The Impact of Hospital-Based Palliative Care on Hospital Costs and Health Care Spending

Health care spending is a topic of great interest in the United States. Reining in costs and improving outcomes while respecting patient preferences is challenging. Hospitals want to provide their patients with the best care, but they also face very real financial constraints; this is particularly true for institutions that serve rural and safety-net populations. In our literature review, we found a mixed picture about financial incentives and barriers to hospital-based palliative care. To further complicate the picture, some studies examine costs, and others examine charges.

Several studies suggest that the cost of providing the full range of services for a patient is lower when the patient receives palliative care than when the patient does not receive such services. In a study of four diverse urban New York State hospitals, researchers calculated that Medicaid patients receiving palliative care between 2004 and 2007 incurred on average $6,900 less in hospital costs for a given admission than those who received usual care (Morrison et al., 2011). However, it is not entirely clear whether costs associated with palliative care are subtracted from that total. In the study, a patient was identified as receiving palliative care if the attending physician requested a palliative care consultation, the patient was seen and evaluated by the palliative care team at least once, and at least one set of recommendations was made by the palliative care team to the patient’s primary team. Compared with a matched group of patients not receiving palliative care services, hospital patients receiving palliative care services incurred, on average, $4,098 less in costs per admission for live discharges and $7,563 less per admission for patients who died in the hospital. The average total cost per admission was $36,741 for a patient discharged alive who received usual care and $32,643 for a matched patient who also received palliative care. For just ICU costs, patients discharged alive who received palliative care incurred on average $2,678 lower costs than those who did not receive palliative care.

In another study of data from 2002 to 2004 of eight geographically and structurally diverse hospitals, laboratory and ICU costs for patients over the age of 18 receiving palliative care who were discharged alive were significantly lower than the laboratory and ICU costs of matched patients not receiving palliative care (Morrison et al., 2008). The hospital costs of patients receiving palliative care who died in the hospital were lower in laboratory, ICU, and pharmacy fields. Patients discharged alive incurred $1,696 less in costs per admission, and those who died incurred $4,908 less in costs per admission compared with matched patients receiving usual care only. Investigators excluded patients with stays less than 7 and more than 30 days for two main reasons: the unlikelihood that short-stay patients would receive palliative care consultations and the difficulty of generalizing data on the long-stay patients.

Researchers have also found that cost reductions extend beyond the hospital stay. A three-site randomized trial of hospitalized patients enrolled in the study between June 2002 and December 2003 found that mean health costs in the 6 months post-discharge were $6,766 lower per patient among those who had received hospital-based palliative care compared with usual care only (Gade et al., 2008). The primary driver was a reduction in readmission costs; while there was no difference in the number of hospital readmissions, patients receiving palliative care had significantly fewer ICU stays on readmission. Intervention patients readmitted to the hospital were again followed by the palliative care team.

Costs reflect all health services provided in the 6 months after discharge. With palliative care staffing costs of $1,911 per patient, as determined by the time spent by each team member multiplied by their hourly wage and benefits, the net savings were $4,855 per patient (Gade et al., 2008). The interdisciplinary inpatient palliative care consultative service consisted of a palliative care physician and nurse, hospital social worker, and chaplain. At all sites, the entire team met prior to every consultation, met with the patient and family, and then met again to develop a plan and organize follow-up. The team was available Monday through Friday, and a palliative care physician was on call after hours. Upon discharge, the team provided a palliative care discharge plan to the primary care physician. Study participants were limited to those 18 years of age and older who were hospitalized with at least one life-limiting diagnosis and whose attending physician noted he or she “would not be surprised if the patient died within 1 year.”

In a study of a public hospital in Indiana in 2005, patients who received a palliative care consultation service incurred $6,907 less in hospital charges than demographically and clinically similar patients who did not receive a consultation service (Bendaly, Groves, Juliar, & Gramelspacher, 2008). Note, however, that this particular study examined hospital charges rather than hospital costs, and that the study only included patients aged 50 or older who died during a hospitalization of at least 3 days.

Achieving meaningful cost reductions while promoting quality is clearly possible when hospitalized patients receive palliative care together with their standard treatment, which reduces costs by an average of $4,098 per admission for patients discharged alive (Morrison et al., 2011). If just 2 percent to 6 percent of Medicaid patients discharged from a hospital received palliative care while in the hospital, annual Medicaid hospital spending in the State of New York could be reduced by $84 million to $252 million, respectively. For this particular calculation, the investigators assumed that every New York hospital with 150 beds or more had a fully operational palliative care team.

