Specific psychiatric symptoms during cannabis use and their relation to psychotic disorders

Abstract

Background and objectives:

Cannabis use is an increasing phenomenon worldwide and represents a controversial issue, as it has direct implications in the development of psychosis. There are no widely accepted tools of evaluating the risk for users and therefore no prevention strategies. As psychosis and schizophrenia are two major psychiatric disorders, developing a screening tool is of great need. The current study evaluates the feasibility of the cannabis experience questionnaire 6 (CEQ-6) questionnaire as a screening tool for cannabis users that are at high risk of developing psychosis or schizophrenia.

Methods:

In this study, 400 patients were selected from the Drug Addiction Department of the Cluj County Emergency Clinical Hospital. The cohort was divided into three groups: patients with cannabis use, but with no current or previous history of cannabis-induced psychotic disorders; patients with current or previous history of cannabis-induced psychotic disorders, and patients diagnosed with schizophrenia having a positive history of cannabis-induced psychotic disorders.

Results:

There are several items of the CEQ-6 questionnaire that can be linked with the risk of developing psychosis and schizophrenia—auditory hallucinations during use, feeling disinhibited after use, and the sensation of understanding the world better during use. Furthermore, some items can be used for assessing the risk of schizophrenia in patients with cannabis-induced psychosis.

Conclusions:

The study shows a possibility of utilizing CEQ-6 as a screening tool for identifying users at high-risk of psychosis. This is an important step toward prevention in the general population as the scale can be applied outside the psychiatric setting.

INTRODUCTION

Cannabis use has a long history worldwide and is becoming more widespread due to the international legalization of medical and recreational use. Given recent trends, cannabis cultivation is expected to continue growing. Given the possible harmful effects of its long-term use,[1] alongside the correlations between mental health disorders and cannabinoids,[2] thorough research is imperative for a better understanding of the risks associated with intake.

Studies to date show a proven link between cannabis use and the development of psychotic disorders and schizophrenia. Users are at higher risk than the general population, due to the psychoactive compounds of the plant that stimulate the endocannabinoid receptors CB1 above the physiological amount and pace during a single session of use and thus disrupting dopamine signaling.[3] There are several risk factors known so far, such as: age of use onset, quantity used,[4] Tetrahydrocannabinol (THC) and Cannabidiol (CBD) concentration in the plant[5,6] and even a genetic susceptibility.[7,8] Moreover, the cross-links between these risk factors and the development of Psychiatric disorders among patients are not well-enough studied, and could counterbalance the benefits of use as a recreational drug.

Cannabis-induced psychotic experiences share notable phenomenological overlap with symptoms typically observed in primary psychotic disorders, such as schizophrenia. Specifically, users experiencing cannabis intoxication commonly report perceptual disturbances, paranoia, auditory hallucinations, depersonalization, and derealization—symptoms that closely mirror psychotic symptoms present in schizophrenia and related disorders. Historically, distinguishing transient, cannabis-induced psychotic phenomena from chronic psychotic disorders has relied primarily on clinical history and symptom persistence beyond periods of intoxication and was focused mainly on factors such as symptom onset, duration, and context—the temporal association between substance intake and symptoms. However, objective tools or standardized screening measures for identifying subtle differences in these overlapping experiences remain limited.

Additionally, given the approximately 40% risk of developing schizophrenia after an episode of cannabis-induced psychotic disorder,[9] establishing prognosis and accurately highlighting risk play an even more important role.

Currently, it is possible to make an empirical estimation of the prognosis solely based on the clinical presentation,[10] but it is not reliable and can only be done by a specialist. Thus, a screening tool such as a questionnaire can help the process by identifying individuals at a higher risk of psychosis even outside specialized institutions and can be of use in public health.

However, there is still no clear method to establish a prognosis or an individual risk level for developing a psychotic disorder. Considering the significant impact on the patient’s life, but also on the healthcare system in general, a screening tool would be useful.

