Ellen Meara, PhD

 Alternative payment models, such as accountable care organizations, hold provider groups accountable for an assigned patient population, but little is known about unassigned patients. We compared clinical and utilization profiles of patients attributable to a provider group with those of patients not attributable to any provider group.

 Cross-sectional study of 2012 Medicare fee-for-service beneficiaries 21 years and older.

 We applied the Medicare Shared Savings Program attribution approach to assign beneficiaries to 2 mutually exclusive categories: attributable or unattributable. We compared attributable and unattributable beneficiaries according to demographics, dual eligibility for Medicaid, nursing home residency, clinical comorbidities, annual service utilization, annual spending, and 1- and 2-year mortality. We estimated multivariate regression models describing correlates of attribution status.

 Most beneficiaries (88%) were attributable to a provider group. The remaining 12% were unattributable. Beneficiaries unattributable to any provider group were more likely to be younger, male, and from a minority group; to have disability as the basis for enrollment; and to live in high-poverty areas. Unattributable beneficiaries included 3 distinct subgroups: nonusers of care, decedents, and those with healthcare service use but no qualifying evaluation and management visits. Many unattributable Medicare beneficiaries had minimal use of healthcare services, with the exception of a small subgroup of beneficiaries who died within the attribution year.

 Attribution approaches that more fully capture unattributable patients with low service use and patients near the end of life should be considered to reward population health efforts and improve end-of-life care.

Given the reliance of alternative payment models on assignment of patients to provider groups, we compared clinical and utilization profiles of patients attributable to a provider group with those of patients unattributable to any provider group.

We estimated multivariate regression models describing correlates of attribution status.

Most beneficiaries (88%) were attributable to a provider group, whereas 12% remained unattributable.

Beneficiaries unattributable to any provider group included 3 distinct subgroups: nonusers of care, decedents, and those with healthcare service use but no qualifying evaluation and management visits.

Attribution approaches that more fully capture unattributable patients with low service use and patients near the end of life should be considered to reward population health efforts and to improve end-of-life care.

By holding networks of healthcare providers responsible for the total cost and quality of care for a designated population, the accountable care organization (ACO) model creates incentives to coordinate care across providers, reduce unnecessary spending, and improve the quality of care.

 and by January 2017, Medicare held ACO contracts with 525 organizations serving more than 10 million beneficiaries.

 Medicare ACO contract participants that meet quality benchmarks are eligible to share the savings that they generate.

 An ACO’s performance is evaluated based on Medicare methods of measurement and attribution. Researchers have evaluated outcomes of ACOs among attributable beneficiaries,

 but no study has described unattributable beneficiaries.

In Medicare ACO contracts, beneficiaries are attributed to organizations based on their use of primary care services from eligible providers,

 so attribution can influence an organization’s performance under an ACO contract. Among organizations serving patients with complex clinical and psychosocial needs, for example, quality metrics may be hard to achieve, giving organizations incentives to avoid accepting such patients. Conversely, ACO participants that are effectively managing population health may not achieve shared savings for patients who, appropriately, do not use

the primary care services that would make them

eligible for attribution. ACOs have little financial incentive to deliver preventive care that might decrease the chance that healthy patients are attributed to an ACO.

 As advanced payment models mature to include downside risk, and as models like Comprehensive Primary Care expand, it is important to understand which patients are left out so that policy makers can develop regulations that encourage participation and improved care in new payment models.

To date, we have no information on the composition of patients who are not attributable to any provider group under Medicare Shared Savings Program (MSSP) regulations. This paper examines beneficiary characteristics associated with attribution and compares hospitalization, mortality, and spending across attributable and unattributable beneficiaries. Results from our analyses can guide policy on whether additional actions are necessary to adequately give provider participants incentives to improve population health and to ensure that vulnerable beneficiaries—who may benefit the most from improved care coordination&mdash;are not excluded from new payment models.

Using Medicare claims and CMS attribution rules for the most widely adopted model, the MSSP,

 we categorized beneficiaries into 2 mutually exclusive groups: patients who were attributable to provider groups and patients who were not attributable to any provider group. We included only beneficiaries with full parts A and B coverage, limiting the sample to those for whom traditional fee-for-service (FFS) Medicare was the primary payer. We combined all beneficiaries attributable to provider organizations (ie, MSSP, Pioneer ACO, and non-ACO medical groups) into a single category because the purpose of our research was to investigate the characteristics of those falling through the cracks of the current attribution methodology. In 2012, Medicare ACO contracts included 32 Pioneer program participants and 114 MSSP participants (over the first performance period from April 2012 or July 2012 through December 2013) that were responsible for more than 2 million beneficiaries.

 Although Pioneer program participants, which were responsible for nearly 700,000 beneficiaries in 2012,

 faced downside risk for spending, virtually all MSSP participants (n = 110) were in Track 1 and eligible for upside savings only.

 We conducted cross-sectional analyses examining characteristics of beneficiaries according to their attribution status.

Beneficiary Attribution to Provider Organizations 

Following MSSP’s 2-step attribution process, a beneficiary who has at least 1 face-to-face outpatient evaluation and management (E&M) visit is assigned to the provider group that has the highest allowed charges by primary care clinicians (ie, general practice, family practice, internal medicine, and geriatric medicine practitioners) for those visits.

 Patients not seeing primary care clinicians are attributed based on visits to qualifying non—primary care clinicians (ie, physicians in other specialties, nurse practitioners, physician assistants, and clinical nurse specialists).

 Beneficiaries who received care only from nonqualifying clinical providers (eg, interventional cardiologists or certified registered nurses) or in nonstandard settings (eg, emergency departments [EDs]) or who had no visits at all were not able to be assigned to any provider group using current MSSP attribution rules; they formed our “unattributable” group.

For beneficiaries in parts A and B who were 21 years or older, data were drawn from a 40% random sample of 2012 Medicare FFS claims and 2012-2013 beneficiary summary files. Beneficiary summary files provided patient demographics, enrollment status, and date of death. Attribution status, diagnoses, utilization, and spending were obtained using claims. We chose 2012 because that was the most recent year of claims available without the suppression of claims indicating substance use disorder in research files

 and within the period of Medicare ACO implementation.

Long-term residence in nursing homes was determined from the 2012 Minimum Data Set (MDS). Patients’ zip codes were used to identify hospital referral regions (HRRs) of residence; HRRs represent regional healthcare markets in the United States defined by the Dartmouth Atlas. We further used beneficiaries’ zip codes to identify rural and urban areas

 and whether they lived in high-poverty Census tracts (≥20% of residents below the federal poverty line).

