Perceived risk of LSD use in the United States from 2015-2019: Are Americans assessing LSD’s risk profile more favorably?

doi:10.21203/rs.3.rs-2667155/v2

Download PDF

Article

Perceived risk of LSD use in the United States from 2015-2019: Are Americans assessing LSD’s risk profile more favorably?

https://doi.org/10.21203/rs.3.rs-2667155/v2

This work is licensed under a CC BY 4.0 License

Version 2

posted

You are reading this latest preprint version

Background: Most Americans perceive lysergic acid diethylamide (LSD) to be a high-risk drug, despite infrequent serious adverse events associated with its use. Though LSD use is rising in the United States, little has been written about whether perceived risk of LSD (prLSD) is changing or factors influencing prLSD.

Methods: Using National Survey on Drug Use and Health data from 2015-2019, we investigated correlates of and temporal trends in prLSD.

Results: On multivariable modeling for respondents 18 years of age and older, lower prLSD was associated in a statistically significant manner with later survey year, personal LSD use, younger age, higher education level, male gender, identifying as a sexual minority, having less self-influential religious beliefs, past year psychological distress, and other factors. Higher prLSD was associated in a statistically significant manner with identifying as Black or Hispanic, past year suicide attempt, and having children in the home. From 2015-2019, there was a statistically significant linear decreasing trend in proportion of respondents who perceived “great” risk in trying LSD [from 70.5% to 64.8%; (p<0.0001)].

Conclusions: We observed important differences in prLSD among subgroups, suggesting factors such as gender, race, ethnicity, and personal history of LSD use could significantly influence prLSD. Our findings also indicate that Americans as a whole appear to be assessing LSD’s risk profile more favorably in recent years.

Health sciences/Medical research/Epidemiology

Health sciences/Health care/Public health

Lysergic acid

LSD

psychedelics

risk perception

survey

National Survey on Drug Use and Health (NSDUH)

Risk perception is one of several factors influencing decision-making around drug use (1–5). Greater risk perception is associated with lower lifetime use of many drugs, and decreasing risk perception has previously coincided with increased use of various drugs at the population level (6). Risk perceptions of cigarette smoking (7,8) and marijuana use (9) have recently declined in the United States (US), while those of alcohol use have remained relatively stable (10). Although past year LSD use in the US rose 56% from 2015 to 2018 (0.55% to 0.86%) (11), surprisingly little scholarly work has thoroughly explored perceived risk of LSD use (prLSD) in the US and whether it is changing.

According to data from the National Survey on Drug Use and Health (NSDUH), most Americans perceive LSD use as carrying a “great” risk of harm (12). NSDUH data reveal that the percentage of Americans 12 and older perceiving “great” risk from trying LSD fell from 73.7% in 2002 to 68.0% in 2014 (13). Relative to other drugs with routinely gathered national data on risk perceptions, LSD sits approximately in the middle of the risk perception hierarchy in the US. According to the 2019 NSDUH (14), 64.8% of respondents considered trying LSD a “great” risk, compared to 85.1% for trying heroin. In addition, 79.5% believed using LSD weekly poses a “great” risk, compared to 93.7% for heroin, 85.7% for cocaine, 44.8% for binge drinking, and 29.0% for marijuana.

These perceived risk data contrast with previous research on actual risk, which indicate LSD has extremely low addictive potential and minimal direct physiological toxicity (15). This likely results from the fact that the public’s risk perceptions of drugs, and governmental regulatory classifications of drugs based on risk perceptions, are often inconsistent with actual evidence of harm potential, particularly in the case of alcohol (16). For example, in a prominent 2010 study, substance use experts in the United Kingdom rated commonly used drugs according to their harm potential for users and others. The analysis included an assessment of nine categories of harm potential for users (four categories of physical harm, three categories of psychological harm, and two categories of social harm). In analyses for both harms to users and others, LSD was ranked as one of the least harmful drugs, despite its class A status in the United Kingdom [similar to Schedule I in the US]. Alcohol, heroin, cocaine, and methamphetamine were deemed most harmful (17). Furthermore, in a survey of 1,501 respondents published the same year, these expert rankings were highly correlated with harm ratings of people who had used or knew someone who had used a particular drug (18).

With this in mind, it appears that Americans may be overestimating the risks of LSD use. Given increasing LSD use and the limited scholarship on prLSD in the US, we sought to determine demographic correlates of prLSD and how prLSD has changed among the general population and subpopulations in recent years using NSDUH data from 2015 to 2019. We hypothesized that race, age, gender, education level, and personal history of LSD use would be influential determinants of prLSD and that prLSD would fall among the cohort during the study period.

Data Source Description

The NSDUH is conducted annually nationwide by the U.S. Substance Abuse and Mental Health Services Administration (SAMHSA) (19). It is administered in person to randomly selected, noninstitutionalized civilians and inquires about substance use, mental health conditions, and other health-related issues, as well as treatments received for mental health conditions and substance use disorders. To improve generalizability, the NSDUH employs a sample-weighted design with weight adjustments for demographics, non-response, and other factors (20).

We used pooled data from NSDUH survey years 2015 to 2019 on all respondents (age 12 and older), which included 282,768 people. Although 2020 survey data were available, data collection that year was disrupted due to COVID-19 and comparing that year’s data to previous years’ has been advised against (21). The primary outcome for this study was prLSD, as ascertained by respondent answers to the question: “How much do people risk harming themselves physically and in other ways when they try LSD once or twice?”. The four responses ranged from “no risk” to “great risk”, and were coded on a 4-point ordinal scale. 8,951 respondents did not answer this question and were excluded.

Statistical Analysis

Statistical analysis was conducted in SAS 9.4 (SAS Institute, Cary, NC). Unweighted frequencies by perceived risk of trying LSD (no risk, slight risk, moderate risk, great risk) are reported, but unweighted frequencies for individual variables are not reported, similar to other publications that utilize NSDUH data. Continuous variables are summarized as weighted medians and interquartile ranges, and univariable comparisons between levels of risk perception were evaluated with log-transformed linear regression. Categorical factors are summarized using weighted percentages and 95% confidence intervals, and univariable comparisons between levels of risk perception were evaluated using Rao-Scott chi-square tests.

Several cumulative logistic regressions were performed with prLSD as the outcome variable to calculate adjusted odds ratios for the increase in prLSD across the four risk responses. Two sets of multivariable models were constructed. The smaller multivariable models were constructed with individual predictor variables of interest (e.g., cocaine use, ever assaulted anyone), and were adjusted for survey year (continuous), age (continuous), education level (continuous), employment (full time, part time, unemployed, other), race (White, Black, Asian, Other), religious beliefs, marital status (never been married, married, divorced, widowed), gender (male, female), and criminal arrest history (yes/no) (11,14). Three separate larger multivariable models were constructed to identify factors associated with higher perceived risk of LSD use: 1) respondents under the age of 18 with appropriate variables removed [see below], 2) respondents 18 and older with all variables, 3) all respondents with some factors adjusted to incorporate the non-response of respondents under the age of 18. Larger models included most of the individual variables and all of the adjustment variables listed above. Individual variables that were not included were strongly associated with a variable that was included, and therefore judged as represented in the final model. Historical drug use for specific substances were also combined into drug classes (hallucinogen use, stimulant use, sedative use) for the larger models. Note that hallucinogen use excluded LSD use, which was analyzed separately. The SAS procedure PROC SURVEYLOGISTIC with orthogonal polynomial contrasts was used to evaluate trends across survey years (22–26). This SAS procedure accounts for the complex survey design of NSDUH, including respondent weight, replicate, and variance strata. Orthogonal polynomial contrasts test the parallel-lines assumption by simultaneously testing the equality of separate slope parameters for each variable. The variances of the regression parameters were estimated with Taylor series (linearization) methods. This approach was used for all trend analyses. Tests about trends in prLSD were adjusted for all respondent variables in the “larger” models described above. Tests for trends in perceiving specific drug use to be “great” was adjusted for all respondent variables in the “smaller” models as described above. Tests for trends in perceiving LSD use to be “great” within each categorical variable of interest was adjusted for respondent age only. Trends were also evaluated visually to inspect fit.

Of note, respondents under age 18 were not asked survey questions about mental health and certain demographics such as sexual identity. Additionally, some questions were answered overwhelmingly with a single choice in this age group: married status (99% never been married), education (99% some high school but no diploma), employment status (70% unemployed or other), and number of children present in the household (either 0 or 1 only). Analyses on questions restricted by age (over/under age 18) were appropriately stratified.

Models were simplified wherever appropriate to reduce complexity and address multicollinearity and confounding. Given associations between drug use and perceived risk, as well as lack of research on factors associated with prLSD, independent variables in our multivariable logistic regression models consisted primarily of factors associated with LSD use. These are detailed in the Supplemental Appendix 1.

Survey analysis procedures were used due to the complex survey design, utilizing the NSDUH respondent weight, replicate, and variance strata. Since this study utilizes five years of survey data, the respondent weight was divided by five for analyses that included all study years, as recommended by the NSDUH (26). Significance was assessed at p<0.05.

Ethics

This study was approved by the Cleveland Clinic Institutional Review Board.