This study is grounded in the observed experiential differences between cannabis users who remain unaffected and those who develop cannabis-induced psychosis or schizophrenia. While many users experience mild, transient effects, others report symptoms such as paranoia or hallucinations, which may signal underlying vulnerability to psychosis. The cannabis experience questionnaire 6 (CEQ-6)[11] was selected to capture these subjective experiences, as it specifically assesses psychosis-like symptoms during cannabis use, and allows for structured comparison across groups.

The existing literature relies on clinical evaluations conducted by mental health professionals but they are limited in early identification. The CEQ-6 questionnaire fills the critical gap by enabling a standardized assessment of the symptoms outside specialized clinical setting and thus contributing to improved preventive strategies and public health interventions.

Although CEQ-6 offers a practical and targeted approach, it has limitations. It lacks validated subscales for diagnostic differentiation and relies on self-report, which may be influenced by recall bias. Nevertheless, it remains a promising exploratory tool for identifying high-risk individuals, warranting further validation in longitudinal studies.

OBJECTIVES

This study aims to investigate the utility of psychiatric symptoms in identifying individuals at risk of developing cannabis-induced psychotic disorders and schizophrenia. It evaluates specific symptom profiles that may serve as reliable indicators for heightened susceptibility. Furthermore, the study proposes the CEQ-6 as a reliable screening tool for the symptoms, among the population.

MATERIALS AND METHODS

Study participants

Four hundred patients were selected from the Psychiatry III clinic, belonging to the Drug Addiction Department of the Cluj County Emergency Clinical Hospital between October 2020 and December 2023. They were distributed into three groups: Group 1: patients with cannabis use, but without current or former history of cannabis-induced psychotic disorders; Group 2: patients with current or former history of cannabis-induced psychotic disorders; and Group 3: patients with schizophrenia having a positive record for cannabis-induced psychotic disorders. The approval is obtained - AVZ38/14.12.2021.

Patients were diagnosed on admission by psychiatrists, and their history data were taken from the hospital database. Also, at admissions, the patients are screened for substance use using urine drug tests. Participants with positive results for other substances were excluded from the study. After taking informed written consent for participation in the study, the diagnose was confirmed by the psychiatrists involved in the study according to the DSM-V diagnostic criteria, and the patients completed the scales with help from the interviewer. The patients were instructed to reflect primarily upon their most recent cannabis experiences not accompanied by active psychotic symptoms. The symptoms were evaluated in a general trend of subjective experiences across multiple episodes of cannabis use rather than a single event. In the case of patients with induced psychotic disorder or exacerbation of schizophrenia, they were included in the study after remission of psychotic symptoms.

The exclusion criteria for both groups were other associated psychiatric pathologies, neurological pathologies, and the use of psychoactive substances besides cannabis.

Questionnaire

Two subscales of the CEQ were used to assess symptomatology. This questionnaire, developed by Barkus et al.[11] has two subscales composed of 56 items that consider a range of symptoms that occur during cannabis use or immediately after. The items are scored on a Likert scale, which assesses the frequency of symptom occurrence from 0 to 4 (0: never, 4: almost always).

For the use of the CEQ-6 questionnaire, written consent was obtained from the author.

Statistical analysis

The data obtained were processed using R with dplyr, ltm, and logbin packages. For statistical analysis, the t-test was applied for comparison of means using R.

The means of the scores were compared between the three groups, based on the results gathered from the individual items of the scale. This procedure was done using the t-test for comparison of means.

RESULTS

General characteristics

400 inpatients were included with at least one cannabis use disorder in the study. They w divided into three groups: Group 1–Cannabis users without a history of psychotic disorders (n = 132), Group 2–Cannabis users with a history of cannabis-induced psychotic disorders (n = 128), and Group 3–Patients diagnosed with schizophrenia and a history of cannabis-induced psychotic disorders (n = 140).

Most of the patients were male (81%) with an average age of 26.44 years. Most patients came from urban areas (75.8%), were unmarried (71.6%), and lived with their families (59.6%). In terms of educational status, 17.9% completed at least higher education and 58.5% considered themselves religious.

Regarding cannabis use, the average duration of use is calculated at 7.8 years with the average age of onset being 17.96 years. Patients consumed about 1.4 g of cannabis/day on average and 34% consumed cannabis daily. The data are presented in detail in Table 1.

Table 1.

