Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE (oxfordre.com/economics). (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 14 October 2019

Pay-for-Performance and Long-Term Care

Summary and Keywords

Pay-for-performance programs have become a prominent supply-side intervention to improve quality and decrease spending in health care, touching upon long-term care, acute care, and outpatient care. Pay-for-performance directly targets long-term care, with programs in nursing homes and home health. Indirectly, pay-for-performance programs targeting acute care settings affect clinical practice for long-term care providers through incentives for collaboration across settings.

As a whole, pay-for-performance programs entail the identification of problems it seeks to solve, measurement of the dimensions it seeks to incentivize, methods to combine and translate performance to incentives, and application of the incentives to reward performance. For the long-term care population, pay-for-performance programs must also heed the unique challenges specific to the sector, such as patients with complex health needs and distinct health trajectories, and be structured to recognize the challenges of incentivizing performance improvement when there are multiple providers and payers involved in the care delivery.

Although empirical results indicate modest effectiveness of pay-for-performance in long-term care on improving targeted measures, some research has provided more clarity on the role of pay-for-performance design on the output of the programs, highlighting room for future research. Further, because health care is interconnected, the indirect effects of pay-for-performance programs on long-term care is an underexplored topic. As the scope of pay-for-performance in long-term care expands, both within the United States and internationally, pay-for-performance offers ample opportunities for future research.

Keywords: pay-for-performance, value-based purchasing, long-term care, nursing homes, home health, quality of care, financial incentives, health economics

Linking Pay to Performance in Long-Term Care

Pay-for-performance programs have become a prominent supply-side intervention to improve efficiency in health care, touching upon long-term care, acute care, and outpatient care (Norton, 2018; Doran, Maurer, & Ryan, 2017; Golden & Sloan, 2008). It is a payment mechanism to pay providers based on quality rather than solely based on quantity or intensity. The ultimate goal of pay-for-performance programs is to use pecuniary rewards to incentivize healthcare providers to improve on quality and spending outcomes. As a pioneer of pay-for-performance programs, the United States has used pay-for-performance programs targeting nursing homes since 1979 (Briesacher, Field, Baril, & Gurwitz, 2009; Norton, 1992). Since then, public insurers have expanded the scope of pay-for-performance programs in long-term care to also target home health care. Together, there are approximately 4.9 million home health and 1.4 million nursing home users in the United States each year (Harris-Kojetin et al., 2016). By 2022, pay-for-performance programs will directly affect all nursing homes and home health agencies in the United States. Because pay-for-performance is ubiquitous across the other parts of the healthcare system, pay-for-performance programs also indirectly affect long-term care providers. Pay-for-performance in hospitals, for example, lays the groundwork for cross-setting collaboration that may alter clinical practice for long-term care providers. As the scope of pay-for-performance programs in long-term care expands, pay-for-performance offers ample opportunities for researchers and policy makers.

This article outlines the theoretical rationale for pay-for-performance programs generally. It also focuses on issues specific to the long-term care sector, with particular emphasis on formal care delivered by rehabilitative skilled nursing facilities, home health agencies, and custodial-care nursing facilities. The theoretical rationale serves as the conceptual background to motivate the various elements common to all pay-for-performance programs. Specifically, all pay-for-performance programs have a policy purpose, performance measures, methods to assess provider performance, and ways to incentivize behavior with financial rewards. Using existing pay-for-performance programs as a guide, the article describes elements common to all pay-for-performance programs, assesses the trade-offs embedded within each element, and focuses on issues specific to the long-term care sector. It also describes why it is prudent to consider the indirect effects of pay-for-performance programs on long-term care because health care is interconnected.

A summary of empirical results is presented. While there is limited rigorous empirical evidence on the effectiveness of pay-for-performance programs in long-term care, the literature can help illuminate gaps in research. Finally, the article provides an overview of select pay-for-performance programs in long-term care because these programs are large in scope, have potentially sizable effects on long-term care, and offer opportunities for research.

Conceptual Rationale for Pay-for-Performance

There has been tremendous optimism regarding the potential for pay-for-performance to improve healthcare functioning (Institute of Medicine, 2007). This section describes the conceptual basis for pay-for-performance in health care generally, using Medicare as an example. The use of pay-for-performance in health care has become commonplace among public insurers, where insurers use financial incentives to reward (or penalize) healthcare providers on measurable outcomes. Public insurers such as Medicare and Medicaid are public stewards and thus act on behalf of both patients and taxpayers (Golden & Sloan, 2008). The rewarded measures in pay-for-performance have typically included healthcare quality—outcomes important to patients—and more recently healthcare spending—outcomes important to taxpayers. Proponents argue that pay-for-performance programs can better align the competing objectives between payers and healthcare providers. Without pay-for-performance, misaligned objectives have contributed to suboptimal quality and overspending. For example, while Medicare would like to maximize quality and minimize costs to society, healthcare providers may make treatment decisions that maximize profits, which could lead to suboptimal quality and higher spending.

