Prospective identification of medication harm in geriatric inpatients using a modified trigger tool

Abstract

AbstractBackgroundMedication harm (MH) causes patient morbidity and is a major healthcare burden. Research into MH is growing, but effective methods to identify MH are debated. The prevalence of MH is often based on an incomplete, retrospective chart review or spontaneous reporting, reliant on busy clinicians. A practical and clinically relevant method to detect MH is required. A trigger tool (TT) offers a solution.AimTo evaluate a modified TT to prospectively detect MH and determine the prevalence and severity of MH in a geriatric population.MethodAn international TT was peer evaluated and modified for use in a geriatric ward of a quaternary hospital. Patients were recruited over a 6‐month period. The TT was applied to prospectively help identify MH, which was assessed for causality and severity. Positive predictive values (PPV) were estimated for each trigger to determine its sensitivity in identifying MH. Ethics approval was granted by the Metro South Human Research Ethics Committee (Reference no: HREC/2022/QMS/80654) and the study conforms to the Australian National Statement on Ethical Conduct in Human Research. Informed consent was obtained from all participants through completion of a written consent form, after a full explanation of the protocol design.ResultsFifty patients consented, of which 16 (32%) patients experienced one or more MH events. A total of 257 triggers were activated (mean of 5.14 per patient) and 31 (12%) predicted an event. Of the 31 events, 19 (61.3%) events were rated as mild and 12 (38.7%) events were rated as moderate to severe. The most common events were bleeding/large bruising, major constipation, diarrhoea, and vomiting. The triggers with the highest PPV included triggers T5 (bleeding/bruising), T9 (gastrointestinal disorders), and T11 (major constipation) with PPVs of 0.455, 0.238, and 0.286, respectively.ConclusionA modified TT helped to detect MH in a geriatric population and will aid in identifying events in future studies.