ECONOMIC EVALUATION Hospitalization beyond 3 days following thrombolysis in patients with uncomplicated acute MI was not economically attractive Newby LK, Eisenstein EL, Califf RM, Thompson TD, Nelson CL, Peterson ED, et al. Cost effectiveness of early discharge after uncomplicated acute myocardial infarction. N Engl J Med 2000; 342: 749d755 OBJECTIVE To assess the cost-effectiveness of prolonging hospitalization beyond 72 hours after thrombolysis in patients with uncomplicated acute MI. DESIGN Decision-analytic model based on the GUSTO-1 database. SETTING Multinational trial. PATIENTS 22 361 of the eligible cohort of 41 021 participants who had an uncomplicated course through the initial 72 hours (i.e. the absence of death, reinfarction, congestive HF, recurrent ischemia, shock, stroke, emergency angiography or angioplasty, CABGS, intra-aortic balloon pumping, or cardioversion or defibrillation). Patients with evidence of three-vessel disease ( ' 75% stenosis) or left main disease, or who had elective CABGS more than 3 days after thrombolysis were excluded, regardless of clinical course. MAIN COSTS AND OUTCOME MEASURES Costs included key hospital resources (e.g. room and monitoring) and physicians’ fees derived from the Medicare fee schedule. Costs were expressed in 1997 US $. The main outcome measure was the cost-effectiveness associated with an additional day of monitored hospitalization expressed in terms of dollars per year of life saved. Incremental survival attributable to one additional day of monitored hospitalization was based on the rate of resuscitation following cardiac arrest between 72 and 96 hours. MAIN RESULTS Of the 22 361 patients with an uncomplicated course over the initial 72 hours following thrombolysis, 16 (0.07%) suffered ventricular arrhythmias during the next 24 hours. Of these 16, 13 (81%) survived for at least 24 hours. Those with arrhythmias on day 4 were older, more likely to be female, have diabetes or prior CAD, to have undergone revascularization, and to have had lower baseline systolic BP and a higher heart rate. The mean incremental survival attributable to one additional day of hospitalization was 0.006 years of life gained. The cost associated with an additional day of monitored hospitalization was $624. The cost-effectiveness ratio was $105 629 per year of life saved. A sensitivity analysis on the basis of variations in the incidence of ventricular arrhythmia on day 4 yielded cost-effectiveness ratios ranging from $65 777 to $183 525 per year of life saved. If the cost could be dramatically reduced or if a high-risk subgroup could be identified, the cost of the fourth day of monitored hospitalization may become economically attractive. CONCLUSION Hospitalization beyond 3 days following thrombolysis in patients with uncomplicated acute MI was not economically attractive. Commentary Clinicians face many pressures to maximize the efficiency of inpatient care, thereby decreasing length of hospital stay and reducing costs to their institution. In the management of acute myocardial infarction, the accepted minimum length of stay has decreased steadily. 1 Technical innovations, such as primary angioplasty, have been shown to increase the ability to discharge patients safely as early as 3 days after MI. 2 Newby et al have given further impetus to early discharge by finding that a policy of an extra day of monitored hospitalization after thrombolysis and an uncomplicated course was not economically attractive compared to routine discharge after 3 days. Applying the results of this model to routine clinical practice requires careful examination of its assumptions. The most questionable is that any testing or care not completed by 3 days can be efficiently shifted to the outpatient setting with no impact on cost or outcome. Early discharge may, in fact, require mobilization of substantial institutional resources, with higher costs, to accomplish the goals of care in three days; these costs are not included in the model. Furthermore, medical management may not be easily completed after discharge, and may thereby have poorer outcomes, at least in the short term, and higher societal costs. In addition, patients’ satisfaction with care and quality of life after discharge are important modifying factors in the practical implementation of this strategy, and this model does not incorporate them. There are no clear data on patient preference for length of stay after MI; however, we suspect that patients may prefer to stay longer because of their need for education and adaptation to a new diagnosis. Finally, 57% of patients in the GUSTO trial were potential candidates for early discharge; the proportion may well be lower in the general population with MI, because of comorbid disease and a greater likelihood of complications. We find this study useful in pointing out how little absolute benefit is gained from an extra hospital day. Clinicians and policy-makers should apply its results carefully to their own patients and settings. Kenneth A. Locke, MD Allan S. Detsky, MD, PhD University of Toronto, Toronto, Ontario, Canada Literature cited 1. Antman EM, Kuntz KM. N Engl J Med 2000; 342: 808d809 2. Grines CL, Marsalese DL, Brodie B, et al. J Am Coll Cardiol 2000; 31: 967d972 ^ 2000 Harcourt Publishers Ltd Evidence-based Cardiovascular Medicine (2000) 4, 53 doi:10.1054/ebcm.2000.0294, available online at http://www.idealibrary.com on 53