The challenge of aligning objectives between an insurer and provider is evident from the trade-offs in the compensation systems between Medicare and providers. For instance, cost-based payment systems protect against low effort from providers but may also encourage overtreatment (McGuire, 2008). Prospective payments—predetermined rates for a unit of care such as a hospitalization—may encourage more judicious use of resources during the unit of care but may also encourage under-provision of effort and avoidance of less profitable patients (Ellis, 1998). Prospective payment for hospitalizations means that discharging patients quickly is more profitable than lengthening a hospital stay to find the most ideal post-discharge setting (Kane, 2011). Furthermore, payments based on individual units of services, like a hospitalization, provide no incentive to coordinate across care settings (Robinson, 2001). For long-term care patients who suffer from multiple chronic conditions, care can be fragmented (Van Cleave et al., 2016). Treatment may be spread across several providers, including primary care physicians, specialists, hospitals, post-acute care, and custodial care. The involvement of multiple providers makes care coordination more difficult (Cebul, Rebitzer, Taylor, & Votruba, 2008). Lack of care coordination can lead to high readmissions rates, a commonly used measure indicative of low quality and unnecessary resource use (Naylor et al., 1999; Coleman, Parry, Chalmers, & Min, 2006; Coleman, Mahoney, & Parry, 2005). Traditionally, rather than being incentivized to prevent readmissions, hospitals actually had a financial incentive to readmit patients because subsequent readmissions can be reimbursed as additional admissions (Mor, Intrator, Feng, & Grabowski, 2010). Nursing homes and home health agencies can also benefit financially from a hospital readmission, which may lead to additional post-discharge care (Konetzka, Stuart, & Werner, 2018). Conflicting objectives between health insurers and providers resulting from compensation designs lead to lower quality and higher costs than society would otherwise choose to bear.

In theory, market forces could mitigate the adverse effects from misaligned objectives between insurers and providers. If consumers demand high quality and low spending, then providers should compete on quality and costs. Unfortunately, market forces in health care are weak and do not lead to this outcome.

First, it is difficult for patients to be fully informed consumers, even on issues of high importance to patients, such as quality (Arrow, 1963). Even if a patient could perfectly observe treatment effort, the channels by which high quality of care are produced are largely unknown (Greenhalgh, Howick, & Maskrey, 2014). Further, because a patient’s health trajectory is difficult to determine, teasing apart the effects of treatment from natural condition progression is difficult. Because patients are typically not medical experts, they cannot determine whether their providers’ actions are in their best interests or even whether the providers are sufficiently skilled to deliver the proper treatments. Because the patient faces significant barriers to distinguishing high quality from low quality, the market for true quality is weak.

Second, health insurance insulates patients from true healthcare costs because insured patients do not pay the full cost of their care (Pauly, 1968). Particularly for Medicare patients, out-of-pocket spending is low and does not vary by the quality of providers (Chandra Finkelstein, Sacarny, & Syverson, 2016). Therefore, patients face little, if any, financial consequences of their treatment choice. Thus, the market forces for low spending are also weak in health care.

Together, characteristics of providers and patients have led public insurers to use pay-for-performance as a supply-side tool to align incentives between insurers and providers (Golden & Sloan, 2008; Rosenthal, 2006). It assumes that providers respond to financial rewards for improving measurable dimensions of quality and spending. The literature has found that providers respond to these incentives (Prendergast, 1999; Golden & Sloan, 2008; Norton, Li, Das, & Chen, 2018). Having reviewed the general motivation for pay-for-performance, the subsequent sections explore each of the key elements in a typical pay-for-performance program.

Elements of Pay-for-Performance

The elements of pay-for-performance need to be carefully calibrated for it to reward true dimensions of quality without introducing even greater sources of inefficiency. This section discusses four components of pay-for-performance programs: the problems that it seeks to solve, measurement of the dimensions it seeks to incentivize, methods to combine and translate performance to incentives, and application of the incentives. For each of these, issues that are especially pertinent for long-term care are highlighted.

Determining the Problems

While the general goals of pay-for-performance may be applicable to all sectors, each pay-for-performance program should have a set of specific aims that are custom to the needs and inefficiencies of that sector (Richardson, 2012). Each pay-for-performance program should be clear about the specific sources of inefficiencies that it would address. Using information asymmetry as an example, how it affects quality in long-term care is examined in three ways: when to use pay-for-performance to target long-term care indirectly, when to target long-term care directly, and which dimensions to reward.

First, patients may be under-informed about quality when making long-term care decisions at hospital discharge. Barriers, such as time constraints during hospital discharge, make it difficult for patients to use quality information in decision making even if information were broadly available. Medicare has provided publicly available quality information for nursing homes since 1998 and for home health agencies since 2003. Therefore, information is generally available. However, public reporting does not eliminate barriers present during a hospital discharge. A significant proportion of long-term care users have complex physical and cognitive health needs (Harris-Kojetin et al., 2016). These social and clinical vulnerabilities may complicate their ability to take advantage of publicly available quality information when they must also deal with the pressures of acute health issues. The availability of information also does not address the lack of incentives for hospitals to help patients identify high-quality post-discharge healthcare providers. To address suboptimal long-term care placement after hospital discharge, pay-for-performance can reward discharging hospitals for helping patients select better post-discharge providers. Pay-for-performance could directly target hospitals and indirectly target long-term care providers.

