Current State of Value-Based Purchasing Programs
The US healthcare system is rapidly moving toward rewarding value. Recent legislation, such as the Affordable Care Act and the Medicare Access and CHIP Reauthorization Act, solidified the role of value-based payment in Medicare. Many private insurers are following Medicare’s lead. Much of the policy attention has been on programs such as accountable care organizations and bundled payments; yet, value-based purchasing (VBP) or pay-for-performance, defined as providers being paid fee-for-service with payment adjustments up or down based on value metrics, remains a core element of value payment in Medicare Access and CHIP Reauthorization Act and will likely remain so for the foreseeable future. This review article summarizes the current state of VBP programs and provides analysis of the strengths, weaknesses, and opportunities for the future. Multiple inpatient and outpatient VBP programs have been implemented and evaluated; the impact of those programs has been marginal. Opportunities to enhance the performance of VBP programs include improving the quality measurement science, strengthening both the size and design of incentives, reducing health disparities, establishing broad outcome measurement, choosing appropriate comparison targets, and determining the optimal role of VBP relative to alternative payment models. VBP programs will play a significant role in healthcare delivery for years to come, and they serve as an opportunity for providers to build the infrastructure needed for value-oriented care.
- costs and cost analysis
- outcome assessment (health care)
- reimbursement mechanisms
- value-based purchasing
US healthcare has reoriented toward quality and value, incorporating both the health outcomes and the resources allocated to achieve those outcomes. The need for value is clear. With healthcare consuming almost one-fifth of the US economy, the burden of healthcare expenditures continues to crowd out funds for other society essentials such as education, infrastructure, and social security programs.1 Despite the fact that the United States ranks the highest in the world on healthcare spending, the US ranks the lowest on health performance indicators among 11 comparable nations.2,3 Although provider organizations, insurance companies, and government payers have attempted to improve quality and lower cost since the 1990s, the Affordable Care Act launched unprecedented reforms to improve healthcare value. A wide spectrum of payment models has been introduced that balance financial rewards and risks based on provider performance on specific measures, such as clinical quality, patient experience, and cost (Figure 1).
The shift toward value-based payment has only accelerated: in January 2015 the US Department of Health and Human Services announced their intent to tie 85% of all traditional Medicare payments to quality or value by 2016 and 90% of payments by 2018.4 The April 2015 passage of the Medicare Access and CHIP Reauthorization Act (MACRA) repealed the long-standing, unsuccessful Sustainable Growth Rate formula for Medicare and widely adopted alternative payment models, thus demonstrating the bipartisan and far-reaching commitment to valued-based payment solutions.5 With this growing traction, it is critical for clinicians to better understand the programmatic structure, the data about the strengths and weaknesses, and the prospects for the future of value-based purchasing.
Cardiovascular care has been and continues to be at the forefront of these changes. Acute myocardial infarction and heart failure have been the focus of policy efforts to improve quality and value for 3 main reasons. First, cardiovascular disease and its associated risk factors impact millions of Americans with an ensuing large morbidity and mortality burden. Second, cardiovascular disease is costly both in the direct healthcare costs and the indirect costs of the illnesses on society.6 Last, the cardiovascular community has been a leader in establishing an evidence base of randomized controlled trials that have led to rigorous guidelines supporting clinical care. These evidence-based guidelines naturally lend themselves to the establishment of quality measures by which performance on delivering guideline-based care can be evaluated. Thus, cardiovascular disease has been integral to quality measurement and incentive programs from their very inception and will continue to be. Cardiologists and other cardiovascular disease providers must understand these programs to succeed in these programs and advocate for program designs that are fair, rational, and best serve the needs of providers and, most importantly, patients.
Concept of Value-Based Purchasing
Traditionally the US healthcare system has relied on fee-for-service compensation, where reimbursements were provided for each service provided, regardless of the resulting patient outcomes or costs. The paradigm has shifted to paying for value, defined as the health outcomes or quality achieved in relation to the costs of the care provided.7 Value can be increased by improving outcomes and quality, reducing costs, or both.
Although value-based purchasing (VBP) can refer to a wide variety of payment strategies that link provider performance and reimbursement, this article will focus on VBP programs where providers are paid fee-for-service with payment adjustments up or down based on value metrics, a structure also known as pay-for-performance. This article will not address payment models that have been covered by other pieces in this series such as Patel et al8 on accountable care organizations (ACOs) and Shih et al9 on bundled payments. This pay-for-performance VBP mechanism serves as an introductory model that allows providers to gain experience at relatively low risk with the essentials of alternative payment models: quality metrics, performance improvement, and payment incentives and risks. This could be the critical transitional step to allow organizations to adopt more advanced alternative payment models involving greater potential financial risk and benefit, such as shared savings models for ACOs.
