Sarasota Health News

Hospitals say flawed formula causes poor mortality scores

David Gulliver - posted 7:30 am Monday, July 27

Two local hospitals scored poorly in the latest federal report on mortality rates, but both challenge the reporting, saying the data does not take into account problems specific to their patients.

Bradenton’s Manatee Memorial Hospital scored significantly worse than the national average for heart attack, heart failure and pneumonia mortality, and Venice Regional Medical Center scored worse than average for pneumonia.

The results place them in a small, bottom tier of hospitals nationwide, because as the report lumps most hospitals in a broad “average” category. The report lists only 2 percent as significantly below average for heart attack, 4 percent for heart failure and 6 percent for pneumonia.

Two local hospitals posted better-than-average mortality scores: Sarasota Memorial Hospital, for heart failure, and Charlotte Regional Medical Center for heart attack. And Sarasota Memorial and Venice Regional did well in a different quality measure, readmission rate, as Sarasota Health News reported previously.

The findings are in Hospital Compare, an online research tool produced by Medicare’s parent agency. It uses a mix of Medicare patients’ case information and statistical models to calculate the percentage of patients who died within 30 days of being discharged, and compares that to national averages.

(For a chart of Suncoast hospitals' mortality rates, click here.)

The data is “risk-adjusted,” meaning it attempts to compensate for the severity of the cases at each hospital, so those treating the most complex cases are not unfairly penalized.

But officials at both Manatee Memorial and Venice Regional say that the adjustments are inadequate for hospitals that deal with a high proportion of aged and often poor patients who opt for hospice care or have a “do not resuscitate” order.

At Manatee Memorial, more than 60 percent of patients in each diagnosis category signed such an order within 48 hours of admission, said Moody Chisholm, its chief executive officer. That made it more likely that those patients would die within the 30-day window.

The hospital, owned by Pennsylvania’s Universal Health Services, also reviewed each patient chart to look at circumstances, and found five heart attack patients who died in the emergency room who were considered inpatients, perhaps improperly.

Venice Regional, owned by Naples-based Health Management Associates, has ranked below national average for both years Hospital Compare has reported pneumonia mortality. In those two years, 26 pneumonia patients died within 30 days of being discharged, but 16 were discharged into hospice care or to a nursing home with a DNR order, officials said.

“I will absolutely stand behind the impeccable care that we provide here,” said Melody Trimble, Venice’s chief executive officer.

The hospital’s own review, performed by a division of Thompson Reuters, found that its in-hospital mortality is 30 percent less than expected given the severity of the patients’ conditions. In other words, statistics suggested 19 more patients should have died than actually did die.

“That speaks to the kind of care we provide,” Trimble said. “We discharge to an appropriate level of care. And that is what we will continue to do.”

Officials at the Centers for Medicare and Medicaid Services, which administers Hospital Compare, said the system cannot -- and probably should not -- adjust for DNR orders.

“The mortality measures do not exclude patients who have DNRs. CMS does not currently have the ability to identify all patients who have a DNR or choose comfort care upon admission,” the agency said in a statement prepared for Sarasota Health News.

“The mere existence of a DNR order is an indication of patient preferences not to attempt resuscitation after respiratory or cardiac arrest and may not reflect the severity or non-severity of the clinical condition or prognosis. Many patients have existing DNR orders placed while they are relatively healthy. A DNR order is not synonymous with “do not treat.”

“However, there are several ways the measure accounts for patients who are at the end of their lives and opting to receive comfort care only,” the statement said.

The risk-adjustment process also looks for diagnoses like dementia, metastatic cancer, and other conditions, which indicate patients may be at the end of their lives. That means hospitals treating those patients would not be overly penalized when calculating the mortality rate

The system excludes patients who were in the Medicare Hospice program in the year before the hospital admission, or chose to enroll into hospice care on the first day of admission.

“CMS chose not to exclude patients who transition to comfort care after the first day of their hospital stay,” the statement said.

Patients may choose hospice care later, when treatment and diagnosis indicate care might be futile, the agency acknowledged. But it’s not a problem that affects some hospitals unfairly, the agency said: “This is a reality of medical practice that applies to all hospitals.”

Trimble disagrees, noting the median age in Venice is 69, and the median age of the hospital’s nine pneumonia deaths last year was 88. No other HMA hospital serves as aged a population, she said.

And Chisholm recalled a doctor telling him last week of the case of a 92-year-old woman who had a heart attack, and chose hospice care -- after 36 hours, far past the Hospital Compare cutoff. Few families make such a difficult choice in the first 24 hours, Hospital Compare's window, he said.

Excluding a handful of cases from the data could be significant. Hospital Compare calculates a percentage “range” for a hospital’s risk-adjusted mortality rate. If the entire range is above the national average, the hospital is considered worse than average.

A Sarasota Health News analysis found the low end of Manatee Memorial’s range missed the mark by one-tenth of one percent for both heart attack and pneumonia -- meaning a slight reduction in deaths would have moved it to “average” status for both measures. But its range was well above the benchmark score for heart failure.

But Chisholm said that being at the edge of those ranges still leaves room for improvement. “There are lots of things we’ve noted, in reviewing the data, where we need to do a better a job,” he said.

The hospital is hiring an advanced practice nurse to track patients after discharge and keep them on their recovery plans. “Historically, hospitals have not been expected to reach out and manage a patient’s life after discharge. But that’s what we’re going to have to do,” Chisholm said.

Venice Regional’s range was well above the pneumonia benchmark, but ruling out hospice cases would have moved it up to an average score.

Trimble has been working with her parent company to assess everything from care to coding, how it labels its patients in Medicare data. “I am peeling the onion back even further,” she said. “If we have to go all the way to CMS, I will, because I know the care here is exceptional.”





