Cannabis Dependence and Psychiatric Comorbidities: Associated Factors and Severity among Patients Followed at the Neuro-Psychopathological Center of Kinshasa

Introduction

Psychiatric disorders result from any alteration in mental health and may manifest at the emotional, cognitive, and/or behavioral levels. They constitute a significant burden on healthcare systems worldwide, particularly in low-resource countries where access to specialized care remains limited [1,2]. Furthermore, the use of psychoactive substances, especially cannabis, is frequently observed among individuals with psychiatric disorders [3,4].

Cannabis is currently one of the most widely used illicit drugs in the world, and its consumption continues to increase in several regions, particularly in sub-Saharan Africa [5,6]. However, behind this widespread use, numerous studies show that it is not without consequences, especially for mental health. Cannabis use may contribute to the onset or worsening of certain psychiatric disorders, including psychotic disorders, mood disorders, and anxiety disorders [7,8].

Among individuals with pre-existing psychological vulnerability, cannabis use often complicates management: treatment adherence decreases, relapses occur more frequently, and hospitalizations increase [9,10].

Several studies indicate that individuals who have used cannabis have approximately twice the risk of developing psychotic disorders compared to non-users. This risk appears to be higher when use begins before the age of 15 or among individuals with a family history of psychiatric disorders [11-14]. A study published in 2012 also estimated that cannabis could contribute to 8 to 15% of new cases of schizophrenia [15].

Furthermore, research by Maggu et al. reports a statistically significant association between cannabis use and bipolar disorder, with an odds ratio of 2.6 [16]. Similarly, Danielson et al. observed links between cannabis use and certain anxiety disorders (OR between 1.3 and 1.4), as well as with depression (OR of 1.2) [17].

However, most available data are generated in Western countries, which limits the generalizability of these findings to African contexts, where sociocultural, economic, and healthcare realities differ significantly [18,19]. In the Democratic Republic of Congo, research on cannabis use among patients with psychiatric disorders remains limited, despite a notable prevalence of psychoactive substance use among patients receiving psychiatric care [20,21].

General Objective

To establish the relationship between cannabis addiction and psychiatric comorbidities among patients with psychiatric disorders.

Specific Objectives

• To describe the sociodemographic and clinical characteristics of patients followed at the CNPP.

• To identify factors associated with cannabis dependence.

• To assess the severity of cannabis use and dependence.

• To study the prevalence of psychiatric comorbidities and their association with the severity of cannabis dependence.

Materials and Methods

2.1. Type and Study Design

This was an observational, cross-sectional, descriptive, and analytical study conducted among patients followed in psychiatric settings for disorders related or associated with cannabis use.

The descriptive component allowed us to characterize the sociodemographic, clinical, diagnostic, therapeutic, and outcome profiles of cannabis-using patients.

The analytical component enabled the identification of factors associated with cannabis addiction according to DSM-5 diagnostic criteria, as well as the determinants of consumption severity assessed using the CAST score.

The cross-sectional nature of the study implies that data were collected at a single point in time during patients’ clinical follow-up.

2.2. Study Setting

The study was conducted at the Neuro-Psychopathological Center of Kinshasa (CNPP/Kinshasa), a specialized facility for the management of psychiatric and neurological disorders.

The CNPP receives patients with various mental disorders, including psychotic disorders, mood disorders, anxiety disorders, substance use disorders, as well as associated psychiatric and somatic comorbidities.

This setting was chosen due to its role as a reference center specializing in the evaluation, diagnosis, and follow-up of patients with psychiatric and addictive disorders, and due to the availability of clinical records containing the necessary information for studying the profile of cannabis users and associated factors.

2.3. Study Period

The study covered patient records from the Neuro-Psychopathological Center over the period from January 2019 to December 2024.

2.4. Study Population

The target population consisted of all cannabis-using patients followed in psychiatric hospital settings, represented by patients treated at the CNPP whose clinical records were available, complete, and usable for the variables selected in the study.

2.5. Sampling

A non-probabilistic convenience sampling method was used. It consisted of including all available, complete, and usable records that met the inclusion criteria.

This method was chosen due to the clinical, institutional, and retrospective nature of the study, based on the use of available medical record data.