Sociodemographic and cannabis use characteristics of participants by diagnostic group

		Group 1 (Cannabis Users, No Psychosis)	Group 2 (Cannabis-Induced Psychosis)	Group 3 (Schizophrenia + History of Induced Psychosis)	Study population
Gender, masculine (%)		79.6	76.5	93.8	80.8
Urban (%)		81.3	71.9	66.7	75.8
Marital status (%)	Not married	62.5	75	93.3	71.6
	Living with partner	22.9	9.4	6.7	15.8
	Married	8.3	6.3	0	6.3
	Other	6.3	9.4	0	6.4
Religious affiliation (%)		61.7	65.6	33.3	58.5
Higher education (college degree or higher) (%)		10.4	18.7	40	17.9
Age (mean±SD)		27.54±6.51	25.65±4.19	26.06±4.83	26.67±5.62
Years of cannabis use (mean±SD)		7.86±4.52	7.44±3.06	8.4±3.83	7.8±3.961
Age of starting cannabis use (mean±SD)		18.78±6.71	17.09±3.12	17.27±3.73	17.96±5.34
Daily cannabis quantity (g) (mean±SD)		1.71±1.29	1.29±1.09	0.88±0.57	1.42±1.16
Still using cannabis (%)		62.5	55.9	46.7	57.7
Positive urine tests (%)		25.8	20.3	14.3	20
Days since last cannabis use (mean±SD)		15±5.4	21.6±7.2	32.2±9	23.1±8.2

Overall symptom profile differences across groups

Analysis of the CEQ-6 questionnaire revealed significant differences in symptom profiles between the three groups. A total of 17 items showed statistically significant differences in mean scores, with 15 items related to symptoms occurring during cannabis use and two items related to symptoms occurring after use. These symptoms fell into two broad categories: affective symptoms: happiness, fear, anxiety, excitement, and feelings of being “full of plans.” and psychotic symptoms: paranoia, auditory hallucinations, and perceptual disturbances.

After applying FDR (Benjamini-Hochberg) correction, three items remained statistically significant with adjusted P values below 0.05. The item referring to auditory hallucinations during cannabis use, had a raw P value below 0.0001 and an FDR-adjusted P value of 0.0000. The item referring to feeling disinhibited after cannabis use, had a raw P value of 0.0001 and an adjusted P value of 0.0028. Finally, the sensation of understanding the world better during cannabis use, had a raw P value of 0.0002 and an adjusted P value of 0.0037.

These three items showed the most robust differences between cannabis users without psychosis, those with cannabis-induced psychosis, and those diagnosed with schizophrenia. All other items, although initially significant at the uncorrected level, did not retain statistical significance after correction and are therefore presented as exploratory findings. The multiple regression analyses revealed that cannabis use variables, particularly daily quantity, and age of onset, were the most consistent predictors of CEQ-6 symptom severity. Daily quantity was significantly associated with 80% of the symptom items, while age of onset reached significance in 56% of cases. In contrast, the diagnostic group indicators were significant in only 16%–18% of the models after adjusting for covariates.

These findings suggest that the variation in cannabis-related experiences across individuals is more strongly explained by patterns of use than by diagnostic classification. The attenuation of group-level effects after adjustment further supports the role of cannabis exposure as a key factor influencing symptom profiles, independent of diagnostic status.

Comparison between group 1 (no psychosis) and groups 2 and 3 (psychosis and schizophrenia)

In comparing the means obtained on individual items between the group of cannabis users without psychotic disorders and the groups with psychosis, 11 items were identified as having significant mean differences. Out of these, none were present in the after-use period, all of them appearing during the effects of cannabis [Table 2].

Table 2.

