthermore, we suggest that educational and harm reduction messaging is unlikely to be effective if it fails to connect with what users report experiencing, or misconstrues what they may get out of MDMA use beyond a short-term high.

Relatively few studies have investigated broader perceptions of MDMA effects among large samples of recreational users. In one survey of 466 MDMA users, respondents noted both negative and positive long-term effects of MDMA, most notably impaired concentration, periods of depression, and being more open towards people (Verheyden et al., 2003). In a further web-based sample of 626 polydrug users, MDMA was perceived as having either no effects, or beneficial effects, on mental health problems (Carhart-Harris & Nutt, 2010). The present study aimed to build upon this work by investigating a broader range of psychologically important perceived long-term effects, giving users equal opportunity to endorse both positive and negative effects.

Obtaining a more detailed picture of both possible positive and negative perceived long-term effects of MDMA use can prove informative regardless of one’s perspective on the safety of MDMA use: User perceptions may point to positive or negative effects that can be investigated in more controlled settings, or suggest avenues for communication about risk-reduction among users based upon a greater understanding of their perceptions. Furthermore, our additional questions regarding motivations for use, acute effects, sub-acute effects, demographics, and polydrug use and frequency may be used to highlight potential relationships among variables that can help reveal risk factors for negative effects.

Methods

Ethical approval and consent

All procedures for the study were approved by the University of Amsterdam institutional review board (2020-COP-11936). All participants gave informed consent.

Participants and inclusion/exclusion criteria

The survey was run from April 2^nd 2020 to April 23^rd 2020. Participants took part voluntarily, with a €50 lottery prize available for a randomly selected participant. Alternatively, students of the University of Amsterdam were able to take part in exchange for .5 research credits. Participants were recruited from a range of sources, including word of mouth/snowball sampling, sharing of the survey on social media by the researchers, as well as outreach to several organizations/groups with a known or probable interest in MDMA (e.g. Trimbos Institute – a Dutch mental health and addiction non-profit, MAPS, social media groups in various countries linked with the dance music or rave scene, groups interested in safe/responsible drug use, such as Dancesafe and Sesh Safety). Sharing of the research by these groups was in the form of clickable links in newsletters and social media feeds of the respective organizations/groups. A landing page for the study also encouraged prospective respondents to share links to the survey with people who they thought might be interested, thereby aiming for a snowball sampling effect. The survey was filled in completely digitally/online, using Qualtrics survey software.

Participants were required to be fluent in either English or Dutch, and aged 18 or older. Responses were excluded if the respondent failed attention checks (n = 25), if they noted using MDMA as part of therapy (n = 13), or if they did not complete a sufficient number of questions to reach the long-term effects questions which included the attention checks (n = 339). Participant numbers and demographics are presented in the results section, but briefly, information on perceived long-term effects of MDMA was available for 886 respondents.

Materials

The questionnaire was composed of several sections in the following order: Demographic information; MDMA use profile; Motivations for use; Acute effects; Subacute effects; Long-term effects; Mental health issues; Context of use; Polydrug use; and finally, Indicators of addiction. Sections on Mental health, Context of use, Polydrug use, and Indicators of addiction are detailed in the supplementary materials and do not feature in any analyses for this paper. The full survey can be found on the OSF page (https://osf.io/8n43e/). There were 91 questions in total, with the questionnaire expected to take between 10 and 20 minutes to complete. Question formats, described below, were mostly sliders for numeric ratings, and multiple choice for categorical/ordinal outcomes.

Demographic information

Participants provided their Country of Residence, Age, Gender, Employment Status, and Education Level. Demographic information is presented in Table 1.¹

Table 1. Sample demographic information, including country of residence, education, employment status, age, and gender.