To better conceptualize, develop, implement, and evaluate VBP programs, several frameworks have been created to outline the primary components and interacting factors.10–13 The design and evaluation of VBP programs depends on 3 main influences: the external environment, provider characteristics, and program features. (Figure 2) The external environment includes regulatory changes, payment policies, patient preferences, and other quality improvement initiatives that can either promote or thwart the potential success of VBP programs. Provider characteristics include structure of the healthcare system, organizational culture, available resources and capabilities (especially in information technology), and patient population served. Program features include defining the targeted patient population, the program goals, measures, financial incentive, and risk structure. Specific program structure considerations include the level at which the data are analyzed and incentives are provided – is it individual physicians, physician groups, or the entire organization. Understanding the multiple factors influencing VBP programs is essential to designing, evaluating, and importantly succeeding in these programs. Moreover, minute details of the measures, formulas for calculating performance, and rules for defining rewards and penalties can have substantial impact on who wins and who loses, and on unintended consequences such as potentially worsening healthcare disparities.14,15
Major VBP Programs
There are numerous VBP pay-for-performance programs both internationally and within the United States, encompassing a wide variety of healthcare delivery settings, and implemented by both commercial insurance companies and by the Centers for Medicare and Medicaid Services. This article will detail several key historical and current VBP programs and specific evaluation studies, ranging from the outpatient to the inpatient settings. Cardiovascular care has been a foundational aspect of many programs, especially in the treatment of acute myocardial infarction, and the secondary prevention of cardiovascular disease and heart failure, as well (specific measures are detailed in the Table).
Outpatient VBP Programs
California Integrated Healthcare Association
One of the first and largest private, nongovernmental multipayer VBP programs is the California Integrated Healthcare Association (IHA) program. The IHA program assesses quality in the 4 domains of clinical quality, meaningful use, patient experience, and appropriate resource use, including cardiovascular health-related metrics: process measures on secondary prevention such as blood pressure control and statin therapy.16 These quality measures translate into 3 forms of incentives: financial, public reporting, and public recognition. From 2004 to 2013, the IHA awarded a total of nearly $500 million in bonus payments; however, to put these rewards in context, the total incentive payment in 2007 was $52 million, which represented only ≈2% of total base compensation.17 The public reporting was an online quality report card on overall and individual measure performances, and 5 public recognition awards were developed. The concept is that the financial, public reporting, and public recognition incentives complement each other to drive improvement.
A 2010 IHA white paper report showed that although physician organizations improved in all measurement areas, the performance range was wide. The average state performance was <75th percentile on national benchmarks for Health Effectiveness Data and Information Set measures, whereas the top quartile of physician groups scored comparable to the 90th percentile. State performance improved an average of 3% annually in overall clinical quality, with improvements in individual measures ranging from 5.1% to 12.4%. Results showed that physician groups in the lowest performance quartile had greater absolute improvement than those that started in the highest quartile, but the resulting variation decrease was minimal because of the small size of groups in the lowest quartile. There was marginal improvement in patient experience, with regional variation in a pattern similar to clinical quality regional variation.18
State Medicaid Programs
The Commonwealth Fund surveyed state Medicaid programs to determine the prevalence of VBP programs and found that, as of 2006, more than half of all state Medicaid programs had at least 1 existing VBP program and >70% had plans for new programs.19 The most common Medicaid VBP programs were managed care organizations that incorporated incentives based on Health Effectiveness Data and Information Set measures, with the majority providing rewards for achieving an attainment threshold. Because children represented half of all Medicaid beneficiaries, most programs focused on addressing the populations of children, adolescents, and women. Given the individualized nature of state Medicaid programs, there is anecdotal, qualitative evidence that these VBP programs are improving quality of care, but no large quantitative study analyzing the success of these VBP programs exists.
