Document resource
Objectives This study aims to identify readily available admission variables associated with unintentional medicinal discrepancies (UMDs) in psychiatric settings and to develop a feasible prioritisation rule with robust psychometric performance.Methods This observational study collected information on items available at admission for patients admitted to six French psychiatric hospitals between June and August 2021. Based on the international literature and personal experience feedback, a multidisciplinary committee network and a psychiatry-mental health research federation elaborated a list of items potentially valuable for prioritisation and proposed a specific tool for psychiatric settings, named O-PsyConcil. Medication reconciliation (MR) was performed, and UMDs were identified. Items associated with UMDs were investigated through a backward stepwise logistic regression model, calculating adjusted ORs and their 95% CIs. The prioritisation rules were proposed using the more efficient combination in terms of feasibility and psychometric characteristics.Results The study included 513 patients (median age 48 years, IQR 33–68; 48% women). Factors associated with UMDs were comorbidities (OR 1.49 for one more comorbidity; 95% CI 1.23 to 1.82), psychiatric high-risk medications (HRMs) (2.64; 1.68 to 4.16), and somatic HRMs (3.43; 1.70 to 7.06). Based on these results, the prioritisation rule combining one comorbidity and/or one psychotropic HRM and/or one somatic HRM led to assessing 61% of the total population, with median performances of 82% (IQR 78-85%) for sensitivity and 48% (43-52%) for specificity.Conclusions The O-PsyConcil tool demonstrated satisfactory performance, suggesting it could enhance the efficiency of psychiatric MR compared with standard non-prioritised MR.