Residence	n	Percent	Education	n	Percent
USA	283	31.9%	High school or less	138	15.6%
Netherlands	255	28.8%	Some college	327	36.9%
Canada	105	11.9%	Completed undergraduate	268	30.2%
UK/Northern Ireland	79	8.9%	Completed postgraduate	153	17.3%
Australia	21	2.4%
Slovenia	17	1.9%	Gender
Germany	14	1.6%	Woman	435	49.1%
Spain	9	1.0%	Man	432	48.8%
Sum of above		88.4%	Other identification	11	1.2%
All other		<1% each	Prefer not to say	8	0.9%
Employment status			Age
Full time	420	47.4%	18-24	308	34.8%
Student	297	33.5%	25-34	370	41.8%
Part time	174	19.6%	35-44	140	15.8%
Looking for work	55	6.2%	45-64	65	7.3%
Homemaker/carer	24	2.7%	65+	3	0.4%
Not employed, not seeking work	22	2.5%
Unable to work	18	2.0%		N
Retired	8	0.9%	Total	886

MDMA use profile

Participants indicated their age of first use (in years) and how long ago their last use was (in months, up to 24+ months ago). Participants also indicated if, when taking MDMA, they ever combined use with Alcohol (‘No’ vs. ‘Yes, sometimes’ vs. ‘Yes, often’), with Cigarettes/nicotine products (‘No’ vs. ‘Yes, sometimes’ vs. ‘Yes, often’), or with other drugs (‘No’ vs. ‘Yes, sometimes’ vs. ‘Yes, often’). If they answered yes to the other drugs question, then they were presented with checkbox options for a range of other recreational drugs. Participants were additionally asked to indicate whether they had ever taken MDMA as a pill/tablet, and if so, what a typical dose (in tablets or fractions of a tablet) would be. The same question was asked regarding taking MDMA as crystals/powder (in milligrams). As participants were provided with a textbox in which to give their answer, there was considerable variation in the interpretability of responses, and so these numbers are not included in any analyses. Finally, participants indicated the proportion of times they used for which they had tested their MDMA (using a 0-100 slider, with 0 = ‘Never,’ 50 = ‘About half the time,’ and 100 = ‘Always’).

MDMA motivations

Respondents were asked to indicate the extent to which a range of motivations for use were applicable to them using a 0-100 sliding scale (0 = ‘Not at all a motivation for me,’ 100 = ‘A very strong motivation for me’). The eight motivations included were based upon previous research investigating reported motivations for MDMA use (Sumnall et al., 2006; Ter Bogt & Engels, 2005). The motivations were: ‘To give me energy,’ ‘To feel euphoric,’ ‘To gain insight into myself,’ ‘To be more sociable/open,’ ‘To be more flirtatious,’ ‘To enhance sex,’ ‘To help deal with or forget my problems,’ and ‘To fit in or go along with other people’. Each motivation gave some small examples to further convey the intended meaning of each motivation, e.g.: ‘To help deal with or forget my problems (e.g. to have less worries, to feel good for once, to forget my problems)’.

Acute MDMA effects

Respondents were asked to indicate (from 0 = ‘Not at all,’ to 100 = ‘Very strongly’) how much they had experienced several possible acute effects specifically when on MDMA. The seven acute effects were derived from previous research investigating acute MDMA effects (Bershad et al., 2016; Green et al., 2019; Kamilar-Britt & Bedi, 2015; Sumnall et al., 2006; Verheyden et al., 2003). The acute effects were: ‘Euphoric effects,’ ‘Energetic effects,’ ‘Social or empathic effects,’ ‘Aesthetic effects,’ ‘Hallucinatory effects,’ ‘Sexual effects,’ ‘Anxious effects,’ and ‘Unpleasant physiological effects’. Again, each acute effect contained some additional information to clarify the intended meaning, e.g.: ‘Aesthetic effects - feeling like there is another level to what you are seeing or hearing, sights and sounds are more vivid, pleasurable, or beautiful’.

Subacute MDMA effects

Participants were asked to think specifically about how they have felt in the days after taking MDMA, and to indicate to what extent they had experienced either ‘Comedown’ or ‘Afterglow’ effects on a 0-100 slider input (from 0 = ‘Not at all,’ to 100 = ‘Very strongly’). These subacute effects were described as follows: ‘“Comedown” effects - feeling depressed, irritable, low mood, or very tired/fatigued,’ and ‘“Afterglow” effects - feeling peaceful, content, happy, or “cleansed”’.

