Are Lower Readmissions Credited to 30-Day Rules Overstated?

Debra L. Beck

January 08, 2019

The Affordable Care Act's Hospital Readmissions Reduction Program (HRRP) may not be as good as previously suggested at reducing 30-day readmissions, according to a new report.

After adjusting readmission rates to account for revised reporting standards that allowed hospitals to document a larger number of diagnoses per claim, the decline in previously noted risk-adjusted readmission rates for targeted conditions including myocardial infarction (MI), heart failure (HF), and pneumonia after the announcement of HRRP dropped by 48%.

The findings deal yet another blow to the view that targeted financial incentives can bring about significant behavior changes. Another study released in late December showed that 30-day mortality may be increased post HRRP initiation for HF and pneumonia, although not MI.

Christopher Ody, PhD, an applied microeconomist at Northwestern University in Evanston, Illinois, and coauthors compared data from the periods before and after HRRP implementation. They found the readmission rates were overstated by about half because of new rules that allowed hospitals to document a larger number of diagnoses per claim, thereby creating the appearance of increased patient risk.

"When we started digging into the mechanisms behind the decline in readmissions, we found that much of the effect shows up in one specific month and that it is not driven by decreases in readmission rates, but rather by increases in patient risk scores," explained Ody in an interview with | Medscape Cardiology. "So it looks like risk-adjusted readmissions are plummeting in one month and staying lower, but while readmission rates have fallen, they didn't fall as much as previously suggested because predicted patient risk increased."

To be clear, Ody and colleagues do not think this is a case of purposeful up-coding, nor is it representative of a substantive change in patient risk.

In a second analysis that adjusted for differences in pre-HRRP readmission rates across samples, they found that declines for targeted conditions at general acute care hospitals were statistically indistinguishable from declines in two control samples, leading them to conclude that either the HRRP had no effect on readmissions or may have led to a system-wide reduction in readmissions not limited to targeted conditions and targeted hospitals, and that was roughly half as large as prior estimates suggest.

"If we use the analogy of how we assess new medications, for example, our prior study in JAMA [which showed increased 30-day mortality] really was looking at the 'safety' of the HRRP — was its implementation associated with any unanticipated adverse effects," said Robert W. Yeh, MD, MBA, from Harvard Medical School, Boston, Massachusetts, in an email.

"This study, in contrast, examines the 'efficacy' side of things and questions whether the policy actually led to meaningful reductions in readmissions," Yeh said. "Given some uncertainty regarding both aspects of the HRRP, I think the public is right to ask the question — shouldn't we demand stronger evidence for both efficacy and safety of health policies before they are comprehensively rolled out?"

The calls by clinicians and health policy experts for developing a better evidence base to support new health policy before widespread implementation have been getting louder and more assertive.

Don't Throw Out the Baby

Leora Horwitz, MD, MHS, agrees with the authors that coding changes likely account for some but not all of the reduction in readmissions attributed to HRRP.

"There has been a lot of chatter in the literature, media, and on Twitter in the past months about this issue of coding, including a research letter on the topic, and I agree true readmission reductions are likely smaller than what the 2016 New England Journal paper suggested, though even in this paper, there is evidence of genuine reduction in readmissions," she told | Medscape Cardiology.

"But we also know that per capita admission rates have been declining over this whole period of time, so there have been fewer patients getting admitted and those that are admitted are probably sicker, so it's hard to tease out the true effect of the policy," she added.

Horwitz is a hospitalist at NYU Langone Health, New York City, and director of the Center for Healthcare Innovation and Delivery Science. She has worked under contract to the Centers for Medicare and Medicaid Services (CMS) to develop readmission measures and has published data showing no association of hospitals' changes in readmissions with changes in mortality.

"Mortality for patients with heart failure has clearly been rising over time," she noted.

"What it really comes down to is a question of the inferred mechanism. The researchers who are concerned about this worry that the way in which people have reduced readmissions is to somehow prevent patients who should be in hospital from being there, but I have sat on readmissions committees at major academic centers and I see really intense work to improve post-discharge care and to carefully review every readmission, so it's just hard for me to believe that the major mechanism for readmission reduction is ER doctors saying 'I'm worried about a readmission penalty for my hospital so I'm going to put this really sick patient under observation instead of admitting her,'" Horwitz explained. "That's just really not how they behave."

Looking at the bigger picture, Ody suggested the stick approach used in HRRP may ultimately do more harm than good.

"As an economist, I love pay for performance in a lot of situations. I wish that we could pay my kids' best teachers more. I wish we could pay their worst teachers less. I wish that we could give their worst teachers a hint that this may not be the right career for them and give those best teachers incentives to stay in the job, even if we weren't getting any teachers to work harder," he said.

"But with hospitals, I don't think the same thing is true. If we penalize hospitals that are safety net hospitals serving low-income patients, and then those hospitals are not responding and we're just taking money away from them, it's not clear that as a society this is a good thing."

Studies have indicated that safety net hospitals face unique barriers in their efforts to reduce readmissions and suffer higher penalties under the HRRP 30-day readmission rules.

Ody and coauthors have reported no relevant financial relationships. Yeh has reported receiving research support from the National Heart, Lung, and Blood Institute and grants and/or personal fees from a number of industry entities not relevant to this study. Horwitz was part of the group at Yale University that developed readmission measures under contract to CMS, though she did not develop the measures used in the HRRP. 

Health Aff. 2019;38:36-43. Abstract

For more from | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: