During the observation period, a total of 2203 patients were admitted, out of which 885 individuals (40.2%) had a diagnosis of SSD and were included in the final analysis. The mean age of the patients was 40.51 years (SD = 12.44). Among the included patients, the majority were male (n = 557, 62.9%), unmarried (n = 318, 35.9%), and held a high school diploma as their highest vocational degree (n = 570, 64.4%). Notably, patients with a SUD, using single or involving multiple substances, were younger compared to those without a SUD (F(2, 882) = 7.233, p = 0.016).

The average time lapse since the first hospitalization (as a surrogate for duration of illness) was 8.71 years (SD = 8.05). Patients had an average of 10.63 (SD = 15.25) previous hospitalizations, with a right-tailed distribution. At the time of admission, the patients had an average HoNOS score of 20.85 (7.77). The length of hospital stay had an average of 24.62 days (SD = 27.13) and exhibited a right-tailed (skewness: 2.50; kurtosis: 10.88) distribution.

For further details, please refer to Table 1.

For more comprehensive information, please refer to Table 2 and Fig. 1. Among the patients included in the study, approximately two-thirds (n = 586, 66.2%) received a diagnosis of schizophrenia. Schizoaffective disorder accounted for nearly one-fifth of the cases (n = 156, 17.6%), followed by brief psychotic episodes (n = 117, 13.2%) and delusional disorder (n = 26, 3.0%).

Length of stay according to the presence of a substance use disorder. Kaplan–Meier plot of length of stay in patients without SUD (green), single SUD (blue), and multiple SUD (red)

In terms of comorbid substance use, approximately one-third of the patients (n = 328, 37.1%) had at least one SUD. Among those with SUD, almost half (n = 157, 47.9%) exhibited a SUD involving two or more substances. The most prevalent substances among patients with SUD were cannabis (n = 199, 60.7%), followed by cocaine (n = 103, 31.4%), alcohol (n = 96, 29.2%), opioids (n = 71, 21.6%), benzodiazepines (n = 61, 18.6%), and amphetamines (n = 41, 12.5%).

Patients with SUD were predominantly male, constituting almost three-quarters of the group, in contrast to those without SUD (χ2(2, 885) = 20.69; p < 0.001). Notably, patients with a SUD, using single or involving multiple substances, were younger compared to those without a SUD (F(2, 882) = 7.233, p = 0.016). Additionally, patients with multiple SUD had lower education levels compared to those with single or no SUD (χ2(4, 885) = 37.55; p < 0.001). No significant differences regarding marital status were observed.

Patients with multiple SUD exhibited a longer duration of illness compared to those with no or single SUD (F(2, 882) = 18.32; p < 0.001). Furthermore, patients without SUD had a lower frequency of hospitalizations compared to those with SUD (either single or multiple), while patients with multiple SUD had a higher frequency of hospitalizations than those with single SUD (F(2, 882) = 69.06; p < 0.001). The duration of illness was positively correlated with the number of previous admissions (0.55, 95%CI: 0.51–0.60; t = 19.79; df = 883; p < 0.001). Notably, there were no significant differences observed regarding the rate of compulsory admission.

There were no differences regarding the HoNOS score at admission (F(2, 882) = 2.62; p = 0.08). However, patients without SUD had a lower HoNOS score at discharge (F(2, 882) = 4.06; p = 0.02); nonetheless, there were no differences regarding the degree of improvement among the three groups (F(2, 882) = 1.64; p = 0.19). There were no differences at any point when excluding the HoNOS item 3 assessing alcohol and substance use (for further details see Table 2). Regarding the length of stay (LOS), patients with multiple SUD had a shorter LOS compared to those with no or single SUD (F(2, 882) = 9.22; p < 0.001). This relationship persisted even after controlling for the age, sex, education, civil status, duration of illness, previous hospitalizations, and antipsychotic treatment. However, for age we found a regression coefficient of − 0.13 (95%CI: − 0.27–0.01); this indicates that for each additional year of age, there is an average decrease of 0.13 days in the length of stay. This negative relationship between age and length of stay was found to be statistically significant (F(1, 883) = 93.68, p < 0.001). For further information, please refer to Table 2.

The majority of patients (n = 567; 64.1%) received treatment with a combination of two antipsychotic medications. Nearly half of the patients (n = 434; 49.0%) received at least one agent that is potentially available on the market in a long-acting injectable (LAI) formulation. Available agents in Switzerland are risperidone, paliperidone, aripiprazole, and some first-generation antipsychotics. Olanzapine is not approved as LAI in Switzerland. Despite this potential that almost half of patients could have been directly prescribed their agents as LAI formulation without switching the agent, the overall prescription rate of LAIs was relatively low (n = 123, 28.3%). In the group of no SUD, the overall prescription rate of LAI was 9.5% (n = 53) compared to 19.9% (n = 34) in patients with single SUD and 22.9% (n = 36) in patients with multiple SUD (X 2(2, 434) = 23.62; p < 0.001).

Specifically, patients with single SUD had an odds ratio (OR) of 2.62 (95%CI: 1.54–4.43), while those with multiple SUD had an OR of 3.06 (95%CI: 1.80–5.20). After correcting for age, sex, length of stay, previous history (duration of illness, number previous hospitalizations), and severity at admission (HoNOS), patients with single SUD (OR: 1.76; 95%CI 1.05–2.92) and multiple SUD (OR: 2.06; 95%CI: 1.18–3.55) had a higher probability to be prescribed a LAI. Consequently, patients with SUD were much more likely to be prescribed a LAI. For further information, please refer to Table 3.