Medicare Physician Quality Reporting System and Value-Based Modifier
Initially called the Physician Voluntary Reporting Program when enacted in 2006, it began as a mere method of promoting reporting requirements by rewarding a one-time lump sum to physicians who sufficiently reported on at least 3 quality measures.20 The program changed in subsequent years, becoming the pay-for-reporting program Physician Quality Reporting Initiative, with changes to the number of measures, the reporting mechanisms, and potential bonus. Although the physician reporting began as pay-for-reporting, the program evolved under the Affordable Care Act to the pay-for-performance Physician Quality Reporting System, with penalties starting in 2015 for not reporting data, and broadening to performance-based penalties and bonuses in the Value-Base Payment Modifier (Value Modifier).21
Cardiovascular measures include appropriate coronary artery disease and heart failure medications, cardiac rehabilitation referrals, and control of blood pressure and cholesterol, as well. With most cardiovascular disease measures being based on clinical practice guidelines, a recent difficulty has been how to adapt the Physician Quality Reporting System measures for changes in the guidelines. For example, the Physician Quality Reporting System program measures included a low-density lipoprotein cholesterol target of <100 mg/dL for patients with ischemic vascular disease at the time when the 2013 American College of Cardiology and American Heart Association cholesterol guidelines were released removing recommendations for specific low-density lipoprotein cholesterol targets.22 Given the inevitable delay between the publication of evidence-based guidelines and those guidelines being translated into quality measures, a discrepancy existed for some time between the guidelines and the quality measures. Subsequently, the quality measures have been adjusted to reflect the updated clinical guidelines. This example is emblematic of the ongoing challenge of how to keep quality process of care measurement up to date with changing clinical standards of care.
Physician Quality Reporting System lays the foundation for the Affordable Care Act’s Value Modifier program, which mandates payment adjustments based on risk-adjusted quality and cost metrics to Medicare physician payments.21 These programs will remain effective until 2018, at which point the system will be replaced with the merit-based incentive payment system (MIPS) created in MACRA, which combines several federal quality programs. MIPS is a fee-for-service–based VBP program beginning in 2019 with payment linked to a composite score on quality, resource use, clinical improvement, and the use of electronic health records. For providers, substantial amounts of Medicare payments will be in play, with positive or negative adjustments up to 4% in 2019 and increasing over time to up to 9% in 2022.5 Exceptional performers may qualify for an additional 10% positive adjustment annually. The overall program will be budget neutral, so providers receiving bonuses will be offset by providers facing penalties. Under MACRA, physician organizations that participate in alternative payment models, such as ACOs, with more than minimal financial risk at stake will be exempt from MIPS. At the time of writing this article, the MACRA rule has not yet been finalized but will contain key details about the expectations for MIPS and APM participation. Cardiovascular disease is likely to play a large role in the measures available within MIPS and by which alternative payment models will be evaluated. Moreover, cardiologists will need to make important decisions about whether they participate in MIPS, or participate in an APM that involves the potential for financial losses as well as bonuses. For cardiologists practicing in large organizations, the opportunity exists to provide informed input to their organizations on which path to choose. For independent and small-group cardiologists, this decision of participating in MIPS or joining a qualified APM will be critical because the implications can be broad for both Medicare participation and more general practice affiliations.
Hospital VBP Programs
Medicare Premier Hospital Quality Inventive Demonstration
The Premier Hospital Quality Inventive Demonstration (Premier HQID) was a 6-year pay-for-performance demonstration by Centers for Medicare and Medicaid Services from 2003 to 2009 for 3 medical conditions (acute myocardial infarction, congestive heart failure, and pneumonia) and 2 surgical procedures (coronary artery bypass grafting and total hip/knee replacement). Premier HQID hospitals could receive either a 1% to 2% bonus or, starting in 2006, penalty adjustments, based on performance on 33 measures.23 Initially, hospitals were eligible if they achieved the top 20th percentile on metrics, but, in the fourth year, the program introduced bonuses to hospitals with substantial improvements in performance. Although HQID is one of the most widely studied programs, evidence is mixed on its effectiveness. Detailed below, some studies showed it improved quality measures, others showed no effect on outcomes such as mortality, and others showed a temporary effect on quality.
Lindenauer et al24 sought to determine the added effect of pay-for-performance to public reporting. Adjusting for baseline performance and characteristics, a range of 2.6% to 4.1% improvement over the study period was attributed to pay-for-performance. This suggested that public reporting itself provided a strong incentive for quality improvement, although financial incentives can modestly facilitate performance improvement. Interestingly, multivariable analysis results showed hospitals with the lowest baseline performance had the largest improvements in heart failure composite measures. Although this was not observed in measures for acute myocardial infarction or pneumonia, it raises the issue of whether greater bonuses should be awarded to higher absolute levels of performance metrics or improvement in those metrics.
