Medication errors leading to hospitalization of the aged

University Of Pennsylvania, Philadelphia PA

Investigators

Linked publications & trials

Abstract

DESCRIPTION: (PROVIDED BY APPLICANT) Medication errors leading to adverse events are a major cause of preventable hospitalizations, particularly among the elderly. Medication errors that result in dose-related adverse events can occur anywhere along the medication use process. The specific causes of and solutions for medication errors are likely to differ across drugs with different properties. Thus, investigations of errors must focus on specific drugs, particularly those with a narrow therapeutic index, such as warfarin, phenytoin, and digoxin. Error prevention strategies might include programs that improve therapeutic drug dosing and monitoring (phenytoin), improve patient drug adherence (warfarin), or reduce physician gaps in knowledge about the risks and benefits of drugs (digoxin). However, the precise design of interventions requires information on the underlying causes of these errors for both new and chronic users of these specific agents. Moreover, such interventions will only be feasible if they can be targeted to high-risk groups of patients for whom the absolute risk of adverse events due to medication errors is sufficiently high to justify these interventions. The primary aim of this project is to identify predisposing factors for hospitalizations due to errors in medication use among a large, representative cohort of community-dwelling elderly patients initiated or maintained on warfarin, phenytoin, and digoxin. The primary study hypothesis is that uncoordinated medical and pharmaceutical care, inadequate delivery of new medication instructions, visual and cognitive impairment, and psychosocial barriers (depression, coping, and support) are predisposing factors for medication errors resulting in hospitalization. The secondary aims of this study are (1) to develop a prediction rule to identify elderly outpatients at high risk for hospitalization due to errors in use of these drugs, and (2) to estimate the costs associated with hospitalization due to these errors. The hypotheses for these aims are: (1) patient factors can be used to accurately predict groups at high risk for these hospitalizations, and (2) the costs of these errors outweigh the costs of potential targeted interventions. The proposed study is a prospective cohort study enrolling members of the Pennsylvania Pharmaceutical Assistance Contract for the Elderly (PACE). Subjects will be sampled across the State to ensure geographic diversity in the study sample. At the time of enrollment in the cohort, subjects will undergo a detailed baseline interview to identify key psychosocial, behavioral, and clinical predictors. Outcomes will be identified over a two-year follow-up period by regular phone contact with cohort members using a screening instrument to identify hospitalizations. Medical records will be reviewed, and clinical findings, along with admission drug or anticoagulation levels, will be used to identify hospitalizations that are likely due to medication errors. Analyses will focus on the identification of risk factors and development of a prediction rule to identify subjects at high risk of hospitalization due to medication errors.