Univariable associations between participant factors and prLSD

The final unweighted sample size was N=273,817 with 165,942 respondents who perceived “great risk” in trying LSD; 58,989 who perceived “moderate risk”; 37,038 who perceived “slight risk”; and 11,848 who perceived “no risk.” On univariable tests, “great” prLSD was associated with, among other subgroups, disproportionately higher numbers of respondents who were: female, age 35 or older, Black race or Hispanic ethnicity, heterosexual, living in a non-metropolitan area, married, divorced/separated, or pregnant, as well as those with: a high school diploma/GED or less education, children in the household, lower income, or no suicide attempts or serious psychological distress in the past year; p<0.001 for all. Full results are reported in Table 1.

[Table 1 here]

Lifetime use or misuse of all psychoactive substances assessed, including LSD, was associated with disproportionately lower numbers of respondents with greater prLSD (Supplemental Table 1). Reporting that LSD was “probably impossible” to obtain was associated with disproportionately higher numbers of respondents reporting greater prLSD. While lifetime LSD users made up 10.0% of the sample, their distribution among levels of prLSD was: 37.5% “no risk”, 27.3% “slight risk”, 12.8% “moderate risk”, and 5.0% “great risk.” Additionally, perceiving binge drinking, cocaine use, heroin use, and marijuana use as higher risk was associated with disproportionately higher numbers of respondents perceiving “great” risk of LSD use, p<0.0001 for all (Supplemental Table 2).

Multivariable analysis of participant factors and prLSD

From the cumulative logistic regression models, notable factors associated with higher prLSD (higher odds ratio) on multivariable analysis among those 18 and older were increasing age (OR: 1.14, 95% CI: 1.13-1.14), identifying as Black (OR: 1.6, 95% CI: 1.52-1.68) or Hispanic (OR: 1.39, 95% CI: 1.32-1.46), having children in the home (OR: 1.04, 95% CI: 1.01-1.08), being divorced or separated (OR: 1.10, 95% CI:1.04-1.16), arrest history (OR: 1.19, 95% CI: 1.15-1.23), history of alcohol use (OR: 1.11, 95% CI: 1.04-1.18), past year suicide attempt (OR: 1.35, 95% CI: 1.16-1.58), and past year hospitalization (OR: 1.08, 95% CI: 1.02-1.14). Notable factors associated with lower prLSD in that group were later survey year (OR: 0.74, 95% CI: 0.71-0.78), never being married (OR: 0.77, 95% CI: 0.74-0.80), increasing education level (OR: 0.88, 95% CI: 0.87-0.89), lack of full-time employment (OR: 0.94, 95% CI: 0.88-0.99), male gender (OR: 0.68, 95% CI: 0.66-0.69), identifying as Asian American (OR: 0.91, 95% CI: 0.83-1.00), living in a metropolitan area (OR: 0.79, 95% CI: 0.76 -0.83), considering one’s religious beliefs to be less influential in decision making (OR: 0.54, 95% CI: 0.52-0.56), history of serious psychological distress over the past year (OR: 0.89, 95% CI: 0.85-0.92), identifying as a sexual minority (OR: 0.83, 95% CI: 0.75-0.92), history of LSD use (OR: 0.38, 95% CI: 0.36-0.40), history of other hallucinogen (OR: 0.61, 95% CI: 0.58-0.64) or marijuana use (OR: 0.67, 95% CI: 0.64-0.70), and history of stimulant (OR: 0.83, 95% CI: 0.80-0.87) or sedative misuse (OR: 0.82, 95% CI: 0.78-0.86). Findings for the under age 18 model were similar, though there was no association between prLSD and education level, history of arrest, or history of alcohol or tobacco use, while male gender was associated with higher prLSD (OR: 1.05, 95% CI: 1.01, 1.09). Figure 1 is a forest plot of the relationship between most of the categorical factors assessed and prLSD in the age 18 and older sample. For full results stratified by age group, see Supplemental Table 3.

[Figure 1 here]

[Table 2 here]

prLSD trend analysis

Using orthogonal polynomial contrasts in survey weighted logistic regression adjusted for all factors in the larger models, we identified a statistically significant linear decreasing trend in prLSD among the all-respondents sample, as well as the under age 18 and age 18 and older subsamples. The weighted proportion of respondents who reported “great” prLSD dropped 5.7% from 70.5% to 64.8% among the entire sample (trend p<0.0001) and 6.3% from 51.7% to 45.4% in the under age 18 sample (trend p<0.0001). For further details see Figure 2 shows descriptive statistics for each year in the study to visualize the trend, and Table 2 reports the results from a trend analysis for prLSD across the study period.

[Figure 2 here]

Orthogonal polynomial contrasts for decreasing linear trends across survey years for each variable of interest revealed statistically significant decreases in the proportion of respondents who believe trying LSD represents “great” risk in all subpopulations assessed except for a small number, which included Native American/Alaskan Native (p=0.12), Native Hawaiian/Other Pacific Islander (p=0.88), and, for factors only evaluated in those 18 and older, not knowing one’s sexual identity (p=0.11), being widowed (p=0.05), and attempting suicide in the last year (p=0.06). Changes over the study period in most subgroups among the 18 and older sample are included on the right side of Figure 1. We have also highlighted data from this trend analysis for racial and ethnic groups in Table 3. Detailed information for all subgroups is in Supplemental Table 4.

We also found statistically significant decreases from 2015-2019 in the proportion of respondents perceiving use of other drugs as “great risk”: weekly cocaine use (-1.5%), weekly LSD use (-5.0%), weekly marijuana use (-7.1%), monthly cocaine use (-2.1%), and monthly marijuana use (-5.3%), all significant p<.0001. No trends were identified for perceived risk of trying heroin (p=0.76), weekly heroin use (p=0.40), or binge drinking twice weekly (p=0.52). For further details see Supplemental Table 5.

[Table 3 here]

This investigation presents two primary notable findings regarding prLSD among Americans. The first is that there is considerable variation in prLSD based on gender, race and ethnicity, education level, religious beliefs, marital status, sexual orientation, urbanicity, personal history of LSD use, and other factors. The second is that Americans’ risk perception of LSD is declining among the general population, as well as most subpopulations analyzed. We will now place some of these findings into context and elaborate upon them.

Our observation of elevated prLSD in people identifying as Black and Hispanic may explain the lower rates of LSD use among these groups compared to Whites (27). These findings are also notable since previous research has found lower risk perception of trying heroin among Blacks and Hispanics compared to Whites (28), but higher risk perception of regular cannabis use (29). In line with our findings, a previous study found US-born Latinos and Latino immigrants had higher prLSD than US-born Whites and non-Latino White immigrants (30). One explanatory factor for higher prLSD in Blacks may be the significant research abuses carried out against this group in early LSD studies (31).

With approval by the US Food and Drug Administration of MDMA for posttraumatic stress disorder (PTSD) and psilocybin for treatment-resistant depression expected by 2024 (32) and 2026 (33), respectively, our findings on prLSD raise concerns that these novel treatments may not be as widely embraced by Black and Hispanic patients compared to White patients due to safety concerns. While LSD likely carries more stigma than most psychedelics given the legacy of the 1960s, it is still the most commonly used classic psychedelic in the US (34), so it may serve as a suitable proxy for perceived risk of psychedelics as a group. Of course, patients may view psychedelics as less risky when administered in a medical setting than a naturalistic one. However, the possibility of higher concerns about risks of psychedelic therapy is already being suggested by diversity-related recruitment challenges facing these trials in the US. Despite Whites making up 77% of the US population, they constituted 89% of psychedelic therapy trial participants in a 2018 study (35). Given that PTSD prevalence is highest among Blacks in the US (36) and MDMA has demonstrated impressive efficacy for severe PTSD (37), poor acceptability of MDMA-assisted therapy by Black patients would be particularly tragic. In what may be a hopeful finding on this front, we did identify linear decreasing trends in prLSD for nearly all racial and ethnic groups analyzed, including Blacks and Hispanics.

Our finding of an association between greater prLSD and older age is consistent with previous findings (6,16,38). In general, perceived risk of drugs increases with age. However, it seems likely that some of the higher prLSD among older Americans is attributable to the legacy of inflammatory social discourse on LSD in the 1960s, which centered around highly publicized claims about terrifying purported LSD side effects such as chromosomal damage, cancer, birth defects, and long term psychosis in users, which have not been supported by subsequent research (39–43).

The association we observed between lower prLSD and identifying as a sexual minority is in line with previous findings among people who have never used LSD (44). Our identified association between higher education level and lower prLSD contrasts with previous studies showing positive associations between education level and perceived risk of marijuana and heroin (28). The relationship we observed between past year serious psychological distress and lower prLSD is intriguing since level of current psychological distress was not associated with perceived risk of marijuana, tobacco, or alcohol use in a previous study (45). However, medical marijuana users have higher rates of past month psychological distress than recreational users (46), and daily marijuana use is more common in people with past month psychological distress (47). LSD use is more common among people with past year psychological distress (48,49), and it is possible that the lower prLSD in this group might result from self-treatment of mental illness with LSD and associated learning about its risk profile to maximize efficacy and safety. However, this finding could also indicate decreased risk perception secondary to psychological distress and desperation for relief without high levels of knowledge about risk profile. Unfortunately, the NSDUH does not collect data on frequency of LSD use or reasons for use. Notably, we found people with past year suicide attempts were more likely to have higher prLSD, suggesting possible awareness among these individuals that LSD use could be potentially dangerous for them given rare reports of suicide following use (50).

