Investigators at Cedars-Sinai in Los Angeles are using electronic health records to identify hospitalized patients likely to have dementia to help medical staff tailor care to best serve these patients.
To identify these patients, investigators created an algorithm to search patients’ electronic health records for a diagnosis of dementia and for prescriptions for medications approved by the Food and Drug Administration to treat dementia. When the algorithm detects a patient with possible dementia, a yellow banner pops up on the patient’s chart to make hospital staff aware.
The method they developed is detailed in a study published in the peer-reviewed Journal of the American Geriatrics Society.
“People with dementia or cognitive impairment can be especially vulnerable in the hospital if their care team is unaware,” said Zaldy Tan, M.D., M.P.H., medical director of the Jona Goldrich Center for Alzheimer’s and Memory Disorders at Cedars-Sinai, in a statement. “Our study is the first to investigate the feasibility of utilizing the electronic health record to identify these patients and alert the hospital team to help guide clinical care.”
If a patient with dementia is hospitalized for an unrelated condition, such as a fall or infection, they might not be able to accurately describe their medical history or safely make decisions about their medical care, Tan said. Patients with dementia might also need help to understand discharge instructions or just to stay calm in the hospital environment.
“Diagnoses such as Alzheimer’s disease, dementia or cognitive impairment are often not documented in a patient’s medical records,” said Tan, who is also director of the Memory and Healthy Aging Program and the C.A.R.E.S. Program at Cedars-Sinai. “And if providers are not aware that their patient has dementia, they may not call a loved one who can provide critical information, help with decision-making, and provide support.”
“The biggest challenge in creating the algorithm was the variety of clinical scenarios that led to a potential diagnosis of dementia,” said Cameron Escovedo, M.D., M.S., physician leader of Enterprise Information Services at Cedars-Sinai and co-author of the study, in a statement. “We had to account for multiple scenarios to ensure the algorithm was complex enough to capture everyone.”
“Given the poor patient outcomes currently associated with dementia care in the hospital setting—including increased risks for falls, use of restraints, and prescription of antipsychotic medications—there was a need for a method to accurately identify these patients,” said Nancy Sicotte, M.D., chair of the Department of Neurology at Cedars-Sinai and senior author of the study, in a statement. “Our algorithm alerts the hospital team to the presence of cognitive impairment so that they can employ targeted interventions and ultimately improve outcomes for vulnerable hospitalized patients.”
To help ensure that medical staff understand how to respond to these patients once identified, a team of nurses and physicians at Cedars-Sinai created and tested a training program and published their results in the peer-reviewed journal Geriatric Nursing.
Tan said that the identification system will be expanded to all medical and some surgical units, and that the system and the training—currently in use only at Cedars-Sinai—could easily be deployed at other institutions as well.
