Associations between lifetime classic psychedelic use and markers of physical health

Classic psychedelics—a subclass of psychedelics that act primarily as 5-HT2A receptor agonists—have attracted renewed scientific attention for their psychological effects. The class includes tryptamines (for example DMT and ayahuasca), psilocybin, lysergamides (LSD), and phenethylamines (mescaline and mescaline-containing cacti). Earlier research, including clinical trials and population studies, indicates low physiological toxicity in controlled settings, immunomodulatory and anti-inflammatory properties, and therapeutic potential for depression, anxiety and substance use disorders. Large surveys such as the National Survey on Drug Use and Health (NSDUH) have established population-level patterns of lifetime classic psychedelic use (around 13–14% of adults) and reported associations with lower psychological distress, suicidality and some forms of substance misuse and criminal behaviour. Using pooled NSDUH data, the present study set out to examine whether lifetime classic psychedelic use (defined as having tried one or more classic psychedelics at least once) is associated with markers of physical health. Specifically, the investigators tested associations with three outcomes: self-reported overall health, body mass index (BMI) category, and a combined indicator of having a heart condition and/or cancer in the past 12 months. The a priori hypothesis was that lifetime classic psychedelic use would be associated with better physical health status.

Data were drawn from four years (2015–2018) of the National Survey on Drug Use and Health (NSDUH), pooled and weighted to represent the civilian non‑institutionalised US adult population. The unweighted analytic sample comprised 171,766 adults aged 18 or older. The NSDUH sampling and questionnaire methods are documented by the survey and were used as provided. Three dependent variables were analysed. Self‑reported overall health (HEALTH2) was recoded into four ordered categories (1 = Fair/Poor to 4 = Excellent). BMI (BMI2) was recoded following NIH/NHLBI guidelines into six categories from underweight to extreme obesity. Heart condition and/or cancer in the past 12 months combined affirmative responses to HRTCONDYR and CANCERYR into a binary variable (1 = yes to either; 0 = no to both). The primary exposure, lifetime classic psychedelic use, was a binary indicator coded positive if respondents reported ever having used any of a specified set of substances: DMT, ayahuasca, psilocybin, LSD, mescaline, or peyote/San Pedro. The authors excluded an NSDUH item that conflated DMT with other tryptamines because it did not permit identification of DMT use alone. Analyses adjusted for an extensive set of covariates chosen to mirror prior investigations and to control for potential confounding by sociodemographic and substance‑use factors. Covariates included age, sex, sexual orientation, ethnoracial identity, educational attainment, household income, marital status, self‑reported engagement in risky behaviour, lifetime use of multiple classes of illicit drugs (cocaine, stimulants, sedatives, tranquilizers, heroin, pain relievers, marijuana, PCP, MDMA, inhalants), lifetime use of various tobacco products (smokeless, pipe, cigar, daily cigarettes), and age of first alcohol use. A dichotomised Kessler Psychological Distress Scale (K6) variable was included only in the ordered logistic model for self‑reported overall health to reduce confounding by current psychological distress. Statistical methods comprised descriptive statistics and multivariable regression models incorporating NSDUH sampling weights. Ordered logistic regression was used for the ordinal overall‑health outcome, multinomial logistic regression for BMI categories (reference: normal weight), and binary logistic regression for the heart condition/cancer outcome. The authors relied on NSDUH’s built‑in imputation procedures for select variables and did not perform additional imputation; remaining missing values were treated as missing. No a priori outlier removal was performed. Although multiple comparisons were not formally controlled, the paper reports exact p‑values to four decimal places to enable readers to apply conservative corrections if desired. The extracted text does not clearly report the statistical software used.

Descriptive results indicated that approximately 14% of the pooled sample reported lifetime classic psychedelic use; the authors extrapolated this to suggest nearly 34 million US adults have used a classic psychedelic at least once, based on NSDUH population weights. In adjusted regression analyses, lifetime classic psychedelic use was associated with higher odds of reporting better overall health on the ordered scale. The multinomial logistic models showed that lifetime classic psychedelic users had significantly lower odds of being in overweight or obese BMI categories compared with being normal weight. For the combined outcome of heart condition and/or cancer in the past 12 months, lifetime classic psychedelic use was associated with lower odds, but this association only approached conventional levels of statistical significance. The authors note that odds ratios were similar when heart condition and cancer were analysed as separate dependent variables (results provided in an appendix referred to in the text). Detailed numerical effect sizes, confidence intervals and p‑values are referenced generically in the extracted text but specific values from the regression tables are not clearly present in the provided extraction.

Simonsson and colleagues interpret the pattern of associations as consistent with the possibility that classic psychedelic use may be linked to better physical health in the population. They propose several non‑exclusive mechanisms that could account for the associations: long‑term changes in health behaviour following acute psychedelic experiences (for example, improvements in diet, exercise or substance use), broader improvements in mental‑health dimensions beyond the absence of distress (including increased prosociality, trait mindfulness and a greater sense of purpose), direct immunomodulatory and anti‑inflammatory effects reported in some preclinical and early clinical research, and pharmacodynamic actions at receptor subtypes such as serotonin 2A that may be implicated in certain physical disorders. The authors explicitly caution against causal inference because the study is cross‑sectional. They acknowledge that despite adjustment for many covariates, residual confounding or response bias might explain the associations—for example, a latent trait that predisposes individuals both to trying psychedelics and to adopting healthy lifestyles. Additional limitations described include the lack of data on frequency, dose and context of psychedelic use, inability to assess individual‑level harms even if population associations are neutral or favourable, absence of information on regular anti‑inflammatory medication use (which could confound immunological pathways), and the limited utility of BMI as a marker because it does not capture fat distribution. The authors also note that some associations were modest in magnitude; nevertheless, they argue that small effects can have meaningful public‑health implications at scale and provide a hypothetical illustration using annual mortality from heart disease and cancer in the United States. Given these caveats, the investigators recommend more rigorous research to examine causality and mechanisms, including experimental studies and longitudinal designs that can assess dose, frequency and context effects as well as biological mediators.

The study concludes that lifetime classic psychedelic use is associated with higher odds of better physical‑health markers in a large, nationally representative US sample. The authors emphasise the need for more rigorous research to determine causal pathways and to evaluate whether classic psychedelics can influence physical functioning beyond their established mental‑health effects.