Over the study period there was a statistically significant linear decreasing trend (-5.7% from 2015-2019) in respondents reporting “great” prLSD. This trend was also observed in nearly all subgroups assessed. There was also a similar trend in respondents who believed using LSD weekly represents “great risk” (-5.0% from 2015-2019). Of the other substances for which perceived risk of weekly use was measured (binge alcohol use, cocaine, heroin, and marijuana), only marijuana experienced a larger drop in respondents reporting “great risk” over this period (-7.1%). The results presented here suggests LSD and marijuana are experiencing uniquely large reductions in societal risk perception.

Risk perception can be significantly impacted by both social context and objective drug knowledge (51), and there appear to be significant social changes occurring around LSD use, while information about LSD is easier than ever to access. It is unclear to what degree decreasing prLSD may reflect wider knowledge around the importance of set and setting (52) in reducing psychedelic-related harms. However, a relationship between risk perception and such knowledge has been observed in MDMA users (51). Notably, the number of “bad trips” experienced by LSD users fell dramatically from the mid-1960s through the 1970s (53), possibly due to reduced expectations of having a bad trip after heated societal discourse shifted from LSD use to other topics after LSD possession was made illegal. However, this reduction could also reflect increased dissemination of harm reduction knowledge over time.

Given the ongoing reduction in prLSD, it is important to consider whether Americans are more realistically judging LSD’s risk profile. Though Americans perceive the risk of weekly use of LSD as being closer to that of heroin and cocaine than marijuana and alcohol, expert consensus has ranked LSD as among the least harmful drugs to users and others (17). The vast majority of LSD users do not experience serious adverse medical consequences, with 0.2% of use events resulting in emergency medical attention in a recent survey of more than 10,000 past year LSD users (54). In that study, the most common adverse events requiring medical attention were anxiety, panic, and confusion. Most LSD-related adverse events are mild or moderate (55), with the majority resolving within 24 hours (54). LSD, like other serotonergic psychedelics, also has low addictive potential (15). It does not produce repetitive self-administration in animal models (56) and physical dependence is extremely rare (57). Lifetime prevalence of hallucinogen use disorder, which also includes use of hallucinogens with more addictive potential such as 3,4-methylenedioxy-methamphetamine (MDMA), ketamine, and phencyclidine, is extremely low at 0.60% in the US (58). While accidental deaths (59,60) among people taking LSD have been reported rarely, there is little evidence of lethal LSD overdose potential in physically healthy individuals, even at extremely high doses (61,62). Despite LSD use by tens of millions of Americans (63), less than a handful of toxicity-related deaths have been reported, usually amid exacerbating circumstances such as massive overdose or use of police restraints (15,64,65). Legal risks of LSD use are also low considering how few people are arrested or imprisoned for LSD possession or trafficking (66,67). There still may be social repercussions for LSD use though. We are unaware of longitudinal data on social disapproval of LSD use among American adults. However, among American 12^th graders, social disapproval of people 18 years of age and older trying LSD remains high, though it has been in steady decline since 2009, when it was 88.2%, falling to 68.7% in 2021, down significantly from the last measurement of 76.1% in 2019 (68). Since psychedelic use has reached record levels in the US recently (69), it seems probable that social acceptance of LSD use among American adults is also increasing, like that of marijuana (70). For further details on LSD-related adverse events see Supplemental Appendix 2.

Given the infrequent and usually transient nature of LSD-related harms, it is surprising that perceived risk of weekly LSD use remains only slightly lower than that of cocaine and heroin and nearly double that of weekly binge drinking, particularly when considering the significantly higher medical risk profile for these other drugs (71–73). One factor driving this discrepancy in regards to marijuana and alcohol might be LSD’s more consistent illicit status since illicit status of a drug may contribute to an unrealistically high perceived risk, which tends to decrease over time (74,75). In line with this, residents of states with medical marijuana laws have lower risk perceptions of marijuana than residents of states without them (76).

With LSD use in the US growing recently (11), it is intriguing to consider the potential contribution of decreased prLSD to this phenomenon. Though past studies indicate inverse associations between prLSD and rates of LSD use (38), this relationship is not as clearcut as it may initially appear. Risk perception is only one of multiple factors contributing to decision-making about drug use. Availability, as well as descriptive norms (perception of frequency of use among peers), injunctive norms (peer approval), and sensation seeking are other important factors affecting drug use with varying levels of influence for different substances. Supporting the importance of other factors besides prLSD in decision-making around LSD use is a previous study that found 63% of young women who used LSD in the past year reported half or more of their peers used LSD, versus 3% for those who used marijuana but not LSD in the past year, and 1% for those who used neither (38). That study also found that young women who used LSD were less disapproving of people trying LSD than those who had used only marijuana or neither drug.

Changes in risk perception predict use of tobacco, alcohol, and cannabis (77). Previous population level studies of American adolescents suggest that changes in prLSD are a leading indicator of inverse changes in use (78,79). However, one survey of college students, which did not inquire about LSD, found that for designer drugs such as MDMA, injunctive norms, sensation seeking, and perceived availability were associated with past year use, while risk perception was not (80). Increased rates of use following decrease in risk perception (a phenomenon termed the “motivational hypothesis”) has been observed for multiple drugs (77). However, for marijuana, there is also evidence for the “risk reappraisal hypothesis”, in which users may readjust unrealistic risk perceptions about a drug following its use. People have also been observed to change expectancies around alcohol use following initial consumption (81). This may be the case with LSD. Strikingly, we found that lifetime LSD use conferred a 62% decrease in likelihood of respondent report of higher prLSD, the largest reduction of any factor analyzed. However, we do not know the prLSD of LSD users in this sample prior to first use, and we are unaware of research analyzing the directionality of the relationship between prLSD and LSD use at the individual level. Intriguingly, previous research on marijuana indicates that though there are differences in risk perception between users and non-users, risk perception among users is associated with neither frequency of use or experiencing drug-related consequences (82). Unfortunately, the NSDUH does not collect data on frequency of use or history of LSD-associated adverse events to allow for a similar investigation here.

In addition to prospectively gathering data on how prLSD changes following first LSD use, future studies on prLSD and its relationship with LSD use should gather data from participants on descriptive norms, injunctive norms, what they consider the most important risks of LSD to be, and whether they know someone who has experienced an LSD-related adverse effect. Additionally, data on LSD use frequency, history of adverse effects, and reasons for use should be collected from LSD users to better understand what bearing these factors have on prLSD. Finally, studies should prospectively investigate the relationship between prLSD in a cohort of people before and after initiation of naturalistic LSD use.

Strengths and Limitations

The primary strength of this study is its use of data from multiple administrations of a large, rigorously conducted survey employing a nationally representative sample. Limitations include use of retrospective self-reported data. However, the NSDUH’s substance use self-report measures have high concordance with drug testing results (83). The NSDUH’s restriction of respondents to the civilian non-institutionalized population of the US is another important limitation. While this represents 97% of the US population (21), the NSDUH excludes people in institutional settings such as hospitals, prisons, nursing homes, and addiction treatment centers, who are likely to have important demographic and prLSD differences.

Using data from a large national survey, we found that Americans continue to perceive LSD to be nearly as risky as cocaine and heroin, despite adverse effects associated with LSD use being infrequent and typically mild to moderate in severity. However, among respondents 18 and older there were significant differences in prLSD based on survey year, age, gender, education level, sexual orientation, race and ethnicity and other factors. Lifetime LSD use was the examined factor associated with lowest odds of reporting higher prLSD. Importantly, from 2015 to 2019 there was a statistically significant linear decreasing trend in prLSD, which was also observed in nearly all subgroups assessed, including most racial and ethnic groups. This suggests that amid rising LSD use, Americans are viewing LSD’s risk profile more favorably.

Acknowledgements: This work was supported by a non-monetary scientific collaboration award from the Cleveland Clinic Center for Populations Health Research awarded to Dr. Barnett.

Funding: This work was unfunded.

Disclosure of relationships and activities: Dr. Barnett holds stock options in CB Therapeutics. He is an advisory board member of CB Therapeutics and Compass Pathways. He also receives monetary compensation for editorial work from DynaMed Plus (EBSCO Industries, Inc) and from Cerebral for consulting services. He is primary investigator on a clinical trial funded by MindMed. The other authors report no potential conflicts of interest.

Data availability statement: Data are publicly available

Author Contributions

BSB conceived and designed the study and wrote the first draft of the article. AA designed the study, interpreted data, and substantively revised the article draft. END designed the study, analyzed the data, and substantively revised the article draft. DS interpreted data and substantively revised the article draft. SMN interpreted data and substantively revised the article draft. DB interpreted the data and substantively revised the article draft. JW designed the study, interpreted data, and substantively revised the article draft. All authors approved the submitted version and have agreed both to be personally accountable for their own contributions and to ensure that questions related to the accuracy or integrity of any part of the work are appropriately investigated, resolved, and the resolution documented in the literature.