2.6. Sample Size

The minimum sample size was estimated using the Schwartz formula, commonly applied in cross-sectional studies aimed at estimating a proportion:

n = Z² × p × q / d²

Where:

• n = required minimum sample size

• Z = standard normal deviate for a 95% confidence level (1.96)

• p = expected proportion of the phenomenon under study

• q

• d = margin of error (5%, i.e., 0.05)

In the absence of sufficiently documented local prevalence of cannabis addiction among CNPP patients, a theoretical proportion of 50% was used. This value maximizes the sample size and ensures a conservative and robust estimate.

The calculation was as follows:

n = (1.96)² × 0.5 × 0.5 / (0.05)² = 384

Thus, the minimum required sample size was 384 patients. To improve estimate precision and strengthen the power of bivariate and multivariate analyses, the final sample size was increased to 500 patients.

2.7. Selection Criteria

2.7.1. Inclusion Criteria

Patients were included if they met the following criteria:

• Followed at the CNPP during the study period;

• Aged at least 18 years;

• Had documented cannabis use in the clinical record;

• Had usable sociodemographic, clinical, diagnostic, and therapeutic data;

• Had an available assessment of cannabis addiction based on DSM-5 criteria and/or consumption severity assessed using the CAST score.

2.7.2. Non-inclusion Criteria

The following were excluded:

• Records incomplete for key variables of interest;

• Records lacking any usable information on cannabis use;

• Records with major inconsistencies after verification;

• Duplicate records;

• Patients whose age, primary psychiatric diagnosis, or essential clinical data were not documented.

2.8. Study Variables

2.8.1. Dependent Variables

The main dependent variable was cannabis addiction according to DSM-5 criteria, coded as present or absent.

The DSM-5 (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition) is an international reference for diagnosing mental disorders. In this study, it was used to identify patients with cannabis use disorder. This disorder is based on behavioral, cognitive, and physiological criteria reflecting problematic use, including: loss of control, persistent desire or unsuccessful attempts to reduce use, significant time spent obtaining or using the substance, craving, continued use despite negative consequences, tolerance, withdrawal, and social or functional impairment.

The secondary dependent variable was the severity of cannabis use, assessed using the CAST score (Cannabis Abuse Screening Test).

CAST evaluates multiple dimensions of cannabis use, including frequent use, solitary use, memory impairment, remarks from others, attempts to reduce or stop, and difficulties related to consumption. It provides a global score reflecting the level of risk or severity. In this study, it was used as a quantitative indicator and to classify patients into three levels: mild, moderate, and severe.

The combined use of DSM-5 criteria and CAST score allowed assessment of both the presence of addiction and its severity.

2.8.3. Independent Variables

Independent variables were grouped into four categories:

• Sociodemographic/ variables: age, sex, marital status, education level, occupation.

• Cannabis use variables: age at initiation, duration of use, frequency of use, type of cannabis use.

• Clinical and diagnostic variables: primary psychiatric diagnosis, schizophrenia, depression, bipolar disorder, generalized anxiety disorder, acute psychosis or schizophrenia, psychiatric comorbidity, somatic comorbidity.

• Therapeutic, outcome, and history variables: treatment adherence, previous hospitalization, suicide attempt, family psychiatric history, current treatment, clinical outcome.

2.9. Data Collection

Data were collected using a digital form via KoboToolbox using the ODK Collect application.

Information was extracted from clinical records, then coded, entered, verified, and prepared for statistical analysis.

Data quality management was a crucial step. After collection, data were checked for entry errors, missing values, duplicates, coding inconsistencies, and outliers.

Qualitative variables were harmonized to standardize response categories. Some variables were recoded into binary form to facilitate analysis, including DSM-5 addiction, psychiatric comorbidity, somatic comorbidity, previous hospitalization, suicide attempt, and family psychiatric history.

Quantitative variables (age, age at initiation, duration of use, CAST score) were checked for extreme or inconsistent values. Particular attention was given to consistency across cannabis-related variables (frequency, type of use, DSM-5 diagnosis, CAST score).

This process resulted in a clean, coherent, and analyzable dataset for descriptive, bivariate, and multivariate analyses.

2.10. Statistical Analysis

Data entry, cleaning, coding, and initial preparation were performed using Excel 2016, and statistical analyses were conducted using SPSS version 27.

2.10.1. Descriptive Analysis

Qualitative variables were summarized as frequencies and percentages. Quantitative variables were described using mean and standard deviation. When relevant, median and range were also reported.

2.10.2. Bivariate Analysis

Bivariate analysis explored associations between cannabis addiction (DSM-5) and explanatory variables.

