The Bounce Back Effect

Critically ill Medicare patients, who are battling for stable health at the end of life, are victims of repeated hospitalizations, especially after being discharged to a skilled nursing facility (SNF).  The cycle of hospitalizations is an indicator of poor care coordination and discharge planning – causing the patient to get sicker after every “bounce back” to the hospital.  Total spending for SNF care was approximately $31 billion in 2011; with an estimated one in four patients being re-hospitalized within thirty days of discharge to a SNF.[1]

Each readmission leads to further test and treatments, higher health care costs, and most importantly, patient suffering.  It is hard to imagine that patients would prefer to spend their last few months of life shuttling from one healthcare setting to another and receiving aggressive interventions that have little benefit to their quality and longevity of life.  The heroic potential of medical care should not compromise the patient’s opportunity to die with dignity.   A hospital is not a place to die.

Medicare beneficiaries are eligible to receive post-acute care at SNFs, after a three day hospital admission stay.  SNFs provide skilled services such as post-medical or post-surgical rehabilitation, wound care, intravenous medication and necessities that support basic activities of daily living.  Medicare Part A covers the cost of SNF services for a maximum of 100 days, with a co-payment of $148/day assessed to the patient after the 20th day.  If a patient stops receiving skilled care for more than 30 days, then a new three day hospital stay is required to qualify for the allotted SNF care days that remain on the original 100 day benefit.  However, if the patient stops receiving care for at least 60 days in a row, then the patient is eligible for a new 100 day benefit period after the required three day hospital admission.[1]  It is evident that the eligibility for the Medicare SNF benefit is dependent on hospitalizations – many of which may be a formality and a source of unnecessary costs.

A large number of SNF patients have limited functional status and are dependent on skilled nursing services.   Patient characteristics to consider:  1) the average age is close to 80 years; 2) one-third of the population is affected by cognitive diseases; 3) 50% of SNF residents have at least three to six limitation in their basic activities of daily living, such as dressing, bathing, mobility or eating; 4) half of the population is being managed for five or more chronic diseases – many of which are considered terminal.   Considering these factors, SNF residents may be receiving aggressive and unnecessary care during a terminal and hopeless stage of their life.  Medicare data from the Dartmouth Atlas has shown a strong correlation between number of physician encounters and SNF re-hospitalizations during the last two years of life. [2] Furthermore, a recent report by the Institute of Medicine this year, assessing the geographic variation of Medicare spending, found that variation in post-acute care utilization, which includes SNF services, accounts for the majority of unexplained variation in total utilization across regions. [3]   Are SNF patients receiving the type and quality of care they desire towards the end of life?  There is no definitive answer.  But what we do know is that advance care planning is inadequate – only one out of three SNF patients has a “do-not-resuscitate” (DNR) directive and less than two percent have a “do-not-hospitalize” (DNH) directive documented. [4] The inability of providers to actively engage patients and their families in a dialogue about the patients’ prognosis and their goals of care is a major concern. Therefore, the timely introduction of palliative care needs to be emphasized.  Attributes of palliative care include the documentation of advance care directives and protection of the patient’s right to establish and maintain control over the type and intensity of medical care they receive.  It also supports the seamless transition to hospice care as the patient continues to get sicker and enters the terminal stage of life.  Advance care planning is often times neglected at nursing facilities and can lead to distressing end-of-life experiences for the patients and their families.  During the last few months of life, comfort measures may be more appropriate than aggressive life sustaining interventions at the hospital ICU.

The economics of the nursing home services pose a challenge for appropriate end of life care.  Higher Medicare reimbursements provide an incentive to nursing homes to hospitalize Medicaid nursing home patients, who are receiving long term care.  After a minimum three day hospitalization, these patients can return back to the nursing home under the Medicare SNF benefit – not to mention the financial benefit of receiving Medicare reimbursement over payment stingy Medicaid.   Furthermore, under the SNF Medicare benefit, regulations prohibit simultaneous enrollment in the hospice benefit – unless the hospice care diagnosis differs from the diagnosis for which SNF care is being provided, which is rarely the case.  Additionally, since Medicare reimburses nursing homes a higher rate for skilled services, patients who transition to the hospice care benefit are not considered “favorable” patients and must pay for room and board out of pocket or through enrollment in Medicaid, which many patients don’t qualify for.[4]  Therefore, access to hospice and comfort care services is mitigated by these financial considerations and affected patients continue shuttling between hospitals and SNFs – and in many cases, succumbing to death in either location.

Let us look at some fee-for-service Medicare spending data from the Dartmouth Atlas to illustrate the correlation between total reimbursement and service specific reimbursement during the patient’s last six months of life, using per beneficiary data from various regions around the country.  Each dot on the graph represents a Hospital Referral Region (HRR).   The price-adjusted reimbursement rates have been further adjusted based on the age, sex and race of the patient population being considered.