The main outcome of interest was whether a beneficiary was attributable to a provider group or not. We estimated correlations between attribution and beneficiary characteristics, including age, sex, race/ethnicity (ie, non-Hispanic white, non-Hispanic black, Hispanic, Asian or Pacific Islander, or other race), residence in a high-poverty Census tract or a rural area, dual eligibility for Medicaid, residence in long-term nursing home (≥100 days in a nursing home according to MDS), disability as a basis of original Medicare entitlement, and date of death.

International Classification of Diseases, Ninth Revision

 codes on claims to identify beneficiaries’ comorbid conditions in 2011 using Hierarchical Condition Categories developed by CMS

]). Chronic conditions included dichotomous variables for cancer, cerebrovascular disease and stroke, chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), connective tissue disease, coronary artery disease, dementia, diabetes, hematologic/thrombotic disease, liver disease, mental health conditions (severe mental illness, substance use disorder, and depression), paralysis (not stroke), Parkinson’s/Huntington disease, peripheral vascular disease, and renal disease. We identified these conditions in 2011 to account for beneficiary diagnosis history and for their chronic nature and association with mortality and costs. We categorized patients with 0 observed chronic conditions, with multiple observed chronic conditions, and with unknown diagnoses in 2011 due to lack of medical claims. We reported utilization (ambulatory care—sensitive condition admissions, admissions to acute care or critical access hospitals, E&M visits, ED visits not leading to admission) and 2012 spending (Medicare payments to all providers, total and by category: inpatient, physician services, home health, and hospice).

To distinguish those who were unattributable because attribution rules may exclude end-of-life care, we examined mortality in 2012 and 2013 (among those alive January 1, 2013). This approach disentangles the mechanical relationship between attribution and death from clinical complexity.

We compared characteristics of attributable and unattributable Medicare enrollees—based on a 40% random sample to minimize computing time&mdash;using multivariate logistic models. We modeled attribution status as a function of demographics, dual eligibility for Medicaid, nursing home residency, disability, measures of chronic condition history, and HRR fixed effects.

We considered the attribution status of those with dual eligibility for Medicaid, who totaled more than 10 million people as of 2016, due to their low socioeconomic status, complex clinical needs, and elevated medical spending.

This study included about 13,000,000 FFS Medicare beneficiaries in 2012. Most of these FFS Medicare beneficiaries (88%) were able to be attributed to a provider group, leaving 12% unattributable (

Compared with attributable beneficiaries, unattributable beneficiaries tended to be younger (65 vs 71 years), male (60% vs 43%), from a minority race or ethnic group (28% vs 18%), living in a high-poverty area (25% vs 22%; 

), and less likely to be dually eligible for Medicaid but no less likely to reside in a rural area (

). Unattributable beneficiaries had fewer observed chronic conditions and were less likely to have 2 or more chronic conditions in 2011 (5.3% vs 22.3%; Table 1) compared with attributable beneficiaries; however, people with no service use or very few visits have no or fewer opportunities for diagnoses to be observed. Mortality was higher in the unattributable group in 2012 (6.6% vs 4.3%; Table 1) but lower in 2013 (2.8% vs 4.9%). More beneficiaries younger than 65 years were unattributable than those older than 65 years (20% vs 10%; Figure 1). We do not report 

 values of differences in variables’ means and proportions across attribution groups, which were all statistically significant (

In terms of geographic distribution, the proportion of unattributable beneficiaries ranged from 6% to 22% across HRRs (

); HRRs with the fewest unattributable beneficiaries (less than 10%) were located primarily in parts of the Midwest, South, and Northeast regions.

not shown, we examined 2011 physician supply in the market—a proxy for access&mdash;and found no evidence for differences across attribution categories.

In the unattributable group, we identified a subpopulation with no healthcare service use or spending in 2012 (63% of the unattributable) and a group that died in 2012 (6%). The remaining beneficiaries (31%) used healthcare services, including primary care services, that did not qualify for attribution (

). Among the unattributable nonusers of care in 2012, few (3.6%) had 1 or more reported diagnoses in 2011. However, spending averaged $9000 in 2011 in this group, primarily from ED visits and hospitalization (not shown).

Unattributable patients with healthcare service use who died (user-decedents) spent $19,000, on average (including $12,000 of inpatient spending and $5000 of hospice spending; Table 2). More than half (60%) of the unattributable decedents died within the first 3 months of 2012, compared with 20% of decedents attributable to provider organizations (

). Of the more than 80,000 unattributable user-decedents, 67% had 1 or more chronic condition diagnoses in claims in 2011; dementia, CHF, renal disease, diabetes, and COPD were common.

The third group of unattributable beneficiaries (with healthcare use but no qualifying primary care visits) averaged more than $3000 in annual spending, which was one-third of the spending of attributable patients. Very few of these beneficiaries had nonqualifying E&M visits (n = 57), whereas others had other healthcare use in nonattribution settings. Of the unattributable nondecedent healthcare users, 31% had reported clinical conditions in the prior year, with severe mental illness, diabetes, renal diseases, dementia, and COPD being most common.

In multivariate regression analyses, older and female beneficiaries were less likely to be unattributable (0.1 percentage-point decrease per year of age and 4.5 percentage points less likely among women; 

). Minorities and those living in high-poverty Census tracts were more likely to be unattributable. Nursing home residents and dual-eligible beneficiaries were less likely than others to be unattributable to provider groups (4.3 and 4.2 percentage points lower probability, respectively; 

 <.001). Disability as original reason for entitlement was associated with a 4.5 percentage-point increase in the likelihood of being unattributable. Beneficiaries with 0 observed chronic condition diagnoses and those with unknown chronic condition history (nonusers of care and those not in both parts A and B FFS Medicare in 2011) in 2011 were more likely to be unattributable in 2012 relative to those with observed chronic conditions (2.4 and 20.4-22.3 percentage points higher probability, respectively; 

Because Medicaid coverage complements Medicare benefits for certain low-income adults, we explored how attribution status differed in the 22% of our sample who were dually eligible for Medicaid benefits. Although they were less likely than other beneficiaries to be unattributable (Figure 1), results from multivariate analyses of dual-eligible enrollees were consistent with those found in the overall sample (

), with 2 exceptions: Asian/Pacific Islander ethnicity and disability status were both associated with a decrease in the likelihood of being unattributable.