Long-term MDMA effects

For long-term effects, respondents were informed that we would like to know how they think MDMA has affected them. We explicitly stated that we were not asking about short-term effects or sub-acute effects, but rather about how MDMA may have affected them over the long-term, when not on the drug or recovering from taking it. Participants were asked to indicate the extent to which 26 statements indicating possible long-term effects were true or not true for them, using a 0-100 sliding scale (0 = ‘Not at all true for me,’ 100 = ‘Completely true for me’). Long-term effects were based upon several sources, including reported long-term effects that might occur in therapy (Mithoefer et al., 2019; Wagner et al., 2017), long-term effects of classic psychedelic drugs (Elsey, 2017), extrapolations of acute effects (Bershad et al., 2016; Green et al., 2019; Kamilar-Britt & Bedi, 2015; Sumnall et al., 2006; Verheyden et al., 2003), long-term effects reported in previous studies of MDMA (Carhart-Harris & Nutt, 2010; Parrott, 2013; Verheyden et al., 2003), anecdotal reports, and psychological factors of general clinical relevance (e.g. several items may be recognized as tapping into aspects of Beck’s ‘cognitive triad,’ reflecting views of the self, the world, and the future – Beck 1979).

Most items included both a positive and negative version regarding the same general effect (e.g. ‘My use of ecstasy/MDMA has enabled me to have richer and deeper emotional experiences in my daily life, even when not on the drug’ vs. ‘My use of ecstasy/MDMA has caused me to have more blunted and shallower emotional experiences in my daily life, even when not on the drug’). Four items reflected more overarching impressions of use and experiences of it (participants indicated the extent to which they were happy that vs. wish they had never tried MDMA, and also indicated the extent to which their best/worst experiences with MDMA had been among the most positive/unpleasant experiences of their lives).

Thirteen items were positive effects, and thirteen items were negative effects. Items were presented in random order, with the exception of the final questions asking about whether the respondent was happy that they had/wish they had never tried it. One further item gave respondents the option to provide their own long-term effect as an open comment. The gist of each item is presented in Figure 1, and exact wording of all questions can be found in the supplementary materials. To ensure that only valid responses were reported, two attention checks were interspersed among the 26 questions.

Figure 5. Categorisation of ratings for positive (+ve) vs. negative (-ve) items within each type of long-term effect, with the respective Probability of Superiority (PSup) estimate. Numbers reflect the posterior mean and [95% highest density interval]. Items are ordered from greatest (‘view on trying MDMA’) to smallest (‘drug use’) difference in endorsement for positive vs. negative outcomes.

Procedure

Participants arrived on a purpose-built landing page that gave basic information about the study and participation requirements. This page and all questionnaire items were available in both English and Dutch. Participants opened a link to the questionnaire, presented using Qualtrics. After providing informed consent, participants proceeded through the questionnaire sections in the order presented above. At the end of the questionnaire, participants were given an email address and code to send to the researchers if they wished to be entered into the prize draw or receive research credit compensation. The code was not uniquely associated with respondents’ answers, so the only information shared by sending an email was that the person had completed the survey.

Data and analysis plan

The study was exploratory and intended to generate a useful data set for investigating user perceptions of MDMA, and to highlight possible relationships of interest among variables that might be explored in more detailed analyses or with alternative methods in future research. The results below present a mostly descriptive overview of the key findings regarding perceived long-term effects of MDMA. Analyses and data visualizations were performed/generated using R (R Core Team, 2013).

To determine whether respondents more strongly endorsed positive or negative long-term effects of MDMA, we used Bayesian categorical regression to generate within subjects Probability of Superiority (PSup) estimates for positive vs. negative effects within each type of effect (e.g. becoming more hopeful vs. more hopeless about one’s future). PSup is appropriate for skewed and ordinal data (Ruscio, 2008), and used for analyses of slider-response data elsewhere (Elsey & Kindt, 2021). For each type of long-term effect, pairs of positive and negative ratings were categorized as indicating a higher positive rating, a higher negative rating, or equal ratings. Bayesian categorical regression using brms v2.17.0 with weakly informative priors (normal(0, 2) for each intercept) was used to assess PSup based on these categorisations. PSup indicates the expected percentage of a sample giving a higher rating for one item vs. another (with equal scores given ‘half credit’). It ranges from 0-100 when given as a percentage, where in this case 50 indicates equality, scores over 50 indicate higher ratings for the positive item, and scores under 50 indicate higher scores for the negative item. Ratings for positive items were considered reliably higher/lower than ratings for negative items if the 95% highest density interval (HDI) for PSup fell above/below a range of practical equivalence (ROPE) of 45-55%.