Comparisons between qualitative variables were performed using the Pearson chi-square test. When conditions were not met, Fisher’s exact test was used. The significance level was set at 5% (p < 0.05).

The strength of associations was assessed using Phi coefficient and Cramer’s V.

2.10.3. Multivariate Analysis (Linear and Binary Logistic Regression)

Linear regression was used to identify factors associated with cannabis use severity (CAST score as the dependent variable).

Results were expressed as beta coefficients with 95% confidence intervals and p-values. A positive beta indicated an increase in CAST score, while a negative beta indicated a decrease.

Binary logistic regression was used to identify factors independently associated with cannabis addiction (DSM-5).

Results were expressed as adjusted odds ratios (AORs) with 95% confidence intervals and p-values. An AOR > 1 indicated increased likelihood of addiction, while AOR < 1 indicated reduced likelihood. Associations were considered statistically significant at p < 0.05.

2.11. Ethical Considerations

The study was conducted in accordance with ethical principles applicable to biomedical research.

Approval was obtained from the National Ethics Committee (reference number …), as well as administrative authorization from CNPP before accessing clinical records.

Confidentiality was strictly maintained. No identifying personal data were collected or reported.

All records were anonymized using codes, and access to data was restricted to study personnel.

2.12. Operational Definitions

• Cannabis addiction: presence of cannabis use disorder according to DSM-5 criteria.

• Severity of cannabis use: measured by CAST score.

• Psychiatric comorbidity: presence of an additional mental disorder.

• Somatic comorbidity: presence of a non-psychiatric medical condition.

• Treatment adherence: degree of compliance with prescribed treatment.

• Clinical outcome: categorized as improvement, stability, or deterioration.

• Recreational use: cannabis use for pleasure, relaxation, or social reasons without necessarily meeting addiction criteria.

• Abuse: problematic use causing negative consequences.

• Dependence: use characterized by loss of control, continued use despite harm, tolerance, withdrawal, or significant impairment.

Results

3.1. Sociodemographic Characteristics

The most represented age group was 36–45 years, with 179 patients (35.8%), with a mean age of 31 ± 7.6 years. This was followed by the 26–35 age group, which included 161 patients (32.2%). Regarding marital status, 334 patients (66.8%) were not living with a partner.

In terms of education level, primary education was the most frequent, with 194 patients (38.8%). Finally, regarding employment status, 367 patients (73.4%) were employed, compared to 133 (26.6%) who were unemployed.

3.2. Characteristics Related to Cannabis Use

The most represented age group for initiation of cannabis use was 16–20 years, with 216 patients (43.2%), and a mean age of 19.2 ± 3.4 years.

Regarding duration of use, the most frequent category included patients who had consumed cannabis for 15 years or more, accounting for 197 cases (39.4%), with an average duration of 12.5 ± 7.7 years.

The frequency of use was relatively balanced between occasional users (179 patients; 35.8%), weekly users (162; 32.4%), and daily users (159; 31.8%). Finally, recreational use of cannabis was reported in 176 patients (35.2%) of the sample.

3.3. Clinical, Diagnostic, and Therapeutic Characteristics

Analysis of clinical and diagnostic characteristics showed that 262 patients (52.4%) met the DSM-5 criteria for cannabis addiction, while 238 (47.6%) did not.

The most frequent primary psychiatric diagnosis was schizophrenia, found in 185 patients (37.0%). This was followed by depression with 102 cases (20.4%), bipolar disorder with 84 cases (16.8%), and generalized anxiety disorder with 39 cases (7.8%).

Psychiatric comorbidity was identified in 233 patients (46.6%). Treatment adherence was considered satisfactory in 252 patients (50.4%) and poor in 248 patients (49.6%).

Additionally, 259 patients (51.8%) had a history of suicide attempts, 242 (48.4%) had a family psychiatric history, and 240 (48.0%) had a somatic comorbidity.

Regarding ongoing treatments, 136 patients (27.2%) were receiving antipsychotics, 124 (24.8%) mood stabilizers, 120 (24.0%) anxiolytics, and 120 (24.0%) psychotherapy.

Clinical evolution was considered stable in 185 patients (37.0%).

Finally, analysis of the CAST score showed a predominance of the moderate level, observed in 282 patients (56.4%), with a mean score of 6.9 ± 2.4, a median of 7 [5-9], and values ranging from 2 to 12.

