Using the Electronic Medical Record to Identify and Screen Patients at Risk for Delirium

Delirium is a common condition for hospitalized older adults. It is often associated with high health care costs and negative outcomes, including: increased rates of nursing home placement, increased likelihood of developing dementia, increased hospital length of stay and costs, and increased risk of death. Systematized screening can increase the recognition of delirium and improve care of hospitalized patients. Electronic medical records (EMRs) have been used to identify patients at risk for various co-morbidities and to standardize tremeatment for complex medical conditions. Likewise, EMRs could be used to extract potential risk factors for demenita for early identification of those patients.

The objective of this project was to use EMRs to implement a standardized screening program in hospitalized patients at high risk for delirium. The research team hypothesized that a screening tool would improve health outcomes for patients by early recognition and treatment.

The specific aims of the project were as follows:

• Use the EMR to identify delirium risk factors and develop a delirium prediction rule. 
• Use the EMR to generate a list of patients at risk for delirium in real time. 
• Use the EMR to improve documentation of delirium in the problem list. 

The researchers conducted a literature review and identified risk factors highly predictive of delirium including age, cognitive status, nutritional status, renal dysfunction, medication usage, and infection. They then used data already captured in the EMR of an academic hospital to create a prediction model to identify patients at high risk for developing delirium. The risk model was integrated into the EMR to allow for real-time screening of at-risk patients and included clinician prompts to improve documentation of delirium in the problem list.

A comparison of diagnoses in the problem list found that, while not statistically significant, there was an overall increase in documentation of delirium over the study period. This study successfully developed and implemented a prediction model and has the potential to improve delirium diagnoses and lead to improved prevention and treatment options. In the future, researchers may build off the risk model to design and implement clinical decision and support tools for prevention and treatment.