To address questions about how VBP affected processes of care in the context of other quality improvement initiatives, Glickman et al25 analyzed data for patients with acute non–ST-segment–elevation myocardial infarction enrolled in the national CRUSADE initiative. No significant difference was found in the rate of quality composite score improvement or hospital mortality between HQID and control hospitals. Only 2 individual incentivized measures (aspirin at discharge and smoking cessation) and 1 individual nonincentivized measure (lipid-lowering medication at discharge) had slightly higher rates of improvement.
Noting that previous studies were mainly on process measures and not outcome measures, Ryan26 evaluated patient mortality, cost, and outlier classification from >6.7 million patients from 3570 hospitals with principal diagnoses of acute myocardial infarction, pneumonia, or coronary artery bypass grafting procedures across 4 years. Controlling for baseline differences, confounders, and time variants, the analysis showed that HQID had no significant effect on risk-adjusted 30-day mortality or on cost growth for the studied diagnoses. To evaluate a change in HQID incentive design to award for improvements, Ryan et al27 reevaluated relative quality improvement of HQID hospitals to matched samples over a 5-year period. Although the initial incentive structure led to better improvement rates in the HQID hospitals, the incentivizing of improvement on top of attainment failed to create higher performance. It is noteworthy that the incentive structure provided the strongest incentive to hospitals just above the median instead of to the lowest performers.
Although these earlier studies suggested HQID had a small early effect on measures of quality, Werner et al28 specifically studied whether the effect was sustained and what programmatic, provider, or external factors may have contributed to the program’s success. In a comparison of Premier HQID hospitals with matched control hospitals, they found a significant initial difference in performance improvement in HQID hospitals in comparison with controls. However, this effect difference decreased after 3 years and was no longer statistically different after 5 years. Subgroup analysis showed that HQID hospitals eligible for larger incentives had larger improvements than control hospitals in comparison with the HQID hospitals eligible for smaller incentives. Further analysis also showed that larger improvements were seen in HQID hospitals in less competitive markets and in better financial health. This suggests the design of future VBP programs may need to incorporate larger incentives, more frequent feedback, and greater focus on measurement of the individual clinician. The results also suggest that hospitals in poorer financial health may need greater up-front resources to invest in quality improvement initiatives.
Although the aforementioned studies showed no early effects on mortality, Jha et al29 studied the longer-term effects on mortality by comparing HQID hospitals with control hospitals in public reporting only. Multivariate regression analysis showed that 30-day risk-adjusted mortality rates for the 5 specified conditions were similar at baseline, with similar rates of decline and no significant difference by the end of the 6-year study period. There was also no difference found when considering factors such as potential financial incentive, institutional financial health, and market competitiveness.
Statewide, All-Payer Inclusive System
Since 1977 Maryland has had a prospective hospital payment system with specific payments for all inpatient and outpatient services for all public and private payers. Two VBP programs were implemented in Maryland – the 2008 Quality-Based Reimbursement Program, which provides financial rewards and penalties based on processes measured, and the 2009 Maryland Hospital-Acquired Conditions Program, which rewards and penalizes based on a hospital’s risk-adjusted rate of preventable complications or hospital-acquired conditions. Calikoglu et al30 assessed the Maryland VBP programs by studying the state average and hospital variation in performance rates over 4 years. Although the study results are limited by lacking a robust control group, process measures improved and hospital-acquired conditions decreased by >15% over 2 years, or an estimated cost savings of $110.9 million. This study suggested some VBP program success when all payers have aligned value-based incentives and the potential of a statewide value system.
Medicare Hospital VBP Program
Implemented as part of the Affordable Care Act, the Hospital VBP Program built on Hospital Inpatient Quality Reporting Program pay-for-reporting program for hospitals that started in 2003.31 Beginning in 2012, the Hospital VBP Program used the Hospital Inpatient Quality Reporting Program’s infrastructure to introduce Medicare payment adjustments for acute inpatient services. The program creates an incentive payment fund by reducing inpatient Medicare payments by a percentage that increases over time, then distributes that fund to hospitals by applying a value-based adjustment factor that is determined by the hospital’s Total Performance Score. The Total Performance Score initially consisted of 2 domains of clinical processes of care and a survey in patient experience. In later years the program evolved their measures, standards, weighting, and structure such as including additional domains in outcomes, safety, and efficiency.32–34 Outcome measures include mortality for acute myocardial infarction and heart failure. Performance is scored based both on attaining a certain achievement level and on improvement in comparison with each hospital’s baseline and benchmark.
