Ever have the urge to shop around for a hospital? To see which is ranked best for, say, cardiac care?
According to a New York Times study of data on more than 3,000 hospitals assessed by the Hospital Quality Alliance and published last August, if you were shopping in Arkansas in 2004 for cardiac care, you were looking at the lowest-performing hospitals in the country.
But using more recent data, Arkansas hospitals beat the national average on two simple heart care performance measures (metrics, in hospital lingo): giving cardiac patients discharge instructions and advice on how to quit smoking.
The University of Arkansas for Medical Sciences, Baptist Health, St. Vincent Infirmary Medical Center and the Arkansas Heart Hospital are showing higher Quality Alliance scores than the national average on two other measures: giving heart attack patients an aspirin on arrival and at discharge.
In fact, Arkansas hospitals, on average, score better than the national average on nine of 12 cardiac measures and five of six others (addressing pneumonia and surgical infection) gathered by the Quality Alliance members and published by Centers for Medicare and Medicaid Services (CMS) on www.hospitalcompare.hhs.gov.
But not everyone thinks the information you'll find there or at www.qualitycheck.org, performance reports from hospitals accredited by the Joint Commission on Accreditation of Healthcare Organizations, tells the complete story about hospital quality. (JCAHO measures are virtually the same as CMS data, but CMS data is on Medicare patients alone; JCAHO's is on all patients.)
“Measuring patient care has become an industry,” UAMS associate dean R.T. Fendley says, and the industry is “not ready for prime time.”
UAMS had the lowest scores of any major hospital in Little Rock in seven out of 17 areas it reported to CMS. Fendley acknowledges that his criticism of the survey tools sounds like sour grapes coming from UAMS. But Fendley noted that scores currently on line are out of date, and that improvements have been made, if not posted. (Current information online on Hospital Compare is for the first half of 2005; JCAHO's postings are through September 2005.)
Many of UAMS' low scores come, surprisingly, in giving smoking cessation advice to patients admitted for heart failure, heart attacks or pneumonia. For example, 89 percent of Heart Hospital's heart attack patients on Medicare get smoking cessation advice, compared to UAMS' 76 percent.
Fendley acknowledged that UAMS isn't pleased with that outcome. “That one does have our attention because UAMS has been on the leading edge to going to smoke-free [campuses],” he said.
Fendley added that “any suggestion that we're not doing our job to the best of our ability gives us pause.” But UAMS says current yardsticks can't measure UAMS' delivery of health care. “We are struggling to find that set of metrics that really does contribute to truly improved patient care, and when we find that we'll throw [ourselves at it] with vigor.”
JCAHO uses a generalized, consumer-friendly grading system: plus signs (for “above the performance of most accredited organizations”), checks (“similar to the performance of most”) and minuses (“below the performance of most”). It also offers details on why a hospital scored as it did.
If you go to JCAHO's website, you'll find that the Arkansas Heart Hospital has earned a minus in the treatment of heart attack. If that makes your heart skip a beat, consider that it gets a plus in the treatment of heart failure.
“Numbers don't mean we don't give good care,” the Heart Hospital's Mary Willis noted. The Heart Hospital's grade on heart attack was lowered by its low percentages of patients provided aspirin on arrival and beta blockers at discharge. The information can be skewed, Willis said, when a chart fails to note that a patient took an aspirin immediately before arrival, or when a doctor fails to note his discharge instructions. “I know a physician takes good care of patients [even if] he leaves out little things,” Willis said. Some survey questions have changed to correct ambiguities, she said.
But by and large, she believes the performance checks will bring improvement where needed. The Heart Hospital's goal is to be “100 percent for everything,” she said. They've reached that goal on smoking cessation advice (the online data, three quarters old, put the number at 83 percent); “I truly think we're making an impact.”
Dr. Phillip Mizell of Baptist Health thinks the surveys offer good information, but notes the caveat offered on the websites: Consult your doctor before making any decisions on where to go for health care.
The CMS (Hospital Compare) and JCAHO data can be searched by hospital name or location. The websites sites explain why each measure — for example, whether heart attack patients were given certain medicines — is considered the standard of care.
Besides the three areas mentioned previously, Arkansas hospitals exceeded the national average in the way they administer antibiotics to pneumonia and surgery patients, in providing thrombolytic medication to heart attack patients on arrival, and getting blood cultures from pneumonia patients before prescribing antibiotics.
Baptist Hospital had the highest score among Little Rock area hospitals for cardiac care when 10 common data sets (out of a possible 12) were averaged. (Some hospitals didn't report or didn't admit enough patients to have a statistically significant result in some areas.) St. Vincent Infirmary Medical Center had the second highest average and the Heart Hospital the third. In a comparison of nine common cardiac data sets, Heart Hospital was second, St. Vincent was third and UAMS was fourth. (Rebsamen Regional Medical Center reported in only four cardiac areas and Southwest in only three.)
St. Vincent scored highest in pneumonia care, Southwest Hospital second and UAMS third when five common data sets were compared. Baptist, which did not include data on one of the five (appropriate initial antibiotic given), scored third when averages for four common data sets were compared, pushing UAMS to fourth.