Skip to main content

Preventable Deaths Report Not Ready for Prime Time

 |  By cclark@healthleadersmedia.com  
   June 19, 2014

A report identifying scores of potential inpatient complications, meant to alert hospitals to their magnitude and frequency, is getting a tepid reception. It is "an interesting idea, but I don't yet know what it tells me," says one hospital quality expert.

Kudos to Premier Inc., the group purchasing and quality collaborative, for telling us that nearly 50,000 patients in 540 hospitals participating in a project may have needlessly died from 86 preventable and very expensive complications last year.

In an eye-popping report published last week in The American Journal of Medical Quality, Premier boldly suggests that if the hospitals had just tried a bit harder, they might have avoided the complications that are the most common causes of inpatient death and cost, such as hypotension, respiratory failure, aspiration pneumonia, and acute renal failure.

It's a scary report.

I don't want to boggle you with numbers, but in addition to the nearly 50,000 people who might have lived, some $4.3 billion might have been saved, and 1.8 million inpatient days might have been avoided. Premier calls this group of 86 conditions "PICs," or "potentialinpatient complications."

The key word here is "potential."

Poking CMS in the Eye
In addition to calculating potential harm and costs, the report also pokes payment policies issued by the Centers for Medicare & Medicaid Services right in the eye.

It shows that only 22 of the PICs are included in any of the three CMS hospital-acquired condition financial penalty programs, two of which kick in on Oct. 1. Though hospitals may scramble to avert those 22 complications, most of them are not the biggest killers or drivers of cost.

They're peanuts compared to many of the remaining 64 PICs on Premier's list, such as respiratory failure.

Don't Get Distracted
"What we're saying [to hospitals] is, 'yes, continue to focus on the CMS [22] conditions,'" explains Richard Bankowitz, MD, Premier's chief medical officer. "'But don't focus exclusively on them… because that may cause you to overlook some other common and very important hospital-acquired conditions. Don't get overly distracted.'"

Take stage III/IV pressure ulcers, which is on the CMS list. In Premier's 540 hospitals, there were only 936 patients with this complication and none died, though it did cost hospitals $5.9 million for additional care.

Iatrogenic pneumothorax, an often avoidable injury to the lung resulting from medical care, is also on the CMS list. This adverse event occurred in nearly 10,000 patients and none died, but costs mounted to $72 million.

But more effort on preventing respiratory failure, which is noton the CMS list, would save lives and a lot more money. The number of potentially avoidable deaths in the participating hospitals from this complication was more than 16,000, and the amount of potentially avoidable spending to treat 119,000 patients was $940 million.

Acute renal failure is another heavy hitter not on the CMS list. This complication occurred in nearly 144,000 patients and killed nearly 3,000. Had those complications been avoided, $490 million would have been saved.

Premier didn't name its participating hospitals or their rates, although it could have. After all, that was the purpose of the report: to eliminate "unjustified variation" in healthcare quality and cost, and preventable complications associated with higher mortality.

If there is variation, we should be able to see it. But we can't. There were too many limitations in how these numbers were counted that could not be overlooked.

Pneumonia, the Old Man's Friend
For starters, when patients go to the hospital, they are often very sick with multiple medical conditions, frailty, and dementia. For many of these patients, death is not just the inevitable outcome, it is the desired one that ends a patient's suffering. As the often-quoted phrase attributed to Sir William Osler, goes, "pneumonia is the old man's friend."

In its report, Premier did not adjust for multiple medical conditions or for advanced age, or for the presence of active "do not resuscitate" orders. We don't know how many patients had end-stage cancer or had suffered a life-threatening trauma. About 5% of patients were over age 90, but we don't know how many of them were among those who died.

And neither did Premier factor in how many previous hospitalizations these patients had endured during the previous year, or whether they came from a skilled nursing home or not.

This was an all-payer mix of patients, so we don't know how many were young and healthy before they came to the hospital.

That's part of a concern raised by Ashish Jha, MD, a hospital quality expert in the Department of Health Policy and Management at Harvard School of Public Health and a general internist at the Boston VA Medical Center.

"I can't tell whether they're identifying actually bad things that happen to the patient as a result of medical care, or whether they're identifying things that are the natural consequence of disease," Jha says.

"If someone comes in with pneumonia and you get them started with antibiotics right away, some proportion of those people will still go on to develop sepsis, and a certain proportion will still die."

Jha says the Premier concept is "an interesting idea, but I don't yet know what it tells me."

"Patients do ultimately die," acknowledges Bankowitz. "And on that way toward death, some of these conditions we have identified are on that pathway."

Although the Premier report does not conclude that each and every one of these deaths was preventable, Bankowitz acknowledges that, "it's not reasonable to say that none of them was preventable. It's more of a spectrum. All we want to do is alert hospitals to the magnitude and frequency of some of them."

"They need to assess what their priorities are by looking at their own data and see where they want to dive more deeply."

Preventable Deaths, or Not?
In the end, we don't know how many, or if any, were truly preventable. So I think the Premier report either reached too far, or is very premature.

In other words, this report is not anywhere near ready for prime time. When it is, it should refocus CMS's payment policies to the really important complications that kill many and cost lots.

That said, it's a bold start.

Bankowitz says that answering those questions—pairing evidence-based interventions to these 86 PICs and taking a deeper dive into the charted abstracts—will better define which complications that hospitals should sink more resources into preventing.

"We need to understand exactly how many of these might have been prevented through evidence-based protocols. Ideally, if we can determine that, then this could be a good way to measure quality," Bankowitz says.

Well said.

Pages

Tagged Under:


Get the latest on healthcare leadership in your inbox.