Inpatient hospitalizations account for 32 percent of the total $2.9 trillion spent on health care in the United States. In the majority of cases, it is necessary and appropriate to admit a patient to the hospital. However, patients returning to the hospital soon (e.g., within 30 days) after their previous stay account for a substantial percentage of admissions. For example, nearly 18 percent of Medicare patients were readmitted in 2013., Research has shown that many factors—including a patient’s socioeconomic status, clinical conditions and their communities’ characteristics—can influence hospital readmissions. Readmissions are costly, potentially harmful, and often preventable.
In 2013, CHRT published an issue brief on the major programs aimed at reducing hospital readmissions, including the Hospital Readmissions Reduction Program (HRRP) established under the Affordable Care Act (ACA). The following is an update on the HRRP and other programs previously highlighted.
CMS Hospital Readmissions Reduction Program (HRRP)
Under the HRRP, acute care hospitals with high readmissions rates for certain conditions could have lost up to 1 percent of their total Medicare inpatient reimbursement payments in fiscal year 2013 (FY2013), up to 2 percent in FY2014, and up to 3 percent in FY2015. The penalties are calculated based on a hospital’s readmission rates over a three-year period.
Beginning in FY2013, the U.S. Centers for Medicare & Medicaid Services (CMS) began reducing Medicare payments to hospitals with high readmission rates for heart attack (AMI), heart failure (HF), and pneumonia.,, In FY2015, CMS added two readmission measures: chronic obstructive pulmonary disease (COPD) and elective hip or knee replacement. CMS plans to add another measure, coronary artery bypass graft (CABG), in FY2017. Over the HRRP’s first three years, hospitals nationwide have been penalized an estimated $945 million in total (see Figure 1).
Figure 1: CMS Hospital Readmission Reduction Program: 3-year Implementation Period, U.S
|Year penalty applied||FY2013||FY2014||FY2015
|Performance (measurement) period||June 2008-July 2011||June 2009-July 2012||June 2010-July 2013
|Diagnoses of initial hospitalization||Heart attack|
Hip or knee replacement
|CMS estimate of total penalties||$290 million||$227 million||$428 million
Source: Kaiser Family Foundation
Some studies indicate that the HRRP is showing initial success in reducing readmission rates. A recent study of New York hospitals found that readmission rates for the first three conditions targeted by the HRRP—AMI, HF, and pneumonia—fell between 2008 and 2012. An analysis of CMS’ Hospital Compare database, which publishes hospital readmission rates, shows readmissions rates for AMI, HF, and pneumonia fell nationwide between 1 and 2 percent by mid-2013. A 2013 MedPAC analysis showed that reductions in readmission rates for these conditions were greater than the reduction in overall readmission rates.
While the HRRP may be having an effect on readmission rates for the target conditions, Medicare payment reductions nearly doubled from FY2014 to FY2015. The increase in total penalties could be due, in part, to the addition of new conditions in FY2015, and the fact that the maximum penalty increased from 2 to 3 percent per hospital stay.
Some hospital executives have expressed concern about the HRRP’s method for determining improvement in readmission rates. That is, a hospital with high readmission rates could reduce its rates from one year to the next, but still lose Medicare reimbursement because penalties are assessed based on a hospital’s readmissions rate compared to the national average which hospitals say unfairly penalizes them even if some improvement is made. For example, Beaumont Health System in Royal Oak, Mich., says it has reduced its readmissions rates in recent years, but could lose a projected $3.86 million in Medicare reimbursement in FY2015 because the system’s rates were higher than the national average.
HRRP Implementation in Michigan
Fifty-four Michigan hospitals were penalized 0.42 percent on average in FY2013, while 55 hospitals were penalized an average of 0.39 percent in FY2014 (Figure 2). In FY2015, when CMS added two measures to the HRRP, 71 Michigan hospitals, three-quarters of the state’s eligible hospitals, were penalized an average of 0.64 percent.
In each of the program’s first three years, Michigan’s average penalty was equal to or higher than the national average (Figure 2). According to Kaiser Health News, only 20 states had higher average penalty rates than Michigan in FY2015.
Figure 2: Michigan Hospital Penalties, FY2013 - FY2015
| ||# of MI Hospitals Penalized||Average Penalty||% of Eligible Hospitals Penalized||
Source: CHRT analysis of data from the Centers for Medicare & Medicaid Services, accessed through Kaiser Health News
According to MPRO, Southeast Michigan hospitals have higher readmission rates than the rest of the state. In total, Michigan hospitals are being penalized an estimated $25 million in FY2015, but those penalties are regionally concentrated: hospitals in Southeast Michigan will lose $21 million (84 percent of Michigan’s total penalties) in Medicare reimbursement, according to an MPRO analysis. MPRO reported that the average FY2015 penalty for hospitals in Southeast Michigan is 0.72 percent, up from 0.42 percent in FY2014.
