Special report: Measuring the clinical performance of hospices
Hospice-care providers to begin reporting on two quality measures, an incremental approach that the industry supports
For the more than 5,000 hospices across the country, success is often measured by hard-to-pindown metrics, such as whether a patient feels comfortable or is able to help to plan his or her funeral, or whether family members feel included in the end-of-life care process.
Quality indicators that are commonly used in other settings, such as measures of improved functional status, are not usually applicable, and even if they are, they’re often not part of a hospice’s overarching goal.
“The difference in hospice is that we’re not focused on a medical diagnosis, as in, ‘We treat with these five things and we expect a resolution,’ ” says Pat O’maley-lanphear, a former hospice quality director who now owns an Arlington, Wash.-based consulting company that works with hospice providers. “We’re trying to make sure patients’ symptoms are managed and they are as informed as possible. We’re trying to support their decisions.”
For instance, some patients with swallowing disorders may not want to have a feeding tube. In those cases, care might include hydration and comfort measures, O’maley-Lanphear says.
Other patients’ hospice goals may include reconnecting with loved ones or staying alive until a particular event, says Patty Luptak, director of professional services at Sivitz Jewish Hospice & Palliative Care, Pittsburgh.
“We treat the whole family,” Luptak says. “We control symptoms, provide spiritual support, help with community resources—it’s a whole different set of dynamics in hospice.”
Based on perception
Instruments that use numeric ratings, such as pain scales, can be problematic, too, because so much of hospice quality is based on patient perception, says Carol Spence, vice president for research and quality at the National Hospice and Palliative Care Organization, a membership organization based in Alexandria, Va.
“You can’t have a pain measure that says all patients should be at a three because one person’s three is another person’s six,” Spence, says. “Hospice has to work for each patient.”
Given the individualized nature of hospice care, many providers say it can be difficult to use standardized quality measures to gauge performance and facilitate side-byside comparison.
In just a few months, though, that’s exactly what the CMS will be requiring hospices to do.
“Yes, hospice is customized to individual patients and families, but how well they provide care can be standardized,” says Robin Dowell, a nurse consultant in the CMS Office of Clinical Standards and Quality. “There are expectations of quality that we can identify.”
Mandated by the healthcare reform law, the CMS Hospice Quality Reporting Program will require hospices to gather data and report on specific quality measures in order to avoid a 2% cut to their annual payment update.
For fiscal 2013, hospices have to report on two measures—a pain-management indicator and a structural measure that evaluates whether hospices have a quality assessment and performance improvement program in place in the fourth quarter of 2012—to avoid a cut in 2014. The program requires hospices to only report, and it does not penalize them based on their score or the content of their submissions.
The data-collection period for the first phase of the program will run from Oct. 1 until Dec. 31. The submission deadline for the structural measure is Jan. 31, 2013, while the
due date for submitting the pain-management measure data is April 1.
Hospice groups mostly praised the CMS for taking what they called a careful, incremental approach to introducing quality reporting to a field that has little, if any, experience with it.
“We believe they have been very cautious in trying not to overload hospices with too much at one time while also getting them ready for broader-scale reporting,” says Theresa Forster, executive vice president of hospice programs for the National Association for Home Care & Hospice, a Washington-based trade association. “We think it’s been done in a judicious way.”
O’Maley-Lanphear says the program will give hospices an opportunity to demonstrate the value of their services while also providing a way to identify and share best practices. “When you look at hospices that have each developed their own internal quality measures, you’re comparing apples to oranges,” she says. “By coming up with some clear guidelines, we can start to compare apples to apples.”
Assessing pain management
The two measures chosen for the first year of the hospice quality program were finalized in the Hospice Wage Index for 2012, published Aug. 4, 2011.
The first, developed by the National Hospice and Palliative Care Organization, measures how many patients who were in pain at their initial assessment after admission to hospice had their pain brought to a comfortable level within 48 hours.
That measure, endorsed by the National Quality Forum in August 2009 and reapproved on Feb. 14, is a source of concern for some hospice providers. Lin Simon, director of quality for Gilchrist Hospice Care, Hunt Valley, Md., contends that the pain-management measure applies only to a very small percentage of hospices’ patient populations.
More than a decade ago, when the measure was developed, hospices served a much larger percentage of cancer patients, Simon says.
“Back then, many of them arrived in hospice in terrible pain, coming from settings that didn’t treat pain well,” Simon says. “Things have really changed and now we have many more patients with dementia, (chronic obstructive pulmonary disease) and heart failure. And people are coming to us from settings that are much better at addressing patients’ pain.”
Based on data from Gilchrist’s internal quality program, less than 10% of the hospice’s patients are admitted in pain, she adds. “This represents only a tiny, tiny subset of who we serve.”
