Summary and Comments – Psychedelics and Mental Health: A Population Study

Psychedelics and Mental Health: A Population Study

by Krebs, T. S., & Johansen, P-Ø. (2013)

(PLoS ONE, 8(8), e63972. DOI:10.1371/journal.pone.0063972)


Psychedelic substances, such as psilocybin-containing mushrooms, DMT-containing brews (ayahuasca), mescaline-containing cacti, ibogaine-containing roots and seeds with LSA, have been used for religious, shamanistic, divination and healing purposes by various cultures for thousands of years. Since the 1960s especially, the use of psychedelics, most commonly lysergic acid diethylamide (LSD-25), psilocybin and mescaline, has also been relatively widespread in the “Western” world for various purposes, including recreational, spiritual and religious uses.

Much research into the potential therapeutic uses of (some) psychedelics was carried out in the 1950s and 1960s, but due to political, legislative and other reasons, a long hiatus followed. In recent years, starting with DMT studies in the 1990s, and over the last 5-10 years in particular, there has been increasing renewed interest in such uses. In light of this renewed interest, research over the safety of these substances appears of great importance.

It is known that the “classic” serotonergic psychedelics, that is, LSD, psilocybin and mescaline, are physically very safe substances, non-addictive, and do not lead to violence or other major social problems in the way some psychoactive substances do. When considering the possible risks of these substances, the main concern has focused on potential mental health effects, in particular on possible links to long-term psychotic, phobic or PTSD-like symptoms. Psychedelics can indeed elicit very intense experiences in their users, often positive and even ecstatic or rapturous but sometimes negative and terrifying. In light of these extreme effects, and some (though limited) similarities in states produced by psychedelic use and those experienced during mental illness, it is not surprising that many have wondered about the connections between psychedelic use and mental health, and some case reports of long-term mental problems following psychedelic use exist.


To study possible associations between psychedelic use and mental health issues at a population level, the study this article reports on used data from the The National Survey on Drug Use and Health (NSDUC) carried out in the US between 2001-2004, arriving at a representative sample of 130,152 individuals. The participants had been asked about their lifetime use of LSD, psilocybin and mescaline, as well as past-year use of LSD. For mental health indicators, (self-reported) data was available on serious psychological distress during the worst month of the past year (measured on the K6 scale), mental health treatment (receiving in-patient or outpatient care or psychiatric medication, and feeling the need for but not receiving treatment), as well as symptoms of panic disorder, major depressive episode, mania, social phobia, general anxiety disorder, agoraphobia, posttraumatic stress disorder, and non-affective psychosis (as measured by the Composite International Diagnostic Interview, CIDI-SF). Further, as the main interest was on possible links to psychotic symptoms, the authors looked at each of the seven symptoms of non-affective psychosis included in the CIDI-SF individually.

A number of sociodemographic, psychological and drug use control variables known from previous research to have associations with mental health were selected to be adjusted for in the analyses. These variables were age, gender, race/ethnicity, propensity for risky behavior, level of education, level of income, marital status, lifetime exposure to an extremely stressful event, and lifetime use of 10 other non-psychedelic drugs.

Multivariate logistic regression was used to calculate the associations between the aforementioned past year mental health indicators and the use of psychedelics, including lifetime use of any psychedelics, lifetime use of LSD, psilocybin, peyote/mescaline, and past year use of LSD. Further, the associations between these mental health indicators and lifetime use of psychedelics in the presence or absence of other risk factors in stratified subgroups (sex, age, past year illicit drug use, lifetime exposure to an extremely stressful event) were analysed. The estimated associations were presented as adjusted odds ratios (aOR), 95% confidence intervals (CI), and p-values.

Results & Discussion

Out of the 130,152 respondents, 21,979 (13.4% weighted) reported lifetime use of any psychedelic. When compared to respondents who had never used any psychedelic, respondents with lifetime use of any psychedelic were more likely to be “younger, male, white, Native American, or more than one race, have somewhat higher income and more education, not be married, like to test self by doing risky things, experienced an extremely stressful event, and to have used all classes of illicit drugs.” Before adjusting for any of these differences, those with lifetime psychedelic use had higher scores on all the studied indicators of mental health problems.