Key Readmission Reduction Initiatives in Michigan
The HRRP has spurred a significant amount of activity to curb hospital readmissions. In 2013, CHRT identified 10 readmissions initiatives used by hospitals and health plans nationally. Six of these initiatives have been implemented in Michigan (Appendix A provides an update on the other four programs). Those programs implemented in Michigan included:
- Care Transitions Intervention® (CTI): Transitions Coaches® (e.g. advance practice nurses, registered nurses, and social workers), trained through the CTI program, review a patient’s discharge plans at the hospital, visit the patient at home within 48 to 72 hours of discharge, and call the patient three times within the first 28 days after discharge.
- Project Re-Engineered Discharge (RED): Nurses coordinate patients’ transitions home, while pharmacists call patients after discharge to review medications and communicate any problems to the primary care provider.
- Transitional Care Model (TCM): Advanced practice nurses provide home visits to high-risk elderly patients for three months, and are available by phone seven days a week.
Hospital to Home (H2H): A central clearinghouse provides hospitals and cardiovascular care providers with information and tools for improving care transitions and reducing readmission rates among patients who experienced heart failure or a heart attack.
- Project BOOST (Better Outcomes for Older adults through Safe Transitions): A toolkit that offers hospitals and primary care providers evidence-based clinical intervention tools for improving care transitions.
- STate Action on Avoidable Readmissions (STAAR): A pilot program that focuses on building community-based and state-based partnerships to improve care transitions.
Each of the six initiatives target one of three levels for intervention—patient, system, and community levels—and are supported by varying degrees of evidence. The following is a summary of their implementation in Michigan, and an introduction to BCBSM’s new initiative to help reduce hospital readmissions in the state.
To date, the initiatives that have been most successful in reducing hospital readmissions—CTI, Project RED, and TCM—focus on transitioning patients from the hospital to the home setting. In randomized controlled trials for each of the three initiatives, data showed reductions in readmissions and overall health care costs.,,
Each model is used by payers, hospitals and other providers in Michigan, but CTI appears to be the most used and serves as the model for BCBSM’s Care Transitions to Home and Care Transitions to Skilled Nursing Facility programs. In these BCBSM care management programs, nurses call or visit high-risk BCBSM members and their clinicians to provide the following services post-discharge:
- Medication reconciliation
- Access to services such as home care and durable medical equipment
- Coaching on signs of worsening symptoms
- Assistance making follow-up primary care appointments
- Triage for referral to other programs
- Assessment of caregiver support
Two of the six initiatives, H2H and Project BOOST, have developed tools, such as standardized forms and methods for transmitting patient information from a hospital to primary care and specialty providers, to help providers implement evidence-based practices for improving the hospital discharge process. Many of these tools are based on principles found in the more-widely used initiatives CTI, Project RED, and TCM.
H2H, an initiative of the American College of Cardiology, helps hospitals and cardiac providers achieve attainable quality improvement goals within their organization by helping to develop goal statements for projects, define measures of success, and assess their progress. An example of H2H’s quality improvement projects is See You In 7, which focuses on helping providers develop a process for scheduling follow-up appointments within seven days of discharging patients. However, no data has demonstrated that the program is associated with reducing hospital readmissions. Approximately 600 hospitals nationwide have implemented H2H, including hospitals in Michigan.
In comparison, Project BOOST offers evidence-based clinical interventions for a wide array of conditions, including risk assessment tools and risk-specific discharge preparations. Data has shown that the program is helping hospital units reduce readmissions. As of June 2015, the program is only implemented in 196 hospitals nationwide.
In 2013, Project BOOST was implemented in 24 hospitals and provider organizations in the Michigan, the majority of which implemented the program through the Michigan Transitions of Care Collaborative (M-TC2). The M-TC2, a collaborative quality initiative (CQI) of Blue Cross Blue Shield of Michigan (BCBSM), was based on Project BOOST and rewarded physician organizations for taking steps to improve care transitions and reduce readmissions. At the end of 2014, BCBSM retired the M-TC2, reducing Project BOOST’s presence in the state. The project coordinators for the CQI found it had limited impact because it only provided financial incentives for physician organizations and not hospitals.
STAAR, which concluded in 2013, aimed to reduce hospital readmissions by improving care transitions but not at the patient level. Launched in 2009 by the Institute for Healthcare Improvement (IHI), the goal of the four-year pilot program (2009-2013) was to form relationships among providers and other stakeholders to effectively coordinate patient care across settings. IHI selected Michigan and three other states—Massachusetts, Ohio, and Washington—to participate. Michigan’s program, led by the Michigan Health and Hospital Association’s Keystone Center for Patient Safety and Quality and the Michigan Peer Review Organization (MPRO), was required to implement two interventions:
- Cross-continuum teams: Community-based teams of hospitals, providers, support services, and patients worked together to align policies and practices across settings to improve the hospital discharge process.
