Imperfect measure: Harvard researchers weigh problems with hospital readmission rates
They have been warning for months that the readmissions measure is faulty. Last April, Drs. Karen Joynt and Ashish Jha wrote in the New England Journal of Medicine that the federal policy penalizing hospitals where patients are more often rehospitalized within 30 days was “misguided,” and unfairly targets those located in areas that are poor and with large minority populations.
In a research letter published Tuesday by the Journal of the American Medical Association, the researchers from the Harvard School of Public Health added some heft to their argument. They looked at penalties assessed against 2,189 hospitals and found that the largest hospitals and those that are teaching hospitals or safety net hospitals, meaning they care for the highest portion of poor patients, were most likely to see their payments docked.
Forty percent of large hospitals and 44 percent of major teaching hospitals received the highest penalties, compared with 28 percent of small hospitals and 33 percent of non-teaching hospitals. Forty-four percent of safety-net hospitals were highly penalized versus 30 percent of those without such a designation.
Joynt said the mathematical models used to predict which patients will have to return to the hospital soon after discharge are not very good. In fact, she said, they’re “barely better than a coin flip.” Those models are used to assess penalties by measuring how hospitals are doing versus what is expected of them.
They don’t fully account for the difference in severity of illness between patients at community hospitals and at those hospitals to which the sickest patients are referred, Joynt said. And they don’t factor in variables in patients’ lives once they leave the hospital grounds.
The likelihood of a person returning to the hospital after having a heart attack depends a lot on housing, financial stability, and other socio-economic factors, she said.
Joynt offered this example: One man has a stent implanted at Boston Medical Center after a heart attack. Another has the same procedure at Newton-Wellesley. But the first is homeless and the second has a strong family support system, with relatives checking in on him at regular intervals to make sure he is taking his medications and has what he needs.
Of course those men are not representative of all of the patients at those two facilities, but a homeless man may be a rare patient in Newton and a far more common one at Boston Medical.
Still, couldn’t the threat of a readmissions penalty incentivize those facilities serving more poor patients to better address their challenges after discharge? Maybe, Joynt said, but the penalty leaves them with less money to do that work.
“We should be thinking about ways to fix all of those things, but penalizing the hospital doesn’t actually get at the core of the problem,” she said. “It’s a useful signal, but I don’t think its nuanced enough to recognize just how complicated a problem this is.”
The penalty program, created under the Affordable Care Act, is well underway. The Centers for Medicare & Medicaid Services calculates penalties based on how many Medicare patients treated for certain conditions -- pneumonia, heart failure, and heart attack -- are rehospitalized within 30 days. Hospital leaders and even some policymakers who worked to implement the program have called for changes.
The formula for measuring unnecessary readmissions could take into account the income levels of the patient population that a hospital serves, Joynt said. Or, hospitals could be assessed against their own performance in the prior year or against other hospitals of similar size and patient mix.
“Does it really make sense to compare what happens outside of the hospital for Boston Medical Center patients and Newton-Wellesley patients?” Joynt said. “It really doesn’t.”
Nearly all of the attention on readmissions has been focused on adult patients. In a separate study also published Tuesday by JAMA, a team of researchers mostly from Boston Children’s Hospital examined admissions at 72 children’s hospitals in the United States.
The group found that 6.5 percent of children treated at the sample hospitals were rehospitalized unexpectedly within 30 days of discharge. Congenital heart defects, respiratory conditions such as asthma, and neurological diseases were among the chronic conditions most likely associated with the readmissions.
The group found significant variation across hospitals, both in the overall readmission rates and in the condition-specific rates.Chelsea Conaboy can be reached at firstname.lastname@example.org. Follow her on Twitter @cconaboy.