1. Nawi AM, Ismail R, Ibrahim F, Hassan MR, Manaf MRA, Amit N, et al. Risk and protective factors of drug abuse among adolescents: a systematic review. BMC Public Health. 2021 Nov 13;21(1):2088.

2. Nargiso JE, Ballard EL, Skeer MR. A systematic review of risk and protective factors associated with nonmedical use of prescription drugs among youth in the United States: a social ecological perspective. J Stud Alcohol Drugs. 2015 Jan;76(1):5–20.

3. Zimmerman GM, Farrell C. Parents, Peers, Perceived Risk of Harm, and the Neighborhood: Contextualizing Key Influences on Adolescent Substance Use. J Youth Adolesc. 2017 Jan;46(1):228–47.

4. Feldman M, Boyer B, Kumar VK, Prout M. Personality, drug preference, drug use, and drug availability. J Drug Educ. 2011;41(1):45–63.

5. Abadi MH, Shamblen SR, Thompson K, Collins DA, Johnson K. Influence of risk and protective factors on substance use outcomes across developmental periods: a comparison of youth and young adults. Subst Use Misuse. 2011;46(13):1604–12.

6. Lipari R, Ahrnsbrak RD, Pemberton MR, Porter JD. Risk and Protective Factors and Estimates of Substance Use Initiation: Results from the 2016 National Survey on Drug Use and Health. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2017 p. 1–32.

7. Pacek LR, McClernon FJ. Risk perceptions regarding cigarette smoking in the United States continue to decline. Drug and alcohol dependence. 2020 Apr;209:107887.

8. Pacek LR, McClernon FJ. Decline in the perceived risk of cigarette smoking between 2006 and 2015: Findings from a U.S. nationally representative sample. Drug and alcohol dependence. 2018 Apr;185:406–10.

9. Levy NS, Mauro PM, Mauro CM, Segura LE, Martins SS. Joint perceptions of the risk and availability of Cannabis in the United States, 2002-2018. Drug and alcohol dependence. 2021 Sep;226:108873.

10. Waddell JT. Age-varying time trends in cannabis- and alcohol-related risk perceptions 2002-2019. Addictive behaviors. 2022 Jan;124:107091.

11. Yockey RA, Vidourek RA, King KA. Trends in LSD use among US adults: 2015–2018. Drug and Alcohol Dependence. 2020;212:108071.

12. Substance Abuse and Mental Health Services Administration. 2019 National Survey on Drug Use and Health Detailed Tables. 3.1B [Internet]. 2019 [cited 2023 Feb 14]. Available from: https://www.samhsa.gov/data/report/2019-nsduh-detailed-tables

13. Lipari R, RTI International, Kroutil L, Pemberton M. Risk and Protective Factors and Initiation of Substance Use: Results from the 2014 National Survey on Drug Use and Health. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2015 p. 1–42.

14. Substance Abuse and Mental Health Services Administration. 2019 National Survey on Drug Use and Health Detailed Tables. 3.2A-3.4B [Internet]. Rockville, MD; 2019 [cited 2022 Aug 17]. Available from: https://www.samhsa.gov/data/report/2019-nsduh-detailed-tables

15. Gable RS. Comparison of acute lethal toxicity of commonly abused psychoactive substances. Addiction. 2004 Jun;99(6):686–96.

16. Cheeta S, Halil A, Kenny M, Sheehan E, Zamyadi R, Williams AL, et al. Does perception of drug-related harm change with age? A cross-sectional online survey of young and older people. BMJ open. 2018 Nov;8(11):e021109.

17. Nutt DJ, King LA, Phillips LD. Drug harms in the UK: a multicriteria decision analysis. Lancet (London, England). 2010 Nov;376(9752):1558–65.

18. Morgan CJA, Muetzelfeldt L, Muetzelfeldt M, Nutt DJ, Curran HV. Harms associated with psychoactive substances: findings of the UK National Drug Survey. Journal of psychopharmacology (Oxford, England). 2010 Feb;24(2):147–53.

19. Substance Abuse and Mental Health Services Administration. 2019 National Survey on Drug Use and Health (NSDUH): Methodological Resource Book, Section 8, Data collection final report. [Internet]. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2020. Available from: https://www.samhsa.gov/data/report/nsduh-2019-data-collection-final-report

20. Substance Abuse and Mental Health Services Administration. 2019 National Survey on Drug Use and Health (NSDUH) methodological resource book: Section 2: Sample design report [Internet]. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2020. Available from: https://www.samhsa.gov/data/report/nsduh-2019-sample-design-report

21. Substance Abuse and Mental Health Services Administration. 2020 National Survey on Drug Use and Health Public Use File Codebook. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2021.

22. Clarke TC, Black LI, Stussman BJ, Barnes PM, Nahin RL. Trends in the use of complementary health approaches among adults: United States, 2002-2012. Natl Health Stat Report. 2015 Feb 10;(79):1–16.

23. Hubert LJ. The Use of Orthogonal Polynomials for Trend Analysis. American Educational Research Journal. 1973;10(3):241–4.

24. Schauer GL, Berg CJ, Kegler MC, Donovan DM, Windle M. Differences in Tobacco Product Use Among Past Month Adult Marijuana Users and Nonusers: Findings From the 2003-2012 National Survey on Drug Use and Health. Nicotine Tob Res. 2016 Mar;18(3):281–8.

25. Schauer GL, Peters EN. Correlates and trends in youth co-use of marijuana and tobacco in the United States, 2005-2014. Drug Alcohol Depend. 2018 Apr 1;185:238–44.

26. Center for Behavioral Health Statistics and Quality. An Overview of Trend Testing Methods and Applications in NSDUH and Other Studies. [Internet]. Substance Abuse and Mental Health Services Administration; 2017 [cited 2023 Apr 15]. Available from: https://www.samhsa.gov/data/sites/default/files/NSDUH-TrendTestTask1-2015.pdf

27. Jahn ZW, Lopez J, de la Salle S, Faber S, Williams MT. Racial/ethnic differences in prevalence of hallucinogen use by age cohort: Findings from the 2018 National Survey on Drug Use and Health. Journal of Psychedelic Studies. 2021;5(2):69–82.

28. Palamar JJ, Le A, Mateu-Gelabert P. Perceived Risk of Heroin in Relation to Other Drug Use in a Representative US Sample. Journal of psychoactive drugs. 2019;51(5):463–72.

29. Pacek LR, Mauro PM, Martins SS. Perceived risk of regular cannabis use in the United States from 2002 to 2012: differences by sex, age, and race/ethnicity. Drug Alcohol Depend. 2015 Apr 1;149:232–44.

30. Ojeda VD, Patterson TL, Strathdee SA. The influence of perceived risk to health and immigration-related characteristics on substance use among Latino and other immigrants. Am J Public Health. 2008 May;98(5):862–8.

31. Strauss D, de la Salle S, Sloshower J, Williams MT. Research abuses against people of colour and other vulnerable groups in early psychedelic research. J Med Ethics. 2021 Jul 12;medethics-2021-107262.

32. Tilley C. MDMA “will be rolled out in US hospitals by 2024.” Daily Mail [Internet]. 2022 Dec 1 [cited 2023 Mar 6]; Available from: https://www.dailymail.co.uk/health/article-11481673/MDMA-approved-MONTHS-Rave-drug-statistically-significant-treating-PTSD.html

33. Lee Y. How a family’s struggle to help their son gave rise to controversial $400 million psychedelics giant Compass Pathways. Business Insider [Internet]. 2023 Feb 15 [cited 2023 Mar 6]; Available from: https://www.businessinsider.com/inside-compass-pathways-psychedelic-psilocybin-origin-story-2023-2

34. Livne O, Shmulewitz D, Walsh C, Hasin DS. Adolescent and adult time trends in US hallucinogen use, 2002-19: any use, and use of ecstasy, LSD and PCP. Addiction. 2022 Dec;117(12):3099–109.

35. Michaels TI, Purdon J, Collins A, Williams MT. Inclusion of people of color in psychedelic-assisted psychotherapy: a review of the literature. BMC Psychiatry. 2018 Jul 31;18(1):245.

36. Roberts AL, Gilman SE, Breslau J, Breslau N, Koenen KC. Race/ethnic differences in exposure to traumatic events, development of post-traumatic stress disorder, and treatment-seeking for post-traumatic stress disorder in the United States. Psychol Med. 2011 Jan;41(1):71–83.

37. Mitchell JM, Bogenschutz M, Lilienstein A, Harrison C, Kleiman S, Parker-Guilbert K, et al. MDMA-assisted therapy for severe PTSD: a randomized, double-blind, placebo-controlled phase 3 study. Nature Medicine [Internet]. 2021; Available from: https://doi.org/10.1038/s41591-021-01336-3

38. Rickert VI, Siqueira LM, Dale T, Wiemann CM. Prevalence and Risk Factors for LSD Use among Young Women. Journal of Pediatric and Adolescent Gynecology. 2003;16(2):67–75.

