A Call for Standards, Auditing of Hospital Quality Data

This article originally appeared on HealthLeaders Media on August 24, 2017

By Tinker Ready

Quality data looks good in ads and on hospital and health system websites. A little too good at times.

A hospital website would seem like an obvious place to deliver healthcare information, especially information about the quality of services. But, for the uninitiated, the marketing piece of the website may be difficult to distinguish from the clinical piece.

Not that hospitals shouldn’t promote themselves. But as more data about infection rates and readmissions becomes available, health systems need to ensure that the information they share actually reflects the quality they offer.

That’s the call from Peter Pronovost, MD, and his team at the Johns Hospital Medicine.

They make their case in a JAMA Viewpoint essay published online earlier this month. This is not Pronovost’s first call for standards and auditing. But this latest call is directed specifically at hospitals and health systems. He thinks they need to take the lead.

“Before healthcare provider organizations ask others to meet standards, we ought to make sure our own house is in order,” Pronovost said in a telephone interview with HealthLeaders.

In some cases, hospitals may be misusing quality data in ways that are at best confusing, and at worst deceptive, he says.

“There is some data that hospitals might be incentivized to make themselves look more positive when they report publicly,” he said. “It wouldn’t surprise anyone that that happens.”

Dubious Claims

Pronovost found one hospital website urging patients to “Come to us, we have no infections.” Without any information about which type of infections, or how and how long infection rates were measured, the claim is essentially meaningless.

But the JAMA piece doesn’t dwell on misleading hospitals ads. Pronovost says his goal is to offer a “positive step forward to begin to mature the field. ”

The opportunities for getting it wrong grow with each data point and every step toward greater transparency.

Hospitals and health systems now have volumes of data they can use (or not use or misuse) to promote the quality of their care, such as mortality rates for specific procedures.

Consumers are now accustomed to comparison shopping and there are plenty of places for them to do that, from Medicare Compare to state-run websites, to Facebook. But, not all are patient-friendly or scientifically accurate.

Emotional Appeals

Yael Schenker, MD, a researcher at the University of Pittsburgh, has explored the issue. She looked at 400 ads from more than 100 cancer centers and found that they promote cancer therapy with “emotional appeals that evoke hope and fear while rarely providing information about risks, benefits, costs, or insurance availability.”

The public trusts information from hospitals, she wrote in response to an email query.

“Inaccurate or imbalanced information on hospital websites should therefore be an area of significant concern,” Schenker wrote.

“The proposal [for standards] represents an important opportunity to clarify the line between quality reporting and advertising, and to provide value to the public.”

One big question is who will hold hospitals to the standards? The Hopkins team proposes an outside entity that would “make transparent the extent to which reporting by each organization adheres to these standards.”

If that sounds vague, it is meant to, says Pronovost. “We wanted to use this piece to initiate a conversation rather than offer a conclusion.”

A full third of the JAMA piece is devoted to the challenges of the approach. For example, the piece acknowledges “numerous calls for less measurement in medicine rather than more.”

The authors argue the standards approach “could help to complement this effort by seeking to improve the accuracy of reported measures.” Again, just how more auditing would mean less measurement is left open for discussion.

Both U.S. News & World Report and The Leapfrog Group invite hospitals to advertise rankings issued by those organizations.

But the aggregation of data into rankings or scores is not the issue. In fact, Leah Binder of the LeapFrog Group and Ben Harper of U.S. News are listed as contributors to the JAMA piece.

It’s the measures themselves, the evidence supporting them, and the way they are presented that are the issue.

For an example, Pronovost returns to the ad from the hospital claiming a zero infection rate.

“On the surface, it is sounds really good. No infections. The public probably trusts that as positive. [But] scientifically, it is kind of a meaningless statement.”

No infections for a week is different than a measure of multiples type of infections over five years.

In the past, and in the JAMA piece, Pronovost has argued that quality reports should be audited in the same way that financial statements are audited.

“Your company’s financial statements are audited in a systematic way, just as IBM’s and Johns Hopkins’ are,” he said. “When people file a financial statement, there is some confidences that what was put in there is accurate. And that doesn’t exist right now in health care quality.”