Top hospitals record a 71% lower mortality rate

Study shows variations continue between regions

Patients who are treated at the nation's top-rated hospitals have on average a 71% lower chance of dying compared with the lowest-rated hospitals as reported in the 10th annual HealthGrades Hospital Quality in America Study. These outcomes were measured across the following procedures and conditions:

  • atrial fibrillation;
  • bowel obstruction;
  • chronic obstructive pulmonary disease;
  • coronary bypass surgery;
  • coronary interventional procedures (angioplasty/stent);
  • diabetic acidosis and coma;
  • gastrointestinal bleed;
  • gastrointestinal surgeries and procedures;
  • heart attack;
  • heart failure;
  • pancreatitis;
  • pneumonia;
  • pulmonary embolism;
  • resection/replacement of the abdominal aorta;
  • respiratory failure;
  • sepsis;
  • stroke;
  • valve replacement surgery.

The study examined patient outcomes at the nation's approximately 5,000 hospitals, covering more than 41 million Medicare hospitalization records from 2004 to 2006. It also found that if all hospitals performed at the level of hospitals rated with five stars by Golden, CO-based HealthGrades, 266,604 Medicare lives could potentially have been saved over the three years studied.

"This is the largest study we've done," says Marigene Hartker, MD, MBA, an Orlando, FL-based senior physician consultant and one of the study's authors. "One of the things that have come up repeatedly is that there is a wide variance in the [quality] differences between the hospitals we rate as five star and those we rate as one star." The 71% mortality differential, she adds, highlights that variance.

This translates into uncertainty for the public, she continues. "When they walk into a hospital there's no surety of the quality they will get," she notes.

If you "drill down" to the state and local levels, Hartker continues, that variance continues. (See box, below.) "What that means for us is that different states have different health care initiatives and that is reflected in their hospitals," says Hartker. In terms of national initiatives, she adds, "they may have been translated differently or perhaps in a different manner."

HealthGrades study finds state-to-state variation

Three keys to quality

In offering an explanation for why some hospitals are outperforming others, Hartker notes there are three key components to quality:

  • Structure: For example, staffing ratios.
  • Process: The different processes a facility puts in place, how it translates the core measures required by the Centers for Medicare and Medicaid Services (CMS), and whether it reviews outcomes such as mortality and complication rates. "I consult on processes with a number of hospitals, and what the best ones do is translate what is required by CMS into something that's practical and works within their hospital," says Hartker.
  • Evaluation: Evaluating performance on those measures, says Hartker, is equally important.

The successful pursuit of quality benefited more than just the individual hospitals, notes Hartker. "What we saw is that at the state level, improvement occurred more rapidly in the states that had a higher concentration of top-performing hospitals," she says.

Where improvement is needed

According to the study, mortality rates at America's hospitals have improved 11.8% during 2004 to 2006, with the nation's top-rated hospitals improving at a faster rate (12.8%) than the lowest-rated hospitals (11.4%). Of the 18 procedures and conditions studied, those that saw the most improvement in mortality rates were pancreatitis (19.2%), pulmonary embolism (17.4%), and diabetic acidosis and coma (16.6%). Those with the smallest improvement were resection/replacement of the abdominal aorta (0.4%), coronary interventional procedures such as angioplasties and stents (0.8%), and treatment of heart attack (8.9%).

What accounts for the varying levels of success? "For [resection/replacement of the abdominal aorta and coronary interventional procedures such as angioplasties and stents] there aren't huge process measures that have been put in place, and not as much attention paid to them," Hartker observes. "For example, CMS is more focused on things like door-to-balloon time."

In other words, she continues, people are motivated by the attention being given to certain disease states and conditions by organizations such as CMS. The relative lack of improvement for heart attack seems to be the exception that proves the rule. "I have to admit that's a little surprising; I don't have a great theory to explain that," Hartker says.

(Editor's note: The full study, along with its methodology and state-by-state hospital quality statistics, can be found at

[For more information, contact

Marigene Hartker, MD, MBA, Senior Physician Consultant, HealthGrades. Phone: (303) 716-6513.]