39. Barnett BS, Ziegler K, Doblin R, Carlo AD. Is psychedelic use associated with cancer?: Interrogating a half-century-old claim using contemporary population-level data. J Psychopharmacol. 2022 Oct;36(10):1118–28.

40. Grof S. The Effect of LSD on Chromosomes, Genetic Mutation, Fetal Development and Malignancy. In: LSD Psychotherapy. Pomona, CA: Hunter House; 1980. p. 318–47.

41. Dishotsky NI, Loughman WD, Mogar RE, Lipscomb WR. LSD and genetic damage. Science (New York, NY). 1971 Apr;172(3982):431–40.

42. Johansen PO, Krebs TS. Psychedelics not linked to mental health problems or suicidal behavior: a population study. Journal of psychopharmacology (Oxford, England). 2015 Mar;29(3):270–9.

43. Krebs TS, Johansen PO. Psychedelics and mental health: a population study. PloS one. 2013;8(8):e63972.

44. Schuler MS, Evans-Polce RJ. Perceived Substance Use Risks Among Never Users: Sexual Identity Differences in a Sample of U.S. Young Adults. Am J Prev Med. 2022 Sep 14;S0749-3797(22)00377-4.

45. Thornton LK, Baker AL, Johnson MP, Lewin T. Perceived risk associated with tobacco, alcohol and cannabis use among people with and without psychotic disorders. Addictive behaviors. 2013 Jun;38(6):2246–51.

46. Lin LA, Ilgen MA, Jannausch M, Bohnert KM. Comparing adults who use cannabis medically with those who use recreationally: Results from a national sample. Addict Behav. 2016 Oct;61:99–103.

47. Weinberger AH, Pacek LR, Sheffer CE, Budney AJ, Lee J, Goodwin RD. Serious psychological distress and daily cannabis use, 2008 to 2016: Potential implications for mental health? Drug Alcohol Depend. 2019 Apr 1;197:134–40.

48. Killion B, Hai AH, Alsolami A, Vaughn MG, Sehun Oh P, Salas-Wright CP. LSD use in the United States: Trends, correlates, and a typology of us. Drug and alcohol dependence. 2021 Jun;223:108715.

49. Han B, Blanco C, Einstein EB, Compton WM. Mental health conditions and receipt of mental health care by illicit lysergic acid diethylamide (LSD) use status among young adults in the United States. Addiction (Abingdon, England). 2022 Jan;

50. Grossman DH, Hendricks PS. Shedding Light on Classic Psychedelics and Self-Harm. J Clin Psychiatry. 2022 Jan 18;83(2):21com14268.

51. Shewan D, Dalgarno P, Reith G. Perceived risk and risk reduction among ecstasy users: the role of drug, set, and setting. International Journal of Drug Policy. 2000;10(6):431–53.

52. Hartogsohn I. Set and setting, psychedelics and the placebo response: An extra-pharmacological perspective on psychopharmacology. J Psychopharmacol. 2016 Dec;30(12):1259–67.

53. Bunce R. Social and Political Sources of Drug Effects: The Case of Bad Trips on Psychedelics. Journal of Drug Issues. 1979 Apr 1;9(2):213–33.

54. Kopra EI, Ferris JA, Rucker JJ, McClure B, Young AH, Copeland CS, et al. Adverse experiences resulting in emergency medical treatment seeking following the use of lysergic acid diethylamide (LSD). J Psychopharmacol. 2022 Aug;36(8):956–64.

55. Leonard JB, Anderson B, Klein-Schwartz W. Does getting high hurt? Characterization of cases of LSD and psilocybin-containing mushroom exposures to national poison centers between 2000 and 2016. Journal of psychopharmacology (Oxford, England). 2018 Dec;32(12):1286–94.

56. Hoffmeister F. Negative reinforcing properties of some psychotropic drugs in drug-naive rhesus monkeys. J Pharmacol Exp Ther. 1975 Feb;192(2):468–77.

57. Modak T, Bhad R, Rao R. A rare case of physical dependence with psychedelic LSD - A case report Tamonud Modak , Roshan Bhad & Ravindra Rao. Journal of Substance Use. 2019;0(00):1–3.

58. Shalit N, Rehm J, Lev-Ran S. Epidemiology of hallucinogen use in the U.S. results from the National epidemiologic survey on alcohol and related conditions III. Addictive behaviors. 2019 Feb;89:35–43.

59. Bodeau S, Bennis Y, Régnaut O, Fabresse N, Richeval C, Humbert L, et al. CASE REPORT LSD instead of 25I-NBOMe : The revival of LSD ? A case report. Toxicologie Analytique & Clinique. 2017;29(1):139–43.

60. Prahlow JA, Ross KF, Lene WJ, Kirby DB. Accidental sharp force injury fatalities. Am J Forensic Med Pathol. 2001 Dec;22(4):358–66.

61. Klock JC, Boerner U, Becker CE. Coma, hyperthermia and bleeding associated with massive LSD overdose. A report of eight cases. The Western journal of medicine. 1974 Mar;120(3):183–8.

62. Haden M, Woods B. LSD Overdoses: Three Case Reports. Journal of studies on alcohol and drugs. 2020 Jan;81(1):115–8.

63. Krebs TS, Johansen PØ. Over 30 million psychedelic users in the United States. F1000Research. 2013;2:98.

64. Fysh RR, Oon MC, Robinson KN, Smith RN, White PC, Whitehouse MJ. A fatal poisoning with LSD. Forensic Sci Int. 1985 Jul;28(2):109–13.

65. Nichols DE, Grob CS. Is LSD toxic? Forensic Sci Int. 2018 Mar;284:141–5.

66. National Drug Intelligence Center. Intelligence Bulletin LSD Trafficking and Abuse. United States Department of Justice; 2004.

67. Bureau of Justic Statistics. Drug Offenders in Federal Prison: Estimates of Characteristics Based on Linked Data. 2015.

68. Johnston L, Miech R, O’Malley P, Bachman J, Schulenberg J, Patrick M. Monitoring the Future National Survey Results on Drug Use 1975-2021: Overview, Key Findings on Adolescent Drug Use [Internet]. Ann Arbor, MI: Institute for Social Research, University of Michigan; 2022 [cited 2023 Mar 5]. Available from: https://monitoringthefuture.org/wp-content/uploads/2022/08/mtf-overview2021.pdf

69. National Institute on Drug Abuse. Marijuana and hallucinogen use among young adults reached all time-high in 2021 [Internet]. 2022. Available from: https://nida.nih.gov/news-events/news-releases/2022/08/marijuana-and-hallucinogen-use-among-young-adults-reached-all-time-high-in-2021

70. Gallup, Inc. Support for Legal Marijuana Holds at Record High of 68% [Internet]. 2021 [cited 2022 Nov 14]. Available from: https://news.gallup.com/poll/356939/support-legal-marijuana-holds-record-high.aspx

71. Bates CK. Medical risks of cocaine use. West J Med. 1988 Apr;148(4):440–4.

72. Theodorou S, Haber PS. The medical complications of heroin use. Curr Opin Psychiatry. 2005 May;18(3):257–63.

73. Rehm J, Room R, Graham K, Monteiro M, Gmel G, Sempos CT. The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: an overview. Addiction. 2003 Sep;98(9):1209–28.

74. Roy A, Wibberley C, Lamb J. The usual suspects: Alcohol, tobacco and other drug use in 15- to 6-year-old school pupils - prevalence, feelings and perceived health risks. Drugs Education Prevention & Policy. 2005 Aug 1;12:305–15.

75. Parker H, Williams L, Aldridge J. The Normalization of ‘Sensible’ Recreational Drug Use: Further Evidence from the North West England Longitudinal Study. Sociology. 2002 Nov 1;36(4):941–64.

76. Choi NG, DiNitto DM, Marti CN. Older marijuana users’ marijuana risk perceptions: associations with marijuana use patterns and marijuana and other substance use disorders. International Psychogeriatrics. 2018;30(9):1311–22.

77. Grevenstein D, Nagy E, Kroeninger-Jungaberle H. Development of risk perception and substance use of tobacco, alcohol and cannabis among adolescents and emerging adults: evidence of directional influences. Substance use & misuse. 2015 Feb;50(3):376–86.

78. Gold MS. The Epidemiology, Attitudes, and Pharmacology of LSD Use in the 1990s. Psychiatric Annals. 1994 Mar 1;24(3):124–6.

79. Johnston L, O’Malley P, Bachman J, Schulenberg J. Monitoring the Future NATIONAL SURVEY RESULTS ON DRUG USE, 1975–2009 Volume II: College Students & Adults Ages 19–50 [Internet]. Ann Arbor, MI: University of Michigan Institute for Social Research; 2009. Available from: https://files.eric.ed.gov/fulltext/ED514367.pdf

80. Kollath-Cattano C, Hatteberg SJ, Kooper A. Illicit drug use among college students: The role of social norms and risk perceptions. Addict Behav. 2020 Jun;105:106289.

81. Aas HN, Leigh BC, Anderssen N, Jakobsen R. Two-year longitudinal study of alcohol expectancies and drinking among Norwegian adolescents. Addiction. 1998 Mar;93(3):373–84.

