Home Analysis Why heart failure patients are really readmitted

Why heart failure patients are really readmitted

June 2008

Published in the June 2008 issue of Today’s Hospitalist

HOSPITALS ARE FACING new scrutiny for their performance on CMS core measures, and physicians everywhere are taking a closer look at how they discharge heart failure patients as one way to prevent these patients from bouncing back within a week or two. But new research shows that in at least some cases, readmissions for heart failure may be partly out of physicians’ control.

When researchers at Albert Einstein Medical Center in Philadelphia looked at heart failure patients readmitted to the hospital, they found that 58 patients racked up a total of 74 readmissions within 30 days over an 18-month period for an exacerbation of the condition. Closer examination, however, revealed several surprises.

For one, the so-called “frequent-flyers” “the clearly non-compliant patients with multiple admissions in the past “were not responsible for the majority of readmitted patients, says Marco Hallerbach, MD, an internal medicine resident who is the study’s lead author.

But researchers also found that roughly half of the evaluated readmissions within 30 days were incorrectly labeled by either medical coders or emergency physicians. Many of the hospital’s heart failure “readmissions,” it turns out, weren’t due to heart failure exacerbations after all.

The wrong diagnosis
The study, which appeared in the January-February 2008 issue of the American Journal of Medical Quality, found through subsequent chart analysis that 10 patients labeled as an exacerbation readmission were really admitted for chest pain. Nine others were admitted for rapid atrial fibrillation, while 10 had either nausea or shortness of breath not caused by a heart failure exacerbation.

According to Dr. Hallerbach, that high rate of mislabeled diagnoses suggests that medical coders may struggle to choose the correct primary diagnosis, especially for patients with multiple underlying comorbidities.

“And readmitted patients are often first treated by emergency department physicians who may prematurely narrow down their differential diagnoses,” Dr. Hallerbach says. “That means that the patient could be discharged later without having been given a medically valid admission diagnosis.”

Dr. Hallerbach says that he finds the mislabeling results “fascinating” because of the implications for using billing and administrative data as a basis for research and to guide quality improvement initiatives.

“Current clinical heart failure trials might fail to demonstrate important patient factors because these data may be diluted with non-CHF patients,” he says. “We should rely more on physician-based CHF databases than on just administrative data.” The medical center, for instance, now maintains a database of heart failure patients who are specifically followed by the heart failure service.

Renal disease a big culprit
While noncompliance is often thought to lead to readmissions, study authors pointed out that noncompliance with medications was documented in only seven of the 74 readmissions, while noncompliance with diet was documented in only one. The researchers did note that because noncompliance is not a part of the hospital’s standardized admission form, “the real rate of noncompliance may be higher.”

The biggest factor behind readmissions that were legitimately due to heart failure exacerbations was the presence of chronic renal insufficiency or failure. Renal disease was found in fully 45% of the 58 patients who “bounced back,” compared to only 26% of patients admitted for heart failure who weren’t readmitted within 30 days.

“That appears to be a reasonable but often under-recognized clinical factor in everyday practice,” Dr. Hallerbach points out. Results suggest, he explains, that physicians need to identify those high-risk patients during their initial hospitalization, something that’s easy enough to do by monitoring plasma creatinine levels.

To improve treatment during patients’ first hospital stay, he adds, physicians should consider continuous renal replacement therapies in patients with chronic renal insufficiency who are headed for hemodialysis.

Physicians should also, he notes, “opt for earlier involvement of nephrology input in these patients.” While the physicians at Albert Einstein haven’t moved to standardize earlier consults as yet, Dr. Hallerbach says they may consider placing those consults sooner “even for patients who aren’t on hemodialysis.

Problems with follow-up
Finally, researchers explored the effect of lack of follow-up. Dr. Hallerbach says that early follow-up with nephrology is mandatory, particularly for patients with chronic renal insufficiency, but researchers found that it didn’t always occur.

While 85% of readmitted patients received a recommendation at discharge to follow up with their primary physician, no such recommendation was documented in 15% of their charts. That could mean the recommendation simply wasn’t documented, but it might also mean that it wasn’t made. None of the initial discharge instructions for any of the patients included appointment dates or times.

Even when discharge instructions were provided, they were not always complete. For 41% of readmitted patients, the initial discharge documentation forms that providers are asked to use with heart failure patients were not completely filled out.

A nurse dedicated to heart failure patients would be a big help, Dr. Hallerbach points out “one that many hospitals don’t have the resources to hire. “With the new DRG system in place, however,” he says, “that might change because readmissions may not be properly reimbursed.”

Bonnie Darves is a freelance health care writer based in Lake Oswego, Ore.