Coprescription of antipsychotics and benzodiazepines at hospital discharge.

INTRODUCTION

Antipsychotics and benzodiazepines (BZD) are known to increase the risk of sedation, dizziness, cognitive impairment and even confusion when used in combination because of their psycholeptic properties. Antipsychotics and BZD coprescription should therefore be minor at discharge from a psychiatric hospital, especially for conventional sedative antipsychotics (CS-AP), such as cyamemazine, levomepromazine, or loxapine which are often used as adjunctive treatments like BZD.

OBJECTIVES

The primary objective was to investigate the prevalence of prescription of CS-AP, long-term antipsychotics (LT-AP), and BZD among patients discharged from hospitalization in psychiatry. The secondary objective was to assess factors associated with CS-AP and BZD coprescription.

MATERIAL AND

METHODS

A retrospective study was conducted by collecting all discharge prescriptions from patients discharged from a psychiatric hospital in 2022 and 2023. To assess prevalence, antipsychotic and BZD use was registered when at least one medication from these classes was prescribed at discharge. Multivariate logistic regression analysis was used to assess factors associated with CS-AP and BZD coprescription. Sensitivity analyses were performed to corroborate our results.

RESULTS

In total, 5,733 psychiatric stays were associated with prescribed medications at discharge, corresponding to 4,131 different patients. Mean age was 39.8 (±18.7) years. Women represented 56.3% of the studied population. The prevalence of psychotropic drugs on discharge prescriptions was 38.7% for CS-AP, 69.1% for LT-AP, and 48.3% for BZD. The most prescribed drugs, respectively for CS-AP, LT-AP, and BZD, were cyamemazine (24.3% of prescriptions), aripiprazole (16.2%), and diazepam (24.7%). Prevalence of CS-AP+BZD and CS-AP+LT-AP+BZD coprescriptions was respectively 14.8% and 10.1%. After adjustment for age, sex, mode of admission, length of stay class, and prescription of at least one LT-AP at discharge, the CS-AP+BZD coprescription at discharge was associated with two factors: length of hospital stay≤14 days and an initial admission to a psychiatric emergency department (aOR: 1.35, 95% CI [1.14; 1.59], and aOR: 1.23, 95% CI [1.05; 1.45], respectively). Sensitivity analyses showed a similar trend, although results were not always statistically significant.

DISCUSSION: Our results highlight the importance of an effective hospital-to-community care transition to optimize psycholeptic treatments and minimize the risk of adverse drug reactions in outpatient settings. Cyamemazine was the most prescribed CS-AP at discharge, probably due to its anxiolytic properties. This trend raises concerns regarding its appropriateness when combined with other anxiolytic treatments, such as BZD. The association between CS-AP+BZD coprescription and length of stay<14 days shows that the short hospitalization may hinder proper adaptation of pharmacological treatments. The association between CS-AP+BZD coprescription and initial admission to a psychiatric emergency department may reflect insufficient re-evaluation of the coprescribed drugs probably introduced during acute crisis management. However, our study has several limitations, including reliance on our classification of CS-AP and BZD, and lack of data on key confounders such as diagnoses, seriousness of the acute episode, prior treatments, and post-discharge care.

CONCLUSION

Prevalence of CS-AP with BZD is high at discharge from a psychiatric hospital, especially in patients with short hospital stays (≤14 days).