To explore relationships among variables in the data set, we conducted an exploratory analysis of two-sided Kendall’s tau correlations, including a correction for tied ranks, between long-term effects and: motives, acute effects, subacute effects, and highest use frequency. Given the large number of correlations there is a high inflation of type 1 errors. A full Bonferroni correction of .05/494 suggests an alpha level of p = 0.0001012, or effectively p <.0001. The correlations are therefore depicted to indicate p <.0001 and p <.00001, reflecting .05 and .005 error rates respectively. Even with the conservative significance level correction, the visual presentation of these findings is intended to highlight broader patterns of relationships that may exist among variables (e.g. if certain negative acute effects are consistently associated with multiple long-term negative effects) rather than to highlight specific individual associations.

Results

Sample characteristics

The total sample of participants giving valid responses up to the point of completing the long-term effects questions was 886. Table 1 provides demographic information on the sample. The respondents mostly resided in the USA and the Netherlands, with sizable remainders also coming from Canada and the UK/Northern Ireland. The sample was about evenly split between women and men, and a small number of respondents indicated being of other gender identifications (non-binary = 3, genderfluid = 3, agender = 2, transwoman = 1, genderqueer = 1, unidentified = 1). The bulk of respondents were aged between 18-34, though substantial proportions were aged 35-44 and 45-64, along with a handful of older participants. Most were either working full-time or students, with a preponderance of respondents undergoing or having completed some form of higher education.

Respondents indicated both how frequently they had used MDMA in their highest 6 months of use, and how frequently they had used MDMA over the last 6 months. Respondents reported having used MDMA in frequencies as follows: rarely (Once in a 6-month period of highest use: 16.1%; in past 6 months: 22.0%), occasionally (About two or three times in a 6-month period of highest use: 37.7%; in past 6 months: 29.1%), regularly (About once or twice a month in a 6-month period of highest use: 28.1%; in past 6 months: 14.4%), very regularly (About once a week in a 6-month period of highest use: 14.8%; in past 6 months: 1.3%), most days (Used the drug more days than not in a 6-month period of highest use: 3.3%; in past 6 months: 1%). Additionally, 32.3% of respondents reported not having used MDMA over the past 6 months. Of the total sample, 13.8% did not reach the point of answering the use frequency questions (8.7%) or appear to have missed MDMA in the list (5.1%: they did not select MDMA from among the list of drugs they had used, but repeatedly indicated having used MDMA throughout the questionnaire). The average reported proportion of times MDMA was tested before use was 31.9%, but the median was only 4.0%, with 55.3% of respondents rarely testing (<=10% of the time) relative to 19.5% usually testing (>=90% of the time) – hence the distribution of responses was skewed towards most people rarely or never testing, but with a minority of respondents usually doing so.

Substantial proportions of respondents reported combining MDMA with other drugs sometimes (52.3%) or often (22.6%), notably alcohol (64.1%), nicotine (62.7%), or cannabis (60.4%), followed by psychedelics (38.6%), cocaine (33.9%), and ketamine (29.0%). All other drugs (for example, laughing gas, speed, 2C-B, poppers) were combined with MDMA by less than a quarter of the respondents. The sample had a relatively large number of heavy (daily or near daily) cannabis users, with 42.0% having used cannabis most days in their highest period of use (16.9% never used) and 27.8% using it most days in the past 6 months (with 30.6% not using it). Information regarding highest use and previous six months’ use for MDMA and a range of other recreational drugs, and drugs that have been taken in combination with MDMA, is available in supplementary section Polydrug Use.