Second, patients are unable to evaluate all dimensions of quality because some dimensions require large sample inference. These dimensions of quality are also not observable ex ante. Health outcomes, for instance, reflect both the quality of the long-term care provider and the patient’s severity of illness. To obtain accurate assessments of quality, a patient evaluating long-term care providers would need to compare her outcomes had she used a different provider, while accounting for her health severity. This comparison requires large data assessments because a patient who wishes to compare providers cannot do so as she has no counterfactual information.

In theory, public reporting could fulfill the information gap and spur demand-side pressures to improve quality. However, pay-for-performance circumvents the reliance on consumer response to public reporting, which has been found to be modest in long-term care and sensitive to the format of information presented and the level of market competition (Werner, Konetzka, & Polsky, 2016; Werner, Norton, Konetzka, & Polsky, 2012; Perraillon, Konetzka, He, & Werner, 2017; Jung, Wu, Kim, & Polsky, 2016; Grabowski & Town, 2011). Using pay-for-performance to directly target long-term care, especially on measures of value to patients, provides incentives for long-term care providers to improve on these dimensions. Therefore, pay-for-performance directly targeting long-term care may be an effective strategy to combat low quality on unobservable dimensions of quality.

The third type of information problem in long-term care relates to dimensions of quality that patients can assess, albeit only after first-hand experience (Pesis-Katz et al., 2013). Unlike the quality dimensions that require large sample inference, these dimensions of quality are difficult to observe in advance but not impossible to determine after care begins. Pay-for-performance programs should not prioritize these dimensions of quality if patients can reasonably assess and respond to them. Instead, policy makers should rely on alternative interventions such as public reporting because providers may already compete on relatively observable dimensions (Richardson, 2012). There are several dimensions of quality that patients can ascertain well. On satisfaction and experience, a patient’s own assessment is the best source of information and can be determined without significant delay. For example, patients can quickly determine whether they are satisfied with the communication style of the long-term care provider. Other dimensions, such as the number of nurses, can also be observed by patients. Patients who face high dissatisfaction on observable dimensions of quality or have relatively low provider switching costs may choose to disassociate with the long-term care provider. Patients dissatisfied on observable dimensions could also give bad reviews to future long-term care users through informal networks or formal report cards.

That is not to say that switching providers is cost free and possible for all patients. For instance, changing providers could lead to information gaps and to interrupted, delayed, or conflicting care. Many studies have documented the prevalence of information gaps during transitions of care among vulnerable populations (Olsen, Hellzén, & Enmarker, 2013; Mesteig, Helbostad, Sletvold, Røsstad, & Saltvedt, 2010). Additionally, not all patients can be placed with alternative providers. Medicaid patients may have more difficulty finding a substitute facility than those insured by Medicare (Hirth, Banaszak-Holl, & McCarthy, 2000).

Not all patients can assess a dimension of quality such as satisfaction; it depends on whether the patient has full capacity to make the evaluations. About 31.4% of home health patients and 50.4% of nursing home residents have Alzheimer’s or other dementias (Harris-Kojetin et al., 2016). In the nursing home sector, residents who switch facilities tend to be younger and more cognitively and physically sound, suggesting heterogeneous health-related barriers to switching providers (Hirth et al., 2000).

Where patients already exert strong influence, pay-for-performance programs have less to gain from directly incentivizing these dimensions, and a pay-for-performance program is better off rewarding dimensions that are harder to observe than dimensions that are easier to observe. However, the extent to which patients can assess and respond to observable dimensions of quality is heterogenous by patients and the long-term care sector. Thus, the appropriateness for a pay-for-performance program to target these dimensions of quality depends on the specific context at hand.

In summary, because all design elements flow from the purpose of each program, clarifying the specific sources of inefficiencies can ensure a set of coherent and sensible design elements. Without clarifying the issues that the pay-for-performance program attempts to resolve within each context, determining whether a particular program achieved meaningful and beneficial effects can be challenging. Clarifying the sources of low quality and high cost will help determine when to target long-term care indirectly, when to target long-term care directly, and which dimensions to reward.

Measuring Performance

After the sources of failures and the problems that pay-for-performance seeks to resolve are identified, measurement is the next important step. Measurement is the basis for incentive payments; it determines the rewards for provider behavior change and it helps to shape societal perceptions of what constitutes a well-functioning healthcare system. However, measurement is challenging in pay-for-performance programs, and special care must be taken to balance the trade-offs between measuring what is important for consumers and what can be measured easily. This section raises points to consider in terms of measurement validity and reliability, the unintended consequences of multitasking, and feasibility.