However, once the control variables were taken into account, multivariate logistic regression analyses showed that lifetime psychedelic use on its own was not significantly associated with serious psychological distress in the worst month of the past year, nor with any of the mental health treatment variables studied. In fact, a number of significant associations with lower rates of  serious psychological distress and receiving or needing mental health treatment and the use of psychedelics were observed. For example, the use of LSD in the past year was associated with lower rates of serious psychological distress (aOR = 0.7, p= 0.01) .

Similarly, lifetime use of psychedelics was not significantly associated with any of the eight past year psychiatric symptom indicators. Again, several negative associations (less symptoms) were observed between lifetime psychedelic use and psychiatric symptoms in the past year.

In multivariate logistic regression analyses stratified by gender, age, any past year illicit drug use, and lifetime extremely stressful event, there were again no significant associations between lifetime psychedelic use and greater risk of any of the mental health outcomes. Here too, twelve different cases of associations between psychedelic use and lower rates of various negative mental health outcomes were observed. Perhaps most interesting among these were the associations found between psychedelic use and reduced psychotic symptoms. For people without a lifetime extremely stressful event, psychedelic users had a greatly lower rate of symptoms of psychosis (aOR = 0.5, 95% CI = 0.3 to 0.9, p= 0.03) and the specific schizotypal psychotic symptoms of “felt force inserting thoughts” (aOR = 0.4, 95% CI = 0.2 to 0.9, p= 0.02) and “felt force steal thoughts” (aOR 0.3, 95% CI = 0.1 to 0.7, p= 0.008). Further, among all women as well as all older people, psychedelic users again had a greatly lower rate of the psychotic symptom “felt force taking over mind” (for women, aOR = 0.5, 95% CI = 0.3 to 0.7, p =0.0005; for older people, aOR = 0.5, 95% CI = 0.3 to 0.8, p= 0.01).

In sum, no relation between lifetime use of psychedelics and any undesirable past year mental health outcomes was found. Further, a number of (relatively weak) associations were observed between the use of any psychedelic or some specific psychedelics and lower rates of mental health problems.

A number of limitations apply to this study. The retrospective, cross-sectional design of the study means no actual causal inferences can be drawn. In addition, information on all potential risk factors was not available, nor was longitudinal data on mental health or other factors. In the authors’ view, the possibility that some individuals or groups feel negative effects from psychedelics that are then counterbalanced by positive effects on mental health for others at a population level cannot be excluded. Also, due to the multiple comparisons carried out, some of the associations with weak statistical significance are likely to be due to chance. Other limitations include the self-report method of collecting data, as well as the fact that people in prisons, hospitals or military service (< 2 % of the population) were not included in the sample.

The results of this study based on a large population survey agree with recent findings from clinical studies in healthy volunteers, where no evidence of lasting adverse effects of psychedelic use have been found. They are also in line with other studies with individuals using peyote/mescaline or ayahuasca/DMT in religious rituals, where such use has been found to correlate with overall better mental health. The authors also discuss the results and limitations of other population studies in light of these results.

When discussing psychedelic use, the issue of “flashbacks”, and more recently “hallucinogen persisting perceptual disorder” (HPPD) often comes up. In this study, lifetime use of psychedelics and past year use of LSD was not associated with past year symptoms of visual phenomena, panic attacks, psychosis, or overall serious psychological distress. In the authors’ view, these findings therefore do not support the idea of “flashbacks” or HPPD as a major outcome of psychedelic use. Indeed, based on a number of studies on these phenomena, e.g., one reporting lower prevalence of LSD use in those suffering from persistent visual symptoms as compared to the general population, the authors are of the opinion that the validity of HPPD diagnosis is questionable, and that the phenomenon seems to better fit within somatic symptom disorders.

Commenting on case reports of long-term psychiatric problems attributed to LSD use, the authors identify a number of problems and point to several issues often not taken into account in such accounts. These issues include the relatively high prevalence of both mental illness and psychedelic use in the population, leading to chance associations; the similar typical onset period of both mental illness and psychedelic use (late adolescence, early adulthood); not ruling out pre-existing conditions; as well as attributing onset of illness to psychedelic use, even when they are separated by a long period of time. In conclusion, the authors state that case reports of mental health problems following psychedelic use can be compared to reports of such problems due to, e.g., intensive meditation or visiting holy sites (“Jerusalem Syndrome”), of which there exist quite a few.