Another issue with the measure, Simon says, is that it requires patients to state verbally after 48 hours that their pain has been brought to a comfortable level. Some patients who were in the original denominator—those who were in pain at the initial assessment—may die during the 48-hour time period. Others may be unconscious and unable to respond, she says.
“What effect will those patients who can’t respond have on the numerator?” Simon says. “We don’t know.”
Virginia White, director of quality and per-
“We treat the whole family. We control symptoms, provide spiritual support, help with community resources—it’s a whole different set of dynamics in hospice.”
—Patty Luptak Sivitz Jewish Hospice & Palliative Care, Pittsburgh
formance improvement for Home & Hospice Care of Rhode Island, Providence, shares some of Simon’s concerns about patients who cannot respond at the 48-hour mark. “To the public who have not read the specifics of this measure, they may see scores that seem really low,” she says.
The CMS, however, says it will gather data specifically on patients who are unable to selfreport their pain level at the 48-hour followup mark.
“The good thing is that CMS is saying to us, ‘Don’t do the math,’ ” Simon says. “They just want us to give them all of the numbers. Hopefully, they’ll take out the people who, for whatever reason, could not respond.”
The second measure for 2013’s quality reporting program was conceived as a way for the CMS to collect data about how hospices are currently measuring quality in their orga- nizations, says Dowell of the CMS.
“The ACA requires that we use endorsed measures in reporting programs unless there are none available and at the time, there were only two,” Dowell says, referring to two NQF measures, the pain management indicator and another that measures patient experience.
In February, the NQF endorsed 14 measures for palliative and end-of-life care, addressing topics such as pain, shortness of breath and care-preference documentation, which the CMS has indicated it will likely choose among for future versions of the hospice reporting program.
In order to fulfill the requirements of the structural measure, hospices will simply have to confirm that they have a quality assessment and performance improvement, or QAPI, program that includes at least three quality indicators related to patient care. Then they will indicate what areas their measures cover by checking boxes next to various quality domains and sub-domains.
“The structural measure gathers information on what hospices are already collecting and what they think are important indicators,” Dowell says.
To help hospices prepare for quality reporting, the CMS arranged a three-month voluntary reporting period when hospices could submit data about their QAPIs. During the voluntary reporting period, which ran from Oct. 1, 2011, until Dec. 31, the CMS received responses from 911 hospices, according to a report from RTI International, a CMS contractor. Of those respondents, more than 95% said they had a QAPI program in place that
included more than three indicators.
The results of that report revealed just how widely varied hospices’ internal quality programs are and how complex efforts to standardize them may prove to be.
In total, participating hospices submitted 6,712 quality indicators during voluntary reporting. Pain assessment and management accounted for more than 1,200 of those measures, while another 812 addressed com- munication with patients and their families. Other common measurement areas included anxiety, shortness of breath, care coordination, nausea, fall prevention and medication management.
“This is an industry that has for years collected information that is discrete to the hospice and to the region, and that data is not necessarily transferrable,” says O’Maley-Lanphear, who emphasized the need for com- mon metrics. “CMS’ first bite—the voluntary reporting cycle—produced thousands of measures.”
Not all measures are created equal, of course. According to RTI International’s report, some hospices are using high-quality indicators that measured “important and actionable areas of quality related to patient care, with appropriate specifications and precise descriptions of the numerator and the denominator.”
On the other end of the spectrum, many hospices’ measures lacked precision, according to the report. Some even submitted indicators that were more like quality-related goals—such as “All patients will be free of pain” —which lacked numerators and denominators altogether.
‘Appropriate and positive’
Ron Fried, senior vice president of public affairs for Vitas, a for-profit hospice provider based in Miami, says hospices are well aware of the need for across-the-board guidelines and measures. Vitas is currently participating in a quality reporting initiative involving more than 40 Florida hospice providers. Using selected measures from the Family Evaluation of Hospice Care satisfaction survey tool, a commonly used performance evaluation tool maintained by the National Hospice and Palliative Care Organization, the hospices are comparing quality data with one another.
“We welcome the CMS program and we think these first two measures are appropriate and positive,” Fried says. “But we do want to make sure the process is as incremental as possible and any implementation challenges are worked out before new measures are introduced. It seems like CMS is sensitive to that.”
For now, the CMS is taking it slowly. For 2014’s program, which will affect 2015’s payment update, the agency is limiting the program to the same two indicators, according to proposed changes published in the Federal Register on July 13.
In future years, the CMS says it would consider expanding the program “to include some additional measures endorsed by NQF.”
“The hospice community has given feedback and told us that they are willing and interested in providing quality data for us,” Dowell says. “I have no reason to believe they are not ready to start collecting data in October.”