MDMA motives, acute effects, and subacute effects

There was considerable variability in the endorsement given to different motives for using MDMA and its acute or subacute effects (Figures 2–4). Feeling euphoric was the most endorsed acute effect and the strongest motivation for taking MDMA, with median endorsement of 95. Using MDMA to be more sociable and connect with people was also rated highly as a motivation (median = 60), with social/empathic effects the second most-endorsed acute effect (median = 90). Many users reported taking MDMA to get better insights into themselves (median = 55), although this motivation seemed to split the respondents, with comparable clusters giving it very high and very low endorsement: 28% of respondents rated this item from 0-19, 35% rated it from 80-100, with each bracket of responses in between receiving at most 14% of responses. Aesthetic and energetic effects were also quite strongly endorsed, both with medians of 80. The bulk of participants did not endorse or only weakly endorsed experiencing hallucinatory effects (median = 20), unpleasant physiological effects (median = 10), or anxious effects of MDMA (median = 7), although these effects clearly occurred among a minority of users. Though not endorsed by the bulk of users, a sizable minority of respondents reported using MDMA to cope with their problems, with a median of 9, but approximately 10% of respondents rating this between 80-100. Two sexual/flirtatious motives similarly received endorsement from only a minority of users, with median ratings of 2 and 1 respectively. Finally, using MDMA to fit in or go along with others was the least-endorsed motivation (median = 0). For subacute effects, afterglow effects received a similar level of endorsement to comedown effects, each with a median of 50.

Figure 2. Bar plots, means, and medians (Med) for endorsement of different motivations for MDMA use. Motivations are ordered from most (‘to feel euphoric’) to least (‘to fit in’) endorsed on average. 0-100 responses have been binned for simpler representation.

Figure 3. Bar plots, means, and medians (Med) for endorsement of different acute effects from MDMA use. Effects are ordered from most (‘euphoric effects’) to least (‘anxious effects’) endorsed on average. 0-100 responses have been binned for simpler representation.

Figure 4. Bar plots, means, and medians (Med) for endorsement for comedown and afterglow effects from MDMA use (i.e. subacute effects). 0-100 responses have been binned for simpler representation.

Perceived long-term effects of MDMA

User-reported long-term effects of MDMA use are presented in Figure 5. Negative long-term effects tended to receive little endorsement from respondents, with the median of the raw responses, from 0-100, never exceeding 1. However, a minority of participants did report experiencing these negative effects. Two responses that received slightly stronger endorsement from users than the other negative effects were the possibility that MDMA had caused them to have problems with concentration or memory, as well as having caused them to have a more negative view of their lives and the future in a way that extended beyond a comedown period – these items respectively had 62% and 64% of respondents rating them between 0-9, whereas all other negative items had 73% or more respondents giving such low ratings.

Figure 1. Bar plots, means, and medians (Med) for endorsement of different long-term effects from MDMA use. 0-100 responses have been binned for simpler representation. Med = median, pos = Descriptive statistics for positive item version, neg = Descriptive statistics for negative item version, +ve = positive, -ve = negative.

There was considerably more variation among the positive effects. The most strongly endorsed positive long-term items reflected positive overarching views taking MDMA: that it had provided them with great memories (median = 100), that they were glad they had tried it (median = 100), and that their most positive experiences on MDMA had been among their best life experiences (median = 80). Many respondents also endorsed more specific positive effects: that their MDMA use had heightened their aesthetic appreciation even when not on the drug (median = 70), helped them develop more or deeper friendships (median = 70), allowed them to learn to let go and express themselves (median = 65), and to develop a more positive view of their lives and the future that extended beyond an afterglow period (median = 60). The only positive effect with median endorsement below 50 was the idea that MDMA might have helped reduce use of other drugs, which had a median of 0.

Positive effects universally received greater endorsement than their respective negative effects, indicated by PSup values consistently above the ROPE of 55% superiority, and all except use of other drugs reliably over 70% superiority (Figure 5). In short, user perceptions of the long-term effects of MDMA in this sample were far more positive than negative.

Exploratory relationships among variables

Several interesting patterns emerged from exploratory analysis of relationships between perceived long-term effects and motives for use, acute and subacute effects, and use frequency (Figure 6). The strongest relationships were for motivation to use MDMA to gain insight into oneself, which was positively associated with all positive long-term effects. Using MDMA to fit in was positively associated with every negative long-term effect, and negatively associated with multiple positive long-term effects. Motivation to use MDMA to cope with problems was also consistently positively associated with negative long-term effects. Two motivations showed somewhat unexpected or difficult to interpret associations: reporting using MDMA to enhance sex was positively associated with both positive and negative long-term effects. Using MDMA to ease romantic encounters/flirting was also positively associated with negative long-term effects. Taking MDMA to feel euphoric, one of the most-endorsed motivations, was positively associated with reporting amazing memories of MDMA use and finding use to have been a very positive experience, as well as with improved friendships.

