Cannabis-Related Hospitalizations Before and After Partial Legalization

An Analysis From Five Psychiatric Clinics in the Rhine-Ruhr Metropolitan Area

Since 1 April 2024, the acquisition, possession, and cultivation of non-medicinal cannabis in adults is no longer illegal under certain circumstances (partial legalization). The epidemiological addiction survey has long been a reliable instrument for determining trends in the consumption of psychoactive substances in adults in Germany (1). At the end of 2025, this survey established an initial trend of increased cannabis consumption in the adult population of Germany after partial legalization—albeit not one that reached significance (1). Similar longitudinal surveys in the USA, where non-medicinal cannabis consumption was legalized on a large scale even before 2020, showed a significant increase in cannabis consumption (including highly potent varieties) in the population and cannabis related disorders in the hospital population (2).

We carried out the present study in the Rhine-Ruhr metropolitan area to investigate whether a trend already exists regarding inpatient admissions for cannabis related disorders after partial legalization.

Methods

Admission diagnoses (main and secondary diagnoses) were retrospectively compared with the ICD-10 codes F12.0-F.12.9 (Mental and Behavioral Disorders due to Use of Cannabinoids) in the first quarter of 2024 (Q1 2024, immediately preceding partial legalization on 1 April 2024) and the first quarter of 2025 (Q1 2025). Five psychiatric hospitals participated in the survey (Langenfeld, Remscheid, Bochum, Essen, Castrop-Rauxel—jointly with a healthcare mandate for 1.7 million population [without any change to the healthcare structures and alternative services in the observation period]).

Our controls were alcohol-related (F10.0-F10.9) and opioid related (F11.0-F11.9) admissions. Furthermore we considered schizophrenia (F202.X), acute and transient psychoses (F23.X), schizoaffective (F25.X), and bipolar disorders (F31.X), since cannabis consumption is considered a relevant risk factor for schizophrenia and manic disorders (2, 3). The diagnostic group F30 (manic episode) was barely used (5 cases each in Q1 2024 and Q1 2025) and excluded from modeling. Outside bipolar disorder (F310–2), F30 is used very rarely.

The study aimed to quantify changes in the rates of diagnoses. To estimate changes in absolute frequencies and relative risks (RR) we used Bayesian logistic regression (Bernoulli model), from whose posterior distribution we calculated median estimators and 95% credible intervals (4) (Figure 1, 2).

Figure 1

Estimated absolute change in risk difference (RD) in per cent of the diagnosis rates in first quarter (Q1) 2024 to Q1 2025. Dots represent the median of the posterior distribution, horizontal the 95% credible interval. The dotted line (0%) marks the absence of any change; F12 is emphasized in orange.

Results

Overall we saw in Q1 2024 versus Q1 2025 in the 5 hospitals the following distribution of diagnoses: cannabis related disorders (diagnostic group F12): 763 versus 1047 cases, alcohol-related disorders (F10): 3895 versus 3503 cases, opioid-related disorders (F11): 731 versus 665 cases; and furthermore, 1020 versus 1008 cases from the subject areas schizophrenia (F20), non-schizophrenic psychoses (F23), schizoaffective disorder (F25), bipolar disorder (F31).

Diagnostic groups—For cannabis (F12) the rate increased substantially (in absolute terms by a median 2.07% (95% credible interval: [[0.47; 4.57]) (Figure 1), which corresponds to an increase in RR by a factor of 1.24 (Figure 2). Simultaneously, the prevalence for alcohol (F10) dropped (risk difference: −9.19%). For opioids (F11) too, a descriptively distinguishable drop was seen (−0.59%). No changes were seen for schizophrenia and bipolar disorder (Figure 1, 2).

Figure 2

Relative risk (RR) for the occurrence of the respective diagnosis in Q1 2024 versus Q1 2025. The dots show the median, the horizontal bars the 95% credible interval of the Bayesian estimate. The vertical dotted line at 1.00 marks the absence of a relative difference between Q1 2024 and Q1 2025.

Individual diagnoses—For cannabis (F12) a clear increase dominated from F12.2 (cannabis dependency; absolute increase by 1.33% [0.34; 3.36]). For cannabis related psychoses (F12.5) too we found an absolute increase by 0.41% [0/08; 0.86]. The relative risk increased by a factor of 1.56 [1.10; 2.25], which corresponded to an increase in the probability of this diagnosis of 56%. Within the group of alcohol related disorders (F10), acute intoxications (F10.0; −1.83% [−3.53; −0.55]), and especially the alcohol dependence syndrome (F10.2; −2.52% [−4.33; −0.95%]) posted notable decreases.

