Mescaline

Psychological and cognitive effects of long-term peyote use among Native Americans

This correlational study (n=176) investigated the long-term residual psychological and cognitive effects of peyote use amongst native American Church members, compared between regular users (n=61), minimal users (n=79), and members with a history of alcohol dependence (n=36). Only members with prior alcohol dependence showed neuropsychological deficits, but there was no link between psychological or cognitive deficits linked to peyote use.

Authors

  • Halpern, J. H.
  • Hudson, J. I.
  • Pope Jr, H. G.

Published

Biological Psychiatry
individual Study

Abstract

Background: Hallucinogens are widely used, both by drug abusers and by peoples of traditional cultures who ingest these substances for religious or healing purposes. However, the long-term residual psychological and cognitive effects of hallucinogens remain poorly understood.Methods: We recruited three groups of Navajo Native Americans, age 18-45: 1) 61 Native American Church members who regularly ingested peyote, a hallucinogen-containing cactus; 2) 36 individuals with past alcohol dependence, but currently sober at least 2 months; and 3) 79 individuals reporting minimal use of peyote, alcohol, or other substances. We administered a screening interview, the Rand Mental Health Inventory (RMHI), and ten standard neuropsychological tests of memory and attentional/executive functions.Results: Compared to Navajos with minimal substance use, the peyote group showed no significant deficits on the RMHI or any neuropsychological measures, whereas the former alcoholic group showed significant deficits (p < .05) on every scale of the RMHI and on two neuropsychological measures. Within the peyote group, total lifetime peyote use was not significantly associated with neuropsychological performance.Conclusions: We found no evidence of psychological or cognitive deficits among Native Americans using peyote regularly in a religious setting. It should be recognized, however, that these findings may not generalize to illicit hallucinogen users.

Unlocked with Blossom Pro

Research Summary of 'Psychological and cognitive effects of long-term peyote use among Native Americans'

Introduction

Hallucinogen use is widespread and increasing, yet the long-term psychological and cognitive consequences remain poorly characterised. Previous studies have been hampered by methodological problems, chiefly that users of one hallucinogen have typically used many other illicit substances, and by potential confounders such as premorbid cognitive differences, concurrent psychiatric disorder, and recent intoxication. The peyote-containing Native American Church (NAC) offers a distinctive opportunity to study long-term exposure to a single naturally occurring hallucinogen, mescaline, because many adherents ingest peyote regularly in a religious context while largely abstaining from alcohol and other drugs. Halpern and colleagues therefore set out to assess whether long-term ceremonial peyote use is associated with residual psychological or neuropsychological deficits. The investigation compares longstanding NAC members with two Navajo comparison groups — one reporting minimal substance use and one comprising formerly alcohol-dependent but currently sober individuals — using a standard mental health inventory and a battery of neuropsychological tests focused on memory and attentional/executive functions. The aim was both to detect any peyote-related impairments and to verify that the test battery could identify deficits in a group expected to show alcohol-related impairment.

Methods

The study recruited Navajo participants aged 18–45 into three groups: NAC members who had ingested peyote on at least 100 occasions (the peyote group); former alcoholics with at least five years of heavy drinking but sober at least two months (the former alcoholic group); and a comparison group reporting minimal lifetime use of peyote, alcohol, or other substances. Recruitment used a Navajo case finder and evaluations were performed off the Navajo Nation reservation. Written informed consent was obtained and the protocol was approved by an institutional review board. Screening included demographic and medical history, a semistructured assessment of lifetime psychiatric disorders, a comprehensive substance-use history, the Rand Mental Health Inventory (RMHI), and the Wide Range Achievement Test-3 (WRAT-3) reading subtest to assess English reading level. A breathalyser was used for candidates in the former alcoholic group and a close contact was interviewed to confirm sobriety. A formal neurological examination was not conducted. The investigators excluded individuals with conditions likely to affect cognition, current psychoactive medication use, substantial lifetime use of other illicit drugs (thresholds specified for cocaine, stimulants, opioids, sedative-hypnotics, hallucinogens other than peyote, hydrocarbons, and cannabis), excessive alcohol history in the peyote or comparison groups, low English reading skills (WRAT-3 <30), or a current DSM-IV Axis I disorder other than simple or social phobia. The text contains a minor inconsistency in reported sample counts during screening (a footnote listing slightly different Ns); the Results section reports the final analysed sample sizes. Neuropsychological testing occurred at least seven days after the most recent peyote ceremony for peyote users, and within four weeks of baseline evaluation for others. A tester blinded to group status administered primarily nonverbal measures to reduce cultural or linguistic bias; the battery included the WAIS-R subtests (Vocabulary, Digit Span, Digit Symbol, Block Design), the Rey–Osterrieth Complex Figure (ROCF), and tests of attentional/executive function including the Wisconsin Card Sorting Test (WCST). Analyses compared the peyote group with the comparison group, and the former alcoholic group with the comparison group, using multivariate linear regression adjusted for age and sex. The investigators also repeated analyses restricted to participants with WRAT-3 scores of at least 40 (approximately eighth-grade reading level), and examined associations within the peyote group between log-transformed lifetime peyote episodes and test scores. Because many outcome measures were correlated, p values were reported without correction for multiple comparisons.

