A 31/2-year effort by the Agency for Healthcare Research and Quality and the CMS to develop a patient survey for publicly comparing hospital satisfaction is on its way toward implementation with the Dec. 7 closure of a 30-day comment period.
The survey, the Hospital Consumer Assessment of Health Providers and Systems, or H-CAHPS, will be tested and put into use next year with public results made available by the end of 2007 (See related story, p. 7).
The program, which has strong support from the American Hospital Association and other industry participants, promises to give patients and administrators unprecedented ability to compare patients' satisfaction for those hospitals participating in the voluntary program.
Existing patient-satisfaction surveys, while widely used, are too tailored to individual hospitals to be used for comparison purposes, industry executives said.
And once available, the H-CAHPS data also may give hospital administrators useful information about their competition. But getting the survey down to its current 27-question format was a huge effort by the AHRQ and the CMS, according to Dennis O'Leary, president and chief executive officer of the Joint Commission on Accreditation of Healthcare Organizations.
O'Leary said officials at both federal agencies have conscientiously listened to critics and tried to incorporate their suggestions into the survey during the development effort, which is now a year and a half behind schedule.
"Neither patient-satisfaction survey work, nor public reporting of the data is new. The JCAHO requires as a condition of accreditation that hospitals do some form of patient-satisfaction surveying," O'Leary said. In addition, voluntary public-reporting efforts at the state, regional and healthcare system levels already exist, with extensive programs in California, western New York and at the Norton Healthcare system in Louisville, Ky.
Costs for hospitals to participate in the H-CAHPS program were estimated by consultant Abt Associates to range from $3,300 to $4,575 as a stand-alone survey or $978 as part of an existing, vendor-managed survey.
Norton's chief medical officer, Dan Varga, said hospital executives decided that publishing overall satisfaction scores on its Web site was "a fundamental piece of our stewardship. We really didn't think we could wait" for H-CAHPS. Norton publishes satisfaction for inpatient and outpatient care, emergency care and outpatient surgery and for 13 clinical measures.
Adds breadth and uniformity
What will be new about H-CAHPS is added breadth and uniformity, said Kenneth Kizer, outgoing president and CEO of the National Quality Forum. While survey firms such as Press Ganey Associates and NRC & Picker use their own approach, "They have proprietary information that the hospital has that doesn't give the consumer a standardized instrument," Kizer said.
"The really important part of it is the ability to compare and contrast information and have it in the public domain. H-CAHPS gives you an instrument where you can look at a hospital in Portland, Ore., or Portland, Maine, and have some confidence the comparisons are meaningful," Kizer said. Separately, mediation is growing as a way to handle patient complaints (See Commentary, p. 22), while some experts urge doctors to use a more congenial manner and apologies to help limit the damage of patient lawsuits (See story, p. 34).
Officials at hospitals with track records of public patient-satisfaction reporting interviewed for this story said there is little to fear in terms of patient backlash or loss of market share, at least in the short term, even from less-than-stellar results. A little preparation goes a long way in preparing the public and the media for the release of the new information.
"Obviously, if you do fairly well, it's always fun to trumpet it," said James Miser, a physician who is CEO and chief medical officer of 153-bed City of Hope National Medical Center in Duarte, Calif., which has participated in three rounds of public reporting of patient-satisfaction survey data since 2001, under a voluntary program run by the not-for-profit California Institute for Health Systems Performance. City of Hope scored well in most of its care categories, and in areas where it didn't, Miser said he "was able to use (the) data to say, `Listen, we need to improve this process.' "
"Theoretically, for me, it shouldn't matter having it published," he said. "It's enough for me to know there is something (wrong) we have to fix it." Still, Miser said, public reporting "does give the community a sense that they can demand a higher standard of care." But from a business standpoint, the more standardized data are published, the better hospitals will fare in their financial tussles with payers-which tend to hold close and selectively use the data they collect, according to Norton executives.
A lengthy battle
Despite expected changes to the survey form and procedures for administering it, the initial version recently passed two procedural milestones. On Nov. 30, the NQF, whose board in May endorsed the survey as a national quality standard, officially published its 76-page report containing both the survey form and its approved methodology. And on Dec. 7, a 30-day public comment period for the survey by the White House's Office of Management and Budget ended.