82. Kilmer JR, Hunt SB, Lee CM, Neighbors C. Marijuana use, risk perception, and consequences: is perceived risk congruent with reality? Addict Behav. 2007 Dec;32(12):3026–33.

83. Harrison L, Martin S, Enev T, Acknowledgments D. Comparing Drug Testing and Self-Report Of Drug Use Among Youths And Young Adults in the General Population. 2007;

Table 1. Demographic Characteristics for NSDUH Survey Respondents by Perceived Risk of LSD Use from 2015-2019.
			Perceived Risk of LSD Use / One or Twice Ever
Factor	Overall		No risk	Slight risk	Moderate risk	Great risk	p-value
Factor	(N=273,817)		(N=11,848)	(N=37,038)	(N=58,989)	(N=165,942)	p-value
Survey Year							*<0.001^c*
2015	55,421	19.7 (19.4, 20.1)	16.1 (15.2, 17.0)	17.6 (17.0, 18.2)	19.0 (18.5, 19.5)	20.4 (20.0, 20.9)
2016	55,158	19.8 (19.5, 20.1)	16.8 (16.0, 17.7)	18.7 (18.0, 19.5)	19.4 (18.7, 20.0)	20.3 (20.0, 20.6)
2017	54,495	20.0 (19.8, 20.3)	19.5 (18.6, 20.5)	20.3 (19.5, 21.0)	19.8 (19.2, 20.3)	20.1 (19.8, 20.4)
2018	54,514	20.2 (19.9, 20.4)	22.3 (21.1, 23.5)	20.5 (19.9, 21.2)	20.6 (20.1, 21.1)	19.9 (19.6, 20.2)
2019	54,229	20.2 (19.9, 20.6)	25.2 (23.9, 26.6)	22.8 (22.1, 23.6)	21.3 (20.6, 21.9)	19.3 (18.9, 19.7)
Male Gender by age	273,817	48.6 (48.3, 48.9)	64.9 (63.4, 66.5)	58.2 (57.4, 58.9)	53.3 (52.7, 54.0)	44.9 (44.6, 45.3)	*<0.001^c*
Male Gender, under age 18	64,082	51.1 (50.5, 51.7)	58.7 (56.1, 61.4)	48.6 (47.3, 49.9)	50.2 (49.4, 51.1)	51.7 (50.9, 52.5)	*<0.001^c*
Male Gender, age 18+	209,735	48.3 (48.0, 48.7)	65.7 (64.0, 67.5)	59.5 (58.6, 60.5)	53.9 (53.2, 54.6)	44.5 (44.1, 44.9)	*<0.001^c*
Age Groups*							*<0.001^c*
age: 12-17	64,082	8.8 (8.7, 9.0)	11.6 (11.1, 12.2)	12.8 (12.3, 13.2)	15.5 (15.1, 15.8)	6.4 (6.2, 6.5)
age: 18-25	68,329	12.7 (12.5, 12.9)	20.9 (19.9, 22.0)	19.5 (19.1, 19.9)	18.1 (17.6, 18.5)	9.8 (9.6, 9.9)
age: 26-34	43,001	14.5 (14.3, 14.8)	21.8 (20.8, 22.9)	19.2 (18.5, 19.9)	16.9 (16.4, 17.3)	12.8 (12.5, 13.1)
age: 35-49	55,416	22.5 (22.3, 22.8)	20.0 (18.8, 21.1)	22.1 (21.4, 22.7)	20.3 (19.8, 20.9)	23.3 (22.9, 23.6)
age: 50+	42,989	41.4 (41.0, 41.9)	25.7 (24.1, 27.2)	26.5 (25.5, 27.5)	29.3 (28.5, 30.0)	47.8 (47.3, 48.3)
Age Groups (over/under age 18)*							*<0.001^c*
under18	64,082	8.8 (8.7, 9.0)	11.6 (11.1, 12.2)	12.8 (12.3, 13.2)	15.5 (15.1, 15.8)	6.4 (6.2, 6.5)
18+ years	209,735	91.2 (91.0, 91.3)	88.4 (87.8, 88.9)	87.2 (86.8, 87.7)	84.5 (84.2, 84.9)	93.6 (93.5, 93.8)
Race/Hispanicity							*<0.001^c*
Non-Hispanic White	160,940	63.3 (62.8, 63.8)	63.8 (62.4, 65.3)	73.1 (72.4, 73.7)	67.3 (66.5, 68.1)	60.7 (60.1, 61.2)
Non-Hispanic Black/African American	34,657	12.0 (11.6, 12.3)	9.5 (8.7, 10.3)	6.5 (6.1, 7.0)	9.2 (8.8, 9.7)	13.7 (13.3, 14.1)
Non-Hispanic Native American/Alaskan Native	3,936	0.55 (0.50, 0.59)	0.84 (0.66, 1.02)	0.44 (0.37, 0.51)	0.50 (0.42, 0.58)	0.56 (0.50, 0.63)
Non-Hispanic Native Hawaiian/Other Pacific Islander	1,344	0.37 (0.33, 0.42)	0.46 (0.30, 0.62)	0.25 (0.19, 0.32)	0.35 (0.29, 0.41)	0.40 (0.35, 0.44)
Non-Hispanic Asian	12,181	5.4 (5.2, 5.6)	5.9 (5.2, 6.6)	5.4 (4.9, 5.9)	6.0 (5.6, 6.5)	5.2 (4.9, 5.4)
Non-Hispanic more than one race	10,482	1.8 (1.8, 1.9)	2.2 (1.9, 2.5)	2.3 (2.1, 2.5)	2.0 (1.9, 2.1)	1.7 (1.6, 1.8)
Hispanic	50,277	16.6 (16.2, 17.0)	17.2 (16.2, 18.3)	12.0 (11.5, 12.5)	14.6 (14.1, 15.1)	17.9 (17.4, 18.3)
Sexual Identity, age 18+							*<0.001^c*
Heterosexual	191,522	94.4 (94.3, 94.6)	87.5 (86.5, 88.5)	90.0 (89.5, 90.4)	92.5 (92.1, 92.8)	95.9 (95.8, 96.1)
Lesbian or Gay	4,574	1.9 (1.8, 2.0)	3.7 (3.1, 4.3)	3.5 (3.2, 3.9)	2.6 (2.3, 2.8)	1.4 (1.3, 1.5)
Bisexual	10,203	3.1 (3.0, 3.2)	7.4 (6.6, 8.1)	6.1 (5.8, 6.4)	4.5 (4.3, 4.8)	2.1 (2.0, 2.2)
Don't know	1,067	0.52 (0.47, 0.57)	1.5 (1.10, 1.8)	0.42 (0.27, 0.58)	0.43 (0.31, 0.55)	0.50 (0.46, 0.55)
County Metropolitan Status							*<0.001^c*
Large Metro	123,260	55.8 (55.3, 56.3)	60.9 (59.4, 62.5)	60.1 (59.0, 61.1)	56.7 (55.9, 57.5)	54.7 (54.1, 55.2)
Small Metro	96,553	30.0 (29.5, 30.5)	28.5 (27.1, 29.9)	29.2 (28.3, 30.1)	30.4 (29.7, 31.1)	30.1 (29.5, 30.7)
Non-metro	54,004	14.2 (13.8, 14.5)	10.6 (9.6, 11.5)	10.7 (10.1, 11.3)	12.9 (12.5, 13.4)	15.2 (14.8, 15.7)
Education level, under age 18							---
some HS, no diploma	63,364	98.9 (98.8, 99.0)	98.6 (97.8, 99.3)	98.9 (98.7, 99.2)	98.9 (98.7, 99.1)	98.9 (98.7, 99.0)
HS diploma/GED	616	0.91 (0.82, 1.00)	1.3 (0.63, 2.1)	0.88 (0.64, 1.1)	0.80 (0.66, 0.95)	0.95 (0.82, 1.07)
some college, no de	84	0.16 (0.11, 0.21)	0.08 (0.00, 0.17)	0.11 (0.02, 0.21)	0.23 (0.12, 0.34)	0.14 (0.09, 0.18)
2-year college degree	3	0.00 (0.00, 0.01)	0	0	0.00 (0.00, 0.01)	0.00 (0.00, 0.01)
4-year college degree	15	0.04 (0.01, 0.06)	0	0.06 (0.00, 0.13)	0.03 (0.00, 0.08)	0.03 (0.00, 0.06)
Education level, age 18+							*<0.001^c*
some HS, no diploma	26,713	12.4 (12.2, 12.7)	13.4 (12.3, 14.5)	7.9 (7.4, 8.5)	8.6 (8.2, 9.1)	13.9 (13.6, 14.3)
HS diploma/GED	55,603	24.8 (24.5, 25.1)	23.0 (22.0, 24.1)	18.6 (18.0, 19.3)	22.0 (21.4, 22.5)	26.5 (26.1, 26.8)
some college, no degree	51,133	21.6 (21.4, 21.9)	22.7 (21.3, 24.2)	22.2 (21.5, 22.8)	23.1 (22.4, 23.7)	21.2 (20.9, 21.5)
2-year college degree	19,516	9.3 (9.1, 9.6)	7.4 (6.7, 8.2)	8.3 (7.8, 8.8)	9.2 (8.7, 9.7)	9.6 (9.3, 9.9)
4-year college degree	56,770	31.8 (31.4, 32.3)	33.4 (31.8, 34.9)	42.9 (41.8, 44.1)	37.2 (36.4, 37.9)	28.8 (28.3, 29.3)
Marriage Status, age 18+							*<0.001^c*
Married	86,219	51.8 (51.3, 52.2)	38.1 (36.3, 40.0)	42.2 (41.2, 43.2)	46.7 (45.7, 47.6)	55.1 (54.6, 55.6)
Widowed	6,466	5.8 (5.6, 6.0)	3.6 (2.8, 4.3)	2.8 (2.5, 3.2)	3.7 (3.4, 4.0)	6.9 (6.7, 7.1)
Divorced or separated	22,634	13.9 (13.6, 14.1)	10.6 (9.8, 11.4)	11.5 (10.7, 12.2)	12.1 (11.5, 12.6)	14.8 (14.5, 15.1)
Never Been Married	94,416	28.5 (28.2, 28.9)	47.6 (45.9, 49.3)	43.5 (42.5, 44.5)	37.5 (36.8, 38.3)	23.2 (22.9, 23.6)
Number of children under age 18 in household							*<0.001^c*
None	202,211	73.8 (73.5, 74.0)	79.2 (77.9, 80.5)	77.9 (77.1, 78.6)	75.3 (74.6, 75.9)	72.4 (72.1, 72.7)
One	29,586	10.9 (10.8, 11.1)	9.9 (8.9, 10.9)	9.2 (8.8, 9.6)	10.5 (10.0, 10.9)	11.4 (11.2, 11.6)
Two	26,376	9.8 (9.6, 9.9)	7.1 (6.4, 7.9)	9.0 (8.5, 9.5)	9.3 (8.9, 9.6)	10.1 (9.9, 10.3)
Three or more	15,308	5.6 (5.4, 5.7)	3.8 (3.3, 4.2)	3.9 (3.6, 4.2)	5.0 (4.8, 5.2)	6.1 (5.9, 6.2)