Results

Baseline interviews were conducted with 311 Navajos; 135 individuals were excluded or failed to return for neuropsychological testing, yielding final samples of 61 participants in the peyote group, 36 in the former alcoholic group, and 79 in the comparison group. The groups differed somewhat in age and gender distribution but were similar in education, reading skill, and English vocabulary; all groups showed mean English vocabulary scores below Western normative medians. Screening for hallucinogen persisting perception disorder ("flashbacks") identified no cases among peyote candidates. On the RMHI, the former alcoholic group reported significantly greater pathology than the comparison group on all nine mental health scales. By contrast, the peyote group showed no significant differences from the comparison group on most RMHI scales and scored significantly better on two scales. Neuropsychological testing revealed no significant differences between the peyote and comparison groups on any measure. The former alcoholic group performed worse than the comparison group on the immediate condition of the ROCF and showed more perseverative errors on the WCST. A direct comparison of former alcoholics with peyote users showed a ROCF immediate-condition estimated difference of −4.0 (SE 1.3), p = .002. Restricting the sample to participants with WRAT-3 scores ≥40 (56 of 61 peyote, 32 of 36 former alcoholic, and 64 of 79 comparison participants) produced little change in the peyote-versus-comparison estimates; deficits in the former alcoholic group were slightly larger in this restricted sample. For example, the Block Design subtest difference between former alcoholics and comparison participants widened to −4.5 (SE 1.9), p = .017, and the Digit-Symbol difference became −4.9 (SE 2.6), p = .062. Within the peyote group, log-transformed lifetime peyote use was not significantly associated with any neuropsychological test measure (all p values >.1). However, higher lifetime peyote use did correlate with significantly better scores on five of nine RMHI scales, including the composite Mental Health Index. The investigators note that significance levels are presented without correction for multiple comparisons and that some modest p values could reflect chance findings.

Discussion

Halpern and colleagues interpret their findings as showing no evidence of residual psychological or cognitive deficits associated with longstanding peyote use in a bona fide religious context. The peyote group did not differ from a low-substance-use comparison group on neuropsychological measures, and in some RMHI domains showed better functioning; lifetime peyote exposure was not linked to poorer neuropsychological performance. By contrast, the formerly alcohol-dependent group exhibited marked psychological distress across RMHI scales and modest neuropsychological deficits consistent with impaired visuospatial memory and increased perseveration, suggesting some persistent frontal-lobe related dysfunction after heavy alcohol use. The authors emphasise several limitations. Selection bias is possible, since individuals with marked cognitive impairment might have been less likely to volunteer, although recruitment methods were the same for all groups. Residual confounding cannot be ruled out, including the possibility of baseline cognitive differences prior to substance exposure. Histories relied on self-report with limited external validation. The multiplicity of correlated outcome measures complicated correction for multiple comparisons, so some findings of modest significance may be chance associations. The cross-sectional design prevents causal inference; sociocultural or psychological factors related to NAC participation could contribute to the observed differences. The investigators also acknowledge the possibility that their test battery may have lacked sensitivity to detect subtle peyote-related deficits, but note that it did detect deficits in the former alcoholic group despite that group's smaller size. Finally, generalisability to other hallucinogens is uncertain, since mescaline-containing peyote may differ pharmacologically and phenomenologically from substances such as LSD or psilocybin. The authors conclude that further studies of residual effects of hallucinogens in populations with minimal exposure to other drugs are warranted.