Because existing evidence on ACOs focuses exclusively on attributable Medicare beneficiaries,

 we studied the excluded group of unattributable patients to understand how these omitted groups might influence performance in Medicare ACO programs.

 Using MSSP attribution methods, nearly 12% of beneficiaries were unattributable in 2012. This group was heterogeneous, composed of decedents, beneficiaries using no services, and those using services ineligible for attribution under common methods. The majority of unattributable beneficiaries had no encounter with the healthcare system at all, which may be due to financial and nonfinancial barriers that limit access to care.

 A small but costly group, 6% of unattributable beneficiaries, died in 2012; among them, inpatient and hospice service use was common. The remaining 31% of unattributable beneficiaries had minimal healthcare use, largely for urgent or hospital-based services.

The distribution of unattributable beneficiaries aligned with regional spending patterns in Medicare.

 Beneficiaries living in low-cost areas, such as Minnesota, had higher proportions of unattributable beneficiaries. This pattern indicates that alternative payment models in these regions, ironically, may not reward organizations for their activities targeting healthier groups who use few services. For example, ACOs in low-cost regions were evaluated against lower financial benchmarks than were ACOs in high-cost regions. Beginning in 2017, cost benchmark calculations were improved to integrate regional factors, making an ACO’s benchmark less reflective of its historical spending and more dependent on FFS expenditures in its region.

 Two changes to attribution might capture more low-utilization beneficiaries. First, Track 3 of the MSSP, with 16 participating organizations in 2016, and the Next Generation ACO model—another demonstration project in the CMS Innovation Center&mdash;are currently piloting attestation, allowing beneficiaries to voluntarily choose an ACO,

 which would allow for accountability for preventive services in healthier populations with little or no observed utilization. Second, “sticky” attribution, or allowing utilization in a prior year to count in the current year for those with no use in the current year, could capture more healthy patients; it could also capture more individuals who are very sick and use non—primary care settings due to death early in the year. Although attestation has the benefit of permitting patients to choose, sticky attribution may be more feasible in existing programs because it can be done using just claims.

Facilitating the attribution of healthy patients and those with minimal use through any method may improve the consistency and continuity of care for patients. Because ACOs are responsible for both cost and quality of care for assigned patients, attribution of healthy patients with no healthcare use, as well as of those with some healthcare use but nonqualifying visits for current attribution purposes, would reward providers for care delivered to these patients, without rewarding unnecessary care.

Attribution of nonusers may more appropriately contribute to an ACO’s shared savings, which are conditional on meeting predetermined quality measure benchmarks, and these benchmarks reflect beneficiaries classified as at-risk populations—for example, in benchmark year(s)&mdash;who currently do not use services.

 Moreover, financial benchmarks are calculated using attributed beneficiaries’ historical healthcare use. The current attribution methodology provides incentives for ACOs to bring in healthy patients for primary care visits, but attributing nonusers through alternative methods may be warranted to appropriately reward providers without inducing potentially unnecessary primary care visits.

 A practical solution that would address those nonusers of care who used healthcare in previous years would be to assign them based on sticky attribution.

Furthermore, healthcare providers and policy makers should give greater attention to the consequential subpopulation of decedents (ie, end-of-life patients) currently falling outside of the responsibility of organizations participating in ACOs. One-fourth of Medicare spending is devoted to care in the last year of life,

 so relative savings on end-of-life care contribute outsized savings to the Medicare program. There are also opportunities for improvements in care quality and preference-based care. Organizations that deliver thoughtful end-of-life care, by using providers or settings not recognized by current attribution methods, are not rewarded for this high-quality care. On the flip side, organizations that fail to coordinate end-of-life care or that deliver resource-intensive care not aligned with patient preferences face no penalty. Additionally, higher hospice spending registered in unattributable decedents signals that involvement of hospice and palliative care in new payment initiatives may improve quality of care and potentially reduce costs

 Alternative attribution methods have different strengths and weaknesses. Retrospective attribution, currently used for most MSSPs and done at the end of the performance year, reflects actual care delivery. It removes patients who no longer receive care from an ACO and captures new patients, such as those aging into Medicare during the year (

 However, with retrospective attribution, organizations cannot plan care for patients assigned to them only after care delivery. In contrast, prospective attribution removes this uncertainty, which might facilitate active engagement with patients and can improve management of patient population health. However, prospective attribution is a forecast based on past healthcare use, and it works best when care-seeking patterns are stable and therefore predictable over time. This attribution, although preferred by ACOs,

 may have important cost implications for ACOs if patient care patterns drastically change from one year to another because patients may see any Medicare provider. This type of attribution would also fail to capture patients who initiated care during the year (eAppendix Table 5).

Most of the gaps in attribution rules that we identified may be addressed with a hybrid attribution method. First, for most populations, the current assignment based on E&M visits is appropriate. Second, for those with limited healthcare use and for the healthy, assignment would be based on attestation over a longer time period, therefore “rewarding” providers for early prevention and early detection care patterns. For those with serious illnesses, assignment could be broadened to include hospice and palliative care to encourage better end-of-life care (eAppendix Table 5). Before patients with serious illness reach the end of life, however, a first priority is to engage them in care. Providers could tailor care management and coordination to the needs of patients with chronic conditions who have accessed very little healthcare

 in order to improve quality and health outcomes.

Gender disparities were noted between attribution categories, with disproportionately larger proportions of younger and male populations among the unattributable beneficiaries. This may signal gender differences in healthcare-seeking behavior.

 The slightly higher proportion of those with dual eligibility for Medicaid attributable to a provider group (90%) may reflect greater coverage for out-of-pocket spending in this group, lowering a barrier to access. New initiatives, such as the CMS Financial Alignment Initiative, which integrates care across primary, acute, behavioral health, and long-term care for dual-eligible beneficiaries, offer another avenue to engage these patients.

This cross-sectional exploratory study in the first year of the ACO payment model has limitations. First, methods other than the MSSP, which was used in this study, may result in slightly different distributions of beneficiaries across categories. However, MSSP is the largest Medicare ACO program. Second, a longitudinal study with additional years of data could examine trends in attribution of Medicare beneficiaries as provider groups gain knowledge and develop better strategies to manage attribution. Third, claims-based measures of patient illness cannot capture the illness burden for those without utilization. Thus, our results may reflect care-seeking behavior, as we cannot attribute or observe clinical comorbidities for those with no or little healthcare utilization and for new enrollees in the preceding year.

