Short‐stay units (SSUs) provide traditional inpatient services in locations commonly adjacent to emergency departments (EDs). Hospitalists are well positioned to staff these units because of their expertise in caring for inpatients. However, little is known about the types of patients admitted to SSUs and whether these patients require unique services.
Our 14‐bed SSU opened in 2003 to reduce overcrowding on the traditional inpatient wards. The unit is geographically part of the ED but is staffed by a core group of hospitalist attending physicians from the department of medicine. From February through May 2006, all attending physicians of record in the SSU prospectively collected data from patient histories, physical exams, and medical records. We then retrospectively reviewed the electronic medical records of all these patients to identify repeat ED visits and hospitalizations (defined as any overnight admission to any inpatient service). We then performed backward stepwise multivariable analyses to explore 5 demographic and 6 clinical predictors of repeat ED visits. We also explored additional predictors specific to the 2 most common diagnoses.
Over a 4‐month period 755 patients were admitted to our SSU. Patients were middle‐aged (mean age 58 years, 25th to 75th per‐centiles 49‐66 years), from diverse backgrounds (29% were non‐English‐speaking, 14% were Hispanic or Latino, and 64% were black or African American), and often lacking a primary care physician (34%). Of the 755 patients, 707 (94%) were triaged into the SSU with cardiovascular diagnoses, mostly chest pain (n = 398), heart failure (n = 229), and syncope (n = 59). The median length of stay was 41.5 hours, and 71 patients (9%) were ultimately admitted to a traditional inpatient service. During a mean follow‐up period of 36 weeks (range 29‐46 weeks) after discharge from the SSU, 276 of 684 patients (40%) who were discharged home from the SSU returned to the ED. Of these 276 patients, 151 (55%) were readmitted, and most (n = 122, 81% of 151) were readmitted for diagnoses similar to the index SSU triage diagnoses. Multivariable analyses of all patients sent home from the SSU found that recent hospitalization (OR = 2.1, 95% CI: 1.5‐2.9) and speaking English (OR = 2.3, 95% CI: 1.6‐3.3) were the only predictors of return ED visits. Among chest pain patients, noninvasive testing predicted less recidivism (OR = 0.5, 95% CI: 0.3‐0.9), but angina classification and ECG findings were not significant. Among heart failure patients, having pulmonary edema predicted less recidivism (OR = 0.4, 95% CI: 0.2‐0.8).
Return ED visits from our hospitalist‐run SSU are very common and not reliably predicted by either traditional demographic markers of health disparity or clinical markers of increased disease severity. Future research should focus on new strategies to help identify patients at high risk for recidivism.
B. P. Lucas, None; S. Tchernodrinski, None.
To cite this abstract:Lucas B, Tchernodrinski S. High Recidivism Rates after Discharge from a Hospitalist‐Run Short‐Stay Unit Preparing to Reduce the “Frequent Flyer” Effect. Abstract published at Hospital Medicine 2007, May 23-25, Dallas, Texas Abstract 41. Journal of Hospital Medicine. 2007; 2 (suppl 2). https://www.shmabstracts.com/abstract/high-recidivism-rates-after-discharge-from-a-hospitalistrun-shortstay-unit-preparing-to-reduce-the-frequent-flyer-effect/. Accessed January 18, 2020.