View full paper sections

SECTION

I llicit use of hallucinogens is widespread and increasing in the United States; between 1999-2001, more than 2 million Americans tried lysergic acid diethylamide (LSD) for the first time. Despite longstanding concerns about the possible toxicity of these compounds, knowledge regarding their long-term psychological and cognitive effects remains limited. Studies of the residual cognitive effects of hallucinogens are conflicting and subject to substantial methodological problems. In particular, hallucinogen users in virtually all studies had also used many other illicit drugs, making it difficult to determine which substances contributed to observed deficits. Other possible confounding variables included subjects' premorbid cognitive deficits, current psychopathology, and acute or recent intoxication with alcohol or other drugs. To assess the residual effects of long-term hallucinogen use, one should ideally examine individuals with extensive exposure to hallucinogens, but minimal exposure to other drugs. To our knowledge, only one large population in the United States offers this opportunity: the 300,000 Native Americans who regularly (and legally) (American Indian Religious Freedom Act Amendments of 1994) ingest the peyote cactus (Lophophora williamsii), which contains the hallucinogen mescaline (␤-3,4,5-trimethoxyphenethylamine), as a religious sacrament during all-night prayer ceremonies in the Native American Church (NAC). NAC members accept peyote as a God-given medicine offering spiritual and physical healing for the betterment of all Native peoples. NAC members may attend prayer ceremonies as often as two or three nights in a week or as infrequently as once a year, but most members attend on average one ceremony a month. Many NAC members will therefore ingest peyote hundreds or thousands of times in their lifetime but, in adherence to their faith, strictly abstain from alcohol or other drugs, except for smoking tobacco at times of prayer. Thus, working with these Native Americans offers a unique opportunity to examine the long-term effects of a hallucinogen in isolation from other confounding substances. Mescaline is a particularly interesting hallucinogen to study, since its structure is partially homologous to LSDand it has historically been used as a reference standard in hallucinogen research, with the psychoactive potency of other hallucinogens expressed in "mescaline units". Though mescaline has the lowest potency of the orally active naturally-derived hallucinogens (1:2500 to 1:4000 mescaline:LSD), a full dose (200 to 400 mg) has a long duration of action, with peak effects 2 to 4 hours after consumption, declining over the next 8 hours). The physiological and psychological effects of mescaline are similar to LSD: both are sympathomimetic, profoundly alter perception of self and reality, increase suggestibility, and intensify emotions. With both substances, some users experience a deeply mystical/transcendental state, while others (especially those ill-prepared or with strong histories of mental illness) experience dysphoric symptoms. In comparison to LSD, mescaline is described as more sensual and perceptual and less altering of thought and sense of self. However, in one double-blind clinical trial, subjects were not able to distinguish mescaline from LSD. A psychoactive dose of mescaline is contained in the amount of peyote typically consumed by an adult member in a NAC ceremony. The mechanism of action of mescaline, like psilocybin and LSD, is hypothesized at the molecular level to result from its effects as a partial agonist of 5-HT2a receptors within the central nervous system. Peyote is also of interest because it is reputed to be an effective treatment for alcoholism when used in the NAC religious context). Of course these reported benefits might be primarily attributable to participation in the NAC religion, rather than to peyote itself. Notably, however, the efficacy of hallucinogens for treating substance dependence is also supported by anthropological reports from other traditional societies, by animal studies using the botanical hallucinogen ibogaine, and by human studies using LSD. Thus it seems possible that peyote and other hallucinogens might have specific pharmacological properties of potential value for treating substance dependence. Before further investigating any such possible clinical effects, however, it would seem important to establish whether hallucinogens produce adverse residual effects of their own -a possibility that has dampened enthusiasm for research in this area over the last few decades. In light of these considerations, we approached members of the Navajo Nation to perform psychological and neuropsychological evaluations of NAC members. This group represented a large group of potential study participants: among 255,000 enrolled members of the Navajo tribe, about one third are NAC members, and almost all of these individuals live on tribal land or close by. We also recruited two comparison groups of Navajosone reporting virtually no peyote or other substance use, and one reporting past alcohol dependence -since alcohol represents a serious problem among many Native Americans. We included the former alcoholic group not only because of its public health importance, but also to assess the sensitivity of our test battery.

