Prediction of hospital utilization among elderly patients during the 6 months after an emergency department visit

Study objective: A simple screening tool, Identification of Seniors at Risk (ISAR), developed for administration in the emergency department for patie nts 65 years and older, predicts adverse health outcomes during the 6 month s after the ED visit. In this study, we investigated whether the ISAR tool can also predict acute care hospital utilization in the same population. Methods: Patients 65 years and older who visited the EDs of 4 acute care Mo ntreal hospitals during the weekday shift over a 3-month period were enroll ed. At the initial (index) ED visit, 27 self-report screening questions (in cluding the 6 ISAR items) were administered. The number of acute care hospi tal days during the 6 months after the index visit were abstracted from the provincial hospital discharge database. High utilization was defined as th e top decile of the distribution of acute care hospital days. Results: Among 1,620 patients with linked data, a score of 2+ on the ISAR t ool predicted high hospital utilization with a sensitivity of 73% and a spe cificity of 51%, the area under the receiver operating characteristic curve was 0.68. The ISAR tool also performed well in subgroups defined by dispos ition (admitted versus discharged) and by age (65 to 74 years versus 75 yea rs and older). Conclusion: The ISAR tool, a 6-item self-report questionnaire, can be used in the ED to identify elderly patients who will experience high acute care hospital utilization as well as adverse health outcomes.