3.4. Clinical History and Comorbidities

Clinical history and comorbidities in this population reflect a high psychopathological burden. More than half reported at least one suicide attempt, nearly one in two had a family psychiatric history, and almost one in two also had a somatic comorbidity.

3.5. Bivariate Analysis Between Cannabis Addiction and Psychiatric Comorbidities

Bivariate analysis showed a statistically significant association between cannabis addiction and several psychiatric variables, including schizophrenia, bipolar disorder, generalized anxiety disorder, depression, acute psychosis, and unspecified category (p < 0.001). However, no significant association was observed with suicide attempts (p = 0.849). Nevertheless, the Phi and Cramér’s V coefficients, all below 0.10, indicate that these associations are weak in strength.

3.6. Multivariate Analysis by Linear Regression

Multivariate linear regression analysis showed a significant association between the severity of cannabis addiction (measured by the CAST score) and several factors, including frequency of use, DSM-5 diagnosis of addiction, and education level.

Compared to weekly use:

• Occasional use was associated with a significant decrease in the CAST score (β = -0.967; 95% CI: -1.196 to -0.739; p < 0.001)

• Daily use was associated with a significant increase (β = 0.944; 95% CI: 0.706 to 1.182; p < 0.001)

Additionally:

• DSM-5 cannabis addiction was associated with a significant increase in CAST score (β = 2.028; 95% CI: 1.842 to 2.214; p < 0.001)

• Higher education level (vs primary) was associated with a significant increase (β = 0.332; 95% CI: 0.098 to 0.566; p = 0.006)

3.7. Multivariate Analysis by Binary Logistic Regression: Factors Associated with Cannabis Addiction (DSM-5)

Binary logistic regression identified independent factors associated with cannabis addiction according to DSM-5 criteria.

After adjustment:

• Secondary education level was associated with a higher

likelihood of addiction compared to primary education (adjusted OR = 1.73; 95% CI: 1.11–2.70; p = 0.015)

• Recreational use was associated with a lower likelihood of addiction (adjusted OR = 0.51; 95% CI: 0.32–0.81; p = 0.004)

Other variables were not significantly associated with cannabis addiction.

Discussion

The present study aimed to describe the sociodemographic and clinical characteristics of patients followed at the CNPP, to identify factors associated with dependence, to assess the severity and dependence on cannabis, and to examine the prevalence of psychiatric comorbidities and their association with cannabis dependence severity. The results confirm the high prevalence of associated psychiatric disorders among individuals with cannabis use disorder, in line with findings from the international literature.

1. Sociodemographic factors

The mean age of participants was 31 years with a standard deviation of 7.6 years. This finding is close to that of a study conducted in Cameroon in a hospital psychiatric setting, which reported a mean age of 23.79 years. That study aimed to assess the frequency of cannabis use among patients consulting for psychiatric disorders and the occurrence of psychiatric symptoms, in order to explore a possible causal relationship or comorbidity [22].

Similarly, a study conducted in South Africa among patients recruited from a university psychiatric service and presenting with a first psychotic episode reported a mean age of 28 years with a standard deviation of 11 years [23]. This research aimed to analyze the association between cannabis use and the occurrence of a first psychotic episode in hospitalized psychiatric patients.

Taken together, these studies suggest that young individuals constitute the main population of cannabis users, both in Africa and in other regions of the world.

Regarding educational level, 38.80% of patients had a primary education level and 31.40% had a secondary education level. Studies conducted in South Africa by Mona et al. and Burns et al., involving psychiatric patients in the KwaZulu-Natal region, also reported a predominance of primary and secondary education levels [24,25].

Likewise, a study conducted in India examined the association between low educational level and problematic cannabis use. It reported an overall low-to-intermediate educational level among patients with cannabis-associated psychosis [26,27].

These consistent findings support the hypothesis that cannabis use is more frequent among individuals with secondary education or lower.

2. Cannabis and psychiatric comorbidities

The prevalence of addictive comorbidities observed in our study was 51.21% in schizophrenic patients, 48.78% in bipolar patients, 53.27% in those with generalized anxiety disorder, 45.09% in depressive patients, and 66.66% in patients with acute psychosis. These results highlight a high frequency of anxiety, depressive, and psychotic disorders among patients with cannabis dependence.