Hospital-Based Palliative Care Reimbursement

Although evidence suggests that further proliferation of hospital-based palliative care is desirable, financial hurdles can be substantial. Currently, no palliative care benefit or clearly defined method of reimbursement exists for hospital-based services. Coverage for palliative care varies widely. Moreover, identifying discrete services as being part of a comprehensive palliative care approach is difficult.

Under Medicare Part A, hospitals do not receive a payment adjustment for palliative care services. Clinicians can bill palliative care consultant services through Medicare Part B and some commercial insurance plans (Center to Advance Palliative Care, 2011; National Hospice and Palliative Care Organization, 2001). However, interdisciplinary team members such as chaplains, social workers, and bereavement counselors may not receive reimbursement from public or private payers for provision of services essential to the success and spirit of palliative care.

As a result of the lack of reimbursement from payers, hospitals typically provide at least half of the overall palliative care program funding (Center to Advance Palliative Care, 2011). A 2014 study of hospital-based palliative care programs in California found that the percentage of programs with chaplains decreased from 77 percent in 2007 to 57 percent in 2011, and the percentage of social workers decreased from 80 percent to 58 percent (Pantilat, O’Riordan, & Bruno, 2014). The percentage of programs comprising clinicians from four or more disciplines also decreased from 61 percent in 2007 to 45 percent in 2011, perhaps reflecting the challenges of providing palliative care services for which the hospital receives no reimbursement.

Small, rural, and safety-net hospitals may face additional barriers in funding and staffing palliative care programs (Morrison & Meier, 2011). These hospitals may be less likely than other hospitals to be part of a health system with sufficient resources to fund a palliative care team.

Financial Research Considerations

The exact degree to which hospitals may benefit financially from implementation of a palliative care program is unclear. This outcome likely varies based on the quality and breadth of the programs and their patient and payer mix. What is evident is the need for additional research into how savings are allocated when the hospital costs associated with a patient who receives palliative care are lower than costs for a matched patient who does not receive palliative care. One key point is the mix of stakeholders who may benefit; these can include patients, hospitals, and payers, depending on the hospital and the type of insurance coverage, if any, the patient has.

Insofar as total hospital costs are lower because of palliative care, depending on the type of insurance the patient may have and the length of stay, the patient may save money. For instance, in 2015, Medicare beneficiaries had to pay a $1,260 Part A hospital inpatient deductible for each benefit period (Centers for Medicare & Medicaid Services [CMS], 2014). The benefit period begins the day a patient is admitted as an inpatient and ends when he or she has not received any inpatient hospital care for 60 days in a row. If the patient is readmitted after a benefit period ends, a new period begins, and the patient must pay the inpatient deductible for the new benefit period.

Patients are responsible for different coinsurance amounts depending on the length of the benefit period. For instance, Medicare patients pay no coinsurance during days 1–60 of the benefit period and pay $315 in coinsurance per day between days 61 and 90. Therefore, a Medicare patient could save money if palliative care reduces his or her total hospital costs below the $1,260 deductible threshold or if palliative care can reduce the length of stay past day 60. If palliative care can reduce the likelihood of hospital readmissions, Medicare patients could also be less likely to have to pay the inpatient deductible for a new benefit period.

Patients with private insurance are responsible, on average, for an 18 percent coinsurance rate for hospital admissions after meeting an average $346 annual hospital deductible (Kaiser Family Foundation and Health Research & Educational Trust, 2013). Table 1 illustrates how palliative care could reduce coinsurance costs for patients who have already met their deductible. The table utilizes the $4,908 average reduction in costs per admission for patients who received palliative care alongside usual care and who died in the hospital (Morrison et al., 2008 ). This simple estimate suggests that families of patients who received palliative care and died in the hospital might save around $883 for such a stay.

Table 1

Hospital costs to patients with coinsurance who died in the hospital.

Determining whether hospitals save money depends on how they are reimbursed by payers. For instance, if total costs are lower because of reduced ICU stays and fewer unnecessary drugs, do hospitals share in any savings? If so, to what extent? Hospital expenses may be reduced if the hospital does not have to conduct as many tests and procedures. Reduced hospital length of stay for patients receiving palliative care could also lead to bed availability for other patients. However, hospital revenue may be lower if the case is consequently assigned a lower-weight diagnosis-related group (DRG).

