VAP: Just how good a benchmark is it really?

VAP hard to measure in 'concrete, reproducible terms'

The headline pulls no punches. It reads: "Ventilator-associated pneumonia — the wrong quality measure for benchmarking."1 This stunning headline, which heralded a "Perspective" article in a recent edition of Annals of Internal Medicine, will no doubt grab the attention of quality improvement experts.

After all, no lesser organization than the Institute for Healthcare Improvement included VAP, or ventilator-associated pneumonia, as a key objective in its "100,000 Lives Campaign," and VAP is also being considered by the Centers for Medicare & Medicaid Services and The Joint Commission. Any number of hospitals have pursued and then trumpeted their successful reduction of VAP rates.

Nonetheless, the authors of the paper are confident in their stance, and cite numerous studies to support their assertions. They point out, first of all, just how difficult it is to accurately diagnose VAP, given the fact that it shares clinical signs with so many pulmonary complications seen in the ICU. They note, for example, evaluations of ventilated patients with fever, pulmonary infiltrates, purulent sputum (or some combination of those symptoms) that indicate only 30% to 40% of them actually had VAP.

No concrete measures?

The authors note that most quality indicators are readily measurable in concrete and reproducible terms — but VAP is not. "When you use the term 'concrete' that means you know exactly what you are measuring," explains Michael Klompas, MD, an infectious disease physician and instructor in the department of ambulatory care and prevention at Harvard Medical School and Harvard Pilgrim Health Care in Boston. "If you say a person has VAP, just how confident are you?" The evidence, he continues, suggests clinical diagnoses have poor correlation with data.

The term "reproducible," he explains, means that if different observers look at the same patient, they will come to the same conclusion — which, again, is not true with VAP. "Each one of the signs has very broad [meaning]," he asserts. "For example, fever could be infection, a blood clot in the brain, or a poor drug reaction. The presence of infiltrates could be dead lung, bruising of the lung, bleeding, collapsed lung, or fluid on the lung."

Examples of much more concrete hospital measures, he offers, would include such things as the use of beta-blockers after heart attack, mortality rates after heart surgery, and bloodstream infections in the ICU.

Why use VAP?

If VAP is such a poor measure, why are so many people using it? "They are driven by high-minded motives — to improve quality of care, and to get some sense of what normal practice should be," says Klompas. "At first glance, VAP seems like a good choice; it's a common problem in ICUs, and a major problem because it causes people to die."

For someone who is not deeply involved in medical diagnosis of these patients, he continues, VAP would appear to be a good measure. "It's more reliable in an outpatient setting, when you go to a doctor with fever, coughing green sputum, and so forth; but unfortunately that same logic does not translate into the ICU, where patients are much more complicated," Klompas explains.

"Measures selected for quality assessment or benchmarking ought to yield consistent results regardless of where or to whom they are applied," the authors write. "They should also closely reflect processes of care that hospitals can modify to improve their outcomes. The current definition of ventilator-associated pneumonia does not meet these standards."

New measures needed

The authors call for the development of new outcome measures to meet the standards that other measures currently meet — they are objectively measurable, they indicate serious complications, and they can "reliably reflect quality of care." These new measures, they argue, should reflect the totality of complications that can affect ventilated patients in addition to pneumonia. Such measures, they add, "should also be relatively straightforward for hospitals to collect."

Until then, suggest the authors, quality managers and others should track evidence-based processes of care measures, including "daily cessation of sedation and appropriate patient positioning."

Quality managers, says Klompas, should also recognize that "there is a real need to involve physicians and research experts in your course of measurement and surveillance. With some other quality measures that had been proposed we ran into exactly that same problem we have with VAP — they were difficult to measure or interpret, or there were unintended consequences."

For example, he notes, the rule developed for community-acquired pneumonia that required patients to receive an antibiotic within four hours "led to a big upswing in people without CAP getting antibiotics; anyone with the possibility of pneumonia got it," he notes. "What researchers found was an increase in antibiotic usage."

The bottom line, he says, is that "you need a thoughtful clinician to be able to say what is or is not a good possibility for a measure, and a researcher to say whether it is a good idea that can translate into reality."

Reference

  1. Klompas M, and Platt R. Ventilator-associated pneumonia — the wrong quality measure for benchmarking. Ann Intern Med 2007;147:803-805.

[For more information, contact:

Michael Klompas, MD, Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, Boston, MA. Phone: (617) 732-5775. E-mail: mklompas@partners.org.]