Validity and Reliability

Measures included in pay-for-performance programs should have high validity and reliability. Specifically, to incentivize provider behavior change, variation in measures must result from quality improvement activities taken by the provider. Measures that inadequately control for patient characteristics unrelated to the care provided can lead to perverse incentives such as avoiding unhealthy patients (Eggleston, 2005; Norton, 2018). For example, rewards for health improvement may disincentivize providers from admitting patients in a trajectory of health decline. This is particularly relevant for pay-for-performance in long-term care given that 41.6% of nursing home and 26% of home health patients are aged 85 and older (Harris-Kojetin et al., 2016). Patients in a trajectory of decline may never improve, even with high-quality care. Other patients may differ in their expected outcomes because of their adherence to treatment instruction. In home health, for example, outcomes are a function of the patient’s adherence to treatment protocols after the provider leaves as well as the safety environment of their homes (Ellenbecker, Samia, Cushman, & Alster, 2008). A patient unwilling to take medications on time or unable to rectify fall hazards may have worse measures regardless of the healthcare provider’s efforts. Therefore, risk adjustment, to the extent that risk is observable, mitigates differences in baseline patient characteristics for more valid measurement of the provider’s effect and can decrease incentives for cream skimming and patient dumping (Robinson, 2001).

To be valid, measures should be unbiased so that changes in measures reflect true improvement rather than gaming by providers. In long-term care, response bias is a concern because of the reliance on self-reported survey data by nursing homes and home health agencies (O’Connor & Davitt, 2012). Home Health Value-Based Purchasing Program measures, such as functional improvement in bathing, require a clinician to observe and assess patient abilities over time. The types of staff used to carry out assessments may differ, staff training can vary, and individual clinicians may fluctuate in their assessments over time (O’Connor & Davitt, 2012). Further, providers targeted by pay-for-performance may have an incentive to document higher levels of performance than actually delivered. To maximize rewards and to avoid reputational backlash, long-term care staff may be pressured to underreport quality deficits and inflate improvements (Rahman & Applebaum, 2009).

Finally, minimizing noise is key to ensuring valid and reliable measurement. If the signal of the provider’s true quality is not differentiable from noise, then any subsequent comparisons made on the basis of these measures are also unreliable. At best, incentive payments are distributed for reasons unrelated to true underlying quality. At worst, patients and providers are misled regarding the distribution of quality in the market, potentially leading to misallocation of patients and wasted quality improvement resources by hospitals (Friedson, Horrace, & Marier, 2016). Low patient volume can increase noise and make it difficult to compare the quality between a high-volume provider and a low-volume provider. In 2015, the median home health agency treated 287 Medicare episodes and 146 distinct Medicare patients; the median nursing home treated 123 post-acute Medicare admissions and 96 distinct post-acute Medicare patients. In choosing measures in pay-for-performance programs, it is important to identify measures that apply to a large proportion of patients and are applicable to a majority of providers. In long-term care pay-for-performance, the chosen measures have typically been those that entail few denominator exclusions and tend to be general across conditions.

Multitasking

A more subtle and fundamental challenge with regard to pay-for-performance measurement relates to the issue of multitasking. Multitasking refers to the response from providers to direct efforts away from the unrewarded services to the rewarded services (Holmstrom & Milgrom, 1991). Multitasking comes about because when supplying one task, the marginal cost to supply another task increases (Sherry, 2016). For example, the time spent by a therapist to help a patient improve walking skills means less time for helping a patient develop upper body strength. When pay-for-performance rewards a single dimension, multitasking is predicted to lead to increased output toward the one rewarded measure. This would also be predicted to reduce effort on most other dimensions unless the rewarded measure is jointly produced with an unrewarded measure, in which case there could be positive spillovers (Mullen, Frank, & Rosenthal, 2010; Konetzka, Skira, & Werner, 2018; Sherry, 2016).

The issue of multitasking becomes more complex when a pay-for-performance program rewards multiple dimensions. When faced with rewards for many dimensions, a provider can determine which, among the many, would offer higher real rewards (i.e., net of costs) from improvement. In assessing the trade-offs, a provider may even be better off performing worse on some rewarded dimensions and allocate resources toward improving dimensions with greater marginal benefits. In these cases, countervailing incentives through multitasking can lead to complex interactions that diminish the incentive effects. In other words, one can no longer expect that rewarded dimensions increase and unrewarded dimensions decrease under pay-for-performance (Sherry, 2016).

The implications of multitasking for pay-for-performance is that, in addition to the possible unintended consequences of reallocation of effort away from unmeasured dimensions, there may also be reallocation among the measured dimensions. Therefore, in setting measures for pay-for-performance programs, it would be prudent to consider not only the measure-specific traits (e.g., validity and reliability), but also these broader interaction effects.

Feasibility

It is also important to balance the societal benefits against the societal costs of each measure. While there are many important dimensions to consider for measurement in pay-for-performance programs, not all are measurable in a feasible way. Some forms of data collection can be resource intensive (Golden & Sloan, 2008; Institute of Medicine, 2007). Data not based on existing sources of information, such as administrative claims, can impose cognitive costs for providers as they try to incorporate data collection into existing workflows. Measuring meaningful outcomes among long-term care patients can be particularly challenging as many are insured by more than one payer. A single payer may have difficulty measuring health outcomes that span across payers because multi-payer information may be unavailable or difficult to obtain quickly. Some data may also require new collection processes, such as survey vendors, that also add to the costs of measurement. In home health care, for example, paperwork is often reported as a key source for staff work dissatisfaction (Shamus, Fabrizi, & Hogan, 2018). Because measurement is not cost free, pay-for-performance programs must weigh the costs carefully as high costs may adversely affect efforts devoted to treating patients and high costs may also mean fewer resources for quality improvement efforts.

