Date: June 28th, 2019

Reference: Cardona et al. Prospective Validation of a Checklist to Predict Short-term Death in Older Patients After Emergency Department Admission in Australia and Ireland. AEM June 2019.

Guest Skeptic:Dr. Justin Morgenstern is an emergency physician and the creator of the excellent #FOAMed project called First10EM.com

Case: An 83-year-old man with early dementia, congestive heart failure (CHF), prior myocardial infarction (MI) with three stents, chronic obstructive pulmonary disease (COPD), and atrial fibrillation is transferred to the hospital because the nursing home thinks he might have a urinary tract infection (UTI). On arrival he is febrile, confused, with an alternating level of consciousness, tachycardia, and a rapid respiratory rate. According to the family with him, he had never had an end of life conversation with his physicians. You think such a conversation is important to guide your care in the next few hours, but you wonder if there is a tool to help you predict this gentleman’s chance of dying during this visit or shortly after.

Background: Discussion about goals of care at the end of life are an essential component of emergency medicine. Such discussions are aided by accurate prognosis, so that life-saving interventions can be provided to those in need, but hopefully without providing overly aggressive management in patients with little hope of recovering.

Such prognosis is difficult, especially in the chaotic and time-limited environment of an emergency department. An accurate decision tool would be welcomed. Unfortunately, available tools are only modestly accurate and have not been rigorously validated.1,2 Basic demographic data alone are insufficient to predict individual patient risk.3

Therefore, the purpose of this study was to validate a personalized risk score – the Criteria for Screening and Triaging to Appropriate aLternative care (CriSTAL) – in older patients presenting to the emergency department.4

Clinical Question: To establish the predictive ability of individual and combined parameters in the CriSTAL tool to predict short-term post-discharge death in an elderly population.

Reference: Cardona et al. Prospective Validation of a Checklist to Predict Short-term Death in Older Patients After Emergency Department Admission in Australia and Ireland. AEM June 2019.

* Population: Patients over the age of 65 who stayed overnight in the emergency department or were admitted to hospital. The derivation population was from five hospitals in Australia, while the validation population was from a single hospital in Ireland.

* Exclusions: Patients with severe cognitive impairment, the critically ill, or those unable to communicate in English were all declared ineligible to participate unless they had a surrogate.

* Intervention:The CriSTAL score, and its various components.

* Comparison:None

* Outcome:

* Primary Outcome(s): Death within three months and CriSTAL’s predictive ability.

* Secondary Outcome: Predictive ability for in-hospital death

This is an SGEMHOP episode and usually we have the lead author on the show. Unfortunately, we were not able to coordinate the recording times with being in three different countries. Our hope is the authors can respond to our nerdy questions on the blog.

Authors’ Conclusions: The modified CriSTAL tool (with CFS instead of Fried’s frailty instrument) had good discriminant power to improve certainty of short-term mortality prediction in both health systems.