- State-level steering committee: A multi-stakeholder, state-level committee charged with (1) coordinating and aligning complementary programs across the state, (2) identifying and mitigating systemic barriers, and (3) promoting a common framing of the issues through provider and stakeholder networks.
STAAR concluded in 2013 due to a lack of continued funding from the Commonwealth Fund. Data is not yet available on the program’s impact in the state.
In February 2015, BCBSM introduced a new CQI, the Integrated Michigan Patient-centered Alliance on Care Transitions (I-MPACT), to reduce avoidable readmissions and post-discharge ED utilization rates. I-MPACT replaces BCBSM’s M-TC2 CQI, which BCBSM retired at the end of 2014. BCBSM found that M-TC2 was implemented in fewer than expected hospital units largely because it did not provide incentives for hospitals, and the staff capacity at participating physician organizations and hospitals to extract data for the CQI was limited.
I-MPACT, which is based at the University of Michigan Health System, develops care coordination partnerships between hospitals, physician organizations, other post-acute care providers and community organizations. Unlike M-TC2, I-MPACT is a hybrid CQI (includes incentives for hospitals in addition to physician organizations) and does not standardize interventions statewide, but is organized regionally (based on 16 hospital referral regions in Michigan), allowing providers to tailor care transition interventions to the community and patient population that they serve.
Participants for the first cohort will be finalized in Fall 2015 and will include three to four regional provider teams. The target population will include adult members (18-64 years), with an emphasis on certain conditions (e.g., health failure). The CQI is expected to launch in 2016.
Providers and payers in Michigan are increasingly investing in initiatives to reduce hospital readmissions, due in part to the HRRP established under the ACA. Many readmission reduction initiatives have reacted to the HRRP by tailoring much of their focus to conditions targeted by HRRP (e.g., heart failure). Future changes to the HRRP could impact where hospitals choose to focus their resources and the strategies they use to reduce readmissions.
One such change that could impact the program is the inclusion of patients’ socioeconomic status (SES), which CMS does not currently adjust for when calculating penalties. For example, Congress has introduced bi-partisan legislation to require the program to risk adjust for SES, but it has yet to pass. This concept is also being explored by the National Quality Forum. In 2015, the organization launched a two-year trial to study the implications of considering patient-level adjustments factors, including SES, in the HRRP and other pay-for-performance programs.,, This change would be particularly relevant for safety-net hospitals, such as those located in Southeast Michigan, that are caring for large low-income populations, and are more than twice as likely to be penalized as hospitals caring for higher-income patients.,
Update on key readmission reduction programs not implemented in Michigan
The following is an update on the four out-of-state programs that CHRT highlighted in a 2013 publication on readmission reduction initiatives. Although these programs are not currently implemented in Michigan, they may serve as a resource as providers and health plans explore opportunities for improving patient care.
1. Collaborative on Reducing Readmissions in Florida (http://collab.fha.org)
The Collaborative on Reducing Readmissions in Florida initiative, staffed by the Florida Hospital Association, provides 75 participating hospitals with a shared learning environment for reducing readmissions.
In August 2015, the collaborative will launch its second phase and begin working with hospitals to develop and implement quality improvement initiatives.
2. Optum CarePlus (www.optum.com/landing/care-management)
Developed in 1987, the CarePlus model (initially known as EverCare) uses advanced practice nurses and care managers to provide individuals in long-term care facilities with in-facility clinical and case management services. The model has been shown to reduce health care spending, hospitalizations, and emergency department visits.,
Today, CarePlus is owned by Optum, a health care services platform under UnitedHealth Group. The vendor provides a broad range of Medicare, Medicaid, and private long-term care delivery and coordination programs to payers based on the CarePlus model.,
3. Guided Care® (www.johnshopkinssolutions.com/solution/guided-care-2/)
Guided Care is a chronic care management program that places nurses in primary care offices. The nurses supplement care to elderly patients with chronic illnesses by providing in-home assessments, coordinating care, and educating patients and their caregivers about self-management skills.
A 2013 study evaluating the program’s effectiveness showed mixed results. A 32-month randomized control trial found the model reduced utilization of home health care by 29 percent for high-risk older patients, but did not significantly improve patients’ health status.
4. Preventing Readmissions through Effective Partnerships (www.ihatoday.org/IHA-Institute/PREP.aspx)
Preventing Readmissions through Effective Partnerships (PREP) is a program used by hospitals in Illinois in partnership with Blue Cross Blue Shield of Illinois and the Illinois Hospital Association. The program aims to improve the aggregate statewide 30-day readmission rate in Illinois from 20.3 percent to below 17.5 percent. Through PREP, hospitals receive technical assistance and access to resources from a number of sources, including Project RED and Project BOOST.
The program concluded at the end of 2014, but an evaluation has not yet been published.