If the OMB has questions or issues with the survey, those will be directed to the CMS for some final negotiation, said Elizabeth Goldstein, director of the Division of Beneficiary Analysis, Beneficiary Education & Analysis Group at the CMS, who has overseen the H-CAHPS project since its inception. After those discussions, or if there are no problems raised by the OMB, the survey will be approved by that office and handed back to the CMS to prepare for what is expected to be about a year-long rollout, with the first public reporting expected in 2007.
Once hospitals are using the program, the CMS will ask most hospitals to get 300 completed per year, Goldstein said. "For the smaller hospitals, we're asking them to provide at least 100 surveys a year," though that's not an absolute requirement, she said.
The survey has been the object of a lot of negotiation, with companies selling customer satisfaction surveys arguing for a shorter H-CAHPS survey. After an initial AHRQ draft of 68 questions, it was eventually whittled down to 25, with the NQF adding back two more in late 2004.
Vendors urged the NQF to have a cost-benefit analysis performed on implementing the additional survey questions, which resulted in the Abt estimates above. "They've come at this every way they know how," O'Leary said.
Press Ganey Associates was one of the survey vendors pressing for a much shorter H-CAHPS survey than even the trimmed-down final version. Deirdre Mylod, director of research and development and public policy for Press Ganey, cited research that the CMS would have done just as well obtaining consumer-useful information with a "more parsimonious survey."
Mylod said Press Ganey's base survey has 49 standard questions, plus 10 optional customized questions. Only nine of the H-CAHPS survey questions overlap the standard Press Ganey questions, she said, making the potential size of the two surveys together rather cumbersome.
Something has to give if the survey is going to fit in the standard four-page booklet that Press Ganey plans to mail to hospital patients, she said. "It will depend on what the hospital wants," Mylod said. "If they want to preserve the breadth of data that they have (from previous surveys), they may have to add onto it," she said, which would result in an increased risk of lower response rates because of length. They could do parallel surveys, running both the hospital's basic vendor surveyor and H-CAHPS separately, but that would increase costs. "My guess is they will do a combination," she said.
Kevin Horne, director of governmental affairs for NRC & Picker, said he doesn't see many hospitals dropping their current vendor-supplied surveys for H-CAHPS because they will still need more department-specific data for internal quality improvement. "We think H-CAHPS is a pretty good tool, although it is a little short for what hospitals need to do," Horne said.
Kizer said it is still unknown how much effect the surveys will have on the decisions of individual consumers on where they will seek hospital care. "A lot of decisions will be made on where the doctor refers you or where your insurance company says you have to go. The bottom line is hospital boards and CEOs are going to be looking at the publicly reported data, too," Kizer said. "And they're going to turn to their CMOs and say, `Why are our docs not being viewed as polite as docs down the road?' I think that's one misperception that consumer information is only going to be used by consumers. Doctors and hospitals are going to be looking at it very closely. The net effect will be the improvement of the quality of care."
Voluntary, for now
Thus far, H-CAHPS is a voluntary program, with no current plans to link it to the Medicare payment incentives, according to a CMS spokeswoman. But that will surely change, predicts Marsha Nelson, president and CEO of the California Institute for Health Systems Performance, which runs a voluntary patient-satisfaction survey and public-reporting program in that state. "It's just a matter of time," Nelson said. "They need to get it rolled out and get the bugs worked out first."
Nelson recommended that national survey planners find a way to invest in subsidizing smaller hospitals to boost program participation. "They'll get better data than they've ever had on their performance and will be able to make changes to improve care."
Another suggestion, based on the California experience with public reporting, is to assist hospitals with low scores to get ready for a media storm. "We helped hospitals prepare with some specific messages and how to respond to some tough questions and I think it paid off. I thought it was very well handled by the media and now they (hospital leaders) realize the sky is not going to fall or the ground won't open up and swallow them."
What do you think?
Write us with your comments. Via e-mail, it?s [email protected]; by fax, 312-280-3183.