In 2016, the Hospital VBP Program increased to representing 1.75% of all Medicare payments to hospitals. In practical application, hospitals will all have 1.75% of all payments withheld, which they can then earn back depending on their Total Performance Score. Some hospitals will earn <1.75% back and have a penalty, and some hospitals will earn >1.75% back and have a bonus. In 2016, about half of hospitals will see a minimal change in the Medicare payments, with the overall adjustment being between –0.4% and +0.4%. The worst performing hospital will earn back none of the 1.75% withheld, and the highest performing hospital will have a net increase of ≈1.25%.35 Although long-term studies have not yet been completed, an evaluation of the first year of the Hospital VBP Program showed no improvement in clinical process or patient experience among hospitals exposed to the program.36 Additional studies are certain with grants having already been awarded for this purpose.
International VBP Programs
The European Observatory on Health Systems and Policies, in partnership with the World Health Organization, published a report on alternative payment models in multiple Organization for Economic Co-operation and Development countries.37 VBP programs exist in primary care, specialist, and hospital categories, with targeted performance measures often overlapping such as in clinical quality, priority services such as immunizations and cancer screenings, and efficiency. Interestingly, other performance domains not commonly seen in the US programs include improving equity or reducing health disparities, although questions exist about whether VBP approaches effectively reduce disparities.38 Although most VBP programs are rewarded or penalized based on either attainment, improvement, or relative ranking, Turkey has an all-or-nothing approach where providers avoid a penalty only if all members reach target performance. The review found that the typical incentive was <5% of total provider income, with the exception of the UK and Turkey programs where general practitioner practices had the potential to increase compensation by ≈25% and 20%, respectively. The evaluation of 12 international case studies showed mixed results. Some programs improved rates of preventative services for some conditions and some programs had improvements in chronic disease management. In general, there were no significant impacts on health outcome measures. In programs that achieved broad significant improvements in processes of care, such as Germany and Estonia, gains in efficiency and cost savings were also reported. Interestingly, cost savings and increased efficiency were not evident in programs with cost reduction incentives, such as targeting brand medications to be replaced with generics.37
Although the 2004 UK chronic disease management pay-for-performance program that could increase family practitioners’ income by 25% (or $40 200) was among those that had mixed results, the National Health Service implemented the Advancing Quality program in 2008 focusing on hospital payments that showed significant improvements in mortality outcomes.39 Analysis of early results showed risk-adjusted mortality for all conditions decreased, and the VBP was associated with a 1.3% reduction in combined mortality.40 However, this short-term success in the first 18 months of the program was not sustained in a longer-term analysis of the next 24 months. Although both participating and control hospitals lowered mortality rates, the initial benefit of program participation disappeared over time, and control hospitals had similar performance to incentivized hospitals.41
Overall Analysis of VBP Approach
VBP programs have been implemented widely, and their impact has been marginal thus far. The challenge of determining the effectiveness of VBP programs is not only complicated by how one might define success, but also by the external factors of simultaneous quality improvement programs such as public reporting. Moreover, the effectiveness of VBP programs stems not only from the process of implementation, but also from specifics of program design and incentive structure. VBP programs essentially remain fee-for-service structures of payment, sustaining all the negative incentives of that system.
In 2014, Damberg et al42 with RAND were commissioned by the US Department of Health and Human Services to evaluate available literature on VBP models including pay for performance, ACOs, and bundled payments. Fifty studies of 1891 articles screened on pay-for-performance programs were reviewed and ranked by study method quality (good, fair, poor) and strength of evidence reflecting a true effect (high, moderate, low, insufficient). Results were mixed. Thirty-nine studies focused on physicians or physician groups, with 7 of the studies determined to be of good quality and 15 of the studies found to be of fair quality. There were modestly positive results of incentives improving treatment, screening, and prevention measures. Eleven studies focused on hospitals, mainly on the Centers for Medicare and Medicaid Services HQID program, with 6 studies having good quality and showing modestly positive results.