	Hospitals say flawed formula causes poor mortality scores David Gulliver - posted 7:30 am Monday, July 27 Two local hospitals scored poorly in the latest federal report on mortality rates, but both challenge the reporting, saying the data does not take into account problems specific to their patients. Bradenton’s Manatee Memorial Hospital scored significantly worse than the national average for heart attack, heart failure and pneumonia mortality, and Venice Regional Medical Center scored worse than average for pneumonia. The results place them in a small, bottom tier of hospitals nationwide, because as the report lumps most hospitals in a broad “average” category. The report lists only 2 percent as significantly below average for heart attack, 4 percent for heart failure and 6 percent for pneumonia. Two local hospitals posted better-than-average mortality scores: Sarasota Memorial Hospital, for heart failure, and Charlotte Regional Medical Center for heart attack. And Sarasota Memorial and Venice Regional did well in a different quality measure, readmission rate, as Sarasota Health News reported previously. The findings are in Hospital Compare, an online research tool produced by Medicare’s parent agency. It uses a mix of Medicare patients’ case information and statistical models to calculate the percentage of patients who died within 30 days of being discharged, and compares that to national averages. (For a chart of Suncoast hospitals' mortality rates, click here.) The data is “risk-adjusted,” meaning it attempts to compensate for the severity of the cases at each hospital, so those treating the most complex cases are not unfairly penalized. But officials at both Manatee Memorial and Venice Regional say that the adjustments are inadequate for hospitals that deal with a high proportion of aged and often poor patients who opt for hospice care or have a “do not resuscitate” order. At Manatee Memorial, more than 60 percent of patients in each diagnosis category signed such an order within 48 hours of admission, said Moody Chisholm, its chief executive officer. That made it more likely that those patients would die within the 30-day window. The hospital, owned by Pennsylvania’s Universal Health Services, also reviewed each patient chart to look at circumstances, and found five heart attack patients who died in the emergency room who were considered inpatients, perhaps improperly. Venice Regional, owned by Naples-based Health Management Associates, has ranked below national average for both years Hospital Compare has reported pneumonia mortality. In those two years, 26 pneumonia patients died within 30 days of being discharged, but 16 were discharged into hospice care or to a nursing home with a DNR order, officials said. “I will absolutely stand behind the impeccable care that we provide here,” said Melody Trimble, Venice’s chief executive officer. The hospital’s own review, performed by a division of Thompson Reuters, found that its in-hospital mortality is 30 percent less than expected given the severity of the patients’ conditions. In other words, statistics suggested 19 more patients should have died than actually did die. “That speaks to the kind of care we provide,” Trimble said. “We discharge to an appropriate level of care. And that is what we will continue to do.” Officials at the Centers for Medicare and Medicaid Services, which administers Hospital Compare, said the system cannot -- and probably should not -- adjust for DNR orders. “The mortality measures do not exclude patients who have DNRs. CMS does not currently have the ability to identify all patients who have a DNR or choose comfort care upon admission,” the agency said in a statement prepared for Sarasota Health News. “The mere existence of a DNR order is an indication of patient preferences not to attempt resuscitation after respiratory or cardiac arrest and may not reflect the severity or non-severity of the clinical condition or prognosis. Many patients have existing DNR orders placed while they are relatively healthy. A DNR order is not synonymous with “do not treat.” “However, there are several ways the measure accounts for patients who are at the end of their lives and opting to receive comfort care only,” the statement said. The risk-adjustment process also looks for diagnoses like dementia, metastatic cancer, and other conditions, which indicate patients may be at the end of their lives. That means hospitals treating those patients would not be overly penalized when calculating the mortality rate The system excludes patients who were in the Medicare Hospice program in the year before the hospital admission, or chose to enroll into hospice care on the first day of admission. “CMS chose not to exclude patients who transition to comfort care after the first day of their hospital stay,” the statement said. Patients may choose hospice care later, when treatment and diagnosis indicate care might be futile, the agency acknowledged. But it’s not a problem that affects some hospitals unfairly, the agency said: “This is a reality of medical practice that applies to all hospitals.” Trimble disagrees, noting the median age in Venice is 69, and the median age of the hospital’s nine pneumonia deaths last year was 88. No other HMA hospital serves as aged a population, she said. And Chisholm recalled a doctor telling him last week of the case of a 92-year-old woman who had a heart attack, and chose hospice care -- after 36 hours, far past the Hospital Compare cutoff. Few families make such a difficult choice in the first 24 hours, Hospital Compare's window, he said. Excluding a handful of cases from the data could be significant. Hospital Compare calculates a percentage “range” for a hospital’s risk-adjusted mortality rate. If the entire range is above the national average, the hospital is considered worse than average. A Sarasota Health News analysis found the low end of Manatee Memorial’s range missed the mark by one-tenth of one percent for both heart attack and pneumonia -- meaning a slight reduction in deaths would have moved it to “average” status for both measures. But its range was well above the benchmark score for heart failure. But Chisholm said that being at the edge of those ranges still leaves room for improvement. “There are lots of things we’ve noted, in reviewing the data, where we need to do a better a job,” he said. The hospital is hiring an advanced practice nurse to track patients after discharge and keep them on their recovery plans. “Historically, hospitals have not been expected to reach out and manage a patient’s life after discharge. But that’s what we’re going to have to do,” Chisholm said. Venice Regional’s range was well above the pneumonia benchmark, but ruling out hospice cases would have moved it up to an average score. Trimble has been working with her parent company to assess everything from care to coding, how it labels its patients in Medicare data. “I am peeling the onion back even further,” she said. “If we have to go all the way to CMS, I will, because I know the care here is exceptional.”