From the Journals

VA readmissions program not linked to increased death


Despite a lack of financial penalties, the U.S. Veterans Affairs Health Care System has seen a steady 15% reduction in 30-day all-cause readmissions for patients admitted for heart failure, with no concurrent increase in 30-day mortality, a large cohort study suggests.

Unlike the Center for Medicare & Medicaid’s Hospital Readmissions Reduction Program (HRRP), whose primary objective is reducing payments to hospitals with excess readmissions, the VA’s efforts to reduce readmissions across their system did not include any financial penalties.

“The intervention focused on encouraging participation in transitions of care programs, such as the American College of Cardiology’s Hospital to Home Initiative and the creation of a heart failure provider network that included more than 900 heart failure providers throughout the VA system,” said the study’s lead author Justin T. Parizo, MD, of Stanford (Calif.) University.

The only measuring sticks the VA used were the public reporting of 30-day readmission rates (starting in 2012) and inclusion of those rates into hospitals’ overall star ratings (starting in 2014).

“The readmissions reductions we saw were similar in magnitude to those seen in patients in CMS fee-for-service categories in the HRRP,” said Dr. Parizo. “And while we had no ability to evaluate causality here, our best guess from what we can see is that there’s been no impact of the readmissions program on mortality,” he added.

Their results were published online June 17 in JAMA Cardiology.

Dr. Parizo and colleagues conducted a cohort study of 304,374 heart failure hospital admissions in 164,566 patients from January 2007 to September 2017. Importantly, he stressed, the researchers were able to do sophisticated risk adjustment for illness trends, something that has been a sticking point in some of the HRRP studies to date.

“We leveraged the robust dataset that the VA provides to adjust for illness severity. Accounting for clinical factors, like blood pressure, weight, creatinine, BNP [B-type natriuretic peptide], and other markers of heart failure severity, but also for changes in coding,” said Dr. Parizo.

Stratification according to left ventricular ejection fraction (LVEF) showed similar results both in terms of 30-day readmission and 30-day mortality for those with LVEF of 40% or greater and those with LVEF less than 40%.

In an interview, Dr. Parizo noted that they actually saw a small but significant uptick in mortality in the 2011-2012 period (compared with 2007-2008) that remains unexplained. “By the 2015-2017 period, 30-day death had returned to baseline levels,” he said.

In contrast, the HRRP, which was rolled out in 2012, has also been shown to reduce readmissions but, in most studies, 30-day mortality had gone up.

Dr. Leora Horwitz of New York University school of public health

Dr. Leora Horwitz

“The VA has a very robust quality infrastructure and a robust mechanism for prioritizing certain quality-improvement goals and getting them accomplished that I think they are underrecognized for,” said Leora Horwitz, MD, MHS, the director of the Center for Healthcare Innovation and Delivery Science at NYU Langone Medical Center, New York.

In an interview, she also noted some concern with the uptick seen in the 2011-2012 period, noting that the increase might be the same signal seen with the HRRP intervention.

“This is around the same time period where other people were writing the HRRP papers that showed an increase in mortality, so that’s something to consider,” she said.

Dr. Horwitz coauthored a study published in 2017 indicating that, on a hospital level (compared with a patient level, the approach most other studies took), reductions in readmissions were only weakly correlated with 30-day mortality rates after discharge.

“So, if you think that a hospital that’s behaving badly and keeping people out of the hospital inappropriately to cut down their readmissions, you’d expect to see increased mortality in that hospital, and in our study there was no correlation whatsoever. So there is still debate as to what is behind the increase in mortality on a patient level with heart failure that we’ve seen in some studies,” she said.

Dr. Horwitz doubts an intervention such as the one undertaken in the VA system – even with its fairly soft-touch “name and shame” component – would work in the non-VA hospital world.

“Those who have been in favor of financial penalties have pointed to the fact that, in general, it’s hard to get health systems to respond without financial alignment, even if it’s not an overt financial incentive,” she said.

“The VA is a unique environment,” she noted. “They have a very strong top-down command control focus where people are kind of used to being told, ‘OK, here are the measures we have to address this year.’ It’s good to see that the system that has worked for them for other outcomes also worked for them for heart failure readmissions too.”

Dr. Parizo has disclosed no relevant financial relationships. Dr. Horwitz has worked under contract to Medicare to develop readmission measures.

A version of this article originally appeared on

Recommended Reading

HFpEF: Gender difference in sacubitril/valsartan response remains mystery
The Hospitalist
SGLT2 inhibitor ertugliflozin shows no CV death or renal benefit
The Hospitalist
New study of diabetes drug for COVID-19 raises eyebrows
The Hospitalist
FDA approves dapagliflozin for low-EF heart failure
The Hospitalist
Even with mild COVID-19, athletes need cardiac testing before returning to play
The Hospitalist
FDA okays emergency use for Impella RP in COVID-19 right heart failure
The Hospitalist
More fatalities in heart transplant patients with COVID-19
The Hospitalist
Mortality differs by LVEF between women and men
The Hospitalist
DAPA-HF: Dapagliflozin slows T2D onset in heart failure patients
The Hospitalist
Cost of preventable adult hospital stays topped $33 billion in 2017
The Hospitalist
   Comments ()