With the current “mini-renaissance” in research on (mainly the therapeutic use of) psychedelics, this major population study is of great interest and importance. As researchers look into the possibilities of using psychedelic substances for therapeutic or other beneficial purposes, it is of utmost importance to study what risks, if any, they might pose to individual mental health.

To be clear, this study does not prove conclusively that the use of psychedelics poses no risks to mental health. Rather, it shows that when other potential confounding risk factors are accounted for, lifetime use of any psychedelic, of LSD, psilocybin or mescaline, and last year use of LSD, does not appear to be associated, at a population level, with increased psychological distress, need or receiving of mental health treatment or the symptoms of many major mental health problems and illnesses. This even in the situation where it is highly likely that most of the use cases in this population sample occurred in “suboptimal” sets and settings, i.e., not in the type of safe and supportive environment therapeutic use would entail.

Together with increasing evidence of no long-term adverse effects on healthy volunteers in clinical studies, these results begin to demonstrate that no major ethical issues or particular risks seem to exist in researching psychedelics, their effects and therapeutic uses further. As for the use of these substances outside therapeutic/religious contexts, the issues are more complex. But in my opinion, no evidence exists of such major risks to public health, safety or order that would necessitate their current strictly illegal status.

As regards this particular population study, it must be noted that before the confounding risk factors were taken into account, associations did exist between psychedelic use and (all) mental health indicators. Clearly, most, if not all, of these associations are explained by other factors, which the authors then took into account in further analyses. As one example, especially in the current situation where the distribution of all illegal psychoactive substances is to some extent handled by the same people, the use of psychedelic and non-psychedelic illegal substances has a high correlation, and the effects of, e.g., stimulant use, had to be separated from those of psychedelic use. It is unfortunate, however, that the study does not report on exactly to which extent each factor they considered as “confounding” explained the initial connection between psychedelic use and mental health issues, as in my view putting in such a wide variety of different control variables all in one fell swoop is a little problematic. The analyses on stratified samples do alleviate some of these concerns, however.

Another issue that was not studied here was the frequency or extent of psychedelic use. There is little to no evidence of actual addiction to psychedelics. But some people do use them repeatedly and sometimes more often than would be prudent. This study featured no data on the number of times people had used psychedelics, and it would be interesting to see the results of a similar study with similar control variables, but looking at light, moderate or heavy use of psychedelics separately. In any case, at least in a religious/spiritual context, even “heavy” use (hundreds of sessions) does not seem to be associated with mental health problems.

As regards psychedelics and psychosis, there is no doubt in my mind that an intense experience with psychedelic use can precipitate/trigger a psychotic episode/break in individuals predisposed to psychotic symptoms. It is my view that any person diagnosed with a psychotic illness or otherwise especially susceptible to such symptoms should avoid psychedelic substances entirely, together with strong stimulants, deliriants, depressants, dissociatives, cannabis and indeed most other strongly psychoactive substances. Those with cases of psychotic illness or similar symptomatology in the immediate family should also exercise great caution. In research and therapeutic use, such individuals should in my opinion be excluded.

That being said, I tend to agree with the authors of this paper in that, in light of the current evidence, the best way to understand the connection, if any, between psychedelic use and the onset/return of psychotic symptoms is to realize that any sufficiently intense, emotional and/or confusing event can trigger or worsen psychotic symptoms. The comparison to the trigger effect of visits to holy sites (so-called “Jerusalem Syndrome”) is particularly apt. Psychedelics can elicit reactions very similar to intense religious experiences. For most healthy people with no particular propensity for psychiatric symptoms, such experiences, even very intense ones, will probably produce no lasting negative effects on mental health. But for, e.g., those predisposed to psychosis, such emotional/powerful experiences, whether induced by psychedelics, deep meditation or religious fervor can act as a trigger for symptoms.