The result was that a lot of patients who weren’t infected got antibiotics, a lot of drugs were wasted, antibiotic resistance was exacerbated, and the cost of care was driven higher. That measure is not in the final VBP list.
“The point is, we are going to get it wrong every now and then,” Wachter says. “But if we wait for absolutely perfect measures, we won’t understand this area well enough, quickly enough. You have to be generous and say every now and then we get it wrong. What we’ll see over time is that the science will get better. But the question is, are we flexible enough to learn the lesson when that happens?”
“The question becomes,” says Rachel Werner, MD, of the Center for Health Equity Research and Promotion at the Philadelphia Veterans Affairs Medical Center, “What happens to the care outside of what’s being measured?’ By focusing resources on a specific thing, like giving PCI [percutaneous coronary intervention] within 90 minutes of arrival, if hospitals have constrained resources,
they may be unable to provide high-quality care in other areas that are not being measured.”
Another concern has to do with whether findings from clinical trials performed with volunteer patients in academic medical centers that determined correlations between process measures and outcome measures can be applied to general patient populations.
Werner’s study, published in the Journal of the American Medical Association in December 2006, found that hospital performance measures predict “small differences in hospital risk-adjusted mortality rates.” She urged researchers to develop “performance measures that are tightly linked to patient outcomes.”
Jan Mathews , assistant vice president of quality management at CaroMont Health, a 385-staffed-bed hospital in Gastonia, NC, a Premier demonstration project participant, said her hospital goes beyond the rule on discharge instructions. “We start giving them the first day of a patient’s hospital stay,” she says. Compliance with the measures in that project decreased CaroMont’s 30-day mortality by 50%, she says.
“With process measures, we always have to worry about these unintended consequences,” Mathews says. “It’s going to be very necessary that evidence-based care is continually evaluated for effectiveness, and if there is any indication there’s a problem, that care is changed. We have to be continually looking at evidence.”
Palomar Pomerado Health, a two-hospital system in northern San Diego County with 426 beds, participated in the Centers for Medicare & Medicaid Services pay-for-performance demonstration project that scored process measures. PPH’s chief quality officer, Opal Reinbold, says PPH knows that the high scores it got for process measure compliance correlates with better outcomes, such as reduced mortality and readmissions.
From that project, she says, PPH has revamped all of its discharge instructions with a clinical nurse who is also a literacy specialist, “to make sure they’re not in medical lingo, that they’re given in a third-grade to fifth-grade language level. The way we talk in healthcare doesn’t make any sense to the patient, and they won’t ask because they don’t want the caregiver to think they’re foolish.”
The system is doing more than the minimum regarding discharge instruction, she says. “We aren’t just checking off the boxes.”