The promise of ACO contracts to lower spending and enhance quality of care, and, ultimately, to improve population health, can be enhanced by improving the attribution of individuals to providers. By rewarding organizations for keeping patients healthy and out of resource-intensive care settings and by appropriately capturing and rewarding compassionate care, such as hospice services, for patients approaching the end of life, Medicare’s continued development of advanced payment models has the potential to move us closer to the goal of improving population health.

The authors thank Kristen Bronner for her help in map designing.

The Dartmouth Institute for Health Policy & Clinical Practice, Geisel School of Medicine at Dartmouth (MHO, EM, C-HC, SRR, JPWB, VAL, CHC), Lebanon, NH; The National Bureau of Economic Research (EM), Cambridge, MA; Department of Internal Medicine, University of Michigan Medical School (JPWB), Ann Arbor, MI; Institute for Health Policy and Innovation, University of Michigan (JPWB), Ann Arbor, MI.

 Supported by the National Institute on Aging (R03AG049360, R33AG044251, P01-AG019783, and 1K01AG049914), a grant from the Commonwealth Fund (20160616), and a grant from the Agency for Healthcare Research and Quality (5U19HS024075-02). The content of the article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging, the Commonwealth Fund, or the Agency for Healthcare Research and Quality.

Articles

This article compares clinical and utilization profiles of Medicare patients who are attributed to provider groups with those of patients unattributed to any provider group in accountable care organization models.

Forgotten Patients: ACO Attribution Omits Those With Low Service Use and the Dying

Aiming to encourage care coordination and cost efficiency, the Center for Medicare & Medicaid Innovation (CMMI) launched the Bundled Payments for Care Improvement (BPCI) initiative in 2013. To help gauge the program’s potential impact and generalizability, we describe early and current participants.

 We examined the cross-sectional association between BPCI participation and providers’ structural and cost characteristics.

 Using data from October 2013 and June 2014, we quantified changes in BPCI participation. We described structural differences between participating and nonparticipating hospitals using 

 tests, and we used the Cochrane-Armitage test to assess whether participants were more likely be in higher 90-day episode cost quintiles than their peers at baseline (2009-2010).

Overall (risk-bearing and non—risk-bearing) participation in BPCI increased from about 400 in October 2013 to more than 2000 in June 2014—attributable, in part, to Model 2, the most comprehensive of the 4 models offered by CMMI for provider participation. Model 2 hospitals increasingly resemble eligible but nonparticipating hospitals. For the most commonly chosen condition of hip replacement, Model 2 hospitals were not costlier than their peers. Hospitals used to make up 97% of Model 2 participants, but physician practices now comprise a substantial number of Model 2 participants. However, most BPCI participants have not yet begun to bear financial risk. Risk-bearing Model 2 hospitals are a smaller and less representative group, with higher baseline costs for hip replacement than their peers.

Growing participation in BPCI suggests strong interest in bundled payments. The long-term impact of BPCI will depend on CMMI’s ability to persuade interested but non—risk-bearing participants to bear risk.

	Growing and more representative enrollment in Medicare’s Bundled Payments for Care Improvement (BPCI) initiative suggests strong interest in bundled payments. The long-term impact of BPCI will depend on the results of risk-bearing participants.

Episode-based bundled payments are one of several proposed healthcare payment reforms aimed at encouraging care coordination, quality improvement, and cost efficiency.

 Historically, hospitals, physicians, and post acute care providers have been paid separately for services occurring during and after hospital admissions. With bundled payments, a fixed lump sum payment is shared among all caregivers, who also share savings when actual expenditures fall below the bundled payment amount. Based on data highlighting wide variation in episode payments across hospitals with many common conditions,

 some estimate that implementation of bundled payments around hospitalizations could save more than $30 billion annually in the Medicare program alone.

A new program by the Center for Medicare & Medicaid Innovation (CMMI) provides early insights about the future prospects of bundled payments. The Bundled Payments for Care Improvement initiative (BPCI) was launched in 2013, with CMMI offering providers 4 models for participation.

 Model 1 includes Part A services for the index hospitalization alone, and thus most closely resembles current fee-for-service payment. Model 2 is the most comprehensive of the 4, encompassing Part A and Part B services for the index hospitalization, readmissions, and all other post acute care. Model 3 includes only post acute care, and Model 4 includes both the index hospitalization and any readmissions.

Although the end results of the BPCI program will not be available for some time, we provide an overview of the program’s general parameters and describe the characteristics of its participants in order to help gauge the program’s potential impact and generalizability. More specifically, we describe national patterns of participation in the BPCI program, as well as the association between participation and providers’ structural and cost characteristics. 

CMMI began accepting applications from hospitals and other providers for the BPCI initiative in 2011. The first participants began bearing financial risk in April 2013 (Model 1) and October 2013 (Models 2-4). CMMI briefly accepted new applications in late 2013 (Model 1) and early 2014 (Models 2-4), with many of these participants not yet having begun risk-bearing (ie, subject to financial gains and losses in the program). Program participants first enter a non—risk-bearing period preparing for implementation (Phase 1), followed by the start of a 3-year risk-bearing period (Phase 2), which is staggered among participants. Even for individual participants, risk-bearing may be rolled out over time; a BPCI participant may begin bearing risk for some of their selected conditions (ie, partial risk-bearing) before doing so for all of their selected conditions (ie, complete risk-bearing).

To describe program participation, we used BPCI participant lists from October 2013 (defined as early participants) and June 2014 (defined as current participants). Data on BPCI participants were first made publicly available in October of 2013.

 Because Model 3 focuses exclusively on post acute care, we examine Models 1, 2, and 4—the 3 models that included the index hospitalization and were open to acute care hospitals.

To describe the structural characteristics of BPCI hospitals versus non-BPCI hospitals, our sample included acute care hospitals eligible for program participation. We utilized the American Hospital Association Annual Survey 2012.

 We defined rural hospitals as those in areas with a population less than 10,000 and we defined teaching hospitals as major or minor (ie, members of the Council of Teaching Hospitals, hospitals with a medical school affiliation, or hospitals with residency or internship training approved by the Accreditation Council for Graduate Medical Education or the American Osteopathic Association). When examining Model 2 hospitals (BPCI participants vs nonparticipants), we defined nonparticipants as all acute care hospitals not participating in Model 2 of BPCI as of June 2014, identified in the 2010 version of the 100% Medicare Provider Analysis and Review File, excluding the Model 2 BPCI hospitals, Maryland hospitals, and federal hospitals.

