Hospice providers earn high marks on CMS satisfaction surveys, but do they reflect reality?
Interim HealthCare has a hard time getting the family members of its hospice patients to complete the CMS’ experience survey. The Lubbock, Texas-based provider works with a third-party vendor to distribute the surveys by mail to families after their loved ones have passed away. But Interim considers itself lucky when just 25% of family members take the time to complete and mail back the survey.
“If we send out 30 surveys and we get five surveys back, there is no telling what the responses will be. Are the five great? Are three bad? It’s a crapshoot,” said Jennifer Bullard, executive vice president for Interim’s Texas and New Mexico division, which operates 14 hospice sites.
Interim is one of more than 4,000 hospice providers across the country required by the CMS since 2015 to send out the Consumer Assessment of Healthcare Providers and Systems Hospice Sur- vey to family members or risk a 2% Medicare payment cut.
The requirement is part of the larger Hospice Quality Reporting Program mandated under the Affordable Care Act in an effort to move the sector to value-based payment. Along with the CAHPS hospice survey, the CMS program requires hospices to publicly report data on quality measures.
In late February, the CMS for the first time posted findings on Hospice Compare from the CAHPS hospice surveys, which were gathered from April 2015 to March 2017.
The results at first glance are a success story. A Modern Healthcare analysis shows that for each of the eight measures the CMS uses in the survey, on average at least 75% of respondents indicated they received a positive experience. The CMS dataset displays “top box” scores to show the CAHPS results, which is the percentage of respondents who gave the most positive responses for each measure.
But experts are questioning the value of the surveys for consumers considering the overwhelmingly positive results and modest response rates. Roughly 33% of family members complete the CAHPS hospice surveys on average.
“It’s unknown how more than two-thirds of family members perceived their care,” said Dr. Kathleen Unroe, associate professor of geriatric medicine at Indiana University. “If everyone is doing well, is that helpful information? You want it to be useful (for patients and family members), but it doesn’t feel that useful.”
The results come after the CMS had to implement a correction to the Hospice Compare site last year because of inaccurate reporting regarding hospice locations and for-profit status.
The new dataset raised more questions about the value of the results. According to the CMS, the 2,795 hospices considered in the analysis were on average strong performers in the categories that involved giving patients and family members emotional and religious support, with about
90% of respondents indicating that they got the right kind of support, and 91% of respondents saying that the hospice team always treated the patient with respect.
Most family members said they would recommend the hospice provider, with nearly
85% of respondents saying they had a positive experience and 80% also indicating providers communicated well.
Hospices did experience comparatively lower scores—although still strong—for categories about helping patients with symptoms and training families to care for loved ones, with
75% of respondents saying they were satisfied.
The Hospice Quality Reporting Program doesn’t yet ding providers if they perform below average compared with their peers, though it likely will soon. If hospices don’t report quality measures in this fiscal year, which ends Sept. 30, they will see a 2% reduction in reimbursement, according to the most recent CMS reimbursement final rule.
In the dataset, there are 1,633 hospices that were not included in the CMS analysis, but the reasons for their exclusion were not provided. The CMS doesn’t yet know how many hospices will receive a penalty.
This period “gives hospices a chance to get familiar with the quality measures and improve on them prior to being held financially accountable for performance,” said Rebecca Anhang Price, a senior policy researcher at the RAND Corp.
Providers argue that the survey results are so strong because the CMS has only just started to gather them. More time is probably needed to analyze the data for trends and variation in performance, said Judi Lund Person, vice president of regulatory and compliance at the National Hospice and Palliative Care Organization.
“I think as we get a little bit further into the CAHPS surveys, we will start to see which questions are popping up” that have more variation, she said. “It might not always be rosy. I think it’s going to take some time.”
But the surveys are only useful indicators of patient experience if the populations that complete them are large enough and diverse, which evidence indicates rarely occurs.
A recent study by Unroe on the Family Evaluation of Hospice Care, a survey hospice providers used voluntarily before the CAHPS survey, shows that whites are more likely to return the survey than blacks, as well as those with family members who were on hospice for longer than six months. The response rate for this survey was also low in the study, 27%. Unroe said it’s important for the CMS to be transparent about the populations represented and those underrepresented in the surveys. “How are they going to compensate (for the underrepresented patients)? Is there going to be additional outreach?” she added.
A CMS spokeswoman said the agency continues to evaluate the findings from the CAHPS hospice surveys “and welcomes suggestions that can make the survey results more meaningful for our beneficiaries.”
The CMS also announced last week that it’s looking to add new quality measures to the Hospice Quality Reporting Program “that address additional identified gaps in hospice quality measurement.” The agency is taking suggestions from stakeholders until April 25.
Unroe said the CMS should not only monitor the surveys for variation in the results but also
provide as much information as possible about the survey respondents and their loved ones. “People are in hospice for different reasons and diagnoses. … It’s important to have as much information as possible to guide patients and families to the right decisions for them,” she said.
However, Price at RAND said she thinks consumers likely see variation between hospices when they use Hospice Compare because they are only comparing one or two providers instead of a large dataset. “CAHPS surveys have been shown to reliably distinguish the performance between hospices,” she said.
Hindering the response rate
Hospice providers said if the surveys could be done by family members electronically, it might boost response rates. Right now, the third-party vendors send surveys by mail, over the phone or both.
“If we could move up in technology and email the survey, or text it, I think we’d get a much higher return rate—I throw away a lot of mail,” said Dr. Michael Roffers, chief medical officer for Interim’s operations in Texas and New Mexico.
The CMS said it’s currently evaluating alternative modes of CAHPS survey administration, including email.
The CMS also doesn’t allow the vendors to put the logo of the hospice provider on the mailed package, so family members can easily throw away the survey by mistake.
The length of the survey, 47 questions, also could be a contributor to the low response rates, Bullard said.
She added that the questions can easily be confusing for family members if they don’t read them carefully. The survey has similar questions phrased in different ways as a tactic to elicit consistent responses. Bullard wonders if sometimes people just misunderstand how questions are worded and answer incorrectly.
But hospice providers can’t follow up with family members to further discuss their answers. The CMS requires vendors to de-identify the respondents when they send the results back to the hospice providers.
“There are a very specific set of rules by the CMS,” said Tony Kudner, vice president of Seasons Hospice and Palliative Care, based in Rosemont, Ill. “You have to be very compliant and careful.”
Despite the limitations, providers say the survey results can still help them improve care.
For instance, Seasons Hospice used the surveys to improve how it treats patients’ pain after responses showed some family members weren’t satisfied.
After reviewing its scores, Seasons in late 2016 implemented SNAP, which stands for shortness of breath, nausea, anxiety and pain. Each hospice team member is now trained to check for each of these things through conversations with the patient or family member during every visit.
Seasons also ramped up its use of quality improvement specialists shortly after the Hospice Quality Reporting Program went into place.
“We have a director and vice president of quality, and then we have folks at every one of our sites who are responsible for monitoring and interpreting the results so we can put some specific plans in place—continuing to do the good stuff or working on the stuff where folks are saying there is room for improvement,” Kudner said.
The 2% reduction in Medicare payment is a “huge motivation” for hospices to report and review their surveys, Person at NHPCO said, especially because hospices only received a 1% bump in Medicare payments in 2018, down from the 2.1% increase they received in 2017.
At Interim, there’s a growing focus on following up with the family members during treatment to ensure all their needs are being met. Bullard said the new information on Hospice Compare has been especially helpful.
“I do love it because you can see how you’re doing next to your competitors,” Bullard said. “We can say all day long we give great care, but if our numbers don’t add up, it doesn’t
● mean much.”