METHODS AND MATERIALS

With the assistance of a Navajo case finder, we recruited three groups of Navajos aged 18 -45: 1) NAC members who had ingested peyote on at least 100 occasions (the peyote group); 2) a former alcoholic group reporting at least five years of drinking more than 50 12-ounce beers (or equivalent) per week, but currently sober at least 2 months; and 3) a comparison group reporting minimal use of any substance. We performed all evaluations off the reservation of Navajo Nation. After complete description of the study to the subjects, written informed consent (approved by the McLean Hospital Institutional Review Board) was obtained. At a baseline evaluation, a trained psychiatrist recorded demographic information, medical history, comprehensive substance use history, and lifetime history of psychiatric disorders as determined by semistructured questions. We then administered the Rand Mental Health Inventory (RMHI;and the Tan Reading Subtest of the Wide Range Achievement Test-3 (WRAT-3; Jastak. In addition, we administered a breathalyzer (Alco-Sensor IV, Intoximeters, Inc., St. Louis, Missouri) to candidates for the former alcoholic group and interviewed a close contact of each candidate to confirm duration of sobriety. A formal neurological assessment was not conducted. We excluded participants reporting 1) a history of head injury or other medical condition that might affect cognitive function; 2) current use of psychoactive medications; 3) lifetime use of cocaine, stimulants, opioids, sedative-hypnotics, hallucinogens other than peyote, or hydrocarbon inhalants more than 10 times, or cannabis more than 100 times; 4) for the former alcoholic and comparison groups, lifetime use of peyote more than 5 times; 5) for the peyote and comparison groups, consumption of more than 5 alcohol-containing beverages per day continuously for one month or more at any time, or any other alcohol consumption qualifying for a diagnosis of alcohol dependence in accordance with the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association 1994); 6) English reading skills below the third-grade level, as indicated by a score of less than 30 on the WRAT-3; and 7) evidence of a current DSM-IV Axis I disorder (American Psychiatric Association 1994) other than simple or social phobia. c Hollingshead-Redich classes 2 or 3 (there were no participants in class 1). Note that n ϭ 60 for peyote group, 33 for former alcoholic group and 78 for comparison group because of missing data. d "Substance" refers to peyote in the case of the peyote group and alcohol in the case of the former alcoholic group. We specifically screened the 80 potential participants (Figure) in the peyote group for a history of hallucinogen persisting perception disorder ("flashbacks"); none reported this condition. We did not formally ask members of the peyote group about tolerance to peyote, as we are aware of no evidence that tolerance develops with the intermittent pattern of use that described most NAC members; we also did not hear any anecdotal descriptions of individuals requiring progressively larger doses. Participants satisfying all criteria were invited to return for a battery of neuropsychological tests, chosen to assess particularly for impairment in memory and attentional/executive functions. The test battery was similar to the batteries administered in our previous studies of cannabis usersand MDMA users, as well as in published studies of alcoholism. For the present study, however, we focused almost entirely on nonverbal measures, because we doubted the reliability of Western verbal tests in Navajo participants, many of whom had grown up speaking Diné (the Navajo language) rather than English. This problem is illustrated by participants' performance on the vocabulary subscale of the Wechsler Adult Intelligence Test-Revised (WAIS-R;, where the three study groups achieved similar mean scores -but all fell about 1.3 standard deviations below average scores for normative Western populations (Table). By contrast, participants' performance was generally comparable to Western norms on nonverbal tests (see Results below) -suggesting that these tests were less vulnerable to linguistic or cultural bias. Peyote group participants were asked to perform these tests at least seven days after their most recent peyote meeting; all others were tested within four weeks of the baseline evaluation. On the day of testing, an investigator, blinded to group status, first administered a breathalyzer test to ensure that participants were alcohol-free. She then administered a battery of neuropsychological tests that included the 1) Vocabulary Subtest of the WAIS-R (as already noted above); 2) WAIS-R Digit Span Subtest; 3) WAIS-R Digit Symbol Subtest; 4) WAIS-R Block Design Subtest; 5) Rey-Osterreith Complex Figure. WRAT, Wide Range Achievement Test-3; OCD, obsessive compulsive disorder; MDD, major depressive disorder. peyote group with the comparison group, and the former alcoholic group with the comparison group, separately using multivariate linear regression, adjusting for age and sex. We also repeated the analyses while restricting the sample to the 152 participants reading at least at an eighth-grade level (WRAT-3 scores of at least 40). Finally, within the peyote group, we assessed the association between log-transformed lifetime episodes of peyote use and scores on all test measures. Because of correlations between many test measures, it was difficult to calculate an appropriate correction for multiple comparisons. Accordingly, the significance levels of findings are shown without correction. Thus some of the differences, especially those with modest levels of significance, might represent chance associations. We address this issue further in the Discussion section below.

