Evaluating Age as a Significant Predictor for Hospital Admission Through the Emergency Department
Migliore M, Amir M, Gunaga S, Jarski R, Tuttle J, and Moss H. 173 Evaluating Age as a Significant Predictor for Hospital Admission Through the Emergency Department. Ann Emerg Med 2019; 74(4):S68-S69.
Ann Emerg Med
Study Objectives: One of the most central decision points in every emergency department (ED) patient encounter is whether the patient will be admitted to the hospital or discharged home. The decision is made based on the evaluation of complex variables ranging from a patient’s history and physical exam, precise diagnostic studies, risk stratification scales, to gestalt. Despite regularly using risk stratification tools to determine a patient’s pre-test probability for disease, there is currently no such tool for determining an ED patient’s pre-test probability for admission. The focus of this study was to specifically evaluate patient age as an independent predictor of hospital admission or discharge. Methods: We performed a retrospective chart review of patients presenting to the ED of a 400-bed community hospital with approximately 60,000 ED visits per year. Study eligibility included children and adults of any age who presented to the ED between January 1, 2016 and December 31, 2016 with an emergency severity index score (ESI) of 3 or lower. A list of 1000 patients from the 1st of every month was generated and subjects were identified based on age in chronological order of presentation for that month. Five patients from every decade were identified from all 12 months to reduce seasonal biases, until each decade’s cohort was filled with an N of 60. Age groups were broken down into decades from 0-9 years, 10-19 years, etc., continuing to >90 years. A variety of data were abstracted regarding each patient encounter, but the primary variable measured was patient age at presentation and whether the patient was admitted or discharged from the ED. A transfer for higher level tertiary care was considered an admission. Data were compared using the chi-square test for association or Fisher’s exact test. Results: A total of 600 patients were identified with 60 in each decade. Age was found to be a significant predictor for admission (P<0.001). The 0-9 age group was least likely to be admitted (4.2%) and those > 90 were most likely to admitted (68.6%). The probabilities of admission by age are presented in Table 1. Discussion: It is well recognized that most of a hospital’s admissions come through its ED and these patients consume a significant portion of health care resources. We identified 3 cohorts that could best represent low, moderate and high-risk admission groups. Patients between the age of birth to 29 (< 10% chance of admission), those between age 30 to 59 (approximately 25% chance) and those 60 years or older (approximately 50% - 70% chance of admission). Medicare and other governing bodies have already developed emergency physician dashboards that use a variety of variables, including admission rates, to grade a physician’s quality of care. Arbitrary cut-off values are set for what constitutes an appropriate hospital admission rate though should instead be based on objective pre-test predictors of admission, like age. Future research should strive to determine additional significant factors that contribute to hospital admissions through the ED and explore the development of formal pre-arrival ED admission risk stratification tools.