A Systems Approach to Identify Factors Influencing Adverse Drug Events in Nursing Homes

The objectives of the study were to identify the incidence rate and characteristics of adverse drug events (ADEs) in nursing homes (NHs) using the ADE trigger tool and to evaluate the relationships between resident and facility work system factors and incidence of ADEs using the System Engineering Initiative for Patient Safety (SEIPS) model. The study used 2 observational quantitative methods, retrospective resident chart extraction, and surveys. The participating staff included Directors of nursing, registered nurses, certified nurse assistants (CNAs). Data were collected from fall 2016 to spring 2017 from 11 NHs in 9 cities in Iowa. Binary logistic regression with generalized estimated equations was used to measure the association between ADE incidence and resident and facility characteristics. We extracted data from 755 medical charts and conducted 33 staff surveys. There were 6.13 ADEs per 100 residents per month. More than half were fall-related (51.1%), and half of those were due to hypotension. Regression analysis revealed significant associations between ADEs and opioid analgesics, psychotropic medications, warfarin, skilled care, consultant pharmacist accessibility, nurse-physician collaboration, CNA vital sign assessment skills, number of physician visits, nurse workload, and use of electronic health records. Five resident characteristics (skilled care, dementia, use of opioids, warfarin, psychotropics) and variables from 5 domains of the facility work system (organization, task, environment, person, technology) had significant associations with ADE incidence. The SEIPS model successfully identified work system factors influencing ADEs in NHs.