Almost half of critical access hospitals' star ratings went down this year, and about two-thirds of all U.S. CAH's didn't even get a star, raising questions about whether CMS' new methodology is a more accurate picture of quality, and if the system is even working for these small hospitals.
Take Fulton County Health Center in Ohio, which went from four stars last year to one star this year, and is part of a larger trend. Forty-seven percent of CAHs went down at least one star this year, compared to 22% of acute care hospitals.
At FCHC their rate of hip and knee replacement complications was below average in the safety and care grouping. All of the other measures reported were better or at average. FCHC CEO Patti Finn said her quality team is reviewing their data on hip and knee replacements, but they have not found out what exactly caused the problem.
"Nothing has jumped out that we had a pattern, and I'm wondering if because our numbers [of hip and knee replacements] are kind of low, if just one or two caused our percentage to be higher," Finn said. "I'll probably make sure that we're looking at that data more closely moving forward to make sure that we're not surprised."
Volume is a perennial problem for critical access hospitals. And even though the new methodology, which now scores hospitals based on how many measures they're able to report based on patient data, is in theory comparing similarly-sized hospitals, it seems to be missing the mark.
While patients do show up at the Fulton County Health Center emergency room with a heart attack, it's the hospital's job to stabilize that person and transfer them to a higher-acuity facility. One of the measures, the death rate for coronary artery bypass graft surgery patients under the mortality grouping, usually isn't something FCHC could get a score on. The mortality grouping includes seven measures for things like mortality rates for pneumonia and stroke patients. The more care that CAHs transfer out – as they should be doing – the more the procedures they are able to do count. Bad outcomes can carry much greater weight.
"That care that was provided in the rural hospital, since they weren't admitted at the rural hospital, doesn't get counted in the Quality Reporting Program," and instead gets counted toward the larger hospital with more specialized services, said Brock Slabach, senior vice president for member services at the National Rural Health Association. "We need a system that recognizes the type of work that rural hospitals do, and has a measurement system that responds accordingly."
CMS, based on a lot of provider input and reporting from Modern Healthcare, changed the methodology for the first time in decades and eliminated 12 quality measures to 48. Hospitals are now also stratified by the number of measure groups they reported, including mortality, readmissions, safety of care, patient experience and timely and effective care. For larger hospitals that are able to track more data and also have high service volumes, they likely fall into the five-measure grouping. Meanwhile, hospitals with lower volumes and less data are more likely to fall into the three-measure grouping.
CAH involvement is also voluntary – they can pull their rating up until the day before the Care Compare refresh goes live.
"I know we have one hospital for sure that dropped from a 4-star to a 1-star, and just decided it was not in their best interest to have consumers not understanding the backend of the story about the measures, so they decided to withhold from a star rating this year," said Nadine Allen, chief quality officer at the Wisconsin Hospital Association.
In Wisconsin, 32 hospitals didn't get a star rating, up from 14 last year. Experts suggested previously that the new methodology would actually push more critical access hospitals into having enough data to report. But nationwide there were189 CAHs this year that didn't get a star, that had a star last year. That compares to 61 acute care hospitals.
Lexington Regional Health Center in Nebraska didn't get a rating at all this year because it didn't have enough data in either mortality or safety of care groupings. These areas are largely derived from Medicare claims data. LRHC's average acute and swing bed daily census ranges from four to 10 patients a day
"We know volume-based metrics are not small-volume friendly," said CEO Leslie Marsh.
Last year, the CAH received a 4-star rating. Marsh said there's got to be a better way. Perhaps CMS could measure the appropriateness of antibiotic selection, or how often newly-diagnosed diabetics are readmitted to the hospital with out-of-control hyperglycemia.
The new system, however, is just a step in a fairly new field of healthcare quality measurement. And some maintain this is a more accurate way to measure performance.
"If you're rating, you really should have a like-to-like comparison and get away from blackbox methods," said Dr. Bala Hota, vice president and chief analytics officer for Rush University Medical Center in Chicago, in reference to the CMS' previous latent variable model that was largely a mystery. Hota and others at Rush also helped develop the new methodology.
But critical access hospitals, which have 25 or less beds, are quite different from others. They're located at least 30 miles or greater from the closest neighboring hospital and so therefore often are in rural areas.
"It's this question of, is it even possible to have an apples-to-apples comparison between all the different types of hospitals that we have in the United States?" Hota said. "It's a legitimate question."
He suggests that quality could move to regional-based rankings, or be based on specialties. Others say that since every CAH hospital has an emergency department, more measures could be incorporated in that area, or look at how patients present to a higher-acuity hospital after transfer from a CAH. Slabach said there is some movement nationally to have a more specific system for smaller and rural hospitals. The National Quality Forum is currently in the process of developing them.
"It's not that we don't want to be part of the quality movement, we want it to be done fairly," Slabach said.
Finn at the Fulton County Health Center said the quality team will continue to look for the complications within hip and knee replacements. Perhaps they provide better discharge planning, or something in the surgery needs to change. Their performance improvement committee meets monthly and will discuss strategy and look closely at complications moving forward.
"To be honest with you, I saw the preview data, and didn't see any issues," Finn said. "Because our numbers are so small, we may not think that it was significant to be considered a complication and have to do something about it."