Figure 6. Correlation matrix for Kendall’s tau coefficients for perceived long-term effects and multiple correlates. Bold text indicates significant associations at p <.0001, bold text with solid outline for p <.00001, reflecting Bonferroni-corrected .05 and .005 significance levels. Directional pointers and color shading indicate positive (orange, upwards pointing) and negative (blue, downwards pointing) associations, with pointers attached when the association reached significance.

For acute effects, heightened aesthetic appreciation, social/empathic effects, and euphoric effects were positively associated with all long-term positive effects, with the exception of ‘less use of other drugs’. Conversely, negative acute and subacute MDMA experiences (anxiety, unpleasant physiological sensations, and a comedown) were positively associated with negative long-term effects and negatively associated with a handful of positive long-term effects. In contrast, reporting a subacute ‘afterglow’ experience was positively associated with positive long-term effects and negatively associated with several negative long-term effects.

Perhaps surprisingly, MDMA use frequency was not robustly associated with very many long-term perceived effects. However, there was a positive association between highest use frequency and reports of concentration/memory problems, negative views of one’s life or the future, having shallower emotional experiences, and with taking other potentially problematic drugs.

Discussion

This study aimed to broaden our understanding of user perceptions of MDMA, with a particular focus on perceived long-term effects. Our opt-in, self-report survey resulted in a total of 886 valid responses from MDMA users regarding long-term negative effects and – to the best of our knowledge – several largely unexplored long-term positive effects. The survey responses are available as an open data set including a range of correlates of interest (e.g. motivations, acute and subacute effects, polydrug use) for other researchers to explore. Although the sample cannot be taken as a representative sample of MDMA users, we found user perceptions of MDMA’s long-term effects were far more positive than negative. The following discussion delves into the details of these findings and their possible implications for understanding MDMA use.

Perceived long-term impact of MDMA

Overall, user perceptions of the long-term impact of MDMA in our sample were far more positive than negative. Taking a broad perspective on MDMA use, the vast majority of users were glad they had tried it, and would count their best MDMA experiences as among some of the most positive in their lives. Users also gave quite high endorsement to a range of more specific positive effects, such as MDMA enhancing their aesthetic appreciation even when not on the drug, creating deeper friendships, allowing them to learn to let go and express themselves, and having a positive impact on their view of their life and the future. The only positive effect that went largely unendorsed was a reduction in use of other drugs; perhaps unsurprising as MDMA is often taken in environments where other drugs are readily available.

In contrast, the median endorsement for all negative effects did not exceed 1 (from 0-100). Despite the typically low endorsement of negative effects, the minority of participants who did report experiencing such effects should not be ignored: even low proportions of users experiencing problems may be concerning, and rates may increase depending on the user profile, situation, or extent of drug use. Negative impacts of MDMA on concentration and memory, as well as on feelings about one’s life and the future, stand out as receiving slightly more endorsement than other negative items. These types of effects were some of the more consistently reported negative effects of MDMA in previous research (Parrott, 2013). Hence, our findings further highlight that concentration, memory, and mood may be areas of concern for MDMA users.

Beyond looking at overall impressions of MDMA’s impact, we were able to explore a range of associations among variables that may highlight promising areas for further investigation. Users reporting using MDMA to fit in or cope with problems were statistically more likely to report having had negative long-term effects of MDMA use, and for the ‘fitting in’ motivation, lower endorsement of several positive effects, a pattern also observed among recreational MDMA users by Ter Bogt and Engels (2005). These associations are consistent with research in other drugs, in which motivation to use drugs either due to peer influence or to escape/cope with problems has been linked with adverse mental health outcomes (Kuhns et al., 2022a; Robinson et al., 2011; Villarosa et al., 2016). This may also highlight that although therapeutic use of MDMA is seen as a means of dealing with problems, it is unlikely that simply taking MDMA will produce benefits when one’s motivation is just to get rid of negative feelings. In therapy, MDMA is used as a tool to delve into one’s problems. In contrast, the coping motivation was framed more like escapism or avoidance, which is typically counterproductive in the long-term (Holahan et al., 2005). It is therefore of further note that taking MDMA to gain insight into oneself was the strongest positive correlate of beneficial long-term effects.