Research is needed to better understand the relationship between palliative care and reimbursement, considering such hospital variables as size, geography, patient population characteristics, ownership (such as for-profit or nonprofit), and type (such as an academic medical center, charity hospital, rural hospital, etc.). For instance, in the randomized control trial discussed previously that looked at post-discharge cost savings, the three participating hospitals were part of the same integrated delivery health plan, and the study was approved by that health plan (Gade et al., 2008). Policymakers need to consider how characteristics of hospitals could influence the potential for cost savings and how that may affect hospitals’ decisions about implementing such programs.

Researchers also need to determine how the expense of maintaining the palliative care team itself compares with any savings experienced by the hospital. An important concern is that hospitals must often absorb a large percentage of the cost of the palliative care team because of low or no reimbursement for many of their services. Even if hospitals may reduce their cost of treating patients with time, without sufficient reimbursement, they may be unable to begin or continue providing palliative care, at least with the types of disciplines and services that are essential to the palliative care team composition.

Finally, with respect to payers, one would expect the federal Medicare program to save money because its payments to hospitals would be reduced. In particular, Medicare may be able to expect savings on patients who would have been high-cost outlier patients without the intervention of palliative care.

Private payers may save money depending on their arrangements with hospitals. For instance, if an insurer will pay only a fixed sum for a certain diagnosis or treatment, and if palliative care reduces total hospital costs, the cost reduction may not translate into savings for the payer. Alternatively, if the payer covers a certain percentage of hospital costs, and palliative care reduces hospital costs, then the payer could save money. Several research questions remain unanswered regarding whether different payers may share in savings from hospital-based palliative care, given the considerable number of possible variables involved. Payers may have both a quality and a financial incentive to promote palliative care based upon their arrangements with hospitals.

Implementation Research Considerations

Research on the relationship between the nature of the palliative care team and the quality of care they deliver, as well as the health and financial outcomes of that care, is needed to maximize the effectiveness of palliative care. Palliative care teams by nature are interdisciplinary, requiring experts from several fields to be truly successful. Such teams need to work well together, have experienced members, and be willing to coordinate and communicate with a variety of disciplines and providers. The interdisciplinary palliative care teams of the hospitals assessing Medicaid patients received training that was in accordance with the National Quality Forum’s framework and preferred practices for palliative and hospice care quality (Morrison et al., 2011). Although not all palliative care teams may need such training, policymakers should know whether team members must have a certain level of expertise or training to be effective in improving quality of care and patients’ health outcomes and satisfaction.

Additional research is needed on which hospitalized patients might benefit the most from palliative care. For instance, the Morrison and colleagues cost studies referenced earlier were limited to patients with diagnoses of severe and potentially life-threatening ailments, such as cancer, HIV/AIDS, heart failure, and COPD; future studies should examine whether palliative care produces better quality and financial outcomes based on diagnosis and prognosis (Morrison et al., 2008; Morrison et al., 2011). Palliative care may have a more meaningful impact on patients with chronic pain and previously unmanageable symptoms than those with lower levels of symptoms. Based on patient exclusions in the aforementioned cost studies, future research goals should also include understanding the quality and financial benefit of palliative care for short- and long-stay patients (fewer than 7 days and more than 30 days), as well as for pediatric palliative care patients.

Policy Considerations

Although some hospitals must absorb significant palliative care costs under current payment mechanisms, hospital-based programs have been growing. Given the established quality-of-life improvements for patients, however, payment reform is still needed to promote faster and more comprehensive proliferation of services. Restructuring palliative care reimbursement or including palliative care in a public sector benefit program should be considered to ensure that palliative care is readily accessible to all patients who might benefit from such services.

In 2014, the Institute of Medicine (IOM) recommended that government and private health insurers provide financial incentives for including palliative care for patients at the end of life, recognizing opportunities to improve quality and reduce costs (Committee on Approaching Death, 2014). As palliative care can be beneficial to patients and families at any point along the illness trajectory, the IOM recommendation may support movement toward comprehensive reimbursement for all palliative care activities. The creation of reimbursable billing codes, such as for pastoral visits, bereavement counseling, and discussions about treatment options, is one possible avenue for ensuring that palliative care is reimbursed and available for all patients who would benefit.

Experts have previously suggested that payers might not currently offer comprehensive palliative care coverage because they have no existing reimbursement mechanism upon which to model coverage (Lubell, 2010). The Center to Advance Palliative Care (CAPC), an organization at the Icahn School of Medicine at Mount Sinai (New York City) dedicated to increasing the availability of palliative care, has recognized the development of reimbursable billing codes as a best practice of payment innovation (Warner & Gualtieri-Reed, 2014).

