Better Outcomes Associated with Higher Volume and More Specialists in Intensive Care, Confirming Leapfrog Standard
GOLDEN, Colo. (October 17, 2005) – A typical patient has a 65 percent lower chance of dying at the nation's highest-rated hospitals compared with the lowest-rated hospitals, in 18 common procedures and diagnoses, according to a large-scale study released today by HealthGrades. That "quality chasm," the HealthGrades study shows, is growing, as the nation's best-performing hospitals lowered their mortality rates 45 percent faster than the poorest-performing hospitals over the same time period.
The findings are part of the eighth annual HealthGrades Hospital Quality in America Study, which analyzes 37 million Medicare hospitalization records, from the years 2002 through 2004, to rate the quality of care at each of the nation's more than 5,000 nonfederal hospitals. The hospital ratings, for 28 procedures and diagnoses at each facility, are listed free of charge for consumers at www.healthgrades.com.
"There is real reason to celebrate in this year's study of quality at America's hospitals," said Samantha Collier, MD, the primary author of the study and the vice president of medical affairs at HealthGrades, the leading healthcare ratings company. "Overall, mortality rates are declining at our nation's hospitals. However, there's still a lot of work to be done because our findings support that we're not making much headway in closing the 'quality chasm' between the best and worst hospitals. If all hospitals performed as well as the highest rated hospitals, more than a quarter million lives would have been saved over the past three years."
Overall mortality rates improved 12 percent, the study shows, with some of the better outcomes associated with higher hospital volumes and higher numbers of physicians who specialize in critically ill patients, called intensivists, staffing intensive care units (ICUs).
The annual HealthGrades study rates each nonfederal hospital with a one-, three- or five-star rating indicating poor, average or excellent outcomes in each of 28 medical categories. Taken together, the ratings produce the following findings:
- The nation's inhospital risk-adjusted mortality rate improved, on average, 12 percent from 2002 to 2004. But the degree of improvement varied widely by procedure and diagnosis studied.
- Five-star rated hospitals had significantly lower risk-adjusted mortality rates across all three years studied and improved 21 percent more than the U.S. hospital average and 45 percent more than 1-star rated hospitals.
- A typical patient would have, on average, a 65 percent lower chance of dying in a 5-star rated hospital compared to a 1-star rated hospital, and a 45 percent lower chance of dying in a 5-star rated hospital compared to the U.S. hospital average.
- If all hospitals performed at the level of a 5-star rated hospital across 18 of the procedures and diagnoses studied, 273,137 Medicare lives could have potentially been saved from 2002 through 2004. Fifty percent of the potentially preventable deaths were associated with just four diagnoses: Heart Failure, Community Acquired Pneumonia, Sepsis and Respiratory Failure.
- Five-star rated hospitals have higher volumes and higher rates of intensivists staffing ICUs associated with three severe illnesses. Higher volumes are associated with better outcomes for Abdominal Aortic Aneurysm Repair. Higher rate of ICUs staffed with intensivists is associated with better outcomes for Diabetic Acidosis and Coma and Pulmonary Embolism. ICU intensivist staffing is an indicator of quality promoted by the Leapfrog Group, and the data used in this study is part of the Leapfrog Group Survey.
Table: Hospital Improvement by Procedure and Diagnoses: 2002-2004
| Hospitalization Diagnosis or Procedure | Percent of Improvement U.S. Average (2002-2004) | Relative Risk Reduction Associated with 5-Star Hospitals Compared to 1-Star | Relative Risk Reduction Associated with 5-Star Hospitals Compared to National Avg. | Reduction in Deaths if All Hospitals Operated at 5-Star Level (2002-2004) |
| Abdominal Aortic Aneurysm Repair | 13.04% | 74.75% | 52.00% | 428 |
| Acute Myocardial Infarction | 6.44% | 42.13% | 24.90% | 26,431 |
| Atrial Fibrillation | 18.65% | 88.28% | 76.23% | 5,718 |
| Bowel Obstruction | 14.32% | 62.50% | 40.92% | 11,653 |
| Chronic Obstructive Pulmonary Disease | 17.46% | 72.08% | 51.79% | 14,630 |
| Community Acquired Pneumonia | 16.02% | 57.02% | 35.60% | 40,986 |
| Coronary Bypass Surgery | 11.68% | 70.81% | 48.89% | 5,537 |
| Coronary Interventional Procedures | -2.09%* | 64.45% | 44.28% | 7,811 |
| Diabetic Acidosis and Coma | 21.88% | 92.89% | 82.57% | 2,884 |
| GI Bleed | 17.53% | 69.95% | 49.02% | 11,903 |
| GI Surgery and Procedures | 0.55% | 55.75% | 34.77% | 18,371 |
| Heart Failure | 17.38% | 57.92% | 36.27% | 34,380 |
| Pancreatitis | 22.15% | 81.54% | 60.68% | 1,307 |
| Pulmonary Embolism | 16.82% | 75.41% | 55.26% | 1,972 |
| Respiratory Failure | 0.46% | 45.67% | 27.06% | 30,145 |
| Sepsis | 3.44% | 47.46% | 27.42% | 29,874 |
| Stroke | 10.58% | 48.70% | 28.64% | 26,134 |
| Valve Replacement Surgery | 9.97% | 65.43% | 43.79% | 2,973 |
| AVERAGE / TOTAL | 12.13% | 65.15% | 45.56% | 273,137 |
*The negative value was caused by a change in the data. Effective in 2004, drug eluting stents were coded with their own ICD-9 code, and this code was one of the predictors used to create the expected value. This code decreased the "expected" for 2004 only, resulting in a larger ratio for that year.
The full study, along with its methodology, can be found at http://www.healthgrades.com.
The study is available in PDF.
About HealthGrades
HealthGrades (Nasdaq:HGRD) is the leading healthcare ratings company, providing ratings and profiles of hospitals, physicians and nursing homes to consumers, corporations, health plans and hospitals. Millions of consumers and hundreds of the nation's largest employers, health plans and hospitals rely on HealthGrades' independent ratings to make healthcare decisions based on the quality of care. More information on the company can be found at http://www.healthgrades.com.