Miriam E. Tucker

October 16, 2013

WASHINGTON, DC — Hospitals could be underestimating their postoperative readmission rates because they typically do not track patients who are readmitted to other institutions, the results of a Medicare data analysis suggest.

Many hospitals only know about their own readmissions, but Medicare penalizes hospitals for readmission to any institution. Hospital quality-improvement initiatives are being based on incomplete data, said Andrew Gonzalez, MD, from the Center for Healthcare Outcomes and Policy at the University of Michigan, in Ann Arbor.

"We examined whether same-hospital readmissions serve as a good proxy for all admissions, he explained.

Dr. Andrew Gonzalez

Dr. Gonzalez presented the results here at the American College of Surgeons 2013 Annual Clinical Congress.

The Hospital Readmissions Reduction Program (HRRP), part of the Affordable Care Act, took effect on October 1, 2012. HRRP currently penalizes hospitals by 1% for excess readmissions of patients previously admitted for heart attack, heart failure, or pneumonia.

Penalties in Effect

That penalty rises to 2% in 2014 and, in 2015, Medicare will add 2 new conditions to the readmission penalty list — hip or knee arthroplasty and chronic obstructive pulmonary disease.

Approximately 1 in 7 patients are readmitted to the hospital within 30 days, according to a recent report (N Engl J Med. 2013;369:1134-1142). Medicare readmissions alone cost the healthcare system $28 billion a year, Dr. Gonzalez noted.

On the flip side, quality-improvement initiatives are expensive for hospitals, and the current system doesn't provide incentives for investing in them.

Dr. Fergal Fleming

Without knowing their risk for being penalized, hospitals can't plan their approach, study discussant Fergal Fleming, MD, from the University of Rochester Medical Center in New York, told Medscape Medical News.

"The same-hospital readmission rate is not necessarily an accurate reflection of the all-hospital readmission rate, and the all-hospital readmission rate is the one that matters," Dr. Fleming explained.

Although Medicare penalties currently apply to the institution, rather than the individual provider, there are likely to be trickle-down effects, Dr. Gonzalez told Medscape Medical News. "The message for doctors is that you have a potentially unknown liability exposure. If the hospital loses money, all shareholders could lose money."

The solution, he believes, lies in interinstitutional collaboration and a shorter Medicare feedback cycle. Reports are currently issued just once a year.

On Average, 5% Readmitted Elsewhere

Dr. Gonzalez's team used Medicare data from 2007 and 2008 and 3940 hospitals to evaluate 30-day readmission rates for 741,656 fee-for-service Medicare beneficiaries admitted for coronary artery bypass grafting (31%), colectomy (31%), or hip fracture (38%).

The same-hospital readmission rate was 8.4% (range, 2.6% - 15.3%) after adjustment for age, sex, and comorbidities, whereas the all-hospital readmission rate was 12.2% (range, 8.4% - 21.2%).

This means that "if a hospital adds 5% to their readmission rate, it may provide a better estimate of their true readmission rate." However, "given the wide margin of error, it is a suboptimal proxy for the all-hospital readmission rate," Dr. Gonzalez noted.

Therefore, the investigators compared quintiles of same-hospital readmission rates with quintiles of all-hospital readmission rates. Overall, 56.2% of hospitals were in different same-hospital and all-hospital quintiles.

For 60% of the top- and bottom-performing hospitals, there was no difference in readmission-rate quintiles. However, the difference was more evident in the middle quintiles, where 60% to 70% of hospitals were in different same-hospital and all-hospital quintiles.

Thus, a considerable proportion of hospitals in the middle quintiles would fall from Medicare's "no different than" to "below" the average national rate.

"These hospitals are of unique importance for public reporting, because Medicare's measure is very mean-centric. If you fall above or below the line, that completely changes your grade," Dr. Gonzalez explained.

Of course, he noted, one limitation of these data is that they were obtained prior to the passage of the Affordable Care Act in 2010. Therefore, any behavior undertaken in anticipation of the HRRP provision could not be analyzed.

Team Effort Needed

As it stands now, the HRRP poses a dilemma for hospital CEOs, Dr. Fleming told Medscape Medical News. Instituting programs to reduce readmissions is very costly. Low-performing hospitals need to analyze their individual situations to determine their best course of action.

The 3 conditions for which Medicare currently penalizes hospitals comprise just 20% of hospital readmissions. "Are you going to just target those 3 conditions, or apply a readmissions policy to all your patients, which would be a much greater expense? These things have huge financial implications," Dr. Fleming noted.

He had one complaint about the study. The quintiles don't line up with the 3-point scoring system used by Medicare. "The methodology used is just different, so it's a little bit like comparing apples and oranges," Dr. Fleming told Medscape Medical News.

Dr. Gonzalez explained that the median quintile for performance can be used. "If you fall in there, it's a good proxy for the mean."

Hospitals need more real-time feedback than the once-yearly report from Medicare, Dr. Gonzalez told Medscape Medical News. One of his institution's initiatives, the Michigan Surgical Quality Collaborative, tracks patients who are readmitted to any institution within the state's network, and feeds the data back to the hospitals in quarterly reports.

Regardless of the reporting cycle, individual providers will need to be involved in the remediation process, and that won't be easy or cheap. "It's going to have to be a team effort. It's going to require buy-in from all the stakeholders," Dr. Fleming told Medscape Medical News. "It has to be a system organizational change, and that requires resources. You've got to change culture, and you've got to spend money."

Dr. Gonzalez and Dr. Fleming have disclosed no relevant financial relationships. Study coauthor Dr. Justin Dimick reports he is a consultant and equity owner of ArborMetrix, a healthcare analytics and information technology firm that was not involved in the collection or analysis of the data presented.

American College of Surgeons (ACS) 2013 Annual Clinical Congress. Presented October 8, 2013.


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.