Document resource
Objectives Clinical decision support systems (CDSSs) may identify risk prescriptions and prevent adverse drug events (ADEs), but their effectiveness depends on design and implementation. Uppsala University Hospital is developing system-assisted pharmaceutical validation (SAPVAL), a CDSS-based service using clinical rules with clinical pharmacists as first-hand recipients. This proof-of-concept study aimed to assess whether the proposed SAPVAL service could address unmet clinical needs in detecting and managing risk prescriptions by using clinical ‘if-then’ rules based on guidelines and local expertise in a large Swedish hospital.Methods A list of clinical if-then rules was adapted following discussions with local clinical experts from different fields. The final set of rules was then determined based on available clinical data. The rules were retrospectively and manually applied to electronic health record (EHR) data from 765 patients aged ≥65 years in a large Swedish hospital. The generated alerts were analysed descriptively, and their clinical relevance was assessed by a senior clinical pharmacist.Results A total of 50 if-then rules were developed and applied to patient data. The participants had a median age of 76 (IQR 71–82) years. Of the 765 included patients, 145 (19%) triggered at least one alert, resulting in a total of 181 alerts. Nearly half (47%) of these alerts were deemed clinically relevant; and, of these, 72% persisted after 48 hours or until discharge.Conclusions The guideline-based, locally informed clinical rules could identify clinically relevant and persistent risk prescriptions in older adult inpatients in a large Swedish hospital. The findings support the feasibility of SAPVAL as a pharmacist-led CDSS intervention. However, since half of the alerts were assessed to be clinically non-relevant, the study highlights the need for ongoing refinement of rules to improve specificity and service efficiency.