Determinants of resource utilization for patients admitted for evaluation of acute chest pain

Objective:To identify determinants of resource utilization among patients with suspected acute myocardial infarction. Design:Prospective cohort study, with prospective collection of detailed clinical data and retrospective collection of nonclinical data and resource utilization data. Setting:Urban, tertiary-care, teaching hospital. Patient population:992 consecutive patients over the age of 30 years, admitted from the emergency department for evaluation of acute chest pain unexplained by obvious trauma or chest roentgenographic abnormality, were eligible for the study. After excluding patients who had left against medical advice, who had been transferred to another bospital, or who had incomplete utilization data, 903 patients were included in the analyses. Measurements and outcomes:The authors evaluated the effects of 22 clinical and nonclinical factors on resource use. Resource use was primarily evaluated by length of stay; charges were evaluated in secondary analyses. Results:In the entire study population, increased length of stay was associated with a diagnosis of acute myocardial infarction or angina, severity of complications, use of invasive and noninvasive testing, and initial triage to the coronary care unit. In the 424 (47%) patients who had had completely uncomplicated courses after admission, high coefficients of variability were found for length of stay (0.88) and for total charges (0.78). In these uncomplicated patients, increased length of stay was associated with the use of noninvasive cardiac testing (66% longer for patients undergoing echocardiography or radionuclide ventriculography, and 46% longer for patients undergoing exercise tests or ambulatory arrhythmia monitoring), initial triage to the coronary care unit (23% longer), admission at the end of the week (21% longer), and insurance coverage other than Blue Cross/Blue Shield or a commercial carrier (21% for self-pay, 25% for Medicaid, and 48% for Medicare). Conclusions:These findings indicate that after adjustment for important clinical factors, nonclinical factors had a significant impact on length of stay among a large group of uncomplicated patients. Interventions aimed at reducing logistic difficulties in the performance of noninvasive testing and decreasing the number of low-risk patients who are triaged to coronary care unit beds may decrease resource utilization. Received from the Divisions of Clinical Epidemiology and General Medicine and the Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School; and the Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts. Presented in part at the annual meeting of the American Federation for Clinical Research, April 28 – May 2, 1989, Washington, DC. Supported in part by grants from the National Center for Health Services Research (HS 05927), the Robert Wood Johnson Foundation, Princeton, NJ (678105), the John A. Hartford Foundation, New York, NY (83102-2H), and the Agency for Health Care Policy and Research (1-PO1-HS06431-02 and HS 06452-02). Dr. Lee is the recipient of an Established Investigator Award (900119) from the American Heart Association. Dr. Udvarhelyi is the recipient of a Medical Foundation Fellowship award.