Future Directions for VBP Programs
The overall effectiveness of VBP programs has been marginal thus far. Although many studies have examined VBP programs and sought to understand what works and why, the answers still elude us. Several potential reasons exist for VBP’s meager impacts. First, the financial incentives may be inadequate to drive change. Second, the quality measurement systems may be overly complex such that providers are confused as to which measures are most closely linked to the incentives. Third, the delay in time between measure performance and incentives decouples the 2 events such that providers do not closely connect cause and effect. Fourth, incentives are often rolled into standard Medicare payments as a percentage adjustment, rather than being called out as a separate payment to highlight the incentive. Last, the multiple Medicare programs and commercial payer programs collectively create a confounding environment for providers who become uncertain as to what interventions to work on and which incentives link to which performance. At least on the Medicare programs, this aspect should improve with MACRA because multiple programs are being combined into MIPS. Future program designs that more closely adhere to behavioral economic principles may more effectively drive the intended performance. In addition, some specific programmatic elements are worth considering.
Quality is a complex concept with aspects in infrastructure, clinical processes, intermediate outcomes, and true outcomes. Because payments are increasingly tied to measurements, there has been debate whether the current set of measures actually correlate with true outcomes that are clinically significant and that are meaningful for patients. Although some studies have shown that process measures did modestly correlate with risk-adjusted mortality rates, strong data to support a correlation between process measures and outcomes is lacking.43 Outcome measures are the ultimate goal in evaluating performance, although outcomes must be risk adjusted and questions exist about how fully the risk adjustment models account for differences in medical risk. An even more challenging question is whether and how to adjust for socioeconomic factors.44 The outcomes that are calculable from claims data also represent a small set of the true outcomes of clinical interest. Extant measures are limited by the reliance of extractable data from claims. There has been an evolving transition to outcome-based measures, although more development is needed in both disease-specific and overall population-level health measures. An additional consideration is to design VBP programs stratifying measures with different weights based on their relative cost-effectiveness, clinical benefit, and value. This will require greater health economics research in cost-effectiveness and cost-benefit analysis of interventions that may be high value yet high cost. The science of quality measurement will need to quickly and substantially improve to meet the expanding need to assess quality performance in VBP programs.
There is much that is not known about how to best design incentive structures, including the size of the incentive, the recipient (individual versus organization), how reward eligibility is determined (attainment, improvement), frequency of information feedback or incentive, and inclusion of nonfinancial incentives (recognition). With regard to incentive size, studies have shown varied results where programs with incentives of as little as $2 per patient incentive have been effective and others with $10 000 per practice have not.28 Little to no research has been done regarding the impact of more frequent performance feedback. Studies in behavioral economics have shown that people tend to discount future losses at lower rates than gains and larger outcomes more than smaller outcomes, suggesting that a high incentive frequency may be more effective, especially for the risk adverse.45 Behavioral economics also demonstrates loss and risk aversion: people have stronger preferences to avoid losses than acquiring gains, even when the objective value is equivalent.46 In terms of incentive recipients, 1 systematic review showed that targeting incentives directly to providers versus the organization had greater positive results.47 Evaluations of existing and planned VBP programs may resolve questions of the optimal incentive structures for various settings; however, such answers will require operational programs to be designed with rigorous evaluations in mind. In addition, incentive structures are often constrained by legislation (as in the case of budget neutrality for the VBP programs), potentially limiting innovative incentive designs (such as frequent incentive payments). Future considerations of incentive structure creation include categorizing performance such as meeting previous performance, improving from previous performance, meeting a minimum threshold standard, and being a regional best performer.
It has been suggested that VBP programs could exacerbate health disparities by penalizing organizations with large disadvantaged populations already less likely to receive recommended care because of factors including lower health literacy, resource constraints of both patients and facilities, and cultural norms. This concept was supported by Chien et al48 who geocoded quality data of 12 000 practices in the California IHA program, and found a significant association between higher socioeconomic areas and higher performance scores, despite outliers. Including patient satisfaction as a major performance indicator may also be disadvantageous to safety net hospitals. In a study dividing safety net hospitals into quartiles by proportion of patients receiving Supplemental Security Income or Medicaid, there was a significant difference in patient satisfaction scores.49 However, Ross et al,50 in a study of Medicare fee-for-service, showed minimal differences in mortality and readmission rates between safety net and nonsafety net hospitals. Clearly the potential for unintended consequences, such as the exacerbation of healthcare disparities, requires consideration in program design and close ongoing monitoring and evaluation.51,52
Unmeasured Priority Areas
There is concern about teaching to the test where VBP programs may cause providers to focus on incentivized areas of care, and possibly ignore nonincentivized but still essential and valued aspects of care.53 Although VBP programs strive to use a wide range of measures covering multiple clinical areas, unmeasured areas will certainly exist and the possibility for inattention is real. The shift toward incentivizing broad outcomes measures may mitigate some of this risk.