Assessing Relative Performance

Because every pay-for-performance program identifies winners and losers, how relative performance is assessed determines which providers face incentives to improve and, in turn, the expected effect of the program. This section describes trade-offs in three approaches used to assess provider performance in long-term care: target-based incentives, improvement-based incentives, and tournament-style peer-group comparisons.

The target-based incentives using a predetermined threshold awards provider performance based on a threshold that is known in advance. This is an approach that has been consistently used over time. Predetermined thresholds are more appropriate for quality measures that have clear and uncontroversial thresholds and can be used to differentiate providers reliably across the threshold. If the thresholds are established where there is low differentiation among providers, then relative performance represents factors other than true quality (Friedson et al., 2016). Critics of the approach argue that predetermined thresholds alone lead to diminished incentives for low- and high-performing providers (Mullen et al., 2010). Those who are above the threshold automatically qualify for rewards while those far below the threshold face little extrinsic incentive to improve. Only those at the margin of the threshold face incentives for change. If the pay-for-performance program aims to provide incentives across the distribution, then predetermined thresholds alone may not achieve this goal.

Another method of gauging performance is scoring relative to a provider’s own baseline. This setup is designed to provide incentives for all pay-for-performance participants to improve (Cromwell, Trisolini, Pope, Mitchell, & Greenwald, 2011). However, if there exists diminishing marginal returns to quality improvement inputs, then the best providers may achieve smaller performance gains than low-performing providers. Therefore, baseline improvement thresholds implicitly prioritize lower- performing providers as opposed to the entire distribution of providers (Cromwell et al., 2011). The value to society for a given unit of improvement for high-performing providers may also be less cost effective than for low-performing providers if high-performing providers require more resources for quality improvement than low-performing providers. Additionally, this structure may inadvertently create perverse incentives for high-performing providers to game measurement with cyclical declines in performance so that they could later improve and capture improvement-based incentives.

Finally, rewards in long-term care have also been based on tournament-style peer performance distributions. Rather than specifying a threshold, performance is based on concurrent relative ranking (Rosenthal & Dudley, 2007; Prendergast, 1999; Shleifer, 1985). In contrast to predetermined thresholds, where every participant could theoretically be a winner, the tournament-style incentive requires some winners and some losers. The benefit of using a tournament approach is that all providers face uncertainty regarding their relative performance and thus all providers have some incentive to improve in order to not fall behind. Peers have included nonparticipating providers in pilot pay-for-performance programs and concurrent participating peers in mandatory programs (Cromwell et al., 2011). In the Nursing Home Value-Based Purchasing demonstration, for example, relative performance was determined by a randomly assigned comparison group for New York and a propensity matched comparison group for nursing homes in Arizona and Wisconsin.

Like the other performance ranking methods, the tournament-style performance assessment also has drawbacks. For instance, determining appropriate peer groups entails trade-offs (Joynt et al., 2017). Because risk adjustment does not fully account for patient differences, it may not be fair to compare long-term care providers who serve different patient types (e.g., long-term chronically ill vs. short-term rehabilitative). Grouping all providers into one peer group can widen disparities through a reverse-Robin Hood effect (Casalino et al., 2007). For hospitals, there have been concerns that comparing hospitals that serve larger proportions of vulnerable populations to other hospitals could unduly punish the hospitals with more vulnerable and less profitable patients (Gilman et al., 2014; Gilman et al., 2015; Woolhandler & Himmelstein, 2015). The argument is that providers who serve lower-income patients face lower revenue and also have less adherent patients, leading to lower performance on pay-for-performance measures (Casalino et al., 2007).

Another potential drawback of the tournament-style approach is that differentiating providers when the outcome has a continuous distribution can be challenging. If the providers cannot be reliably differentiated across the entire distribution, then relative performance rankings provide misleading information because these rankings likely reflect differences in factors unrelated to quality (Friedson et al., 2016).

The pay-for-performance programs since the 2000s use a blended version of target-, improvement-, and tournament-style comparisons. For example, nursing homes in the Skilled Nursing Facility Value-Based Purchasing Program are awarded both improvement and achievement scores (Centers for Medicare & Medicaid Services [CMS], 2016) on an ordinal scale, based on magnitude of improvement, as compared to baselines, and magnitude of achievement, as compared to concurrent peers. It also incorporates comparisons against a threshold, calculated from the performance distribution in the baseline period, by awarding zero achievement points if a facility performs below the 25th percentile of all facilities’ performance. Proponents of the approach believe that it fosters improvement across all providers (Centers for Medicare & Medicaid Services [CMS], 2011). For low performers, the use of baselines incentivizes improvement relative to self. A minimum threshold ensures a pre-specified level of quality. Further, equity in terms of incentives is also more balanced because high performers are also rewarded (Institute of Medicine, 2007). However, this blended approach may still lead to a substantial number of providers without incentives to improve. In the blended version of target-, improvement-, and tournament-style comparisons approach used in Hospital Value-Based Purchasing, only two thirds of hospitals faced incentives to improve (Norton et al., 2018). Thus, it remains an open question as to how to meaningfully assess relative performance while also incentivizing a large proportion of the market.

