The objective of this study was to determine the efficacy and safety of treatment CIP in patients who were diagnosed with CIP and had positive urine screening for THC at admission. Therefore, we can assure that our patients had psychosis related to the use of cannabis at admission. THC, a psychoactive substance, was commonly found in high amounts in the flowers of cannabis. Cannabis sativa which is the common species grown in Thailand [11]. THC elicits its acute psychoactive effect through the agonistic effect at the CB₁ receptors [12]. However, there were some patients used other substances in combination with cannabis. Some of them (6.9%) had a positive urine test for methamphetamine. A combination of methamphetamine and cannabis may have a synergistic effect on psychosis by increasing dopamine level in the nucleus accumbens shell region [13]. Notably, the mean BPRS score for patients testing positive for both substances (58.6 ± 9.8) had a trend to have a higher mean BPRS score than those in THC-only positive patients (54.8 ± 9.6).

The legal status and perception of cannabis in Thailand have changed due to the withdrawal of dried leaves and flowers of cannabis plants, cannabis extracts and byproducts from extracting process with THC content no greater than 0.2% from the narcotic drugs list in 2020 [14]. The rate of CIP in our study, which was in the period between October 2013 and September 2019 (before delisting cannabis from narcotic drugs), was approximately 5% of total admission at PMNIDAT each year. However, data from the PMNIDAT website during 2019–2023 demonstrated that the prevalence of cannabis-related admission was rising drastically after cannabis legalization, especially in 2022 and 2023 (4.62%, 4.17%, 6.57%, 10.7%, and 16.31% in 2019 to 2023, respectively) [15].

Moreover, the demographic data of our study demonstrated that the average age of patients with CIP was in the range of 19–24 years old, and some of our patients were students (67/317; 21.1%). A comparison with two retrospective studies revealed a similar trend with the majority of patients were male and the average age were in the range of 21.0 years and 28.10 years, respectively [16, 17]. Notably, we found that patient below the age of eighteen accounted for approximately 10% (32/317) of those experienced psychosis from cannabis. The mean age of first cannabis use was 16.3 ± 3.4 years old and the lowest age in our study was 9 years old. Teenagers are more likely to try cannabis due to curiosity and a lack of knowledge regarding its long-term risks, such as cognitive impairment, and mood disorder. The studies on teenagers using cannabis indicated that THC has a more pronounced impact on cognitive decline in adolescents compared to adults [18, 19]. In addition, cannabis use in early adolescence carries a heightened risk of developing psychotic illnesses [20, 21]. THC-induced cognitive decline and poor judgement, which are usually not reversible, are related to effect of THC on CB1 receptors which are widely distributed throughout the central nervous system, especially enriched in the prefrontal cortex and the hippocampus. Thus, any disruption of the prefrontal-hippocampal pathway will negatively impact cognitive functions such as working memory, decision-making and inhibitory control [22]. These consequences highlighted the cautions of making cannabis easily accessible to general population, especially those who are under 18 years old.

For the treatment of CIP, nearly all of the patients received antipsychotics to target symptoms of CIP, especially hallucinations, delusions, irritability, and aggression. Antipsychotics play a crucial role by blocking D₂ receptors in the mesolimbic tract, thereby mitigating dopamine-mediated signaling, which is the underlying cause of these symptoms. Risperidone and haloperidol were the most commonly used antipsychotics during acute and maintenance phase treatment. The median doses of risperidone and haloperidol used in this study were 2–4 mg/day and 5–20 mg/day, respectively. These dose range can effectively block D₂ receptor. Thus, we could observe that the BPRS on day 8 following the administration of these antipsychotics showed a significant decrease of approximately 28% from the first day and subsequently decreased afterwards. However, the average risperidone equivalent dose of all antipsychotics during the maintenance phase in this study was 8.0 ± 5.9 mg/day, which was slightly higher compared to doses used in previous studies (6 mg/day) [8] and those used for treatment of first-episode psychosis (risperidone 3.3 mg/day) [23].