There is extensive geographic variation in quality and costs, often previously attributed to practice variations and cost-of-living adjustments, but recent research has also shown there are significant price charge variations even in 1 geographic region and no correlation between federal Medicare and private insurance prices.54 Hospital comparison targets are often determined nationally. For ACOs, some have recommended using regional financial benchmarks, and this element may be incorporated into the Medicare Shared Savings Program.55,56 Perhaps similar consideration of regional financial benchmarks should be included in VBP programs.
Interaction With Alternative Payment Models
Although it is clear that nearly all Medicare payments will be linked to value and likely many private insurance payments, it is less clear how, or what mix of programs will exist and what programs will dominate. Medicare has signaled that it favors either episode-based payments, such as bundled payments, or global payment models, exemplified by ACOs.5 However, bundled payments may not be appropriate for all conditions or all procedures, and not all providers will have the scale or capabilities to participate in an ACO. Thus, it is likely that a variety of alternative payment models will exist in the years to come, with bundled payments, ACOs, and still fee-for-service for some episodes of care. The fee-for-service that continues to exist will have some incentives linked to payment, likely in forms similar to current and recent VBP programs.
The strength of evidence on the effectiveness of pay-for-performance VBP programs on improving health delivery and patient outcomes is mixed and modest. Many studies were over a short time frame; whereas a longer study period may allow better observation of long-term outcomes, it also makes the study more vulnerable to confounding by other factors. As healthcare payment models increasingly move toward risk and accountability, there needs to be greater understanding of the individual design levers in creating a VBP program and responsiveness to the incentives of individual providers and provider organizations, as well. With recent legislation expanding VBP for organizations not in qualifying alternative payment models, these programs provide opportunities for providers of cardiovascular care to build infrastructure for value-oriented care.
Dr Borden receives salary as the Director of Healthcare Delivery Transformation at the George Washington University Medical Faculty Associates. Dr Wasfy receives salary as assistant medical director of the Massachusetts General Physicians Organization. The other authors report no conflicts.
- © 2016 American Heart Association, Inc.
- 1.↵Centers for Medicare & Medicaid Services. NHE-Fact-Sheet. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet.html. Published July 28, 2015. Accessed August 27, 2015.
- 2.↵Mirror, Mirror on the Wall, 2014 Update: How the U.S. Health Care System Compares Internationally. Commonwealth Fund. June 2014. http://www.commonwealthfund.org/publications/fund-reports/2014/jun/mirror-mirror. Accessed August 27, 2015.
- Squires D
- 5.↵Centers for Medicare & Medicaid Services. MACRA: MIPS & APMs. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/MACRA-MIPS-and-APMs/MACRA-MIPS-and-APMs.html. Published October 15, 2015. Accessed October 15, 2015.
- Ojeifo O,
- Berkowitz SA
- Patel KK,
- Cigarroa JE,
- Nadel J,
- Cohen DJ,
- Stecker EC
- Shih T,
- Chen LM,
- Nallamothu BK
- Dudley RA,
- Frolich A,
- Robinowitz DL,
- Talavera JA,
- Broadhead P,
- Luft HS
- Fisher ES,
- Shortell SM,
- Kreindler SA,
- Van Citters AD,
- Larson BK
- 13.↵Theory and Reality of Value-Based Purchasing: Lessons from the Pioneer. Agency for Healthcare Research and Quality. http://www.ahrq.gov/professionals/quality-patient-safety/quality-resources/tools/meyer/index.html. Published November 1, 1997. Accessed August 27, 2015.
- Borden WB,
- Blustein J
- 16.↵Integrated Healthcare Association. Factsheet: Value Based Pay for Performance in California. 2015. http://www.iha.org/sites/default/files/resources/vbp4p-fact-sheet-final-20150925.pdf. Accessed March 30, 2016.
- Yegian J,
- Yanagihara D
- 18.↵Integrated Healthcare Association. Pay for performance report of results for measurement year 2010. Oakland, CA: Integrated Healthcare Association; 2012.
- Hartman T,
- Kukmerker K
- 20.↵Centers for Medicare & Medicaid Services. Overview - Physician Quality Reporting System. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/PQRS/. Published June 12, 2015. Accessed August 28, 2015.