Incentivizing With Financial Rewards

The effectiveness of financial rewards in improving quality and spending depends on three things. First, the magnitude of the potential rewards matter. Second, reward uncertainty can diminish the effects of a large reward. Third, pay-for-performance incentives can be indirectly affected by interactions with other programs.

Magnitude of Rewards

The magnitude of the maximum potential rewards in a pay-for-performance program overall is a key component to elicit provider response. The size of the rewards relative to the cost of effort, or the net financial returns, must be large enough for the effort to be worthwhile (Christianson, Leatherman, & Sutherland, 2008; Conrad, 2015). In Hospital Value-Based Purchasing, for example, hospitals respond to the rewards, but only for measures with the highest marginal future reimbursement (Norton et al., 2018). In other words, the magnitude of rewards has a direct effect on providers’ willingness to respond.

The magnitudes of incentives in pay-for-performance programs have increased and are expected to increase further. The Home Health Value-Based Purchasing’s reward (or penalty) to home health providers will range from 3% in 2018 to 8% in 2022 of each provider’s total Medicare payment (Centers for Medicare & Medicaid Services [CMS], 2015). For the Skilled Nursing Facility Value-Based Purchasing program, nursing homes can expect payment adjustments of around 1.2% of their Medicare reimbursement (Centers for Medicare & Medicaid Services [CMS], 2017). A similar ramp-up is found for hospital-based programs for Medicare. The Hospital Value-Based Purchasing program, which includes several post-discharge measures that indirectly affect long-term care, began with 1% in the first year and has since increased up to 2% in rewards or penalties. The Hospital Readmissions Reduction Program, which rewards hospitals on condition-specific 30-day readmissions, started with a maximum penalty of 1% in 2013 and is currently up to 3%.

If reward sizes continue to grow, it is possible that greater responses will be observed in the future, but it is also unclear how large the rewards should be. Some have raised concerns that extrinsic rewards may come into conflict with other motivations, such as providers’ intrinsic motivation for improvement. For instance, a reward for quality may lead the long-term care provider to improve quality in the short run, but simultaneously reduce incentives to invest in quality in the long run. If monetary incentives produce a psychological effect on providers, then the concern is that financial incentives may crowd out other underlying reasons for providing high-quality care (Gneezy, Meier, & Rey-Biel, 2011). In health care, the provider may not only have pecuniary motivations, but also be motivated by other factors such as prestige and social or ethical concerns (Newhouse, 1970). Approximately 30.6% of nursing homes and 20% of home health agencies were not-for-profit institutions in 2014, suggesting that there may be a portion of firms that may have other motivations beyond a financial one (Harris-Kojetin et al., 2016). Thus, as the magnitude of rewards increases over time, there may be more potential to observe trade-offs between intrinsic and extrinsic motivation among long-term care providers.

Within programs, the magnitude of rewards could differ across dimensions by design, such as different weights across measured dimensions. These weights signal policy priorities to providers and consumers and directly cap the financial rewards a provider can expect to receive given improvement on a particular dimension (Konetzka et al., 2018). In the Home Health Value-Based Purchasing program, for example, there are two types of measures. The first type is scored for performance. Each measure can score at most 5.63% of the total performance score. The second type is for reporting rather than performance where, by reporting the measure to Medicare, the agency can earn up to 3.33% of the total performance score per reported measure. By attaching different weights to measures, the rewards for dimensions such as improved bathing ability are worth as much as keeping patients in the community at discharge, but more than reporting data on aspects such as advance care planning, conditional on all else being equal.

Another way that rewards are designed to differ by measure is in the populations that the measured dimensions encompass. Mechanically, a narrow measure denominator places more weight on quality for a smaller subset of patients than a broader measure denominator. Again, using the Home Health Value-Based Purchasing program as an example, although measures on improved bathing ability and keeping patients in the community at discharge are weighted the same, the population of patients who are evaluated for improved bathing ability is more exclusive. Bathing ability is not measured for patients who could bathe independently at the start of care. Thus, pay-for-performance on the basis of a measure that is only clinically applicable to some patients is prioritized by design. Therefore, by design, rewards across measures may differ within programs.

Certainty of Rewards

The effects of large rewards can be diminished if there is high uncertainty about whether such a reward will be received (Grabowski et al., 2017). Some pay-for-performance programs are based on shared savings where rewards are shared as a percentage of savings. In these instances, the uncertainty associated with having any savings to distribute may diminish the effects of otherwise large rewards (Hittle, Nuccio, & Richard, 2012; L&M Policy Research and Harvard Medical School, 2013). Particularly in older long-term care pay-for-performance programs, shared savings pools were generated across all pay-for-performance participants. This setup meant that only relatively homogeneous providers with strong desires to improve could reap the benefits from improvement. Past experience shows that the use of shared savings has led to strong dissatisfaction among participants and likely dampened responses (Hittle et al., 2012).

Similarly, many pay-for-performance programs compare performance across providers, such as the case in tournament-style peer performance distributions. Because the reward depends on the performance of other providers, it is difficult to predict whether improvements would yield rewards.

