Relationships between medications used in a mental health hospital and types of medication errors: A cross-sectional study over an 8-year period

Background Particularities in psychiatry care can increase the risk of medication errors (MEs). Objective To analyze the MEs that occurred in a psychiatric hospital and to quantify relationships between the use of certain types of medication and the type of MEs. Methods We conducted a retrospective register based cross-sectional study in a French psychiatric hospital (2014–2021). All MEs were analyzed using ALARM method to identify type, stage of occurring and interception (defenses), consequences, drug involved and root causes. The prevalence-odds ratio (POR) was calculated to estimate relationships between five selected medication situations (long-acting injectable antipsychotics (LAIA), oral liquid dosage forms in a multiple-unit-container (OLDS-MC), psychotropic drugs (PD), controlled medicines (CM) or high-alert drugs) and the type of MEs occurred. Results Among the 609 MEs reported, wrong dose (32.2%), wrong drug (30.3%), omission (14.2%) and wrong patient (12.9%) were frequently observed. The ME occurrence stage were prescribing (55.3%) and administration (30.2%). Medication order review intercepted 77.9% of MEs. CM or LAIA increased the risk of medication omission (POR: 3.9, 95%CI: 1.8–8.4 and 2.5, 95%CI; 1.2–5.1, respectively) while the use of high-alert medications decreased it (0.2, 95%CI: 0.1–0.8). OLDS-MC and PD were more likely to be administered to the wrong patient (6.1, 95%CI: 3.3–11.4 and 16.1, 95%CI: 7.2–35.8). LAIA were associated with an increased risk of wrong dose (3.4, 95%CI: 1.8–6.3). Actual errors risk was lower with high-alert drugs (0.5, 95%CI: 0.3–09), but higher with CM (3.5, 95%CI: 1.5–8.0), OLDS-MC (2.1, 95%CI: 1.2–3.8) and PD (2.5, 95%CI: 1.8–3.5). Patients exposed to high-alert drugs were likely to have a serious error (3.5, 95%CI: 1.2–10.4). Conclusions This study sheds an innovative approach to analyze MEs by demonstrating that certain medication situations were more likely to lead to certain types of error. This enables the most appropriate prevention barriers to be put in place to intercept ME.