Given concerns that providers might “cherry-pick” high-cost conditions for performance measurement (because it may be easier to lower costs for conditions with baseline costs above benchmark), we performed a sensitivity analysis of baseline costs of care for hip replacement—the most popular condition chosen. We describe the distribution of Model 2 BPCI hospital participants who selected hip replacement, across national cost quintiles. These cost quintiles were created by ranking all hospitals (all BPCI eligible nonfederal acute care hospitals that had at least 10 hip replacement cases per year and were not located in Maryland&mdash;which is exempt from the Inpatient Prospective Payment System and Outpatient Prospective Payment System&mdash;with payment rates set by the state) by their average 90-day risk-adjusted, price-standardized, reliability-adjusted total payments, and then dividing this into quintiles. Because almost no physician practices are risk-bearing yet, we did not conduct a similar analysis for hospitalized patients attributed to BPCI physician practices (vs non-BPCI physician practices).

For this work, we utilized Part A and B Medicare claims data from 2008 to 2011. We restricted our sample to acute care hospitals that cared for fee-for-service Medicare beneficiaries 65 years or older who underwent hip replacement (

International Classification of Diseases, 9th revision, Clinical Modification

 codes 81.51 or 81.52, and diagnosis-related group codes 469 or 470) between 2009 and 2010. We included no more than 1 index admission per year per beneficiary. We price-standardized total episode payments from the procedure date through 90 days postdischarge, using a previously described methodology.

 We risk-adjusted payments using age, sex, race, admission acuity, and the 29 Elixhauser comorbidities.

This study was declared not regulated by the University of Michigan Institutional Review Board. We conducted all analyses with SAS version 9.3 (SAS Institute, Cary, North Carolina).

The BPCI program has grown rapidly since its inception. Only 171 acute care hospitals were participating in Model 2 when the earliest participants began risk-bearing in October 2013 (

). Even fewer hospitals were participating in Models 1 and 4. Participation rose sharply after CMMI accepted new applications in early 2014. The number of Model 2 participants increased more than 3-fold. Even risk-bearing participation grew substantially during this same time period. In contrast, the number of Model 1 and 4 participants remained low—and in the case of Model 4, dropped.

Physician groups make up a growing and sizable proportion of BPCI participants (

). While physician practices comprised just 3% of all Model 2 participants in October of 2013, they now make up many of the Model 2 participants. Most practices specialize in orthopedic surgery, hospital medicine, cardiology, or emergency medicine; the growth in physician groups has outpaced that of hospitals. Nevertheless, almost no physician practices have begun risk-bearing. About 10% of all eligible acute care hospitals are currently part of Model 2.

Model 2 hospitals (risk-bearing + non—risk-bearing) differ from nonparticipating hospitals (Table). For example, these hospitals are more likely to be large (500 beds or more) and to be teaching hospitals. Hospitals in the northeast are substantially overrepresented, while those in the south are underrepresented. Participating hospitals are also more likely to be nonprofit than their eligible nonparticipating peers. Risk-bearing Model 2 hospitals are an even less representative group than non–risk-bearing hospitals.

BPCI hospitals and physician groups in Model 2 chose from a menu of 48 eligible clinical conditions for performance measurement. These 48 episodes make up approximately three-fourths of all admissions at acute care hospitals for older fee-for-service Medicare beneficiaries. However, most early entrants limited their participation to a narrow subset of these conditions (Table). For example, nearly half of early Model 2 participants selected only 1 or 2 conditions: the most popular choices were major lower extremity joint replacement (selected by 86% of Model 2 participants), congestive heart failure (60%), coronary artery bypass grafting (CABG) (44%), and chronic obstructive pulmonary disease/asthma (43%). Later entrants were more likely to choose all 48 conditions. For example, about half of the Model 2 orthopedic practices chose all of the nonorthopedic conditions offered, in addition to all of the orthopedic conditions. When participants did limit themselves to a smaller number of conditions, orthopedic and cardiac conditions were favored.

Among current risk-bearing Model 2 hospitals that selected the most commonly chosen condition of hip replacement, more than 5 times as many Model 2 hospitals were in the highest compared with the lowest cost quintile (

). In contrast, the baseline costs of all current Model 2 hospitals (risk-bearing + non—risk-bearing) resemble that of nonparticipating hospitals.

We found that provider participation in the national bundled payment initiative for Medicare beneficiaries has grown substantially in the past year, with participation tripling for the most comprehensive model (Model 2). Although this growth is comprised primarily of non—risk-bearing participants, the number of risk-bearing providers has also increased. Physician practices—especially orthopedic practices&mdash;have increasing interest in the program, as evidenced by the rising number of physician practices in the BPCI’s non–risk-bearing phase. The clinical scope has broadened as well. Whereas many early participants selected 1 or 2 conditions, many current entrants have selected all 48 conditions offered.

Bundled payments are 1 of several “alternative payment models” that were highlighted by HHS Secretary Sylvia Burwell, when she described the federal government’s efforts to move toward paying for value.

 The size of the BPCI program dwarfs prior bundled payment initiatives, and is comparable in size to other CMMI payment reforms such as the Medicare Shared Savings Program (with 343 accountable care organizations [ACOs] as of January 2014). CMMI permits providers to enroll in multiple payment reform initiatives simultaneously (eg, BPCI, Next Generation ACOs). Of note, we found that BPCI hospitals share a number of characteristics with Medicare ACOs.

BPCI builds upon earlier and ongoing efforts around bundled payments. As early as 1991, the Medicare Participating Heart Bypass Center Demonstration tested bundled payments with CABG at 7 hospitals.

 This was followed by efforts to implement bundled payments in other settings or for other populations; for example, Medicare’s Acute Care Episode Demonstration, which ran from 2009 to 2012, targeted orthopedic and cardiac conditions in 5 hospitals or health systems.

 The Geisinger ProvenCare program started with CABG in 2006 and now includes several other conditions and procedures,

 and the ongoing PROMETHEUS payment model offers many different bundles of care.