Comparison of CEQ-6 symptom scores between cannabis users without psychotic disorders and those with psychosis or schizophrenia (Groups 1 vs 2 and 3)

Symptom	Relative risk (RR)	95% Confidence Interval	P
Feeling Happy	1.19	1.07–1.33	<0.01
Lethargic	0.9	0.83–0.98	0.02
Energized	1.16	1.05–1.27	<0.01
Slowing of time	0.86	0.79–0.94	<0.01
Auditory hallucinations	1.27	1.17–1.39	<0.01
Able to understand the world better	1.23	1.12–1.36	<0.01
Being relaxed	1.12	1.01–1.25	0.03
Laid back	1.19	1.07–1.33	<0.01
Full of plans	1.17	1.07–1.28	<0.01
Feeling more creative	1.15	1.04–1.26	<0.01
Feeling full of ideas	1.13	1.03–1.24	<0.01

CEQ-6=Cannabis experience questionnaire 6

Comparison between group 1 (no psychosis) and group 2 (cannabis-induced psychosis)

Upon making a comparison between the group without psychosis and those with induced psychotic disorders, but without schizophrenia, significant results were obtained in 12 items that concern intoxication symptoms [Table 3].

Table 3.

Comparison of CEQ-6 symptom scores between cannabis users without psychotic disorders and those with cannabis-induced psychosis (Groups 1 vs 2)

Symptom	Relative risk (RR)	95% Confidence Interval	P
Feeling Happy	1.20	1.05–1.37	<0.01
Feeling threatened by an unknown force	1.17	1.04–1.30	0.01
Energized	1.16	1.03–1.30	<0.01
Slowing of time	0.85	0.76–0.95	<0.01
Auditory hallucinations	1.19	1.07–1.32	<0.01
Able to understand the world better	1.23	1.09–1.39	<0.01
Having visions	1.13	1.01–1.26	0.03
Being relaxed	1.18	1.03–1.35	0.02
Laid back	1.22	1.07–1.40	<0.01
Full of plans	1.17	1.04–1.31	0.01
Feeling more creative	1.18	1.04–1.33	0.01
Things not feeling “right” on your skin or in your body	1.12	1.01–1.24	0.04

CEQ-6=Cannabis experience questionnaire 6

Comparison between group 2 (cannabis-induced psychosis) and group 3 (schizophrenia)

After comparing the groups with induced psychosis and schizophrenia, significant results were marked on seven items, 2 of which were related to the post-use period [Table 4].

Table 4.

Comparison of CEQ-6 symptom scores between individuals with cannabis-induced psychosis and those with schizophrenia (Groups 2 vs 3)

Symptom	Relative risk (RR)	95% Confidence Interval	P
Feeling fearful	1.25	1.03–1.52	0.03
Feeling threatened by an unknown force	0.78	0.65–0.93	0.01
Auditory hallucinations	1.8	1.38–2.34	<0.01
Sleepy	1.35	1.06–1.72	0.01
Things not feeling “right” on your skin or in your body	0.83	0.70–0.99	0.04
After initial effects—Paranoid without reason	1.22	1.03–1.44	0.02
After initial effects—Feeling depersonalized	1.22	1.04–1.44	0.02

CEQ-6=Cannabis experience questionnaire 6

DISCUSSIONS

This study adds to the growing evidence showing a connection between cannabis use and the development of cannabis-related psychotic disorders, as well as the possible onset of schizophrenia.[12] By analyzing responses from the CEQ-6 questionnaire, we found significant differences across different patient groups, suggesting that certain symptoms might act as early warning signs of psychosis. These results highlight the potential of the CEQ-6 as an important screening tool, especially in everyday settings, where observing these signs early can lead to quicker, more effective interventions.

Regarding socio-demographic and clinical characteristics of the study groups there are differences noted. Residential background differed between groups, with cannabis users without psychosis more frequently residing in urban areas compared to the other two groups. This difference may reflect variation in environmental exposure, healthcare access, or lifestyle factors that could influence substance use patterns and psychiatric outcomes. Social integration also appeared to decline across the groups, as the percentage of unmarried individuals increased progressively, from cannabis users without psychosis to those with schizophrenia. Very few participants with schizophrenia were married or living with a partner, which may indicate the social impact of chronic psychiatric illness.

Religious affiliation followed a similar trend, with a lower proportion of religious individuals among schizophrenia patients compared to the other groups. This may reflect broader disengagement from social and cultural institutions among individuals with more severe psychiatric conditions. In contrast, higher education levels were most frequently reported among schizophrenia patients, which, while counterintuitive, may suggest either a delayed functional decline in this group or a possible association between cognitive functioning and symptom insight.

