Reassessing hospital readmission penaltiesBMJ 2013; 346 doi: https://doi.org/10.1136/bmj.f1043 (Published 14 February 2013) Cite this as: BMJ 2013;346:f1043
- Douglas Kamerow, chief scientist, RTI International, and associate editor, BMJ
The Patient Protection and Affordable Care Act of 2010 is midway through its four year implementation period. The centerpiece of the law is the requirement for near universal healthcare insurance coverage. It becomes effective in 2014.
So much for the patient protection part of “Obamacare.” What about the affordable care aspect: cutting the hugely expensive costs in the US healthcare system? The act has always been weaker in that regard, but it does have some cost cutting measures. One that has already raised a ruckus is section 3025, the Hospital Readmissions Reduction Program.
Under this program the Centers for Medicare and Medicaid Services (CMS) began reducing payments to hospitals with excess readmissions of patients with certain conditions in October last year. CMS began with three sentinel conditions—heart failure, myocardial infarction, and pneumonia—and is penalizing hospitals whose 30 day rates of readmission of patients with these diagnoses are higher than would be expected from each hospital’s case mix. In the first year of the program 2000 hospitals will lose about $300m (£194m; €225m) in payments, it is estimated.1 The penalties will rise next year and again in 2015. Needless to say, this program has received a lot of attention.
The idea that there are many preventable readmissions has high face validity. Medicare generally pays hospitals a fixed rate, on the basis of diagnosis codes, for each admission. Shorter stays thus mean lower costs and higher profits for the hospital, giving them an incentive to discharge patients as soon as possible. If patients are then readmitted, all that usually happens is that they generate another payment. It is thus no surprise that 25% of Medicare patients who have been hospitalized with heart failure, 20% of those with myocardial infarction, and 18% of those with pneumonia are readmitted within 30 days of discharge.2
But there are a number of concerns about using the readmission metric as a quality measure and cost reducer.
Large referral hospitals and those that have a high proportion of poor and underserved patients complain that the case mix adjustment that CMS applies does not adequately account for the severity of their patients’ illnesses and the paucity of their social supports. One doctor at a charity hospital said, “For us it’s not a readmissions penalty; it’s a mission penalty.”1 Some data bear this out. An analysis of more than 3000 hospitals’ penalties found that large teaching hospitals and safety net hospitals were significantly more likely than other hospitals to be penalized under the program.3
Even if fair adjustments could be made, however, critics have described other problems with the readmissions penalty.4 5 One concern is that most readmissions may not actually be preventable. The vast majority of diagnoses among patients readmitted within 30 days are not the same as those on initial admission.2 Patients, especially elderly and disabled patients who receive Medicare, become deconditioned and vulnerable on hospitalization. They then acquire other illnesses and injuries and are readmitted to hospital at a high rate. Estimates of the proportion of rehospitalizations that are truly preventable range from 10% to 30%.4
Another problem is that some readmissions may, paradoxically, reflect unusually good care (keeping patients alive who may have died in other hospitals, resulting in a sicker patient population at discharge) or better access to hospitals (permitting such patients to be rehospitalized rather than dying outside hospital).
Further, it seems unfair to hold the discharging hospital accountable for all the services needed to minimize the possibility of acute readmission. Certainly hospitals should work to ensure a safe transition to home or a care facility on discharge, with accurate transfer of records and medication reconciliation, but it is hard to imagine that hospitals alone can create the financial, social, and health support systems that are necessary to optimize outcomes.
Some people are concerned that so much effort is being expended trying to reduce the penalties associated with readmissions that other, possibly more important and valuable programs are being ignored. If only a small proportion of 30 day readmissions are preventable, then a lot of effort is being expended for a very small potential benefit. One doctor noted the fact that the readmission program has the largest penalties in Obamacare, “and that’s just crazy.”1 And if hospitals are obsessed with reducing readmission rates, will that just increase use of emergency departments, where patients can be boarded for days at a time so that their treatment won’t “count” as a readmission?
Still, everyone involved closely with the US healthcare system can see that it has become increasingly fragmented, with patients seeing many specialists inside and outside the hospital.5 As a result, transitions from hospital are often botched, resulting in poorly coordinated, duplicative, and sometimes missing and dangerous care. No doubt preventable hospital readmissions result from all this. The question is whether focusing on the 30 day readmission rate is the best way to lower costs and improve outcomes.
This is, of course, the central issue of the quality improvement and patient safety movements in general. How do you use the relatively blunt instruments of quality measures and financial penalties over widely disparate settings to achieve desirable outcomes? It is a huge challenge, but it seems that punishing hospitals for readmissions is not the most effective way to go about it.
Cite this as: BMJ 2013;346:f1043
Douglas Kamerow is the author of Dissecting American Health Care (www.kamerow.com/Dissecting_American_Health_Care.html).