Policymakers could encourage payers and hospitals to improve identification and differentiation of palliative care practices through billing codes. In July 2015, the Center for Medicare & Medicaid Services (CMS) proposed paying physicians for optional end-of-life discussions with Medicare patients by activating two CPT billing codes (Lowes, 2015). However, CMS did not set a rate for the codes, so Medicare administrative contractors, who are normally private health insurers, will generally be responsible for the decision to reimburse physicians when processing Medicare claims. While these discussions are advance-care planning specific for patients at the end of life, we believe this remains positive movement toward reimbursement for additional care planning discussions.

Expanding the existing Medicare hospice benefits to allow Medicare patients who are not terminally ill to receive palliative services without forgoing curative care is another potential route to consider. Twenty percent of Medicare beneficiaries have at least five chronic conditions, and two-thirds of Medicare spending covers these beneficiaries’ care (Belluck, 2014). These types of patients in particular, despite not necessarily being terminally ill, could benefit from the improved oversight and care coordination that are expected to accompany palliative care.

The Medicare Care Choices Model demonstration may provide valuable insights into the joint provision of palliative and curative care. In this program, Medicare and dually eligible (Medicare and Medicaid) beneficiaries can receive care from hospice providers while concurrently receiving curative treatment, not necessarily while in the hospital. Hospices will receive $400 per beneficiary per month to provide services. The demonstration is restricted to patients with certain diagnoses (e.g., COPD, advanced cancer, HIV/AIDS), although patients with other illnesses may also benefit from palliative services. Furthermore, the model is restricted to patients who meet hospice eligibility requirements, not patients at any stage of their illness. Depending on the quality and financial outcomes of the demonstration, the evaluation will inform the costs and benefits of reimbursing palliative care and reimbursement strategies. Due to robust interest, CMS has already expanded the model from an originally anticipated 30 Medicare-certified hospices to over 140 hospices and extended the model from 3 to 5 years, targeting around 150,000 Medicare beneficiaries. The first phase of the model began in January 2016, and the second phase of participating hospices will begin in January 2018 (CMS, 2015).

An increase in palliative care might be a natural outgrowth of new payment and delivery systems. One example is accountable care organizations (ACOs) that are participating in CMS initiatives that seek to achieve cost savings and improve their performance in areas such as readmissions, care coordination, and concordance with patient preferences. As previously noted, additional research is needed on the distribution of any savings or costs among stakeholders, but ACOs with a financial incentive to reduce costs while maintaining quality care may be in a stronger position than traditional health systems to accommodate any reduced revenue.

The prevalence of palliative care teams in hospitals across the country is not uniform. The CAPC found that while 90 percent of hospitals with at least 300 beds had a palliative care team in 2015, only 29 percent of hospitals with fewer than 50 beds had teams (Morrison & Meier, 2015). Interestingly, for-profit hospitals of any size are less likely to have a palliative care team than nonprofit hospitals. The CAPC has noted the reasons for this discrepancy are unknown. Perhaps for-profit hospitals are more hesitant than nonprofits to absorb costs for the palliative care team despite potential future savings.

Geography also plays an important role in determining accessibility of palliative care to patients and in distributing potential savings. In 2011, in the Northeast, 73 percent of hospitals with 50 or more beds had palliative care teams, as compared with 51 percent in the South (Morrison & Meier, 2011). Policymakers may wish to consider existing barriers when crafting policy to promote palliative care, as lack of reimbursement may constrain smaller and rural hospitals in particular from providing services to all who might benefit.

Limitations

A limitation of our literature review is that it did not meet the standard requirements for a formal systematic review, as doing so was not our purpose as part of a larger background report providing context about hospice care. We believe that we identified the most pertinent and recent scientific literature and research organization reports on palliative care financing. Although literature exists on the possibility of cost savings from hospital-based palliative care, a substantial number of research questions remain outstanding, largely regarding which stakeholders may benefit from savings under different financial reimbursement arrangements among patients, payers, and hospitals. In-depth research on how the hospital costs of palliative care teams compare with hospital savings must be undertaken in order to inform any policy decisions and business cases. We also note that costs and savings as discussed here do not include the monetized value of life quality or quantity or utilize quality-adjusted life years (QALYs), and thus may not include the full impact of palliative care interventions. As other researchers have found, this type of cost-effectiveness is challenging and may lead to studies assessing costs separately from outcomes (Smith, Brick, O’Hara, & Normand, 2014).