These earlier efforts have highlighted the difficulty of implementing bundled payments,

 and thus, the rapid growth of BPCI participation is somewhat surprising. Administrative and logistical challenges to bundled payments include establishing provider networks that share and distribute risk, constructing the legal and regulatory framework to support these arrangements, and modernizing information and billing systems to accommodate episodes of care. Recent data suggest that these are continuing challenges. For example, a recent effort to establish bundled payments for orthopedic procedures in California broke down during the implementation phase, with disagreements such as how a bundle should be defined.

 In contrast, CMMI delineated what services would be included in a bundle, had an established risk-adjustment methodology, and designated benchmarks for episode costs; it is too early to tell though whether or not these aspects of the program will be enough to minimize implementation challenges. Finally, it is possible that many of BPCI’s current participants will drop out of the program prior to beginning to bear risk.

The rising interest of physician groups is another notable finding from our descriptive work, and one that makes sense given the central role that physicians play in an episode of care. For example, post acute care makes up a substantial proportion of total episode costs for major joint replacement.

 While orthopedic surgeons may not dictate length of stay in a skilled nursing facility (SNF), they are able to influence which patients are discharged with the less-costly option of home health (vs an SNF). Similarly, hospitalists help decide which patients require sub-acute care on discharge, and may help triage patients presenting for readmission. Furthermore, some hospitalist groups have assumed the care of patients in skilled nursing facilities in addition to acute care hospitals. Guidelines for BPCI encourage participation from physician groups, as only those inpatients cared for by a participating physician are attributed to participating physician groups. How BPCI participants will distribute risk among the various providers that share in a patient’s care (eg, attending physician, consulting physician, acute care hospital, post acute care provider) remains to be determined, however.

Although participating providers increasingly chose a wide range of conditions, the most popular conditions remain orthopedic and cardiac. This may reflect the confidence gained from prior bundled payment initiatives around these same conditions (eg, the ACE Demonstration, Geisinger’s ProvenCare). In addition, unlike many chronic medical conditions, procedures have discrete start and end points, lending themselves to episodes of care that are measured in days rather than years. Finally, surgical inpatients such as those undergoing a hip replacement, are on average more clinically homogeneous then medical inpatients. CMMI offered risk corridors to mitigate the unexpected effects of outliers, but providers may remain concerned about the inability to reliably predict the clinical course for patients with multiple comorbidities. 

Any interpretation of our results should account for the fact that BPCI is an ongoing program. For example, as of June, most Model 2 participants were still in the non—risk-bearing preparatory period (Phase 1), with the 3-year risk-bearing period (Phase 2) slated to begin by January 2015. It is possible that a substantial number of participants will drop out of the program before they begin to bear financial risk. Between October and June, one-fifth of Model 2 enrollees dropped out. It is also possible that fewer conditions will be selected for the risk-bearing phase of the program. Nevertheless, we believe that the magnitude of the trends— toward greater participation and more physician group involvement for a wider range of conditions&mdash;make these results important to note. Moreover, while we focused on those models that were open to acute care hospitals, the same general patterns were found among Model 3 (post acute care bundling) participants.

Interest in the national Medicare bundled payment program is growing, as evidenced by rising and increasingly diverse provider participation for a broad array of clinical conditions. With increased participation, the potential for generalizability of the program’s eventual outcomes has increased as well. However, the magnitude of BPCI’s long-term impact will depend on the still uncertain results of risk-bearing participants; success by a large risk-bearing group would make it easier for CMMI to convince other providers to join the program.

The authors thank Haiyin Liu, MA; Mary Oerline, MS; and Anne Cain-Nielsen, MS, for the analytic support they provided. They were compensated for their work.

 Division of General Medicine, Department of Internal Medicine (LMC) and Center for Healthcare Outcomes & Policy (LMC) and Institute for Healthcare Policy and Innovation (LMC), University of Michigan, Ann Arbor, MI; VA Ann Arbor Healthcare System (LMC), Ann Arbor, MI; Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth (EM, JDB), Lebanon, NH; National Bureau of Economic Research (EM), Cambridge, MA.

 This work was supported by funding from the National Institute on Aging (Grant No. P01AG019783). This work was also supported by a Career Development Grant Award (K08HS020671) from the Agency for Healthcare Research and Quality.

 Dr Meara received a grant for the preparation of this manuscript from the National Institute on Aging (P01AG019783). Dr Birkmeyer has equity interest in ArborMetrix, a company that profiles hospital quality and episode cost efficiency; the company played no role in the manuscript. Dr Chen reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (LC, EM, JDB); acquisition of data (JDB); analysis and interpretation of data (LC, EM, JDB); drafting of the manuscript (LC); critical revision of the manuscript for important intellectual content (LC, EM, JDB); statistical analysis (LC); obtaining funding (EM, JDB); administrative, technical, or logistic support (JDB); and supervision (EM, JDB).

 Lena M. Chen, MD, MS, University of Michigan Division of General Medicine, North Campus Research Complex, 2800 Plymouth Rd, Bldg 16, Rm 407E, Ann Arbor, MI  48109-2800. E-mail: lenac@umich.edu.

1. Hackbarth G, Reischauer R, Mutti A. Collective accountability for medical care--toward bundled Medicare payments. 

2. Hussey PS, Sorbero ME, Mehrotra A, Liu H, Damberg CL. Episode-based performance measurement and payment: making it a reality. 

3. Bundled Payments for Care Improvement initiative fact sheet. CMS website. http://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-Sheets/2014-Fact-sheets-items/2014-01-30-2.html. Published January 30, 2014. Accessed May 26, 2014.

4. Cutler DM, Ghosh K. The potential for cost savings through bundled episode payments. 

5. Bundled payments for care improvement (BPCI) initiative: general information. CMS website. http://innovation.cms.gov/initiatives/bundled-payments/. Updated August 20, 2015. Accessed October 4, 2015.

6. Calsyn M, Emanuel EJ. Controlling costs by expanding the Medicare acute care episode demonstration. 

Expanding and more representative participation in Medicare's Bundled Payments for Care Improvement initiative suggests potential for large impact, pending the results of risk-bearing participants.

Medicare's Bundled Payments for Care Improvement Initiative: Expanding Enrollment Suggests Potential for Large Impact

To assess the impact of 2005 and 2006 reductions in chemotherapy reimbursement, mandated in the Medicare Modernization Act, on patterns of chemotherapy receipt in the last 14 days of life.