Indirect Rewards

The indirect effects from other programs may strengthen the incentives from pay-for-performance rewards. When the measures rewarded by one pay-for-performance program overlap with other programs or are correlated with other sources of incentives, there may be indirect effects from other programs.

One of these alternative sets of incentives comes from patients. Since 1998, Medicare has provided various report cards to inform patients about quality of nursing homes and home health agencies (Institute of Medicine, 2007). These report cards are intended to increase demand for quality and, in turn, provide market incentives for providers to improve on quality (Werner et al., 2012). The latest format of report cards for nursing homes, for instance, has been found to increase market share among high-performing nursing homes (Perraillon, Konetzka, He, & Werner, 2017; Werner et al., 2016). These report cards are based, in part, on measures also used in pay-for-performance programs. For example, hospital readmissions during a nursing home stay are measured in the Skilled Nursing Facility Value-Based Purchasing program and reported on Nursing Home Compare; in total, 80% of measures used for the Home Health Value-Based Purchasing program are also reported on Home Health Compare. Some state pay-for-performance programs, such as the Texas Quality Incentive Payment Program for Nursing Homes, directly build upon report card measures by either exclusively or partially tying payments to performance (Texas Health and Human Services, n.d.). Therefore, the overlap of pay-for-performance measures with report cards could serve to increase incentives to improve as well as to reinforce the long-term effects of pay-for-performance incentives.

Indirect Effects of Pay-for-Performance on Long-Term Care

Pay-for-performance can affect long-term care indirectly because healthcare sectors are interconnected. For the long-term care population, patients are in poorer health and can have multiple complex health needs requiring care from a variety of provider types. For many, the transition between independence to high dependence can occur rapidly and without preparation following hospitalizations (Kane, 2011). The rates of discharges to nursing homes and home health care following hospital stays are high, at roughly 20% of inpatient stays across all payers and 38% across Medicare (Tian, 2006). Therefore, there are many avenues for hospitals to significantly affect the experience and outcomes facing patients transitioning into and throughout long-term care. The large and frequent overlaps in patient populations indicate that the effects of pay-for-performance programs for hospitals can spill over to long-term care.

The indirect effects of hospital pay-for-performance on care coordination with long-term care providers are intended. The Hospital Value-Based Purchasing measure on total 30-day episode payments evaluates Medicare spending for all services provided during the episode period, including the initial inpatient admission as well any subsequent services such as readmissions or admissions to a nursing home or home health following hospital discharge (Centers for Medicare & Medicaid Services [CMS], 2011). Similarly, the Hospital Readmissions Reduction Program evaluates each hospital on hospital readmissions following the initial inpatient stay to any hospital within a 30-day period. Again, the intent is to incentivize hospitals to consider the quality and spending of post-discharge settings when discharging patients. Incentives for measures such as readmissions, for example, may lead hospitals to make more careful selections of post-discharge long-term care providers.

Facing these incentives, hospitals may steer more patients toward preferred nursing homes and home health providers (Rahman, Grabowski, Mor, & Norton, 2016). To do so, hospitals may use more narrow network choices through formal efforts such as integration or informal efforts such as forming preferential patient sharing arrangements (Mor, Rahman, & McHugh, 2016). The indirect effects of hospital pay-for-performance could mean that nursing homes and home health agencies receive a different patient case mix. For example, to decrease episode spending, a hospital may choose to discharge patients away from higher- to lower-cost post-discharge care. Therefore, a nursing home may receive patients who otherwise would have gone to an inpatient rehabilitation facility. Hospitals could also mitigate readmissions risk by increasing the number of patients who use lower-cost care, such as home health to bridge patient needs during transitions (Kripalani Theobald, Anctil, & Vasilevskis, 2014). Therefore, a home health agency may face more patients who traditionally would not have received home health care.

In total, the effects of pay-for-performance on long-term care include both direct and indirect effects. To the extent that policy makers explicitly reward measures that encompass multiple providers, the indirect effects are broadened. As of 2019, most research has focused on the direct effects of pay-for-performance. Expanding research to better understand the indirect effects of pay-for-performance in long-term care will become more important as pay-for-performance proliferates.

Empirical Evidence of Pay-for-Performance

Empirical Findings

State Medicaid programs have traditionally led the pay-for-performance movement in long-term care. Since 1979, individual states have implemented their own versions of pay-for-performance programs for nursing homes (Briesacher et al., 2009). A survey of state Medicaid agencies found nine out of 50 states had implemented nursing home pay-for-performance programs between 2000 and 2007, covering one fifth of all nursing homes (Werner, Konetzka, & Liang, 2010).