Acute and subacute effects of MDMA were also associated with long-term effects in sensible directions: generally negative effects – feeling anxious, having unpleasant physiological experiences, or a comedown – were positively associated with multiple negative long-term effects, and associated with lower endorsement of several positive long-term effects. In contrast, feeling euphoric, having heightened aesthetic experiences, connecting with people, and having an afterglow experience were associated with higher endorsement of all positive long-term effects except reduced use of other drugs.

If we take user perceptions of the long-term impact of MDMA at face value as indicating genuine effects, then it is worth considering how MDMA might induce such changes. It is increasingly recognized that drug effects can be understood as operating not only in terms of altered neurochemistry, but with reference to the subjective experiences this neurochemistry may enable or facilitate (Roseman et al., 2017, though see Olson, 2020). Several of the highly endorsed positive effects appear to be extensions of acute effects (e.g. seeing things in a positive light, heightened aesthetic appreciation, building deeper connections with others). Intense acute experiences induced by MDMA may transport the user into a subjective state that they have not previously experienced – for example, being far more intensely affected by touch and music than usual, or being unusually unguarded and open with people. Simply knowing that it is possible to inhabit such a subjective state may change the user’s perspectives and what they are able to appreciate. Having had such experiences may also help the user to move towards such states even when not on the drug.

These ideas remain speculative, but are supported by the observed associations among acute/subacute and long-term effects, and align with anecdotal reports from others who experienced profound perspective changes after taking MDMA (e.g. Harris, 2014).

Implications for prevention and harm reduction

Positive relationships between being motivated to use MDMA to fit in or to cope with problems and a plethora of negative long-term effects highlight that particular types of MDMA users may be especially susceptible to negative consequences of use. In environments where initiation of recreational drug use is common, such as college campuses (Arria et al., 2008), it may be particularly beneficial to ensure adequate access to mental health resources so that such maladaptive coping strategies are less likely to be pursued. Creating environments where a host of activities beyond partying provide opportunities for peer validation may also help prevent initiation or reduce use.

When communicating about risks associated with MDMA use, our findings also underscore some inadequacies of a ‘deficit model’ (cf. Joffe, 2003) approach to public health communication. As noted, a deficit model orientation often assumes users are simply biased towards choosing a short-term high over concern about long-term consequences. Our findings suggest that users perceive, and may therefore expect or be motivated by, a range of positive long-term effects. Moreover, even the acute effects they experience may be remembered as some of the most amazing experiences they have had, and facilitate closer friendships and new perspectives on life. Failing to appreciate these more nuanced considerations may convey to MDMA users that the communicator simply does not understand what they are doing and why, undercutting the perceived authority and validity of the message.

Limitations

The present findings speak to user perceptions of the long-term impact of MDMA, and do not confirm the existence of such effects – either positive or negative. Users may misattribute both positive and negative effects that are actually caused by other drugs: both stimulant drugs and alcohol have been linked to issues with memory and concentration (Gouzoulis-Mayfrank & Daumann, 2006; Kuhns et al., 2022b), and frequent cannabis use is associated with mood problems (Kuhns et al., 2022a). Classical psychedelic drugs have been associated with reported increases in aesthetic appreciation (Elsey, 2017). Perceived effects may also be explained by factors associated with drug use (e.g. going to parties and socializing a lot, sleep deprivation), and even the simple passage of time. Nevertheless, such perceptions can prove important both directly, in understanding user perspectives on drug use, and as sources of inspiration to further investigate the validity of such effects. Beyond the effects being user perceptions, the intake of MDMA itself is also based on user assumptions: many users did not test their MDMA before use (as has been found in previous research on MDMA use – Winstock, 2021), such that some effects could be attributable to impurities or other drugs.

