Hospital ratings show sharp differences in safety, chance of dying

A surgical tech prepares an operating room in a file photo. REUTERS/Jonathan Bachman

By Sharon Begley NEW YORK (Reuters) - An analysis of government data on hospital safety, including how likely patients are to die of avoidable surgical complications, shows that hospitals vary markedly on these measures and that patients are at higher risk in some nationally-known facilities than at tiny hospitals little known outside their rural communities. The safety ratings of 2,591 hospitals, released by Consumer Reports magazine on Thursday, come at a time when estimates of the number of Americans killed by hospital errors is soaring. According to the 1999 report by the Institute of Medicine that first put a spotlight on the issue, the death toll from medical mistakes in hospitals was at least 98,000 then. In 2010, however, the Department of Health and Human Services (HHS) inspector general said that poor hospital care contributed to 180,000 deaths every year - and that was only among Medicare patients, those 65 or older. And a 2013 study estimated such deaths at a minimum of 210,000 annually and as many as 440,000. If the highest number is correct, poor hospital care would be the country's third leading cause of death, after heart disease and cancer. In 2011, 722,000 annual hospital-acquired infections alone killed 75,000 patients, the U.S. Centers for Disease Control and Prevention reported on Wednesday. The Consumer Reports hospital safety analysis comes eight months after it released ratings of the quality of surgical care at 2,463 hospitals, based on the percentage of Medicare patients who died in the hospital during or after their surgery and the percentage who stayed in the hospital longer than expected. For the current analysis, Consumer Reports compiled data on readmissions (often a sign of poor initial care or follow-up, and something Medicare now penalizes hospitals for), overuse of CT scans (which can cause cancer years later), hospital-acquired infections, communication (on, for instance, medication doses after a patient is discharged) and mortality. The latter was composed of patients who had a heart attack, heart failure, or pneumonia and died within 30 days of entering the hospital, plus surgery-related deaths, meaning patients who had treatable but ultimately fatal complications after an operation. Those include blood clots in the legs or lungs, or cardiac arrest. All of the data were adjusted so that hospitals were not penalized for having sicker patients. Combining the raw data yielded a safety score of 0 to 100. Miles Memorial Hospital in tiny Damariscotta, Maine, came out on top with a safety score of 78, while Bolivar Medical Center in Cleveland, Mississippi, brought up the rear with an 11. The data all came from the Centers for Medicare & Medicaid Services, part of HHS, and were as recent as 2012-2013 (for bloodstream infections) and as old as 2009-2011 (for adverse events in surgical patients). Data on deaths, readmission, and CTs were from patients 65 or older, while that on hospital-acquired infections was for patients of all ages. RATERS TEND TO DISAGREE The differences between hospitals at the top and bottom can be a matter of life and death. Patients who are hospitalized for pneumonia at a low-scoring facility were 67 percent more likely to die within 30 days of admission than pneumonia patients at a top-scoring hospital, according to the Consumer Reports analysis. Of 1,000 surgical patients who develop a serious surgical complication in a top-rated hospital, 87 or fewer die, compared to more than 132 in a low-rated one - a 52 percent higher fatality risk. Consumers are likely to be frustrated if they look up their local hospital on both Consumer Reports and Medicare's Hospital Compare, which is at Medicare.gov/hospitalcompare. In both, many kinds of data are missing from many hospitals. For some hospitals, the results are fairly consistent across ratings. Bolivar Medical Center in Cleveland, Mississippi, got Consumer Reports' lowest safety rating, 11. Medicare shows that its death rate for pneumonia and heart failure patients are worse than the national average, as is its readmission rate. Bolivar declined to comment. In other cases, however, raters disagree. Consumer Reports gave a Nyack Hospital in Nyack, New York, a safety score of 25, tied for ninth worst. But Medicare says its rate of surgical complications is about average, as are readmission and death rates for pneumonia, heart attack, and heart failure patients. A spokeswoman for Nyack declined to comment. There is disagreement at the high end, too. Miles Memorial, which received the highest safety rating in the Consumer Reports analysis, had rates of surgical complications, infections, death from pneumonia and heart failure, and readmission of heart failure and pneumonia patients no different from the national average, according to Medicare. It did very well in avoiding unnecessary imaging, however. One reason is that Medicare might regard a hospital's infection or mortality rate as "average" if it is just a few percentage points below the U.S. average, explained Doris Peter, associate director of Consumer Reports Health, who led the data analysis. But the magazine would see that as below average. The article is available in the May issue of Consumer Reports and online at www.ConsumerReports.org, but accessing the ratings of individual hospitals at www.ConsumerReports.org/hospitalratings requires a paid subscription. (Reporting by Sharon Begley)