- 21.↵Centers for Medicare & Medicaid Services. Value Based Payment Modifier. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeedbackProgram/ValueBasedPaymentModifier.html. Published August 4, 2015. Accessed August 28, 2015.
- Stone NJ,
- Robinson JG,
- Lichtenstein AH,
- Bairey Merz CN,
- Blum CB,
- Eckel RH,
- Goldberg AC,
- Gordon D,
- Levy D,
- Lloyd-Jones DM,
- McBride P,
- Schwartz JS,
- Shero ST,
- Smith SC Jr.,
- Watson K,
- Wilson PW,
- Eddleman KM,
- Jarrett NM,
- LaBresh K,
- Nevo L,
- Wnek J,
- Anderson JL,
- Halperin JL,
- Albert NM,
- Bozkurt B,
- Brindis RG,
- Curtis LH,
- DeMets D,
- Hochman JS,
- Kovacs RJ,
- Ohman EM,
- Pressler SJ,
- Sellke FW,
- Shen WK,
- Smith SC Jr.,
- Tomaselli GF
- 23.↵Centers for Medicare & Medicaid Services. Premier Hospital Quality Incentive Demonstration. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/HospitalPremier.html. Published January 9, 2013. Accessed August 28, 2015.
- Ryan AM,
- Blustein J,
- Casalino LP
- Werner RM,
- Kolstad JT,
- Stuart EA,
- Polsky D
- Calikoglu S,
- Murray R,
- Feeney D
- 31.↵Centers for Medicare & Medicaid Services. Hospital Inpatient Quality Reporting Program. https://www.cms.gov/medicare/quality-initiatives-patient-assessment-instruments/hospitalqualityinits/hospitalrhqdapu.html.
- 32.↵Centers for Medicare & Medicaid Services. Frequently Asked Questions Hospital Value-Based Purchasing Program. Last Updated March 9, 2012. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/hospital-value-based-purchasing/Downloads/FY-2013-Program-Frequently-Asked-Questions-about-Hospital-VBP-3-9-12.pdf.
- 33.↵Centers for Medicare & Medicaid Services. Hospital Value-Based Purchasing Program. March 2013. https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNProducts/downloads/Hospital_VBPurchasing_Fact_Sheet_ICN907664.pdf.
- Wheeler B
- 35.↵Centers for Medicare & Medicaid Services. Fiscal Year (FY) 2016 Results for the CMS Hospital Value-Based Purchasing Program. https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheets-items/2015-10-26.html. Published October 26, 2015. Accessed March 31, 2016.
- Cashin C,
- Chi Y-L,
- Smith P,
- Borowitz M,
- Blustein J,
- Weissman JS,
- Ryan AM,
- Doran T,
- Hasnain-Wynia R
- Damberg CL,
- Sorbero ME,
- Lovejoy SL,
- Martsolf GR,
- Raaen L,
- Mandel D
- 44.↵National Quality Forum. Risk Adjustment for Socioeconomic Status or Other Sociodemographic Factors. Washington, DC: National Quality Forum; 2014. http://www.qualityforum.org/Publications/2014/08/Risk_Adjustment_for_Socioeconomic_Status_or_Other_Sociodemographic_Factors.aspx.
- Tversky A,
- Kahneman D
- Ross JS,
- Bernheim SM,
- Lin Z,
- Drye EE,
- Chen J,
- Normand SL,
- Krumholz HM
- Ryan AM,
- McCullough CM,
- Shih SC,
- Wang JJ,
- Ryan MS,
- Casalino LP
- Cooper Z,
- Craig SV,
- Gaynor M,
- Reenen JV
- 55.↵The Revised Medicare ACO Program: More Options … And More Work Ahead. Health Affairs. http://healthaffairs.org/blog/2015/06/16/the-revised-medicare-aco-program-more-options-and-more-work-ahead/. Accessed March 31, 2016.
- 56.↵Centers for Medicare & Medicaid Services. Medicare Program; Medicare Shared Savings Program; Accountable Care Organizations-Revised Benchmark Rebasing Methodology, Facilitating Transition to Performance-Based Risk, and Administrative Finality of Financial Calculations. Federal Register. 2016:81:5824–5872. https://www.federalregister.gov/articles/2016/02/03/2016-01748/medicare-program-medicare-shared-savings-program-accountable-care-organizations-revised-benchmark. Accessed March 31, 2016.