Employment Status, under age 18							*<0.001^c*
Employed full time	1,747	2.4 (2.3, 2.6)	2.4 (1.8, 3.0)	2.0 (1.6, 2.3)	2.0 (1.8, 2.3)	2.8 (2.6, 3.1)
Employed part time	8,687	12.8 (12.5, 13.2)	11.5 (10.0, 13.0)	12.4 (11.5, 13.3)	12.4 (11.7, 13.1)	13.3 (12.8, 13.9)
Unemployed	3,796	5.7 (5.4, 5.9)	7.2 (6.1, 8.4)	5.1 (4.5, 5.8)	5.2 (4.8, 5.5)	6.1 (5.7, 6.4)
Other	49,852	79.0 (78.6, 79.5)	78.9 (77.0, 80.8)	80.5 (79.4, 81.6)	80.4 (79.5, 81.2)	77.8 (77.2, 78.3)
Employment Status, age 18+							*<0.001^c*
Employed full time	109,572	49.7 (49.3, 50.1)	51.0 (49.4, 52.6)	55.5 (54.4, 56.5)	54.1 (53.3, 54.9)	47.8 (47.4, 48.1)
Employed part time	33,023	13.1 (12.9, 13.4)	14.1 (12.8, 15.4)	16.2 (15.5, 16.9)	14.9 (14.4, 15.4)	12.2 (12.0, 12.5)
Unemployed	12,586	4.3 (4.2, 4.5)	7.3 (6.5, 8.1)	4.9 (4.5, 5.3)	4.9 (4.5, 5.2)	4.0 (3.8, 4.1)
Other	54,554	32.8 (32.5, 33.2)	27.7 (26.0, 29.4)	23.4 (22.5, 24.3)	26.2 (25.5, 26.9)	36.0 (35.6, 36.4)
Number of Employers / 12 months							*<0.001^c*
One	119,864	74.6 (74.3, 75.0)	65.6 (64.0, 67.3)	69.0 (68.0, 70.0)	71.9 (71.0, 72.8)	76.9 (76.6, 77.3)
Two	39,391	18.3 (18.1, 18.6)	22.6 (21.2, 24.0)	21.9 (21.2, 22.7)	20.0 (19.2, 20.7)	17.0 (16.7, 17.3)
Three	10,678	4.4 (4.3, 4.6)	6.9 (5.9, 7.8)	6.1 (5.6, 6.5)	5.3 (4.9, 5.6)	3.8 (3.6, 3.9)
Four or more	5,702	2.6 (2.5, 2.7)	4.9 (4.2, 5.6)	3.1 (2.8, 3.4)	2.8 (2.5, 3.2)	2.3 (2.1, 2.4)
Total Family Income							*<0.001^c*
Less than $20,000	51,836	16.0 (15.7, 16.3)	20.5 (19.4, 21.6)	14.3 (13.8, 14.9)	14.3 (13.9, 14.8)	16.5 (16.2, 16.8)
$20,000 - $49,999	83,300	29.2 (28.8, 29.5)	28.9 (27.7, 30.1)	24.7 (24.1, 25.3)	27.0 (26.3, 27.7)	30.4 (30.0, 30.9)
$50,000 - $74,999	42,510	15.9 (15.6, 16.1)	15.1 (13.8, 16.4)	16.1 (15.5, 16.8)	15.7 (15.1, 16.3)	15.9 (15.6, 16.3)
$75,000 or More	96,171	38.9 (38.4, 39.5)	35.6 (34.1, 37.0)	44.8 (43.8, 45.8)	42.9 (42.1, 43.8)	37.1 (36.5, 37.7)
Poverty Level							*<0.001^c*
Living in Poverty	51,392	14.6 (14.3, 14.9)	19.1 (17.9, 20.2)	12.6 (12.2, 13.1)	13.8 (13.3, 14.2)	14.9 (14.5, 15.2)
Income ≤2X Fed Poverty Threshold	58,930	19.8 (19.6, 20.1)	20.1 (19.1, 21.1)	16.1 (15.6, 16.7)	18.2 (17.6, 18.7)	20.9 (20.5, 21.2)
Income >2X Fed Poverty Threshold	161,336	65.6 (65.1, 66.1)	60.8 (59.6, 62.0)	71.2 (70.5, 72.0)	68.1 (67.4, 68.7)	64.3 (63.8, 64.8)
Num of Skipped Work Days / 30 Days, median [Q1, Q3]	150,892	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	*<0.001^b*
Working Cell Phone	273,550	97.8 (97.7, 97.9)	97.2 (96.5, 97.8)	98.5 (98.3, 98.7)	98.3 (98.1, 98.5)	97.6 (97.5, 97.7)	*<0.001^c*
Participated in Government Assistance Program	273,817	18.4 (18.1, 18.8)	20.5 (19.2, 21.7)	15.0 (14.4, 15.5)	16.9 (16.4, 17.5)	19.3 (18.9, 19.6)	*<0.001^c*
Covered by Any Health Insurance	271,699	90.7 (90.5, 90.9)	88.0 (87.1, 88.9)	90.8 (90.4, 91.2)	91.4 (91.0, 91.7)	90.6 (90.4, 90.9)	*<0.001^c*
Body Mass Index (BMI), median [Q1, Q3]	261,991	26.8 [23.3, 31.3]	25.4 [22.3, 29.5]	25.7 [22.4, 29.9]	25.8 [22.5, 30.5]	27.4 [23.7, 31.9]	*<0.001^b*
Currently Pregnant (Female only, excluding "don't know")							*<0.001^c*
Yes	3,557	3.3 (3.1, 3.4)	2.8 (2.0, 3.7)	2.3 (2.0, 2.7)	3.1 (2.8, 3.4)	3.5 (3.3, 3.8)
No	106,022	96.7 (96.6, 96.9)	97.2 (96.3, 98.0)	97.7 (97.3, 98.0)	96.9 (96.6, 97.2)	96.5 (96.2, 96.7)
Number of Religious Services / 12 months							*<0.001^c*
25 or More	181,714	68.0 (67.7, 68.2)	74.9 (73.8, 76.0)	69.0 (68.2, 69.7)	67.0 (66.4, 67.7)	67.7 (67.4, 68.0)
Less than 25	88,782	32.0 (31.8, 32.3)	25.1 (24.0, 26.2)	31.0 (30.3, 31.8)	33.0 (32.3, 33.6)	32.3 (32.0, 32.6)
My religious beliefs influence my decisions							*<0.001^c*
Strongly Disagree	52,423	17.5 (17.3, 17.8)	37.3 (35.7, 38.9)	28.7 (28.0, 29.4)	20.2 (19.6, 20.7)	14.0 (13.7, 14.3)
Disagree	49,660	15.7 (15.5, 15.9)	19.9 (18.8, 21.0)	22.8 (22.2, 23.5)	20.4 (19.8, 21.0)	13.1 (12.9, 13.4)
Agree	94,605	35.1 (34.8, 35.3)	27.3 (26.2, 28.4)	32.9 (32.2, 33.5)	37.5 (36.8, 38.3)	35.2 (34.9, 35.5)
Strongly Agree	72,099	31.7 (31.4, 32.0)	15.5 (14.1, 16.9)	15.6 (15.0, 16.3)	21.9 (21.3, 22.5)	37.7 (37.3, 38.0)
My religious beliefs are very important							*<0.001^c*
Strongly Disagree	50,442	17.8 (17.5, 18.0)	36.5 (35.0, 38.0)	27.3 (26.5, 28.1)	19.5 (18.9, 20.0)	14.8 (14.5, 15.1)
Disagree	40,040	13.0 (12.8, 13.2)	18.8 (17.7, 19.8)	21.3 (20.7, 22.0)	17.5 (17.0, 18.1)	10.2 (9.9, 10.4)
Agree	93,934	33.5 (33.2, 33.8)	27.7 (26.4, 29.0)	31.9 (31.2, 32.6)	37.0 (36.2, 37.7)	33.1 (32.8, 33.5)
Strongly Agree	84,177	35.8 (35.5, 36.1)	17.0 (15.9, 18.2)	19.4 (18.8, 20.1)	26.0 (25.4, 26.7)	41.9 (41.6, 42.3)
Historical Criminal Record
Committed theft (>$50)	272,814	1.04 (0.99, 1.09)	2.7 (2.2, 3.2)	1.9 (1.7, 2.1)	1.3 (1.2, 1.4)	0.74 (0.69, 0.80)	*<0.001^c*
Sold Illegal Drugs	272,711	1.5 (1.5, 1.6)	5.6 (5.0, 6.1)	3.4 (3.2, 3.6)	1.7 (1.6, 1.9)	0.96 (0.89, 1.02)	*<0.001^c*
Attacked Someone with Intent to Harm	272,856	1.2 (1.1, 1.2)	2.4 (2.0, 2.8)	1.8 (1.7, 1.9)	1.6 (1.5, 1.7)	0.93 (0.86, 0.99)	*<0.001^c*
Arrested	273,045	15.8 (15.5, 16.0)	23.3 (21.9, 24.7)	21.0 (20.5, 21.6)	16.8 (16.2, 17.4)	14.3 (14.0, 14.6)	*<0.001^c*
Number of Arrests / 12 months							*<0.001^c*
None	30,291	88.0 (87.6, 88.4)	82.6 (80.4, 84.9)	87.7 (86.6, 88.7)	86.5 (85.5, 87.5)	88.9 (88.4, 89.4)
One time	5,027	8.9 (8.6, 9.3)	12.5 (10.6, 14.4)	9.1 (8.2, 9.9)	10.2 (9.3, 11.2)	8.2 (7.7, 8.7)
Two times	1,119	2.1 (1.9, 2.3)	2.8 (2.0, 3.6)	2.2 (1.7, 2.8)	2.3 (1.8, 2.9)	2.0 (1.8, 2.3)
Three or more times	629	0.95 (0.83, 1.07)	2.0 (1.4, 2.7)	1.05 (0.74, 1.4)	0.98 (0.67, 1.3)	0.82 (0.68, 0.96)
Health over the past year
Hospitalized overnight	272,560	9.6 (9.4, 9.8)	7.8 (7.0, 8.6)	7.3 (6.8, 7.8)	8.6 (8.3, 8.9)	10.3 (10.0, 10.5)	*<0.001^c*
Serious psychological distress	209,735	11.4 (11.2, 11.5)	18.6 (17.4, 19.7)	18.3 (17.7, 18.8)	14.4 (13.9, 14.9)	9.3 (9.1, 9.4)	*<0.001^c*
Attempted to kill self	208,452	0.56 (0.52, 0.60)	1.2 (0.95, 1.5)	0.79 (0.68, 0.90)	0.69 (0.60, 0.78)	0.46 (0.42, 0.51)	*<0.001^c*
Ever used a needle to inject Drugs	273,817	1.7 (1.6, 1.8)	4.1 (3.7, 4.6)	3.4 (3.1, 3.7)	2.2 (2.0, 2.4)	1.1 (1.06, 1.2)	*<0.001^c*
Frequencies presented are unweighted counts. P-values: a=linear regression; b=linear regression with log transformation; c=Rao-Scott chi-square test. SAS Survey Procedures used for all analyses.