Patterns of cannabis use also differed across groups. Although the duration of cannabis use was relatively consistent, patients with schizophrenia reported slightly longer use histories. The age at which cannabis use began was earlier among those with cannabis-induced psychosis and schizophrenia, aligning with previous research indicating that earlier exposure is associated with greater psychiatric risk. Notably, daily cannabis consumption was highest among users without psychosis and lowest among those with schizophrenia. Similarly, ongoing cannabis use was most common in the group without psychosis, and decreased progressively in the psychosis and schizophrenia groups, possibly reflecting treatment effects or decreased access and motivation in the context of severe psychiatric illness. We analyzed individual items of the CEQ-6 rather than subscales because the CEQ-6 does not have widely validated subscales for grouping its items. Given this limitation, individual symptom items were evaluated separately to explore their specific relevance and potential predictive utility. However, future research could benefit from conducting a factor analysis on CEQ-6 responses to identify coherent symptom clusters or subscales, which might enhance the questionnaire’s interpretability, predictive validity, and clinical applicability.

Auditory hallucinations, feelings of disinhibition, and altered perceptions of understanding the world emerged as the most distinctive symptoms. These findings suggest that such experiences are not characteristic of typical cannabis intoxication but may instead reflect an underlying predisposition to cannabis-induced psychotic disorders or schizophrenia.

The fact that these items remained significant after FDR correction supports their potential relevance as markers for clinical screening. Among them, auditory hallucinations closely resemble core psychotic features, while the other two suggest more subtle affective and perceptual changes that might precede or predict psychosis onset. These symptoms could reflect prodromal traits or heightened sensitivity to cannabis effects in vulnerable individuals.

Although several additional symptoms appeared significant before correction, they did not withstand FDR adjustment and should be viewed as hypothesis-generating observations. Future research should prioritize replicating these findings in larger, more diverse samples and consider factor-analytic approaches to streamline the analysis of symptom clusters. One of the most striking findings was the presence of both affective (e.g., happiness, anxiety) and psychotic symptoms (e.g., paranoia, auditory hallucinations) in cannabis users, even among those without a history of psychotic disorders. This suggests that these symptoms may act as early indicators of vulnerability to psychosis. The combination of these symptoms, particularly during cannabis use, could help identify individuals at risk before the onset of more severe psychiatric conditions.

These symptoms might be early warning signs of vulnerability, because they are more pronounced in people, who do develop schizophrenia or cannabis-induced psychosis. Earlier cannabis related psychological studies and meta-analytic evidence suggest at-risk populations report higher prevalence of psychosis including paranoia, hallucinations, and perception disturbances.[3] In addition, the connection between the frequency and amount of cannabis consumption and the severity of symptoms, which corroborates earlier studies, forms the basis of a dose-dependent relationship.[4]

The persistence of symptoms such as paranoia and depersonalization after the effects of cannabis had worn off was particularly notable in individuals with schizophrenia. These post-use symptoms may represent a critical risk factor for the development of schizophrenia, as they were more pronounced in those who eventually developed the disorder. This aligns with previous research suggesting that cannabis-induced psychotic symptoms can precede the onset of schizophrenia in vulnerable individuals.[9]

Given the notable symptom overlap between schizophrenia and cannabis-induced psychotic disorders, it is possible that cannabis use could both trigger psychotic episodes and exacerbate schizophrenia in people, who are already at risk for the disorder. About 40% of individuals with cannabis-induced psychotic disorder go on to develop schizophrenia, according to Starzer.[9] This suggests that assessments like the CEQ-6 could be helpful in identifying individuals who are at risk before their condition develops.

The adjusted results reinforce the interpretation that cannabis use patterns—especially higher daily quantity and earlier age of onset—are the primary factors driving symptom expression among users. While initial group comparisons suggested differences in CEQ-6 symptoms between individuals with and without psychotic disorders, these associations were largely reduced after controlling for relevant cannabis-related variables.

