Share this article on:

Ignore Performance Measures at Your Own Peril

Welch, Shari J. MD

doi: 10.1097/01.EEM.0000357586.16567.38
Quality Matters

Dr. Welch is a fellow with Intermountain Institute for Health Care Delivery Research, an emergency physician with Utah Emergency Physicians, and a member of the board of the Emergency Department Benchmarking Alliance. She has written two books on ED operational improvement; the latest, Quality Matters: Solutions for the Efficient ED, is available from Joint Commission Resources Publishing.



The National Quality Forum last year endorsed a series of 10 quality and performance measures that were sent to the Centers for Medicare & Medicaid Services. You may think this is not relevant to your practice, but it truly is.

This work by numerous quality organizations and technical expert panels was solicited by CMS to be used in a new reimbursement model. CMS plans to withhold a small percentage of reimbursements for care, and then return some of the money to hospitals to award for performance. You do care about performance measures!

Back to Top | Article Outline

Clinical Quality Measures

Five measures are included, and there is little to take issue with.

Septic shock treatment bundles should begin in the ED: The evidence for septic shock treatment bundles looks irrefutable. The hope is that we see gains in septic shock treatment similar to what we have seen in acute ST segment elevation MI. It is likely that your hospital is already tracking compliance with treatment guidelines and outcomes. One issue to be considered on an institution-by-institution basis is when and where the handoff should take place. For facilities with longer waits for beds, emergency physicians have become de facto intensivists, starting central lines and resuscitation measures, and providing the first few hours of critical care in the ED. Emergency medicine is a zero-sum game, though, and these patients consume a disproportionate amount of ED resources that deprive other ED patients. Increased pressure to get admitted patients upstairs may shift this care burden to the ICU, though care would still be initiated in the ED.

Proper ET tube placement should be confirmed: This clinical measure should have no opposition. Even prior to its endorsement, it has been considered standard of care in emergency medicine. From a risk management perspective, this measure should be embraced. Although the National Quality Forum did not specify actual end-tidal CO2 measurement and documentation with a digital device (as opposed to a colorimetric one), it is moving from the OR to the ED as the standard of care.

Female patients with abdominal pain should have a pregnancy test; treatment for pulmonary embolism should begin in the ED: These two measures were endorsed by the American College of Emergency Physicians, and are measures of adherence to clinical guidelines: pregnancy test in females with abdominal pain (of childbearing age) and the institution of treatment for pulmonary embolism in the emergency department. Kevin Klauer, DO, the director of quality and clinical education for EMP Management Group, a large self-insured emergency physician group in Canton, OH, thinks these are straightforward. At EMP, a pregnancy test is ordered by protocol on all women with abdominal pain. They also have standardized order sets for anticoagulation in the ED once a diagnosis of pulmonary embolus is made.

Pediatric patients should have a kilogram weight documented: The only fact you need to understand about this measure is this: It measure is being promoted and endorsed by the American Academy of Pediatrics. Medication errors are 20 times more likely in children than adults. This focus should help us to improve in this realm.

Back to Top | Article Outline

Performance Measures

The length of stay of admitted and discharged patients form the first two of five National Quality Forum time measures. Note that these should be reported as median numbers, not averages. Many in ED quality improvement have concerns about these measures, including that aging Americans will mean that each ED encounter will require more tests and treatments and therefore longer lengths of stay. All ED volume bands in all parts of the country are seeing increasing lengths of stay, which are in part tied to the fact that we are doing more for patients.

Sometimes a longer ED visit is good care. You can imagine the patient with abdominal pain observed for six hours to see if there is appendicitis brewing or the patient with chest pain who receives a time-consuming stress test and cardiology consultation that ultimately prevents an admission. On the other hand, length of stay does correlate with patient satisfaction and probably has a place as a performance metric. There has been no attempt to stratify EDs according to census and type, and there is no question that the operating characteristics differ from the small community hospital to the large urban tertiary care center.

The third operational measure looks at the time from the decision to admit to departure, which all EPs know as boarding. This has been identified as one of the biggest patient flow problems, correlated with crowding, diversion, and poor clinical outcomes. The solutions to boarding obviously are not in the emergency department, and this focus may help organizations understand hospital-wide flow and hospital-wide solutions.

Finally, door-to-provider time is a time measure, and left without being seen is a percent measure, but they are intimately related. The studies depicting the linear relationship between door-to-provider time and left without being seen are almost too numerous to count, and both correlate with patient satisfaction. Door-to-provider time also has an important clinical link: Clinical entities from acute coronary syndromes and community-acquired pneumonia to septic shock and acute stroke are on the clock. The sooner the physician encounter takes place, the more likely the time guidelines for acute clinical care can be met.

Defining and measuring quality in emergency medicine is a formidable task, and many iterations are to come. These 10 measures for our specialty are an attempt to quantify how well we and our organizations perform on clinical guidelines and operational efficiency. They have been thoughtfully developed so we can move the specialty toward increased efficiency and safety. In the improvement world, there are two slogans that are frequently tossed about in relation to this work: “You can't manage what you can't measure,” and its corollary, “Be careful what you measure!” Even in this first iteration, it would appear that these slogans have been applied.

Back to Top | Article Outline

Measuring Performance

Clinical Quality Measures

  • ▪ Begin septic shock treatment in the ED.
  • ▪ Confirm proper ET tube placement.
  • ▪ All female patients with abdominal pain should have a pregnancy test.
  • ▪ Begin treatment for pulmonary embolism in the ED.
  • ▪ Document the weight of pediatric patients in kilograms.

Operational Performance Measures

  • ▪ Length of stay for admitted patients.
  • ▪ Length of stay for discharged patients.
  • ▪ Time from the decision to admit to departure (boarding).
  • ▪ Door-to-provider time.
  • ▪ Patients who leave without being seen.
Back to Top | Article Outline

Bonus Article by Dr. Welch on

Read all about redesigning patient flow in Dr. Shari Welch's exclusive EMN Online article, “In the Zone.” Go to, and click on EMN Online for July.

Back to Top | Article Outline

EMN Online

Read the National Quality Forum's report, “National Voluntary Consensus Standards for Emergency Care, on Go to EMN Online for July, and click on EMN FastLinks.

© 2009 Lippincott Williams & Wilkins, Inc.