Article Text

Download PDFPDF

Chest pain observation units
  1. Louis Graff
  1. Division Emergency Medicine, Department Surgery, Yale University Medical School, USA
  1. Correspondennce to: Professor Graff (louisgraff{at}

Statistics from

Editor,—Goodacre concluded in the January 2000 issue of the journal “there is no strong evidence that chest pain observation units (CPOU) will improve outcomes” and further evidence is necessary to determine whether this approach can be applied in the United Kingdom.1 He did not reach this conclusion from the cost studies that he listed in table 2. These savings were present in all nine studies reviewed and ranged from $1873 per patient to $567 per patient. He reached this conclusion from examination of mortality and missed pathology that he summarised in table 1. The five reviewed studies included three randomised clinical trials.

The flaws in Goodacre's analysis lies in his failure to examine physician emergency department disposition patterns and his failure to perform power calculations. The missed myocardial infarction (MI) diagnosis rate ranges from 2.8% to 13% in large clinical trials without CPOUs.25 There is a 11% to 25% death rate for those whose diagnosis is missed and the patient released home from the emergency department with false reassurances.2, 3 This is the leading cause of adverse outcomes and malpractice suits in emergency medicine in the United States.6 The rate of missed diagnosis has been shown to be inversely related to the percentage of emergency department patients receiving a “rule out MI evaluation” (performed during hospital admission before the development of CPOUs).5

What sample size is needed to demonstrate a 25% reduction in the missed MI rate? The average miss rate in emergency departments in the United States is 4% with a 60% “rule out MI evaluation” rate.5 At this emergency department disposition rate, over 50% of admitted patients are found after full evaluation to have no serious disease as the cause of their symptoms.7 The study sample size required to demonstrate a reduction in the average missed MI rate from 4% to 3% is 6262 patients per study arm (85% power). The size of the three randomised clinical trials reviewed by Goodacre were much smaller than this requirement with the largest trial having only 212 patients in each study arm.

I agree with the author's suggestion to not be complacent with the present traditional emergency department approach to chest pain evaluation. Examination of present United Kingdom utilisation practices (% emergency department patterns admitted, % admitted with serious disease) and quantifying the quality of patient care (rigorous follow up to identify the per cent of released emergency department patients with missed disease) might lead the author to reconsideration the value of implementing CPOUs.


View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.