The emergency department (ED) is an inherently high-risk setting, especially for older adults. EDs employ a rapid triage and evaluation system that may not capture the complex nature of a geriatric visit. The disposition plan of patients who visit an ED is to be directly discharged home, placed in observation care, or admitted to the hospital. There is mounting concern that recent increased use of observation care may also subject patients to a rapid diagnostic regiment with unknown utility. As older patients continue to rely on the ED and hospitals for their care, the potential for treatment that is not geriatric-oriented and results in undesirable outcomes shortly after discharge is high. A negative outcome within 7-days of discharge from an acute care encounter is temporally related to the visit and has important policy implications. The goal of this research program is to identify patients at risk for a short-term negative outcome, defined as either an inpatient admission or death, within 7 days of discharge from their acute care encounter. The proposal will utilize a randomly selected sample of 5% Medicare Claims data, which amounts to 2.4 million beneficiaries, of which patients age >65 years who visit the ED in 2012 will be analyzed. The multi-disciplinary project team includes experts in health services research, geriatrics, emergency medicine, administrative claims data research, and statistics.
In Aim 1, the study will assess predictors of ED disposition (discharge or observation care) following the ED visit. The team will then conduct a stratified analysis to identify predictors of negative outcomes following discharge from observation care (Aim 2) or the ED (Aim 3). The proposal will assess the association between two sets of primary predictors, ED discharge diagnosis codes developed by the PI as well as the recently created Medicare Chronic Conditions, and outcomes. Additional variables will include patient demographics, comorbidities, facility characteristics, prior inpatient utilization, skilled nursing facility usage, and community demographics. The analysis will include descriptive statistics, assessment for similarities and collinearity between the predictors, and hierarchical random effects logistic regression models pertaining to each Aim. All project aims have sufficient power to complete the analyses. The proposed study will identify characteristics of patients at risk for negative short-term outcomes following acute care encounters and will (1) provide population-level national data to understand negative outcomes following discharge, (2) serve as a foundation for future work in developing a post-discharge intervention that will reduce these events, and (3) help ED managers, geriatricians, and health policy experts develop national guidelines.

Public Health Relevance

The emergency department (ED) is an inherently high-risk setting, especially for older adults. EDs and more recently, observation services, employ a rapid triage and evaluation system that may not capture the complex nature of a geriatric visit and could result in either admission or death within 7 days of discharge. Understanding the characteristics of patients at risk for such outcomes after discharge from the ED or observation care can help improve both ED and follow-up care. Currently, this information is not known. Our study will use national Medicare claims files to identify the patients at risk for these outcomes with the goal of helping improve care for all older adults.

National Institute of Health (NIH)
Small Research Grants (R03)
Project #
Application #
Study Section
Special Emphasis Panel (ZAG1)
Program Officer
Bhattacharyya, Partha
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of California Los Angeles
Internal Medicine/Medicine
Schools of Medicine
Los Angeles
United States
Zip Code