I was fascinated to read a blog post earlier this week from the Centers for Medicare & Medicaid Services (CMS) in which federal officials pointed to a nationwide reduction in avoidable hospital readmissions.
The data from CMS showed that all states but one have seen Medicare 30-day readmission rates fall in a five-year span from 2010 to 2015. The one state that did not lower readmission rates was Vermont, as its rate increased slightly from 15.3 percent in 2010 to 15.4 percent in 2015. What’s more, in 43 states, readmission rates fell by more than 5 percent; in 11 states, readmission rates fell by more than 10 percent, according to the CMS data. Overall, readmission rates fell by 8 percent nationally, the federal agency reported.
This information was of particular interest to me because fresh in my mind was the big Kaiser Health News report from August which revealed that the federal government’s penalties on hospitals for failing to lower their rehospitalization rates have hit a new high, as Medicare will withhold approximately $528 million—about $108 million more than last year. And, the government will punish more than half of the nation’s hospitals — a total of 2,597 —for having more patients than expected return within a month. While that is about the same number penalized as last year, the average penalty will increase by a fifth.
The fines for failure to meet CMS’s avoidable readmissions reduction criteria focus on six conditions: heart attack, congestive heart failure, pneumonia, chronic obstructive pulmonary disease (COPD), elective hip and knee replacements, and for the first time this year—coronary artery bypass graft surgery—and are based on readmissions between July 2012 through June 2015. The fines, which adjust payments for hospitals with higher than expected 30-day readmission rates for these targeted clinical conditions, will be levied in October.
Although that KHN report did note that since the Hospital Readmissions Reduction Program began in October 2012, national readmission rates have dropped, I became curious to how these two reports on the same issue came to somewhat variant conclusions. Regarding the KHN report, a deeper analysis from Healthcare Informatics Editor-in-Chief Mark Hagland came to the only real conclusion one could draw from that data: it’s time for hospitals to realize that performance improvement is no longer an option; rather, it’s a necessity.
As Hagland wrote, “There is simply no longer any place to hide, for hospitals that have been under-performing in terms of core clinical and operational performance. Hospital and health systems leaders are going to need to begin—if they haven’t already done so, which they really should already have been doing—to leverage data and information technology in pursuit of the ‘blessed cycle’ of continuous clinical performance improvement.” Hagland also noted that there were more than 1,400 hospitals that were automatically exempted from readmission reduction program penalties. “Given that there are 3,414 hospitals covered by the readmissions reduction mandate, 2,597 hospitals being penalized this year actually represents fully 76 percent—more than three-quarters of hospitals—seeing pay cuts as a result of readmissions deemed avoidable by CMS officials,” he wrote.
So given all this information in front of us, why was the most recent CMS commentary on hospital readmission rates so positive? In the blog post, Patrick Conway, M.D., principal deputy administrator and chief medical officer, CMS; and Tim Gronniger, deputy chief of staff, CMS, wrote that “The data show that these efforts are working,” referring to the Hospital Readmissions Reduction Program as well as the agency’s Partnership for Patients initiative, which aims to make hospital care safer and improve the quality of care for individuals as they move from one healthcare setting to another.
What Do the Numbers Really Say?
Taking a deeper look into the CMS data on readmission rates, the agency’s readmission rates table shows a state-by-state breakdown of hospital admissions and readmission rates in 2010 versus 2015. In Alaska, for example, in 2010, there were 9,809 hospital admissions for the given conditions in the federal program, with a readmission rate of 14.5 percent. In 2015, Alaska had 9,954 hospital admissions, with a readmission rate of 13.7 percent, resulting in a reduction of 100 readmitted patients—across all hospitals covered in the state, five years later. Is that really reason to celebrate?
Using this one state example, the data reveals a small change in the raw numbers when you boil it down to the actual number of patients being readmitted. Other states have much higher readmission reduction numbers than Alaska of course, but also have more hospitals and more admissions. Thus, it begs the question, Is this data really an indicator of success, especially when some 76 percent of hospitals in the program are seeing pay cuts as a result of too many avoidable readmissions?
This also leads to another question of importance: How relevant are these figures? As Jordan Rau, author of the August KHN report, wrote at the time, “The penalties are the subject of a prolonged debate about whether the government should consider the special challenges faced by hospitals that treat large numbers of low-income people. Those patients can have more trouble recuperating, sometimes because they can’t afford their medications or lack social support to follow physician instructions, such as reducing the amount of salt that heart failure patients consume. The Centers for Medicare & Medicaid Services says those hospitals should not be held to a different standard.”
What’s more, a recent study from Johns Hopkins physicians and researchers suggested that “data on mortality and hospital readmission used by CMS suggest a potentially problematic relationship.” Daniel Brotman, M.D., at John Hopkins, and his colleagues examined three years of CMS’s publicly available data from hospitals across the U.S, looking at nearly 4,500 acute-care facilities’ hospital-wide readmission rates and compared them with those facilities’ mortality rates in the six targeted areas used by CMS.
The researchers found that hospitals with the highest rates of readmission were actually more likely to show better mortality scores in patients treated for heart failure, COPD and stroke. In each case, adjusted odds ratios indicated that patients treated at facilities that had more readmitted patients had a fractionally better chance at survival than patients who were cared for at hospitals with lower readmission rates, according to the research.
“But using readmission rates as a measure of hospital quality is inherently problematic,” Brotman said. “High readmission rates could stem from the legitimate need to care for chronically ill patients in high-intensity settings,” especially, he says, in the cases of medically fragile patients who have been kept alive against the odds. To this end, last year, an American Hospital Association (AHA) report noted that reducing readmissions is a “complex undertaking because not all readmissions can or should be prevented; indeed, some are planned as part of sound clinical care.”
In the end, I don’t think Dr. Brotman or anyone else would argue against the idea that hospitals need to do a better job of keeping patients with these serious conditions out of the hospital. To do this, IT will undoubtedly need to be leveraged. As Hagland wrote in his analysis of the penalties being levied by CMS, “Healthcare IT leaders are clearly going to find themselves in the middle of all the activity needed to continuously improve clinical and operational performance, as data collection, data analytics, and the leveraging of data (through dashboards, continuous reporting, etc.) will help to drive continuous performance improvement.
It’s clear that in healthcare’s future of value-based contracting, caring for patients outside of an organization’s four walls will be critical. Anyone in this industry will acknowledge that’s what's on the other side of the health reform door. But maybe, the government shouldn’t put too much emphasis on using these readmission rates as a sticking point to measure quality—especially when the numbers can tell more than one story.