Further comments on the in-press meta-analysis of ecstasy user memory studies
No cause for alarm.
Ilsa Jerome Ph.D.
June 26 2007
One of the authors of the meta-analysis allowed to view the paper, and we have since corresponded about the study, and I have now viewed the publication itself.
After review and discussion, I can say with certainty that the meta-analysis does the service of examining the bulk of retrospective ecstasy user studies but cannot and should not be used as a means of assessing risks of human MDMA studies.
The current study is intended as an extension of a previous meta-analysis conducted by Verbaten in 2003 that looked at verbal memory in 2003. This meta-analysis examined verbal and visual memory, and includes an addition 16 or so studies from the ten studies Verbaten considered. However, both meta-analyses are based on retrospective studies. Like the original meta-analysis, the paper by Laws and Kokkalis found that ecstasy use was associated with lower performance on tests of verbal memory while having a negligible effect on visual memory. The current meta-analysis failed to find any verbal memory effects of cannabis use.
Despite the headlines, none of the studies included in the meta-analysis contained participants reporting use of ecstasy once or twice. Rather, even the minimum exposure of 16 tablets was greater than the ten-tablet limit set used in two of three NeXT (Netherlands XTC Toxicity) team limits, and average cumulative dose was above 300. Following Verbaten’s methods, when a study contained groups of ecstasy users varying on degree of use, they included the heavier users, and if a study contained current and abstinent users, they examined current users only. Several studies comparing ecstasy users with cannabis user controls were excluded from analysis, while studies with drug-nave or polydrug user controls were included. It is notable that studies using cannabis-user controls also tended to fail to find lower memory scores exclusively in ecstasy users (Dafters et al. 2004; Lamers et al. 2006; Simon et al. 2002) or moderate users specifically (Gouzoulis-Mayfrank 2003). Studies employing cannabis user controls often found the same types of memory problem in both ecstasy-using and cannabis-using participants.
While the lack of a dose effect for verbal memory can be interpreted as if it is a “one hit” problem, as Laws states in one news article, this is by no means the only interpretation. The lack of a dose effect may indicate that something else, such as use of another substance or changes in circadian rhythm, is the cause of this relationship. At the least, since the meta-analysis intentionally selected heavier users, it might have missed a dose relationship that studies comparing moderate and heavy users have found in the past.
A meta-analysis can compensate for small sample sizes common to ecstasy user studies, but it cannot repair problems with retrospective studies or the common confounds existing in these studies. As the authors note in the paper, most if not all findings from retrospective studies are confounded by differences in substance use, and studies that worked hard at matching for substance use tended to find fewer differences between ecstasy users and non-users. The authors could not find enough studies to examine the potential impact of other drugs upon memory. This is unfortunate given the existence of findings examining ecstasy users performing tasks involving decision-making and working memory were affected by other substances as well (see for instance Montgomery et al. 2005; Roiser et al. 2007).
Conclusions from meta-analyses are also weakened when the studies the analyses are based on use different measures or ask different questions. While ecstasy user studies tend to more often than not to use some measures of memory, they do not all sue the same measures nor are they all testing the same hypotheses. Sometimes even minor variations in measures can have strong impact on study results. Even if these variations are random, such variations can muddy interpretation of results.
What’s more, when memory is measured before and after administering one or two doses of MDMA in randomized, controlled settings, researchers do not see differences in memory. Data presented at a conference in 2003 from a study of the effects of 1.5 mg/kg (about 100 mg) MDMA failed to find effects on memory (Ludewig et al. 2003). Even one prospective naturalistic study that did find people who chose not to use ecstasy outperforming ecstasy users still reported that everyone, ecstasy users included, had normal memory scores, and this study included at least one person reporting use of 30 rather than fewer than ten tablets (Schilt et al. 2007).
This report, when published, makes use of a large number of studies of repeated users of street ecstasy to reach stronger conclusions than might be reached from each individual study. However, the report cannot draw conclusions about what effects, if any, a few doses of actual MDMA have upon verbal memory.
References
Dafters RI, Hoshi R, Talbot AC. (2004) Contribution of cannabis and MDMA (“ecstasy”) to cognitive changes in long-term polydrug users. Psychopharmacology (Berl). 173(3-4):405-410.
Gouzoulis-Mayfrank E, Daumann J. (2006) The confounding problem of polydrug use in recreational ecstasy/MDMA users: a brief overview. J Psychopharmacol 20(2):188-93.
Halpern JH, Pope HG Jr, Sherwood AR, Barry S, Hudson JI, Yurgelun-Todd D. (2004) Residual neuropsychological effects of illicit 3,4-methylenedioxymethamphetamine (MDMA) in individuals with minimal exposure to other drugs. Drug Alcohol Depend 75(2):135-147.
Huizink AC, Ferdinand RF, van der Ende J, Verhulst FC. (2006) Symptoms of anxiety and depression in childhood and use of MDMA: prospective, population based study. BMJ 332(7545):825-828. Epub 2006 Feb 24
Lamers CT, Bechara A, Rizzo M, Ramaekers JG. (2006) Cognitive function and mood in MDMA/THC users, THC users and non-drug using controls. J Psychopharmacol. 20(2):302-311.
Laws KR, Kokkalis J (2007) Ecstasy and memory function; a meta-analytic update. Hum Psychopharmacol Clin Exp. Published online before print
Paper under discussion Ludewig S, Ludewig K, Hasler F, Vollenweider FX (2003) No lasting effects of moderate doses of MDMA (Ecstasy) on memory performance and mood states in health humans. Biol Psychiatry 53 (Suppl); 205S. Data presented at the 58th Annual Conference of the Society for Biological Psychiatry, San Francisco, CA.
Medina KL, Shear PK, Corcoran K.. (2005) Ecstasy (MDMA) exposure and neuropsychological functioning: a polydrug perspective. J Int Neuropsychol Soc 11(6):753-765.
Montgomery C, Fisk JE, Newcombe R, Wareing M, Murphy PN. (2005C) Syllogistic reasoning performance in MDMA (Ecstasy) users. Exp Clin Psychopharmacol 13:137-145.
Roiser JP, Rogers RD, Cook LJ, Sahakian BJ. (2006) The effect of polymorphism at the serotonin transporter gene on decision-making, memory and executive function in ecstasy users and controls. Psychopharmacology (Berl). 188(2):213-227
Schilt T, de Win MM, Koeter M, Jager G, Korf DJ, van den Brink W, Schmand B. (2007) Cognition in novice ecstasy users with minimal exposure to other drugs: a pr
ospective cohort study. Arch Gen Psychiatry 64(6):728-736.
Third NeXT team study
sys/w3pb.pl?mode=search&c_pkey=22908&displayformat=allinfo&type=citation
Commentary and report on Schilt et al. 2007
Simon NG, Mattick RP. (2002) Abstract The impact of regular ecstasy use on memory function. Addiction 97(12):1523-1529.
Verbaten MN. (2003) Specific memory deficits in ecstasy users? The results of a meta-analysis. Hum Psychopharmacol 18(4):281-290.
Initial meta-analysis performed on 10 ecstasy user studies
Commentary on news reports concerning the ecstasy user cognitive function meta-analysis
I gratefully acknowledge the insight and comments of Neal Goldsmith Ph.D. concerning the effects of inter-study variation on meta-analyses
No cause for alarm: Despite claims to the contrary, the recently published meta-analysis of studies of memory in ecstasy users does not demonstrate that memory is impaired even after a few doses of ecstasy. Here are our conclusions reached after reviewing the paper and correspondence with one of the study authors.