As expected, higher spending HRRs spend greater amounts on SNF and inpatient care than lower spending HRRs during the last six months of a patient’s life.  However, there is no correlation between total spending and hospice care reimbursement during the patient’s last six months of life.  Even if higher spending HRRs consist of very sick people that need greater level of inpatient and SNF care, then one would think that hospice care utilization would also be greater at these regional units during the patient’s last six months of life.  But the above graph does not show any such correlation – there is no identifiable trend.  However, if we plot the proportion of hospice reimbursement against total spending for a given HRR, this is what we see:

Interestingly, HRRs that utilize a greater proportion of hospice care in relation to total reimbursement tend to have lower overall spending during the last six months of a Medicare beneficiary’s life.   This population level data is proof of how end of life care spending can be controlled with better resource allocation towards the uptake of hospice care.  Given that all patients considered have a life span of six months; it is safe to say that HRRs that have higher inpatient and SNF spending are not necessarily providing better quality care.  Thus, there is an opportunity to avoid repeated hospitalizations and subsequent SNF care in this fragile patient population.

The healthcare delivery system is in the midst of reform – now is the right time to address the issue of SNF resident hospitalizations.  The Patient Protection and Affordable Care Act has enabled the Center of Medicare and Medicaid Services (CMS) to explore various initiatives aimed at patient care coordination and safety – which will encourage hospitals and SNFs to collaborate and enhance the quality of care and efficient use of resources.  Even for the success of Accountable Care Organizations, greater focus must be placed on strengthening the partnership between hospitals and SNFs in order to support the clinical integration of patient care.  Besides improving the outcome of discharged patients and encouraging appropriate mechanisms for end-of life-care, the payment system needs further reform.  Innovative reimbursement models, such as bundled payments for each patient episode, hopes to circumvent the fee-for-service system and prevent overutilization of medical care.  Currently, CMS is penalizing hospitals who have higher than average risk-adjusted readmission rates for certain diagnoses – the maximum penalty is 1% of all Medicare payments made to the hospital.  Furthermore, under the Medicare Nursing Home Value-Based Purchasing Demonstration, participating nursing homes are being rewarded financially if they are able to decrease their threshold of avoidable hospitalizations and meet certain performance measures – incentivizing and building a culture of accountability might be a solution.  Payment reform can help improve efficiency, prevent repeated hospitalizations, and decreases the regional variation in SNF care utilization and spending.  However, prioritizing advance care planning and eliminating barriers that prevent access to Medicare hospice benefit will be essential to maximize the success of these new payment models.   As the American population ages, more patients will require Medicare coverage for acute hospitalizations and continuing care at SNFs.  Thus, it is crucial that the healthcare system efficiently meets this demand and, more importantly, assure that critically ill patients receive “appropriate care” – the care that they genuinely desire.

Dr. Anubhav Kaul is a recent medical graduate from Ross University School of Medicine and a future medicine resident at Lahey Hospital and Medical Center.  He is currently pursuing a Masters in Public Health at The Dartmouth Institute for Health Policy and Clinical Practice.

Sindhu Kubenderan is pursuing a Masters in Public Health at The Dartmouth Institute for Health Policy and Clinical Practice.  She is a prospective medical student. This post appeared first in Healthcare Finance News.


1)      Medicare Payment Advisory Commission (March 2013). Report to the Congress: Medicare and the health care delivery system. Washington, DC: Medicare Payment Advisory Commission.

2)      Mor V, Intrator O, Feng Z, Grabowski DC. The revolving door of rehospitalization from skilled nursing facilities. Health Aff . 2010;29(1):57-64.

3)      Institute of Medicine. Interim report of the committee on geographic variation in health care spending and promotion of high-value care: preliminary committee observations. Washington, DC: National Academies Press, 2013.

4)      Aragon K, Covinsky K, Miao Y, Boscardin WJ, Flint L, Smith AK. Use of the Medicare posthospitalization skilled nursing benefit in the last 6 months of life. Arch Intern Med. 2012 Nov 12;172(20):1573-9.

5 replies »

  1. My 84 year old father was dismissed from a hospital to a nursing home. The ambulance crew and the nursing home nurses took his vitals and the head of the ambulance crew said we are not leaving him like this. They rushed him back to the ER where he had a temp of almost 103 degrees and he had pneumonia. The following day the hospital sent a representative to insult my intelligence by disclaiming all liability in this 20 minute bounce back. Would you please provide me the number I need to call and report this 20 minute Medicare bounce back? I don’t want this overlooked, and I would like my side of the situation heard by someone other than the hospital that committed the 20 minute bounce back. Thank you.

  2. Hi Anubhav, nice blog shared above. Really great information shared about Skilled Nursing Facility. Very useful information shared for patients with prolonged illness. Awaiting for more posts like this.

  3. The reason why don’t these people have “Do certainly not hospitalize” purchases as well as DNR!!!!!!!! Have been they will and their families recommended they can accomplish this? Otherwise, you could start to?

  4. Why don’t these people have “Do not hospitalize” orders in addition to DNR? Were they and their families advised they can do this? If not, why not?

  5. Excellent Blog Post
    I would also clearly attribute much of the problems cited to the lack of understanding of the Social Services in the hospital (who mean well, but are often hamstrung and often working under their own biases) of the needs for the coordination, setting of limits and expectations with patients and family as well as communicating with the Nursing Homes Social Services.
    A vast amount of enducation needs to be done with Nursing Staffs, Social Services in Hospitals as well as in Nursing Homes.
    All additionally with guidance, care coordination with the PMD.
    Without this, this culture of treating everyone for everything, the most expensive way possible, with the least amount of affect on natural processes and with the greatest cost financially to the system and patients and greatest cost to the patients and families emotionally, will fail to address the problems cited in this blog.