Table 2. Trend Analysis for Lower Perceived Risk of Trying LSD Once or Twice
	Percent of Responses within each Year					Percent
Perceived Risk	2015	2016	2017	2018	2019	Change	Trend	p-value
All Respondents
No Risk	2.9%	3.0%	3.5%	3.9%	4.4%	+1.5%	Linear	<0.0001
Slight Risk	10.0%	10.6%	11.3%	11.4%	12.6%	+2.6%
Moderate Risk	16.6%	16.8%	17.0%	17.6%	18.1%	+1.5%
Great Risk	70.5%	69.6%	68.2%	67.0%	64.8%	-5.6%

Under Age 18 Only
No Risk	4.3%	4.2%	4.7%	4.8%	5.3%	+1.1%	Linear	<0.0001
Slight Risk	14.4%	15.7%	16.1%	16.3%	18.4%	+4.1%
Moderate Risk	29.7%	29.8%	29.1%	31.5%	30.9%	+1.1%
Great Risk	51.7%	50.2%	50.2%	47.4%	45.4%	-6.3%

Age 18+ Only
No Risk	2.8%	2.9%	3.3%	3.8%	4.3%	+1.6%	Linear	<0.0001
Slight Risk	9.6%	10.1%	10.9%	10.9%	12.1%	+2.5%
Moderate Risk	15.3%	15.6%	15.8%	16.3%	16.9%	+1.6%
Great Risk	72.3%	71.5%	70.0%	68.9%	66.7%	-5.6%
Trend analysis was conducted using orthogonal polynomial contrasts for survey year in multivariable logistic regression to test for linear trends in the perceived risk of LSD use across the study period and adjusted for all factors in the “large” models as described in the methods section of this paper. All analysis was adjusted for the complex survey design of the NSDUH using respondent weight, replicate, and variance strata.

Table 3. Percent of respondents according to race/ethnicity who perceived the risk of trying LSD to be "great" by survey year with test for linearly decreasing trend over time.
Race/Hispanicity	2015	2016	2017	2018	2019	Avg Ann		p-value
Race/Hispanicity	(N=35,317)	(N=34,308)	(N=33,134)	(N=32,350)	(N=30,833)	%∆	Trend	p-value
Non-Hispanic White	67.9 (67.0, 68.8)	66.9 (66.0, 67.9)	65.4 (64.5, 66.3)	64.0 (63.2, 64.7)	61.6 (60.7, 62.5)	-1.57	Linear ↓	<.0001
Non-Hispanic Black/African American	80.1 (78.9, 81.2)	78.5 (77.0, 79.9)	78.4 (77.0, 79.9)	76.9 (75.5, 78.2)	74.9 (73.2, 76.6)	-1.30	Linear ↓	<.0001
Non-Hispanic Native American/Alaskan Native	76.1 (71.8, 80.4)	65.8 (60.0, 71.7)	68.1 (62.6, 73.5)	69.1 (63.9, 74.4)	70.3 (63.5, 77.2)	-1.44	No trend	0.12
Non-Hispanic Native Hawaiian/Other Pacific Islander	71.5 (63.7, 79.3)	70.4 (61.8, 79.0)	74.2 (66.2, 82.3)	68.6 (62.2, 75.0)	75.1 (69.8, 80.4)	0.88	No trend	0.88
Non-Hispanic Asian	67.8 (64.6, 70.9)	66.9 (64.0, 69.8)	66.4 (63.7, 69.1)	61.9 (58.6, 65.2)	64.8 (61.0, 68.6)	-0.74	Linear ↓	0.03
Non-Hispanic more than one race	66.4 (62.6, 70.2)	66.0 (62.8, 69.3)	64.0 (59.4, 68.6)	65.7 (61.1, 70.2)	55.0 (51.6, 58.4)	-2.85	Linear ↓	<.0001
Hispanic	74.7 (73.5, 76.0)	74.9 (73.5, 76.2)	72.6 (71.0, 74.2)	73.3 (72.2, 74.3)	70.2 (68.9, 71.5)	-1.12	Linear ↓	<.0001
Avg Ann %∆ = average annual percent-change. Frequencies presented are unweighted counts. All trend tests are adjusted for respondent age. Orthogonal polynomial contrasts were used to test for decreasing linear trends, and were adjusted for age only. All analysis was adjusted for the complex survey design of the NSDUH using respondent weight, replicate, and variance strata.

Competing interest reported. Dr. Barnett holds stock options in CB Therapeutics. He is an advisory board member of CB Therapeutics and Compass Pathways. He also receives monetary compensation for editorial work from DynaMed Plus (EBSCO Industries, Inc) and from Cerebral for consulting services. He is primary investigator on a clinical trial funded by MindMed. The other authors report no potential conflicts of interest.

4162023PerceivedRiskLSDSUPPLEMENT.docx
Supplement

Download PDF

Version 2

posted

You are reading this latest preprint version

Perceived risk of LSD use in the United States from 2015-2019: Are Americans assessing LSD’s risk profile more favorably?

Status:

Version 2

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusions

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 2