Skip to main content

Patient Harm Data to Remain on Medicare's Hospital Compare Site

 |  By  
   May 16, 2013

An inaccurate media story about changes to Medicare's reporting on hospital-acquired conditions has sparked a flurry of criticism from patient safety advocates. Here's what's really going on.

Under its "Readmissions, Complications & Deaths" tab, Medicare's Hospital Compare website lists more than two dozen types of avoidable hospital-acquired conditions and the frequency with which they occur at most of the nation's 4,000 acute care hospitals.

One can see, for example, how often staff administered transfusions of mismatched blood, whether doctors lacerated organs during surgeries, the frequency with which lapsed hygiene protocols transmitted lethal infections, and how often surgeons forgot to remove objects like sponges or towels before closing incisions.

Patient advocates and Centers for Medicare & Medicaid Services officials believe such transparency is an important type of peer pressure that helps all hospitals do a better job to keep patients safe. Additionally, the increasing number of third-party hospital quality rating systems use many of these data points to configure safety scores or hospital rankings they distribute for free to the public.

So when a Bloomberg report two weeks ago said administration officials were planning to remove eight key hospital-acquired condition measures from Hospital Compare starting with the update scheduled for July, there was strong opposition from consumer groups.

Leapfrog Group's President and CEO Leah Binder and others expressed concern that the government was relinquishing its commitment to transparency. Leapfrog, a non-profit organization, uses some of those same measures to compile its twice-annual composite score in which thousands of hospitals get an A, B, C, D or F in patient safety.

The problem is, the Bloomberg story was erroneous, CMS officials told me. I don't want to pick on Bloomberg; Lord knows they've got enough problems on their hands this week. And I understand why the error occurred–there's enough blame to go around to numerous entities.

The culprit is complexity
The biggest culprit is the complexity of the rule-making process, and the difficulty in making sure that measures do what they're supposed to do, and aren't just artifacts of coding errors when hospitals submit claims.

Patrick Conway, MD, CMS chief medical director, explained some of the details to me in a lengthy "on background" phone interview last week. But he stressed that none of the data elements in question are being removed. How they are reported may change a bit, but the public will still be able to see rates of patient harm for almost every measure and then some in future reimbursement years.

What is changing, he explains, is that some of these measures will no longer be part of the Inpatient Quality Reporting program, for which hospitals have received a 2% market basket pay increase in exchange for reporting the data. CMS will still get the data, but the hospitals won't get paid for submitting it.

"Despite their removal from IQR, we will continue to calculate these claims-based measures and supply data via," he wrote in an e-mail after the interview. "Moreover, third parties also have the option of requesting claims data through RESDAC, and calculating the claims based HAC measures themselves." 

"Just because a measure is not reported to the IQR program, it doesn't mean that it will be taken off Hospital Compare," he says. These eight measures, of the 11 hospital-acquired conditions for which Medicare denies hospitals reimbursement under provisions of the 2005 Deficit Reduction Act (DRA), and which are now posted on Hospital Compare, "will continue to be displayed and available in the downloadable database after July, 2013."

3 measures go away
Three measures, however, are being removed from the IQR program, and will be removed from Hospital Compare in July. They will no longer be available for download. But they aren't among the DRA HACs referred to in the Bloomberg story. They are:

  • Deaths for certain conditions
  • Death after surgery to repair weakness in the abdominal aorta
  • Deaths after admission for a broken hip

A fourth type of hospital-acquired condition, breathing failure after surgery, is no longer in the IQR and will no longer be displayed on Hospital Compare after July, 2013. But results will be updated in a downloadable database in either October or December. Currently, the information is unavailable on Hospital Compare, CMS says, due to a software problem.

Measuring quality is a tricky business.

Advocates and originators of quality standards must first determine if a type of harm is avoidable and if it occurs often enough to represent a facility's quality. Then they must determine that what constitutes such an event is clear enough for a definition to be universally understood and applied. The measure can not be a reflection of a coding error or an artifact used to submit a claim.

Tricky business.

The NQF's MAP report
What's behind the changes is a 491-page report from the Measures Application Partnership. Published in February, the report documents an effort by the National Quality Forum in contract with CMS to vet hundreds of measures that purport to rate hospital, physician, outpatient, and nursing home care quality.

What CMS includes in Hospital Compare, and what is downloadable from the federal database, is based on the MAP recommendations, CMS officials say.

But all of that complexity was lost on the public when the erroneous media report was published. Patient safety advocates went to various websites to blast CMS for being in the pocket of hospital industry lobbies.

"Please go to the Govt. and tell them WHY these changes are NOT good for patients and families, wrote Jeni D.-Safer Healthcare on the ProPublica Patient Harm Community website. This might WELL be the most important thing that YOU ever do in the name of patient safety in YOUR lifetime!!"

On the Bloomberg story's comments section, one person identified as "gatesballs" wrote "Of course! The lobbyists have won again after they gave Obama a bundle of dough." And someone with the handle Bobo TheGorilla wrote, "What a surprise. Our government doing something that is not in the public's best interest. Everything in America is Rigged."

These reactions illustrate that this brave new world of payment-for-performance is a hot-button issue. It's a fluid process though, an experiment in how to turn the healthcare industry upside down.

Hospitals and the regulators who pay them aren't entirely sure whether they're incentivizing the right things, or just the easiest things. The data may not be all there just yet.

It may turn out that some of these reporting measures and the payment policies that follow may push hospitals to teach to the test, focusing on avoiding some types of more serious harm but not others.

Time will tell if these measures are imperfect, or have to be tweaked or dropped and replaced with others more relevant. But I believe, and I think the officials at CMS believe as well, that we have to start somewhere.


Get the latest on healthcare leadership in your inbox.