Included in the study were Medicare beneficiaries dying with poor-prognosis cancer from 2003 to 2007. We compared pre- and postreform probability and frequency of chemotherapy receipt in the last 14 days of life, a validated quality measure, using linear models. We assessed changes in chemotherapy use in physician offices (where prescribing is often directly linked to physician income) and hospital outpatient departments (where the link is indirect and likely weaker).

Among patients receiving chemotherapy in the 6 months before death in physicians’ offices before the policy implementation (2003 to 2004), 18% received chemotherapy in the last 14 days of life. Those dying after implementation (2006 to 2007) were 3.5 percentage points (95% confidence interval [CI], —5.4 to –1.6; P <.001), or 20%, less likely to receive chemotherapy in the 14 days before death than those dying before implementation. By contrast, there was no significant change in the percentage of patients receiving chemotherapy in the last 14 days of life in hospital outpatient departments between 2003 and 2004 and between 2006 and 2007.

 In physician offices, where drugs generate the majority of revenue, and prescribing patterns can determine physician income, use of chemotherapy at the end of life fell significantly after reimbursement reductions; no concurrent change occurred in hospital outpatient departments. These results suggest that payment reform may be used to better align appropriate financial incentives with better quality of care.

(Am J Manag Care. 2012;18(5 Spec No. 2):e200-e208)

In the early 2000s, Medicare drug reimbursements provided high margins under an average wholesale price system, which permitted physicians to obtain drugs well below reimbursed rates.

 In response to concerns about rapid growth in the use of chemotherapy drugs with high margins, the Medicare Modernization Act (MMA) changed reimbursements to more closely reflect acquisition costs.

 The MMA reduced payments to 106% of manufacturer-reported average sales prices while increasing drug administration fees; the change took effect in physician offices in 2005 and in hospital outpatient departments in 2006. The Government Accountability Office estimated that the MMA reform would reduce the average profit margin on infused chemotherapy drugs from 22% to 6%.

 Payments for chemotherapy and anemia medications dropped by $1.2 billion between 2004 and 2005, after implementation of the reform in physicians’ offices.

 This payment change had a large impact on oncologist practice revenue, because drug reimbursement accounted for 77% and drug administration accounted for approximately 10% of practice revenue during this time period.

Policy makers have noted the potential conflicts of interest inherent in the pre-MMA system of payment for chemotherapy treatments.

 Studies assessing the effect of MMA reimbursement changes on access to and use of chemotherapy in patients with newly diagnosed cancer have generated mixed results.

 Despite fears that physicians would cease to offer chemotherapy services in their offices, restricting access to care and increasing volume in hospital outpatient departments,

2 studies found no meaningful effects on access to chemotherapy as measured by wait time, travel time, or infusion locale.

Using Medicare data 6 and 10 months into the payment change, respectively, the Medicare Payment Advisory Commission

 found increased use of chemotherapy in patients with newly diagnosed cancer and evidence of switching to more costly agents. However, these studies focused on newly diagnosed patients cared for in a short window after MMA implementation and provide little sense of how changes affected quality of care.

The effect of financial incentives on service delivery may be most pronounced in cases of marginal or low value,

 such as chemotherapy at the end of life, where aggressive anticancer therapies may prove to be more toxic than beneficial, use is not related to probability of benefit, and appropriately timed cessation of chemotherapy is integral to a patient’s terminal quality of life.

We augment evidence on the impact of payment reform for Part B drugs and provider response to financial incentives through a study of end-of-life chemotherapy treatment.

 We expand current understanding in 3 ways. We examined a measure of the quality of cancer care—chemotherapy receipt in the last 14 days of life, a validated quality measure and now included as a benchmark for improving clinical practice in the Quality Oncology Practice Initiative (QOPI).

 Second, we examined the effects of payment changes in physician office settings separately from hospital outpatient department settings to leverage differences in financial incentives. In office settings, chemotherapy use is often directly related to physician income and makes up a large portion of practice revenue; in the hospital outpatient setting, drug reimbursement makes up a much smaller proportion of the revenue mix, and the link between physician income and prescribing patterns is indirect and likely weaker. We hypothesized that changes in chemotherapy payment were more likely to alter prescribing decisions in physicians’ offices; modeling the groups separately allowed us to disentangle changes in behavior as a result of reimbursement from broader trends in end-of-life care. Finally, by examining data through 2007, we were able to observe effects of the policy change that occurred over a longer time horizon compared with previous studies and thus did not limit our assessment to immediate policy response.

We used Medicare claims data for beneficiaries dying with poor-prognosis cancer from 2003 to 2007 and assessed pre- (2003 to 2004) and post- (2006 to 2007) reform probability and frequency of chemotherapy receipt in the last 14 days and 3 months of life. We analyzed trends in treatment before and after reform by treatment location (physician offices and hospital outpatient departments).

From the 20% Medicare denominator files spanning 2003 to 2007, we identified fee-for-service Medicare beneficiaries who died between age 66 and 99 years and had continuous Parts A and B coverage in the last 6 months of life. Decedents were included in the study if they had at least 1 hospital claim or at least 2 clinician visits in the last 6 months of life with poor-prognosis cancer.

 Poor-prognosis cancer was defined by Iezzoni et al

 using claim diagnosis codes associated with high rates of death during hospital admissions, thus permitting us to create a cohort for whom providers likely understood prognosis to be poor in the last 6 months of life. We categorized decedents into 1 of 26 cancer types based on their predominant cancer diagnosis.

For each patient, we used billing codes to assess receipt of outpatient chemotherapy (administered by a clinician or facility) in the last 6 months of life, last 3 months of life, and last 14 days of life (codes defined in the 

). Our main outcome of interest was receipt of chemotherapy in the last 14 days of life, conditional on receipt in the last 6 months of life. We chose this measure because it is similar to the quality metric defined by Earle et al.

We determined setting of chemotherapy infusion based on claim file type and place of service code during the last 6 months of life (hospital outpatient department, physician office, or both). We also assessed the number of chemotherapy treatments in each time window among those who received chemotherapy, defined as the number of days of treatment.

From Medicare files, we obtained patient age at death, race (dichotomized as black or nonblack), state of residence, and sex. We used the methods of Iezzoni et al

to assign patients up to 8 noncancer chronic conditions based on 1 inpatient or 2 outpatient

 International Classification of Diseases, Ninth Revision

, diagnosis claims occurring between 6 months and 1 month before death (

 We categorized patients’ metastatic cancer status using the definitions of Iezzoni et al.

 Each patient’s residential zip code was used to assign an estimated household income based on US Census tract data and the proportion of the population in that area in poverty.