Risperidone was the most commonly prescribed medication both during the acute phase (83.6%) and the maintenance phase (85.5%). This was in accord with reviews and studies recommending the preference for atypical antipsychotics over typical antipsychotics in CIP for several reasons. Firstly, atypical antipsychotics demonstrate comparable efficacy to typical antipsychotics but with fewer adverse effects. Two randomized controlled trials found no significant difference in efficacy between typical and atypical antipsychotics, but atypical antipsychotics had a lower rate of extrapyramidal side effects [7, 8]. Secondly, atypical antipsychotics had beneficial effect on long-term cannabis use induced amotivational syndrome [24]. From our study, it was shown that risperidone could significantly improve all symptoms in negative domains of BPRS at day 22 when compared to baseline.

In our study, LAI antipsychotics were used more frequently in the maintenance phase (14.4%) compared to the acute phase (5.4%). Haloperidol decanoate was the most commonly used. However, LAI atypical antipsychotics such as paliperidone palmitate, and aripiprazole monohydrate, should be considered for treatment of CIP during maintenance phase. These drugs might control both positive and negative symptoms of CIP better than LAI conventional antipsychotics. Referring to 6.4% of our patients who were readmitted for CIP, LAI antipsychotics should be chosen which may help to improve adherence and prevent relapse in these CIP patients [25].

In addition to psychosis, high doses of cannabis usage led to the sudden onset of mood symptoms such as depression, fear, irrational panic, anxiety, and insomnia [26, 27]. Therefore, antidepressants and benzodiazepines were prescribed for mood symptoms in these patients. However, the anxiety and tension scores were still persisted at day 22. These may be related to cannabis withdrawal or the reduction of benzodiazepines dose in the maintenance phase.

Following the patients with CIP, we found that approximately 7.0% of patients (21 of 317) developed schizophrenia. A systematic review and meta-analysis by Murrie et al. in 2020 found that cannabis had the highest rate of developing schizophrenia compared to other substances (6 studies and 3040 people, 34%, 95% confidence interval 25–46%) [28]. It is confirmed that CIP could be an important risk factor for developing schizophrenia. In our study, this prevalence of cannabis-related primary psychosis (cannabis-induced schizophrenia) appeared to be lower than previous reports [28]. It might be due to our study design which was retrospective cross-sectional study conducted at a single institution. All data were collected solely from medical charts over a six-year period. Consequently, some cases of cannabis-related primary psychosis may have been missed due to lack of follow-up at our institution. However, the prevalence of schizophrenia resulting from cannabis use in Thailand may be rising, especially after the cannabis was removed from the narcotic drugs list, making it more easily accessible. The underlying mechanism of cannabis-induced schizophrenia may be explained by the impact of THC on CB1 receptors which are widely distributed in the sensory cortical circuits. Stimulation of CB1 receptor promotes neurogenesis and oligodendrogenesis in the central nervous system [29]. However, chronic THC use leads to downregulation of the CB1 receptors, potentially affecting these trophic and repair processes, ultimately contributing to irreversible brain damage associated with schizophrenia [30]. Furthermore, THC may induce a disproportionate downregulation of N-methyl-D-aspartate (NMDA) receptor, leading to hypofunction and implicating in schizophrenia [31]. The binding of THC at CB1 receptors also affect the myelination and synaptic pruning process of prefrontal cortex which is one of the last brain regions to develop during late adolescence. This area is responsible for many executive functions, including problem solving, working memory, abstract thinking, and increased inhibitory control [32]. Additionally, once patients with CIP who are in their teens develop schizophrenia, they can experience permanent cognitive impairment. There will be a tremendous impact in terms of manpower and economic burden. Thus, there is a pressing need to reevaluate the legalization of cannabis, given the observed of CIP or cannabis used-induced schizophrenia in our study. This phenomenon could lead to the potential loss of manpower, an increased burden on the healthcare system, and pose a hazard to families and others due to psychotic symptoms.

Our study had several limitations. First, the data were sourced from medical charts, potentially leading to some missing information. Secondly, changes in the BPRS scores may have been affected by the use of other psychotropic drugs. We were unable to control medication during the study because it was a retrospective observational study in real word data. Lastly, the lack of cognitive function recorded in medical charts may not adequately reflect the cognitive decline in individuals with CUD, which was found to be as high as 80%. In addition, the data from our study may not be extrapolated to other settings which have different context. Further prospective studies investigating the consequences of cannabis use and prevalence of cognitive impairment, and poor judgment, including impact of pharmacological treatment on these symptoms, should be conducted.