It should also be stressed that the sample is not representative of MDMA users broadly, from which the likelihood of different effects among the total population of people who have used MDMA can be estimated. Because the total population of MDMA users is unknown, we also cannot weight or post-stratify the sample. Some of our outreach occurred via groups that typically have a relatively permissive or positive view of some types of drug use, and some such respondents may have been tempted to portray MDMA in a positive light, or may simply be unrepresentative of typical users. This would produce a positivity bias in the responses. On the other hand, large communities of users who are positive about a drug presumably do not arise by chance, and the presence and popularity of such groups may point to generally positive perceived effects of MDMA. We also reached users through standard word of mouth or through channels that raised awareness about risks associated with drug use (e.g. the Trimbos Institute – a Dutch research/policy institute with a focus on drug prevention and harm reduction – and Dancesafe – a non-profit organization that raises awareness of risks associated with drug consumption and provides harm reduction and educational materials). Furthermore, the relatively low endorsement of a priori more unlikely positive effects (e.g. a positive impact on use of other drugs) suggests that respondents were not just indiscriminately positive about MDMA. Future research might consider adding additional and more implausible positive effects to better gauge positivity bias. Even given the possibility of bias, in our view the difference in patterns of endorsement for positive versus negative long-term effects remains striking. The reported presence of the many positive long-term effects also highlights aspects of MDMA user experience that may have been neglected in previous research and could be studied further, even if their rates are lower than in this sample.

Finally, our presentation of a large number pairwise associations among variables was exploratory and, even with a full Bonferroni correction, can give rise to spurious relationships. The intent was to highlight broad patterns in the relationships rather than to pick specific links. We argue that a number of psychologically and mechanistically plausible and consistent patterns do emerge, which align with previous research (e.g. links between several relevant motivations and long-term effects, or negative acute or subacute effects and long-term negative effects). Even so, the pairwise associations do not incorporate information about possible confounds or covariates. Consequently, some of the observed relationships – even those variables with multiple relationships in the same direction – could result from their association with other variables in the data set or unobserved variables. For example, there may be types of people who are keen both to say that they mostly take drugs to gain insight into themselves, and to paint a positive picture of drug use. Or, some may wish to denigrate drug use, and report bad short-term and long-term effects. Both these cases could produce similar associations to those we observe. Different motives might also be related to underlying vulnerabilities or protective factors which influence the effects of MDMA. Nevertheless, the interesting and plausible relationships between variables can help guide future research that might seek to disentangle causal relationships among variables through both experimental and longitudinal approaches, or to identify risk factors.

Conclusion

In summary, among a large sample of recreational MDMA users, perceptions of positive long-term effects of MDMA greatly outshone negative perceived effects. By exploring relationships among motivations, acute effects, subacute effects, and long-term effects, we also highlighted several correlates of longer-term positive or negative effects that may be investigated in future research. This may be particularly valuable in understanding factors that increase the chances of negative long-term outcomes which, though reported by a minority of our respondents, are important. For positive long-term effects, it would be especially interesting to investigate under more controlled settings whether aspects of the acute experience do indeed ‘unlock’ some of the long-term effects - an idea which may also have relevance for clinical practice. Our findings also challenge what might be termed a ‘deficit model’ approach to understanding MDMA use, in which users are simply seen to make a bad trade-off between short-term gains and medium or long-term costs. User perceptions of MDMA indicate that many users believe they experience quite profound and positive long-term effects of use, ranking experiences on MDMA among their top experiences, and feeling that their use has positively impacted their friendships, aesthetic appreciation, perspectives on life, and many other areas. Hence, factors beyond the anticipation of a short-term high likely play into many users’ decisions to use MDMA.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

We would like to acknowledge Emanuella Schuuring, Tola Agbaje, and Bryan Opdam for their help in setting up the early stages of this project. We would also like to express our thanks to the many individuals and organizations that helped spread our questionnaire to potential respondents, and to everyone who took part.

Disclosure statement

The authors report there are no competing interests to declare.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes

1 Information on Nationality was collected but redacted in the dataset to prevent identifiability from rare combinations, as were open comments about drug effects. We have unfortunately been unable to retrieve the unredacted data from storage.