These observations are consistent with those reported by Nora D. Volkow in a study conducted in the United States in a general population, which highlights a frequent association between cannabis use and various mental disorders, particularly psychotic, anxiety, and depressive disorders [28]. This study found a significant association between cannabis use and psychosis, with an odds ratio (OR) of 1.4. The association with depression appeared more modest but variable, with an OR ranging from 1.2 to 1.6.

Furthermore, a meta-analysis including 14 longitudinal studies conducted in five countries (United States, Netherlands, Canada, Australia, and New Zealand) assessed the association between cannabis use and subsequent depression. The results showed a statistically significant but modest association, with an OR of 1.17 [29].

Similarly, Robin M. Murray demonstrated that regular cannabis use, particularly when it has a high THC content, increases the risk of psychosis, especially in vulnerable individuals. The risk is estimated to be approximately 2 to 4 times higher in regular users compared to non-users [30]. The study by Martínez-Aguayo also confirms that cannabis can not only precipitate the onset of a psychotic episode but also worsen the course of pre-existing psychotic disorders [31].

Finally, the similarities observed across studies may be explained by a convergence of methodological, biological, and epidemiological factors.

3. Factors associated with addiction severity

The results of this study show that occasional cannabis use is significantly associated with a decrease in the dependent variable compared to weekly use (β = -0.967; 95% CI [-1.196; -0.739]; p < 0.001), whereas daily use is associated with a significant increase in this same variable (β = 0.944; 95% CI [0.706; 1.182]; p < 0.001). These findings suggest a dose–response relationship between frequency of use and the evolution of the studied variable.

These observations are consistent with the literature. Indeed, the study by Chen et al., conducted among a non-institutionalized population in the United States, shows that low-frequency use (occasional or moderately weekly use) is associated with more stable and less pathological trajectories, whereas increasing frequency is a key factor in the transition to problematic use and dependence [32-35]. Similarly, the meta-analysis by Hall et al. reports that low levels of use, such as occasional or infrequent consumption, are associated with a significantly reduced risk of developing cannabis use disorders compared with weekly or daily use [36].

Overall, these findings are consistent with the existing literature, which emphasizes that the effects of cannabis depend not only on use per se, but mainly on its frequency and intensity [37-39].

The effects of cannabis on the body and brain are therefore dose-dependent and cumulative [40].

Conclusion

The findings highlight a high prevalence of psychiatric comorbidities, particularly psychotic, anxiety, and depressive disorders, among individuals with cannabis use disorder.

Furthermore, the analysis of factors associated with addiction severity reveals a dose–response relationship between frequency of consumption and severity of dependence, suggesting that daily use is a worsening factor, whereas occasional use appears less severe.

Overall, these results underscore the need for special attention to cannabis use among young and vulnerable individuals, as well as the importance of an integrated care approach combining the treatment of addiction and psychiatric comorbidities, in order to improve patients’ overall prognosis.

Abbreviations

DSM-5: Diagnostic and Statistical Manual of Mental Disorders – 5e édition

CAST: Cannabis Abuse Screening Test

CNPP: Centre Neuro-Psycho-Pathologique de Kinshasa

DRC: Democratic Republic of Congo.

Declarations

Ethical approval and consent to participate

This study was approved by the Ethics Committee of the School of Public Health of the University of Kinshasa, in accordance with the principles of the Declaration of Helsinki.

All participants received clear information regarding the study objectives, data collection procedures, the confidentiality of the collected information, and their right to withdraw at any time without consequences. Free and informed consent was obtained from all participants prior to their inclusion in the study.

Consent for publication

Consent for processing and open-access publication was obtained from all study participants or waived where applicable.

All authors confirm that this study did not involve any animal subjects or any animal tissue.

Competing interests

In accordance with the ICMJE uniform disclosure form, all authors declare that they received no financial support from any organization. They report no conflicts of interest.

Availability of supporting data

The data used in this study include questionnaires and interviews conducted with participants. They are stored by the author and may be made available upon request, in compliance with confidentiality and anonymity requirements.

Funding

No funding was received for this article.

Author contributions

Odon Nzuzi Mabiala: Literature review, database management, statistical analysis, writing.

Fiston Mbata: Statistical analysis.

Hergy Bazungula Mumpasi: Data collection, data entry, and statistical analysis.

Christian Kasongo Mwenze: Data collection, data entry, and statistical analysis.

Justine Panzu Mavinga: Data collection.

Alfred Sodi Magudigana: Data collection.