Readmissions are on every hospital’s radar. Hospitals spend significant resources to address what some see as an epidemic of readmissions. The earliest article in the National Library of Medicine on readmissions dates to 1956, discussing the readmission of patients with drug-resistant tuberculosis.¹ In 2020, there were more than 4,900 published articles on readmissions.
As I was writing this article, I received an email invitation to attend the “Fifth National Summit on Reducing Hospital-Acquired Conditions and Readmissions.” The invitation stated, “An estimated $25 billion is spent annually on preventable hospital readmissions that result from medical errors and complications, poor discharge procedures, and lack of integrated follow-up care.”² While no one will argue that preventing avoidable readmissions is appropriate, and I expect this issue of CMSA Today to include many excellent articles by well-known contributors, the regulatory background, my area of expertise, is not so simple.
The Centers for Medicare & Medicaid Services (CMS) began the Hospital Readmission Reduction Program (HRRP) in fiscal year 2013 and began to penalize hospitals based on their readmission rate, with the methodology adjusted over time to add risk adjustments and factoring in the number of dual eligible patients (Medicare and Medicaid) served by the hospital. The statistical methods used by CMS are beyond complex, but suffice it to say that your physicians’ documentation is as important to your readmission rate as it is to your DRG payment. Your actual readmission rate is compared to your expected readmission rate to determine if a penalty will be imposed. The expected readmission rate is based on the likelihood of each patient being readmitted based on the diagnosis codes submitted with the claim. As with the DRG calculation, higher specificity and completeness lead to a more accurate calculation of the expected readmission rate.
It is also worth noting that while the HRRP looks at 30-day readmissions, this was chosen out of convenience rather than based on data. In fact, a study in 2018 found that there is a significant difference in the number of readmissions that are deemed preventable after 7 days.³ So while the HRRP penalizes for almost all readmissions, only a small proportion of those readmissions occurring within a 30-day time frame can be linked to activities by the hospital or physician or any other member of the care team. In fact, as we all learned in grade school, there are more months in a year with 31 days than 30 days, so the choice of 30 days is truly non-evidence-based.
The direct financial penalties from Medicare readmissions are also complex. The HRRP is based on readmissions for several diagnoses – COPD, heart failure, pneumonia, acute myocardial infarction, coronary artery bypass surgery and total joint arthroplasty. The all-cause readmission rate is also calculated for use in other quality programs that may affect hospital payment and quality ratings but does not have any direct effect on payment.
Contrary to common belief, CMS pays for every single readmission with a full DRG payment to the hospital that provides the care during both the index admission and the readmission, with one exception. The exception is that if a patient is readmitted as an inpatient to the same hospital on the day of discharge for a reason related to the prior admission, the hospital is required to combine the two admissions into one, as if the patient was never discharged. But if the second admission on the same calendar day is for an unrelated reason, it is fully paid. At no point do they require or even allow hospitals to combine two admissions unless they occur on the same day. The CMS guidelines on readmissions are found in the Medicare Claims Processing Manual, Chapter 3, section 40.2.5. On the other hand, CMS does allow the Quality Improvement Organizations to review readmissions with the option to deny payment for the readmission. But as stated in the manual, that review is not even limited to 30 days.
It should also be noted that the CMS HRRP, which can have monetary effects, is not a concurrent program. CMS calculates a hospital’s readmission rate each year and compares it to the “expected” readmission rate for the targeted diagnoses. It then uses an incomprehensible calculation to establish a “penalty rate” for the hospital, which is applied to all admissions in future years on a rolling basis. In other words, the hospital is paid for every readmission in a year but if the overall rate is higher than expected based on patient acuity, the hospital may see a reduction in every DRG payment for the next 3 years, up to a maximum of 3%.
While it may be beneficial to prioritize activities, it is impossible to know where a hospital stands in its readmission rate at any one point in time. In other words, a new readmission of a lower-risk patient may be the one that results in the hospital hitting the threshold to now be penalized; that readmission may be a “statistically expected” readmission and have no effect or your cumulative readmission rate may be so low that an additional readmission results in another DRG payment without any effect on your risk of incurring a penalty.
It should be noted that readmissions can occur back at the index hospital, in which case the index hospital gets paid two DRGs but may incur a penalty. Or the readmission could be to another hospital, in which case, the index hospital gets one DRG and then may have a potential penalty but does not get the financial benefit of the payment for the readmission. An analysis of a hospital’s PEPPER readmission data can provide some insight into the percentage of the readmissions that return to the same hospital as opposed to going elsewhere.