This indicates that the symptom profiles observed in this study are not merely consequences of diagnostic status, but are instead closely tied to the characteristics of cannabis use itself. A small number of group-level effects persisted after adjustment, which may point to a potential interaction between cannabis use and underlying vulnerability in certain individuals. However, the dominant influence of cannabis exposure underscores its central role in shaping psychosis-like experiences, particularly those captured by the CEQ-6.

These findings support a risk model in that cannabis use—rather than psychiatric diagnosis—accounts for much of the variation in symptomatology and may contribute to the development of clinically significant outcomes in vulnerable individuals. Future longitudinal studies are warranted to determine whether specific symptom profiles and use patterns can predict the transition to persistent psychiatric disorders.

Limitations and future directions

While this study reveals promising links between CEQ-6-assessed cannabis-related symptoms and psychosis risk, several limitations must be noted. Its cross-sectional design restricts causal conclusions and prevents assessment of CEQ-6’s predictive validity—longitudinal studies are needed to confirm its utility in forecasting chronic psychotic disorders.

Recruiting solely from a psychiatric hospital may have biased the sample toward more severe cases, possibly inflating associations. Although toxicology screens were used, self-reporting may still be affected by stigma or underreporting.

The patients in Groups 2 and 3 answered the questionnaire after the remission of psychotic symptoms. There is the possibility that the illness course, and the treatment effects could have influenced the symptom reporting.

Psychosis is multifactorial, and other influences—such as genetic risk, socioeconomic stress, or comorbid conditions—may also shape symptom patterns. Future studies should include broader clinical evaluations, genetic data, and psychosocial profiling.

Given the exploratory nature of the research, generalizability to non-clinical settings is uncertain. Larger, more diverse samples across settings are needed to validate CEQ-6 as a reliable screening tool in both clinical and public health contexts.

Clinical and public health implications

Applying these findings in real-world settings presents several challenges. Although CEQ-6 is effective in research, its reliance on self-reporting limits its accuracy in general populations due to recall bias, underreporting, and poor symptom insight—particularly in early psychosis. Stigma around cannabis and mental health may also prevent honest responses or help-seeking. Additionally, cultural views of cannabis as safe or therapeutic could hinder acceptance of screening tools like CEQ-6. Successful implementation would require public education, clinician training, and cultural sensitivity. Without longitudinal validation, its predictive use remains preliminary and should be applied cautiously outside research settings.

Nonetheless, integrating CEQ-6 into primary care, addiction services, or digital platforms could enable early detection and intervention for those at risk. Equipping healthcare providers with screening tools supports timely care and may prevent worsening mental health outcomes.

Given cannabis’s cultural significance, public messaging about its psychiatric risks may meet resistance. Involving people with lived experience (PWLE) of cannabis-induced psychosis in awareness efforts can offer relatable, impactful education. Their voices in campaigns and peer-led programs may reduce stigma, encourage help-seeking, and inform responsible cannabis use—especially among youth and frequent users.

CONCLUSIONS

This study underscores the need for proactive measures to address cannabis-related mental health risks, especially as legalization expands. The CEQ-6 questionnaire shows promise as a practical tool for early identification of cannabis-induced psychosis, even outside specialized care. Integrating it into public health programs could improve early detection and support for at-risk individuals.

Raising awareness among the public and healthcare professionals is also essential. Campaigns can reduce stigma and promote help-seeking, while clinician training can improve recognition of cannabis-induced symptoms. Early screening combined with education may lessen the mental health burden linked to cannabis use.

The study also emphasizes the need for deeper insight into cannabis effects. While it may have therapeutic uses, its psychoactive risks—particularly for vulnerable individuals—must be better understood. Identifying biomarkers or genetic risk factors could support more personalized prevention.

Given its cross-sectional design, the study’s findings remain exploratory. Longitudinal research is needed to confirm whether CEQ-6 can reliably predict progression to chronic psychotic disorders and serve as a validated public health screening tool.

Generative AI declaration

This manuscript was reviewed for grammar, clarity, and readability using ChatGPT (OpenAI, version GPT-4o). All intellectual and analytical contributions remain those of the authors.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

This work was supported by the Iuliu Haţieganu University of Medicine and Pharmacy Cluj-Napoca through a Doctoral Research Project (No. 1032110/13.01.2021).