Spending is not linked with big drops in mortality rates from common conditions, according to a study published in the August edition of Health Affairs.

Although mortality rates were falling across most of the seven conditions examined, the amount of money spent to achieve these gains varied widely, with some conditions, such as sepsis, virtual black holes of spending - receiving lots of money but with little to show for it.

"We've never paid attention as a society to the value that we're getting in health care, and that's really one of the messages of this study, to say, we should be looking at value, and not just cost," Dr. Michael B. Rothberg, associate professor of medicine at Tufts University School of Medicine in Boston, and Baystate Medical Center in Springfield, Mass., told DOTmed News.

Unlike most consumer goods, where costs hold steady or are falling, costs in medicine keep rising, according to the paper. There are two hypotheses, Rothberg said: are patients getting more care, or is the same care just costing more?

"It's probably a combination of both," he said. "We know the costs of providing the same care are going up, because [the costs of] all the components are going up."

Rothberg got involved in the research as he noticed over the years a 20 percent reduction in mortality from pneumonia. Although significant, it might not be visible to providers, he said, as the reduction translates into mortality rates of 4 out of 100, down from about 5 out of 100. He was thinking of ways of sharing this information, when he wondered what was spent to achieve that drop in mortality.

In the study, Rothberg and his team measured seven common diagnoses at 122 U.S. hospitals from 2000 to 2004, using inpatient data hospitals report to the Healthcare Cost and Utilization Project.

These included heart attack, chronic obstructive pulmonary disease, pneumonia, congestive heart failure, stroke, sepsis and urinary tract infection.

For all except sepsis, mortality dropped steadily.

Three of these, heart failure, heart attacks and pneumonia, had the biggest relative declines in mortality, with 21, 19 and 20 percent declines, respectively. They were also the diseases targeted by quality improvement campaigns - in 2001, the Joint Commission introduced "core measures" of quality for these conditions.

But money spent on the conditions did not reflect the size of the relative drop in mortality. Pneumonia, which had a 20 percent decline in mortality, witnessed a 26 percent increase in spending, or about $1,619 more per patient, in 2004 dollars. Heart failure, which saw death rates drop 19 percent, experienced the biggest cost increase, up 60 percent, or an additional $3,452 per patient in 2004 dollars.

In terms of cost per life saved, the most cost-effective was pneumonia, at $167,000 per life. The least was sepsis, at $1.8 million per patient.

"Our treatments for sepsis don't seem to work," Rotherbeg said. "We've had lots and lots of treatments for sepsis, and none of them are very good. We're trying to come up with better treatments for sepsis, but it's a really hard disease."

But focusing only on cost per life saved is not the most useful metric, Rothberg argued: The cheapest, and greatest, reduction in mortality was for the oldest patients, those above 85 years old, as they had the higher mortality rates to begin with.

"If you want to drop the mortality rate at your hospital, the best thing to do is concentrate on the oldest patients, because they're the easiest ones to get mortality reduction on," Rothberg said. "But in terms of benefit to society, it's not necessarily the greatest benefit. If you keep the 95-year-old from dying of their heart attack, the next month they're dying from pneumonia. Whereas if you prevent a 50-year-old from dying of pneumonia, they could have a lot of years left."

Life-year Analysis

Breaking the data down to cost per life-year gained, a different picture emerged. For cost per life-year, the most effective for the elderly was treatments for acute myocardial infarction (heart attack), at about $12,000 per life-year. For those ages 18 to 54 for this same condition, the cost was $60,000 per life-year, likely because for younger patients, doctors might be more likely to use aggressive surgical interventions, while elderly patients might just receive medication, which is cheaper, Rothberg said.

But Rothberg is careful to point out that because the data come from individual hospitalizations, the results of the study reflect only in-hospital mortality. For instance, although costs for treating heart failure went up a lot, some of the money could have been spent on implantable defibrillators, whose potential life-saving results would not be seen by this particular hospitalization.

"If I put in an implantable defibrillator, it's to take care of the next two years," Rothberg said. "This [study] only measures death during this hospitalization."

Nonetheless, he feels that the study should prompt a look into ways to better allocate money – such as by putting it in research where it could potentially save more lives.

"Some things we're doing aren't working," he said. “We should stop spending money on fancier treatments that don't work and maybe spend money researching treatments that might work."