A long-simmering debate over how to grade the quality of care provided by hospitals serving vastly different patient populations flared into the open this week, with the nation’s top hospital groups charging that new measures could cost already-strapped hospitals millions and should be scrapped.
In a letter delivered late Tuesday, the American Hospital Association, the Association of American Medical Colleges, the Federation of American Hospitals and America’s Essential Hospitals say the measures are “flawed” because they fail to take into account key social and demographic factors beyond the hospital’s control that can affect patients’ recoveries.
They registered the complaint in a formal appeal to the organization that endorsed the new measures, the National Quality Forum, the non-profit charged by congress with setting standards for hospital performance measurement.
“We will pull together all the groups and think about what the path forward is,” Dr. Helen Burstin, NQF’s chief science officer, said Friday.
Hospitals are especially worried about the new measures because they compare hospitals based on the costs of their care for three common and costly conditions: heart attack, heart failure and pneumonia. Introduced by the Center for Medicare and Medicaid Services, they are the first to be endorsed by NQF in a two-year examination of ways to include social and demographic information about patients.
These measures are not yet incorporated into Medicare pay-for-performance programs. But hospital officials say there is a real possibility that they could eventually cost hospitals millions in Medicare reimbursements. Medicare already imposes financial penalties on hospitals with higher readmission rates, says Dr. Janis Orlowski, Chief Health Care Officer of the Association of American Medical Colleges.
“Hospitals are being dinged millions of dollars,” Orlowski says. “Are hospitals that take care of more disadvantaged people going to be penalized?”
Hospitals have pushed to have these performance measures take into account factors beyond their control that might reflect badly on the quality of their care. For example, newly discharged patients who can’t afford to refill prescriptions or visit primary-care doctors are more likely to end up back in the hospital than people who can get the treatment they need. Hospitals pay a price for these lapses in community care through higher readmission rates and higher costs. Both can negatively affect hospitals’ performance scores.
Perhaps the biggest challenge is to reliably measure those factors, by identifying patients who are poor or who lack social support networks. Researchers have proposed many options, none of which have panned out yet, Burstin said at the group’s annual meeting.
The simplest option involves giving hospitals more credit if they treat more racial and ethnic minorities. Researchers have also suggested identifying hospitals with more patients who are eligible for both Medicare and Medicaid, an indication of genuine financial need. Neither approach has proven reliable, Burstin says.
Other options under consideration include use of nine-digit zip codes or census data to pinpoint people who live in disadvantaged communities, she says.
Safety-net hospitals and academic medical centers are most likely to be judged unfairly because of the people they serve, says Beth Feldpush, senior Vice President of policy and advocacy at America’s Essential Hospitals.
“They have a lot of uninsured patients, they have a lot of Medicaid patients and they provide a lot of services that are unique in their communities,” Feldpush says. “They have trauma centers, burn centers, neonatal intensive-care units and other high-intensity services that community hospitals don’t have.”
But data also show that some hospitals charge considerably more than others and get the same results. For instance, studies have found twofold variations in the cost of care for heart attacks.
“What CMS wants to do is reduce the amount of variation in the cost of care and insure there’s a connection between cost and quality,” says Kristen Barlow, a senior consultant at the health care consulting firm, the Advisory Board. “It’s another indication that CMS is doubling down on the connection between what they pay and outcomes.”
The groups also objected to an analysis that considered using race — “black” or “non-black” — as a crude proxy for social and other factors that affect care.
The NQF’s own expert advisory panel ruled that there is no evidence suggesting that race can serve as a proxy for outside factors that affect patients’ recoveries. Instead, they proposed using dual eligibility for Medicare and Medicaid, federal insurance programs for the elderly, disabled and poor, as a more straightforward indication that patients are impoverishment. Unfortunately, Burstin says, this method too proved unreliable.
Orlowski, of the Association of American Medical Colleges, says that being black doesn’t necessarily mean that a patient is poor or has an inadequate support network. Likewise, being “non-black” doesn’t mean a person is more affluent, especially when it’s a catch-all category including other racial and ethnic minorities.
“If you’re just splitting the world into black and non-black, you’re lumping together folks who are poor and rich in ways that don’t make any sense,” Orlowski says.
On the question of race, Burstin says, “We couldn’t agree more.”
The next and most important step, she says, will be a comprehensive effort to use this approach not only to measure hospital performance but to ensure that patients, no matter what their background or circumstances, get high-quality care.