 Models included these patient characteristics as well as cancer type and age squared.

We conducted 2 sets of analyses. We modeled the distinct individual monthly impacts of the 2005 and 2006 policy implementations and then considered their impact collectively as a single payment change, implementation of which spanned 12 months.

We plotted regression-adjusted mean predicted probabilities that a patient received chemotherapy in the last 2 weeks of life and last 3 months of life by setting of treatment (Fig 1). We estimated ordinary least squares models of the probability of chemotherapy receipt or frequency of chemotherapy receipt as a function of payment change, where the patient was the unit of analysis. We captured payment change with a linear time trend with coefficients to capture changes in the time trend in January 2005 and January 2006. We chose the linear time trend because it fit the raw data accurately and is readily interpretable. In sensitivity tests of the specification, we obtained similar results estimating probability of chemotherapy receipt using logistic regression. All regression models controlled for a linear time trend to capture trends in end-of-life treatment during the full time period. We estimated models for the entire cohort and separate models by site of chemotherapy receipt in the last 6 months (physician office or outpatient department). We also predicted each outcome without payment reform, assuming the time trend from the period before the payment change (2003 to 2004) persisted (ie, as if changes in trend in January 2005 and January 2006 were zero).

We repeated these models after dropping patients who died in 2005 and estimated the cumulative effect of the payment change in January 2005 (physician offices) and January 2006 (hospital outpatient departments) as a single event, controlling for a linear time trend and the same covariates. This model has 2 advantages. It estimates the magnitude of the overall effect of the MMA on probability and frequency of chemotherapy receipt, allowing for a transition period as the payment change in physicians’ offices had been in place for 1 year. Second, it excludes the cohort whose last 6 months before death spanned the pre-2005 and post-2005 periods, yielding a cleaner definition of the pre- and postperiod observations. We adjusted variance estimates for the correlation of observations within state and time period (before January 2005, after December 2005) using Huber-White sandwich estimators. 

Because previous publications studying effects of the MMA have focused specifically on incident lung cancer,

 we repeated our analysis for the subset of patients with lung cancer.

Overall, 235,821 patients met inclusion criteria; 32% received chemotherapy in the last 6 months of life (n = 75,026; Table 1). Mean age at death was 78.1 years; 49% were women; 9% were black. The most common cancer types were: lung cancer (32%), cancer of unknown primary (10%), and hematologic cancer (9%). Among patients who received chemotherapy within 6 months of death, 74% received all chemotherapy treatments in a physician’s office, 21% received all treatments in a hospital outpatient department, and 5% received treatment in both places. Receipt of chemotherapy in hospital outpatient departments increased over the time period (from 18% in 2003 to 25% in 2007). Demographics varied by setting of administration, with those receiving chemotherapy in a physician’s office significantly more likely to be older, nonblack, male, and from a wealthier zip code. Those with more comorbidities were treated in physicians’ offices. Setting-specific population characteristics were similar among the subset of patients with lung cancer, although these patients were more likely to be younger, nonblack, and poorer and have more comorbidities relative to the full sample.

Among those receiving chemotherapy in the last 6 months of life, 77% received chemotherapy in the last 3 months of life, and 15% received chemotherapy in the last 14 days of life over the full 2003 to 2007 time period (Table 1). Overall, chemotherapy receipt near the end of life was significantly more likely for those treated in physician office settings versus hospital outpatient departments. Of those treated in physician offices, 77% received chemotherapy in the last 3 months of life, and those who received chemotherapy in this window averaged 4.1 days of treatment. In physicians’ offices, 16% received chemotherapy in the last 14 days of life and averaged 1.5 days of treatment. Of those using outpatient departments for treatment, 73% received chemotherapy in the last 3 months of life (averaging 3.4 days of treatment), and 12% received chemotherapy in the last 14 days (averaging 1.4 days of treatment). All of these differences across settings are statistically significant, with 

 values below .05. Patterns were similar among the subset of patients with lung cancer.

graphically presents results from models considering the 2 policy implementation events separately. It indicates that within both time windows (14 days and 3 months) before death, the probability of chemotherapy receipt dropped significantly after the payment change in physician offices, whereas it increased slightly in hospital outpatient departments. Patterns were similar among those with lung cancer.

 lists the coefficients on the payment change when transitional data from 2005 are dropped, and the payment change is measured as a one-time shift, controlling for patient demographics, comorbidities, cancer type, metastatic disease, and a linear time trend. Comparing 2006 and 2007 with 2003 and 2004, the rate of chemotherapy receipt in the last 14 days of life fell 2.6 percentage points (or 2.6 per 100 patients), a drop of approximately 20% (95% confidence interval [CI], —4.2 to –1.0; 

 = .002). The drop was driven entirely by reductions in terminal chemotherapy in physician offices, where the probability of chemotherapy in the last 14 days dropped 3.5 per 100 patients (95% CI, —5.4 to –1.6; 

<.001) relative to the mean of 18% in the preperiod.For those who received treatment in a hospital outpatient department, there was no significant change over the same time period (95% CI, —2.4 to 4.4; 

 = .541). We also observed significant reductions in chemotherapy for the last 6 months of life (2.5 per 100 patients; 95% CI, —3.7 to –1.3; P <.001) and the last 3 months of life (4.2 per 100 patients; 95% CI, –6.1 to –2.2; 

 <.001). In the lung cancer cohort, there were no significant changes in chemotherapy in the last 14 days, but estimates of changes in probability in the last 3 and 6 months were similar in significance and magnitude to the full cohort. Because the composition of cancer type varied across the 2 settings, and differences in treatment for lung cancer (such as epidermal growth factor receptor inhibitors as a later line of therapy) could be a possible confounder, we estimated a model excluding patients with lung cancer. The results were unchanged.

In the full cohort who received chemotherapy during the window, the number of treatments per patient declined by 0.95 treatments in the last 6 months (95% CI, —1.3 to –0.6; 

 <.001) and 0.61 treatments in the last 3 months (95% CI, —0.8 to –0.4; 

Payment reform may be used to better align appropriate financial incentives with better quality of care.

Ellen Meara, PhD

Articles

Forgotten Patients: ACO Attribution Omits Those With Low Service Use and the Dying

Medicare's Bundled Payments for Care Improvement Initiative: Expanding Enrollment Suggests Potential for Large Impact

Impact of Payment Reform on Chemotherapy at the End of Life