RESULTS

We performed baseline interviews on 311 Navajos, of whom 135 were excluded or failed to return for the neuropsychological testing visit, leaving final samples of 61 participants in the peyote group, 36 in the former alcoholic group, and 79 in the comparison group (Figure). The groups differed somewhat in age and gender distribution, but were similar in level of education, reading skills, and English vocabulary (Table) -although all groups displayed lower mean English vocabulary scores than the fiftieth percentile of Western populations, as discussed above. On the RMHI, the former alcoholic group reported significantly greater pathology than the comparison group on all 9 of the mental health scales, whereas the peyote group showed no significant differences from the comparison group on most scales and scored significantly better on 2 (Table). The neuropsychological tests yielded no significant differences between the peyote and comparison groups on any measure, whereas the former alcoholic group showed poorer performance on the immediate condition of the ROCF and on total perseverations on the WCST (Table). On the ROCF, the former alcoholic group showed greater deficits when compared directly to the peyote group (for example, on the immediate condition, the estimated difference [SE] was Ϫ4.0 [1.3]; p ϭ .002). The restricted sample of participants with WRAT-3 scores of at least 40 (comprising 56 [92%] of the participants in the peyote group, 32 [89%] in the former alcoholic group, and 64 [81%] in the comparison group) yielded little change in the estimated differences between the peyote and comparison groups on any measures. The former alcoholic group, however, showed slightly greater deficits relative to the comparison group on virtually all of the measures in Tablesand. For example, on the WAIS-R Block Design subtest, the estimated difference between the alcoholic group and the comparison group widened to Ϫ4.5 [1.9], p ϭ .017; on the Digit-Symbol Subtest, the difference became Ϫ4.9 [2.6], p ϭ .062. Finally, within the peyote group, no associations between log-transformed lifetime peyote use and any neuropsychological test measure approached significance (p Ն.1 in all cases; see Table). On the RMHI measures, however, greater lifetime peyote use was associated with significantly better scores on five of the nine scales, including the composite Mental Health Index (see Tablefor details).

