Goldman Chest Pain Protocol

The Goldman chest pain protocol is a computer-derived decision aid and was developed to assist physicians in using routinely collected clinical and test data in the ED in identifying patients likely to be having an AMI who therefore require triage to the CCU. A statistical technique of recursive partitioning was used to divide the study's subjects into subgroups by ED data elements of the history, physical examination, and ECG into having proportions of AMI higher or lower (Fig. 2) (70,71).

The protocol was developed using prospectively collected data on patients presenting to the ED with acute chest pain (71). AMI was used as the outcome on which to base triage to the CCU, given that the risk of emergent complications early in the admission is 17% compared with 0.5% in patients without AMI. Recursive partitioning was used to develop a decision tree with the probability of ruling in for an AMI as the outcome

Chest Pain Protocol

*chest pain protocol divides patients into high and low risks of AMI on the basis of a recursive partitioning model. It uses routinely collected and interpreted history, physical examination, and electrocardiographic data. Reproduced with permission from ref. 57. [Zalenski RJ, McCarren M, Roberts RR, et al. An evaluation of a chest pain diagnostic protocol to excluded acute cardiac ischemia in the emergency department. Arch Intern Med 1997;157:1085-1091. ©1997 American Medical Association.

*chest pain protocol divides patients into high and low risks of AMI on the basis of a recursive partitioning model. It uses routinely collected and interpreted history, physical examination, and electrocardiographic data. Reproduced with permission from ref. 57. [Zalenski RJ, McCarren M, Roberts RR, et al. An evaluation of a chest pain diagnostic protocol to excluded acute cardiac ischemia in the emergency department. Arch Intern Med 1997;157:1085-1091. ©1997 American Medical Association.

of each branch. The protocol was prospectively validated in a population of 4770 patients who presented with chest pain (70). Follow-up of the 2232 patients who were discharged from the ED was performed by either physical examination, follow-up measurement of CK, or telephone to determine whether an AMI had occurred after discharge from the ED. Diagnostic performance for AMI compared with that of physicians for the same patients is shown in Table 7.

These data show the sensitivity of the protocol for predicting AMI with triage to the CCU to be the same as that of physicians, but with higher specificity than physicians. It was projected that 11.5% of patients without AMI would have been triaged elsewhere had the protocol been used.

A prospective trial used a time series study design to determine the impact of the protocol on the triage and outcomes of patients presenting to the ED at the Brigham and Women's Hospital in Boston (72). The time series design used six 14-wk cycles, consisting of a 5-wk control and/or intervention period separated by 2-wk wash-out cycles. Risk estimates and triage recommendations were provided to physicians in a nonobtru-sive fashion. Rates of admissions during intervention and control periods were unchanged in the hospital (52 and 51%, respectively) and in the CCU (10% each). Also, there were no significant differences in hospital length of stay or average total costs (72).

Table 7

Diagnostic Performance for AMI Compared with that of Physicians for the Same Patients

Table 7

Diagnostic Performance for AMI Compared with that of Physicians for the Same Patients

Your Heart and Nutrition

Your Heart and Nutrition

Prevention is better than a cure. Learn how to cherish your heart by taking the necessary means to keep it pumping healthily and steadily through your life.

Get My Free Ebook


Post a comment