No abstract

No abstract

Adverse drug events occurred frequently in routine outpatient care, and many were serious and preventable. Computerized monitoring represents an efficacious approach for identifying ambulatory ADEs, although it needs additional refinement. In addition, site-specific variations need further exploration.

BackgroundDelirium is a life-threatening, clinical syndrome common among the elderly and hospitalized patients. Delirium is under-recognized and misdiagnosed, complicating efforts to study the epidemiology and construct appropriate decision support to improve patient care. This study was primarily conducted to realize how providers documented confirmed cases of delirium in electronic health records as a preliminary step for using computerized methods to identify patients with delirium from electronic health records.MethodsThe Mental Health Consult (MHC) team reported cases of delirium to the study team during a 6-month study period (December 1, 2009 - May 31, 2010). A chart extraction tool was developed to abstract documentation of diagnosis, signs and symptoms and known risk factors of delirium. A nurse practitioner, and a clinical pharmacist independently reviewed clinical notes during each patients hospital stay to determine if delirium and or sign and symptoms of delirium were documented.ResultsThe MHC team reported 25 cases of delirium. When excluding MHC team notes, delirium was documented for 5 of the 25 patients (one reported case in a physician’s note, four in discharge summaries). Delirium was ICD-9 Coded for 7 of the 25 cases. Signs and symptoms associated with delirium were characterized in 8 physician notes, 11 discharge summaries, and 14 nursing notes, accounting for 16 of the 25 cases with identified delirium.ConclusionsDocumentation of delirium is highly inconsistent even with a confirmed diagnosis. Hence, efforts to use existing data to precisely estimate the prevalence of delirium or to conduct epidemiological studies based on medical records will be challenging.