The applicability of the HRRP to Medicare patients is also misunderstood. The HRRP applies only to patients with traditional fee-for-service Medicare and not those Medicare beneficiaries who have elected coverage from a Medicare Advantage (Medicare Part C) plan. Medicare Advantage (MA) plans are not restricted by CMS in how they handle readmissions. Rather, it is governed by the contract between the MA plan and the provider. Section 10.2 of the Medicare Managed Care Manual, Chapter 4, states, “MA plans need not follow original Medicare claims processing procedures. MA plans may create their own billing and payment procedures as long as providers – whether contracted or not – are paid accurately, timely and with an audit trail.”
Some MA plans have policies that will automatically combine two admissions based on the time between the discharge and readmission, the relation between the index admission and the readmission, the determination if the readmission was preventable or any combination of these factors. As one can expect, the variations here are myriad. Some plans will deny payment for all readmissions within 30 days; some will allow a readmission within 30 days to be combined with the index admission and submitted for payment; others will analyze each readmission and determine if the readmission is preventable or related to the index admission. In some cases, payers will deny payment for the index admission and pay the readmission. Finally, some will also look for readmissions to the same facility or to another facility owned by the same entity and deny payment for the readmission or index admission. As noted, this is all based on the hospital’s contract with the payer. If a contract is signed allowing them to deny payment for all readmissions in 30 days, they have the right to enforce that provision.
It should also be noted that almost all payers, including Medicare, will exclude any index admission from their penalty program if the patient was discharged from the hospital against medical advice (AMA). While we often think of an AMA discharge as one that is confrontational with an angry, unsatisfied patient, there is no requirement that the patient be angry and unsatisfied, only that the physician has determined that the patient requires continued hospital care and the patient is insistent on leaving. One can imagine an elderly patient with a heart failure exacerbation requiring hospital care who is the primary caregiver for their spouse with dementia. The patient may insist on discharge to return home to care for their spouse, knowing the potential consequences to their health, rather than remaining in the hospital. In this case, ensuring adequate documentation of the physician’s disagreement with discharge and coding of the claim as a discharge “against medical advice.”
While the relatedness and preventability of a readmission is not considered by CMS in the HRRP, it is considered by many other payers and would be a consideration if the readmission was reviewed as part of a Medicare quality of care review. Because of this, it is crucial that these factors be addressed on every readmission. For many readmissions, the patient is being readmitted for another exacerbation of a chronic disease. It is rare for a patient with chronic heart failure to be discharged from an index admission with their heart failure cured, unless the patient underwent a heart transplant. That means that a readmission after 25 days is much more likely due to the natural progression of the disease or a patient-specific factor such as dietary indiscretion or medication non-adherence rather than to an omission by the medical team at discharge. Documentation should address this, noting that “the patient received dietary instructions, but due to his limited budget, he was forced to purchase salt-rich foods, which led to this exacerbation.” As the cost of medications increases, many patients are forced to choose between their medication and paying other bills. Once again, it would be important to note that “the patient reports she was only taking her heart medication every other day due to cost.”
The increasing shift to Medicare Advantage plans has also led to some avoidable readmissions due to the limited choices available to these patients for home healthcare, durable medical equipment, medication formularies and skilled nursing facilities. If a readmission is precipitated by the patient running out of home oxygen from an unreliable provider, the note should indicate that “the MA plan’s DME supplier did not deliver the patient’s supplies as required.” If the contracted home health agency has staffing issues and misses visits leading to a readmission, documentation should reflect that “patient reports that MA plan’s home health RN missed two visits leading to readmission” to ensure the hospital does not get financially penalized by the poor performance of an agency over which it has no control.
On the other hand, if the chart is reviewed from the index admission and the prescriptions were never provided to the patient, the medication reconciliation was done poorly with the patient sent home on two similar blood pressure medications, or the patient’s medical condition was not stable at discharge, the hospital should own up to that mistake and work to prevent it from recurring.
In summary, the regulatory nuances of readmissions are not simple. Efforts at reducing readmissions are not only costly but reducing readmissions also decreases overall revenue in many cases. Efforts at reducing readmissions are laudable but should have a goal of reducing avoidable readmissions because it is the right thing to do rather than as an attempt to reduce or avoid any adverse financial effects because in many circumstances, reduction of readmissions will also mean reduction in revenue. Likewise, hospital culture should be blame-free, but when attributing that blame means the difference between getting paid or not for a readmission, an objective documentation of the circumstances is certainly warranted.
REFERENCES
, MD, FACP, CHCQM, CHRI, is a vice president of regulations and education at R1 RCM Inc. He is the co-author of The Hospital Guide to Contemporary Utilization Review.
Image credit: ISTOCK.COM/BEERKOFF