Discussion

Similar to the trend in North America (2) we found after cannabis liberalization in our region for psychiatric hospital admissions a notable increase of cannabis related disorders—especially cannabis dependence (F12.2) and cannabis related psychosis (F12.5). This trend was seen for F12.2 in all and for F12.5 in 4 out of 5 participating hospitals (not shown here for reasons of space). An increase in cannabis related psychoses was also recently seen in the psychiatric hospital population in Bavaria. However, simultaneously the number of schizophrenia diagnoses there had fallen (3). In our area, however, the rates of schizophrenic disorders but also of manic disorders remained stable. A rise in cannabis related disorders in the two existing German cohorts did not seem to be associated with an increase in severe psychiatric disorders (such as schizophrenia and bipolar disorder), which would have been problematic from a health political perspective. In North America too, such a trend has thus far not been shown validly (2). Our results and similar ones may be subject to expectation bias (Rosenthal effect), because after the partial legalization we may have been particularly alert as regards F12.2 and F12.5. But we also found in our controls reduced admission rates for opioid related and, even more notably, alcohol related disorders. The question of whether this trend could be explained with self-treatment using cannabis or a shift of addiction in patients with alcohol or opioid consumption has not been investigated. Similarly, it is not unequivocally clear whether cannabis increases alcohol or opiate consumption or reduces it (5). A recent study used a randomized controlled design to study this in patients with substantial alcohol consumption and found that persons who regularly consumed both, alcohol and cannabis, smoking cannabis led to a significantly reduced alcohol consumption compared with placebo over a 3-day observation period (5).

Udo Bonnet, Norbert Scherbaum, Eugen Davids, Bodo Lieb, Benedikt Bernd Claus, Georg Juckel

Conflict of interest statement
UB received a lecture honorarium from Janssen Cilag. He is a member of the AkdÄ (the Drug Commission of the German Medical Association).
NS is the chairman of the board of the DHS (German Centre for Addiction Issues).
BL received a lecture honorarium from Recordati Pharma and a speaker honorarium from the North Rhine State Chamber of Physicians. As an examiner on behalf of the chamber and training peer of the Deutsche Rentenversicherung (the German statutory pension insurance) he received an expense allowance and was reimbursed for travel expenses. He is the chair of the FV-QSAD (Fachverband Qualifizierte Akutbehandlung Drogenabhängiger, the professional association for qualified acute treatment of drug addicts).
The remaining authors declare that no conflict of interest exists.

Manuscript received on 17 December 2025, revised version accepted on 9 March 2026.

Translated from the original German by Birte Twisselmann, PhD.

Cite this as
Bonnet U, Scherbaum N, Davids E, Lieb B, Claus BB, Juckel G: Cannabis-related hospitalizations before and after partial legalization: An analysis from five psychiatric clinics in the Rhine-Ruhr metropolitan area. Dtsch Arztebl Int 2026; 123: 339–40. DOI: 10.3238/arztebl.m2026.0042

References

1. Hoch E, Krowartz EM, Hollweck R, Möckl J, Olderbak S: Cannabis ­consumption before and after partial legalization in Germany: Early trends, consumption patterns, and motives. Dtsch Arztebl Int 2025; 122: 632–7

2. Mekonen Yimer T, Hoch E, Fischer B, Dawson D, Hall W: The adverse public health effects of non-medical cannabis legalisation in Canada and the USA. Lancet Public Health 2025; 10: e148–e59

3. Greiner SK, Jäger M, Schmauß V, et al.: The frequency of psychotic ­disorders since the legalization of cannabis in Germany: A before-and-­after analysis. Dtsch Arztebl Int 2025; 122: 646–7

4. Kruschke JK: Doing Bayesian Data Analysis. Elsevier Academic Press 2015.

5. Metrik J, Aston ER, Gunn RL, Swift R, MacKillop J, Kahler CW: Acute ­effects of cannabis on alcohol craving and consumption: A randomized controlled crossover trial. Am J Psychiatry 2026; 183: 134–43