DISCUSSION

The residual psychological and cognitive effects of long-term hallucinogen use are poorly understood, in part because most previous studies have evaluated hallucinogen users who were also heavy users of other illicit drugs -making it difficult to identify any effects specific to hallucinogens themselves. These effects deserve further study -first, because illicit hallucinogen use is widespread and growing in Western cultures, and second, because many members of traditional cultures, including some 300,000 Native Americans, regularly use hallucinogens as religious sacraments. We administered the RMHI and a battery of nonverbal neuropsychological tests to 61 Navajo longstanding participating members of the NAC. We compared this group to 79 Navajos reporting minimal lifetime use of peyote or any other substance, and 36 Navajos reporting at least 5 years of alcohol dependence, but currently sober at least 2 months. The peyote group showed no significant differences from the comparison group on most scales of the RMHI and scored significantly better on 2 scales. We also found no significant differences between the peyote group and comparison group on any of the neuropsychological measures. Moreover, within the peyote group, log-transformed lifetime episodes of peyote use showed no significant associations with neuropsychological measures and were associated with significantly better scores on several RMHI measures. These findings suggest that long-term use of this hallucinogenic substance, at least when ingested as a bona fide sacrament, is not associated with adverse residual psychological or cognitive effects. By contrast, we found highly significant psychological deficits and a few significant neuropsychological deficits in the former alcoholic group. The latter findings (decreased visuospatial memory on the ROCF and increased perseverations on the WCST) suggest that some deficits in frontal lobe functions may persist long after alcohol consumption has ceased. Although these deficits were modest, it should be noted that the former alcoholic participants were young individuals, reporting a median of 8 months of abstinence from alcohol, and screened to exclude cases with major neurological or psychiatric disorders or a history of any other substance abuse. A less rigorously selected group of former alcoholic individuals might have performed more poorly. Notably, the former alcoholic participants reported more cannabis consumption than the other two groups (see Table), although by design no participant exceeded 100 lifetime episodes of cannabis use. However, cannabis use seems unlikely to explain the neuropsychological and psychological deficits in the former alcoholic group, since we have shown in a previous study that even individuals reporting a median of 20,000 lifetime episodes of cannabis use displayed virtually no detectable neuropsychological test deficits after a 28-day washout. Several limitations of our study should be considered. First, selection effects likely influenced our recruitment efforts; for example, individuals with severe cognitive deficits might have been less likely to volunteer for the study. However, selection bias would not occur unless there were differential effects across groups -a less likely possibility, since all groups were recruited in the same manner. A related consideration is that we excluded a few candidates from each group on screening because of current psychiatric disorders (see Figure). However, these exclusions were infrequent and similar across groups, so that differential effects across groups were again likely minimal. Second, we cannot exclude the possibility of residual confounding, either due to unmeasured confounders or inadequate adjustment for measured confounders. For example, participants in the three groups might have differed in terms of baseline cognitive ability prior to ever ingesting peyote or alcohol. Third, participants' histories were obtained by self-report without external validation, except as indicated above. However, participants were screened without knowledge of the "right" answers needed for acceptance into the study, thus reducing the chances of false responses. These three limitations are inherent to all naturalistic studies of the long-term effects of substance use; we have discussed these methodological issues in detail in previous publications. Fourth, as noted earlier, it is difficult to calculate an appropriate correction for multiple comparisons in the present study, because many of the measures -such as, for example, the subscales of the RMHI, or scores on various cognitive tests of memory -were very closely correlated with one another. Therefore, a simple Bonferroni correction, dividing the alpha level by the total number of comparisons, would tend to over-correct, possibly causing many type II errors (failing to reject the null hypothesis when in fact a genuine difference exists). Given these considerations, we have presented the findings without correction for multiple comparisons; readers should therefore recognize that some findings of modest significance (e.g., p values between .01 and .05) may represent chance associations. However, we would note in passing that all significant differences between the former alcoholic group and the comparison group were in the same direction, arguing that the findings cannot easily be ascribed to chance. Fifth, the cross-sectional study design limits our ability to draw causal inferences. For example, the superior psychological functioning of the peyote group and inferior psychological functioning of the former alcoholic group are probably not exclusively due to pharmacological effects of peyote or alcohol themselves, but also due to sociocultural or psychological factors. Sixth, it is possible that our test battery was not sensitive enough to detect residual deficits from peyote use. Arguing against this possibility is that our tests readily detected significant psychological and cognitive deficits in the former alcoholic group, despite the lower statistical power of comparisons involving this smaller group. Thus, although we cannot exclude a type II error with the peyote group, it seems unlikely that we would have missed a psychological or cognitive deficit of major clinical significance. Of course, more complicated tasks, not performed in this study, might yet reveal differences in functioning not detected by the instruments that we used. Our findings have public health importance for several reasons. Most important, for the Native Americans who use peyote as a religious sacrament, it appears that this practice does not cause residual psychological or neuropsychological deficits detectable in the battery of tests that we administered. These observations also offer reassurance regarding the more than 10,000 NAC members who serve in the United States Armed Services; we find no evidence that a history of peyote use would compromise the psychological or cognitive abilities of these individuals. It is not clear whether our findings with peyote would apply to other types of hallucinogens. Although mescaline resembles other hallucinogens in certain respects, it may differ in other respects; for example, it does not appear to produce "flashbacks" (hallucinogen persisting perceptual disorder) in the manner of LSD. Therefore, we cannot exclude the possibility that long-term use of chemically different hallucinogens (such as LSD or psilocybin) might produce adverse residual effects, even if peyote does not. In any event, further studies of the residual effects of these substances-especially in populations with minimal exposure to other types of drugsare warranted.

Study Details

Your Library