Home Feature A look at what’s wrong with performance measures for AMI and heart...

A look at what’s wrong with performance measures for AMI and heart failure

January 2006

Published in the January 2006 issue of Today’s Hospitalist

Editor’s note: In December 2005, Today’s Hospitalist examined how well hospitals are meeting performance measures from Medicare and JCAHO. In this issue, we take a closer look at the content of those measures.

In the new era of performance reporting, hospitals everywhere are worried how they stack up against their peers on a long “and growing “list of measures. For the physicians who are supposed to deliver care that meets those measures, however, it may not always be easy “or even advisable “to adhere to the indicated treatment approach.

Concern among physicians about how pay-for-performance initiatives will affect American medicine is nothing new. Some subspecialists, however, are now speaking out about specific problems they’ve found in the performance measures that are behind most pay-for-performance programs.

One problem, these physicians say, is that the performance measures that focus on myocardial infarction and heart failure have flaws that could actually inhibit the provision of appropriate care.

There are two concerns: First, some of the recommendations are behind the times and don’t always measure appropriate care. The second is that the black-and-white method of reporting adherence may penalize physicians who decide not to provide guidelines-recommended care to a patient for whom the treatment would be medically inappropriate.

Timing isn’t everything

Duke University cardiologist Matthew Roe, MD, for example, says that some of the “core measures” released by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) are off base or incomplete. These performance measures are quickly becoming the industry standard because most have been adopted by the Centers for Medicare and Medicaid Services (CMS).

Dr. Roe says that while the measures for ST-segment elevation MI address the timing for thrombolytic therapy or reperfusion, they don’t specify who should receive the treatment. “The JCAHO indicators say that [STEMI] patients should get a betablocker within 30 minutes or angioplasty within 90 minutes,” he explains, “but there’s no indicator for the percentage of eligible patients who get reperfused. Timing is important, but it’s not the most important issue.”

If that seems like a minor complaint, think again. Because the measures put timing above all other considerations, Dr. Roe says, physicians can be penalized for what amounts to doing the right thing. “If you don’t reperfuse patients,” he says, “they don’t count in your denominator. That doesn’t compel the right approach.” Dr. Roe similarly takes issue with the fact that the core indicators make no mention of statins, which are known to be beneficial in AMI.

“What we have is measures that are five years out of date,” Dr. Roe contends. “A lot of things are missing. What we need is regular updating and broader consensus on what the measures should be. At least for AMI, the JCAHO and CMS quality indicators “¦ don’t incorporate newer therapies that have strong benefits.”

Excluding effective therapy

Gregg Fonarow, MD, director of the Ahmanson-UCLA cardiomyopathy center, says that the heart failure measures are “a good starting point,” but he is quick to add that there are some real challenges.

In short, he explains, the measures suffer from the same types of drawbacks as the AMI measures: They focus on a single therapy to the exclusion of other treatments that may be as “or more “beneficial for certain patients.

“The only recommended therapy called for in the performance measures is an ACE inhibitor,” Dr. Fonarow explains, “yet there are two other class 1 life-saving therapies that aren’t integrated: beta-blockers and aldosterone antagonists. There’s a lag between the national performance measures and what current guidelines recommend.”

One of the key shortfalls of the current measures, both Drs. Roe and Fonarow agree, is that they don’t do a good job of taking into account critically important issues such as patients’ age, co-morbidities or functional status. And because the measures don’t incorporate patient-specific indications and contraindications, Dr. Roe says, they could have a less-than-desirable effect on medical decision-making.

Wanted: recognition of contraindications

“That’s one of the biggest challenges,” Dr. Roe says, “trying to figure out who is an ideal candidate for a certain therapy, strategy or procedure” that might be called for in various quality measures. To do that, the people creating performance measures need to factor into their decision-making process realistic contraindications, something Dr. Roe says hasn’t occurred yet in AMI and heart failure because of insufficient data.

Most data used to decide on treatment- and procedure-related quality indicators and measures has come from clinical trials, which typically enroll relatively younger, healthier patients and exclude individuals with co-morbidities. “Some people argue, understandably,” he maintains, “that some of the guidelines may not apply to the older, sicker patients” that cardiologists and hospitalists see in their everyday practice lives.

Conversely, the “all-or-nothing” method of performance measure reporting doesn’t allow a physician who decides not to prescribe a beta-blocker to note that the patient has a slow heart rate.

“We need a better evaluation of performance measures as applied to different populations,” Dr. Fonarow says, “so that the measures focus on the things that truly matter in terms of clinical outcomes.”

To move the argument beyond the realm of academic discussion, and to propel use of more up-to-date and flexible performance measures, both Drs. Fonarow and Roe urge physicians, including hospitalists, to tap into registries and data initiatives that seek and report more detailed information.

The ADHERE registry (www.adhereregistry.org), in which Dr. Fonarow is involved, collects a broad range of data on how patients with acute heart failure are managed, for example. And the CRUSADE initiative (www.crusadeqi.com), for which Dr. Roe is a principal investigator, is collecting data on care provided to patients with acute coronary syndromes.

Bonnie Darves is a freelance writer specializing in health care. She is based in Lake Oswego, Ore.

For more information or a listing of the recently expanded Centers for Medicare and Medicaid Services measures reported on the Hospital Compare Website, go to their Web site and click on the tab labeled “data details.” Look for the “for professionals” tab. JCAHO’s core indicators are also online.

What Medicare is telling patients about your AMI care

The following information about AMI care is reported on Medicare’s Hospital Compare Web site for all hospitals that have provided data. JCAHO’s core measures for AMI also track inpatient mortality, and hospitals have the option of reporting three other measures:

  • LDL cholesterol assessment, cholesterol testing within 24 hours of arrival at the hospital, and lipid-lowering therapy at discharge.
  • Percent of patients given ACE inhibitor for left ventricular systolic dysfunction.
  • Percent of patients given adult smoking cessation advice/counseling.
  • Percent of patients given aspirin at arrival.
  • Percent of patients given aspirin at discharge.
  • Percent of patients given beta-blocker at arrival.
  • Percent of patients given beta-blocker at discharge.
  • Percent of patients given PCI within 120 minutes of arrival.
  • Percent of patients given thrombolytic medication within 30 minutes of arrival.