There have been several other programs implemented by states and their managed care plans since 2007 (Libersky, Stone, Smith, Verdier, & Lipson, 2017; Taylor, Dyer, & Bailit, 2017; Lipscomb University, 2014). These pay-for-performance programs tend to target nursing homes and vary in their incentive structure, quality dimensions targeted, and approach to assessing performance. For instance, Colorado’s Nursing Facility Pay for Performance Program began in 2009. The program is voluntary and rewards nursing homes for performance on measures related to quality of life and quality of care. To receive an incentive payment, the nursing home must first meet a minimum quality threshold (Colorado Department of Health Care Policy & Financing, 2018; Public Consulting Group, 2017). In another example, Kansas’s voluntary PEAK 2.0 program focuses on quality of life and cultural change in nursing homes. The program qualitatively determines whether nursing homes meet various structural and process of care requirements for incentive payments, which range from $0.50 to $3 dollars per day (Kansas State University Center on Aging, 2018; Poey et al., 2017). Maryland’s program differs from those in Colorado and Kansas in that it is mandatory and rewards facilities for both improvement and achievement. Focusing on Medicaid patients, the program employs a variety of measures such as consumer satisfaction, staffing, and quality measures. Another output of the program is a publicly available ranking of facilities in the state based on its pay-for-performance measures (Maryland Department of Health, n.d.; Lipscomb University, 2014). See Libersky et al. (2017) and Taylor et al. (2017) for an overview of other state-based programs.

Beginning in 2008, Medicare also began to experiment with pay-for-performance programs in long-term care (see Table 1). Medicare has funded four large-scale, multistate, pay-for-performance initiatives in both nursing homes and home health agencies, three of which were demonstration programs and one of which was a permanent feature of Medicare.

Outside the United States, the majority of pay-for-performance programs affect long-term care only indirectly. An exception are two programs from Japan, formed in 2006 and in 2012, that focused on long-term care explicitly (Norton, 2018). One program rewarded utilization-based measures, such as prevention services, and another rewarded adult day-care centers for outcomes of care. In addition, the Japanese government plans to expand the long-term care incentive program to the entire country (Iizuka, Noguchi, & Sugawara, 2017).

The majority of programs in other countries focus on primary care or hospitals. The United Kingdom has the world’s largest primary care pay-for-performance since 2004, via the Quality and Outcomes Framework. The program targets half of the healthcare workforce in the National Health Service. It links a quarter of family practitioners’ income to quality measures on chronic disease management, healthcare organization, and patient experience (Roland, 2004; Ryan Krinsky, Kontopantelis, & Doran, 2016; Doran et al., 2017). Brazil’s Social Organizations in Health Program represents an example of a hospital-based program. The Brazilian program assesses hospital performance on measures including health outcomes, patient satisfaction, reporting of information, and utilization. It also evaluates hospitals for volume of services, such as admissions and lengths of stay (Chi & Hewlett, 2014). While studies suggest that the program led to more efficient and higher-quality hospitals, they did not examine how the programs affected the long-term care sector (World Bank, 2006; La Forgia & Couttolenc, 2008). For other examples of pay-for-performance outside of the United States, see Cashin, Chi, Smith, Borowitz, and Thomson (2014) and Norton (2018).

Despite a long history of pay-for-performance in long-term care, evidence on their effectiveness remains sparse. The small body of literature focuses on U.S. health care. Even so, most of the state programs were not evaluated, leaving it uncertain whether they were effective, and many of the Medicare programs are too new to generate any published research findings.

Table 1. Overview of Recent Pay-for-Performance Initiatives in Long-Term Care

Overview

Measures

Financial Incentives

Panel A. Nursing Homes

Nursing Facility Pay for Performance (2012–ongoing, state)

Mandatory nursing home participation in Maryland; patient populations: long-stay patients and facility level

4 staffing measures

4 safety measures

2 process of care measures

1 patient survey

Incentive determination: baseline, threshold, and participating peer comparison; incentive size: $2.57–$5.14 per diem for achievement, $0.46–$0.92 per diem for improvement

PEAK 2.0 (2012–ongoing, state)

Voluntary nursing home participants in Kansas; patient populations: facility level

10 process of care measures

5 staffing measures

10 quality of life process measures

Incentive determination: threshold; incentive size: $0.50–$3 per diem

Colorado Nursing Facilities Pay for Performance (July 2009–ongoing, state)

Voluntary nursing home participants in Colorado; patient populations: long-stay patients and facility level

1 safety measure

13 quality of life process measures

5 staffing measures

1 hospitalization measure

8 health status measures

1 patient survey

Incentive determination: threshold and improvement based; incentive size: $1–$4 per diem

Nursing Home Value-Based Purchasing (July 2009–June 2012, Medicare)

Voluntary nursing home participants in Arizona (N = 38), New York (N = 151, 50% randomized into treatment), and Wisconsin (N = 61); patient population: all Medicare-eligible individuals

4 staffing measures

0 or 2 safety measures*

1 citation measure

1 hospitalization measure

3 or 6 health status measures*

Incentive determination: baseline and non-participating peer comparison; incentive pool: bonus contingent on Medicare savings from quality improvement; incentive size: variable by state and year

Skilled Nursing Facility Value-Based Purchasing (January 2017 onward, Medicare)

Mandatory for Medicare-certified skilled nursing facilities; patient population: all Medicare-eligible individuals

1 hospitalization measure

Incentive determination: baseline, threshold, and participating peer comparison; incentive pool: withheld from usual payment, redistributed from low to high performers as penalties and bonuses; incentive size: payments up to ±1.2%

Panel B. Home Health Agencies

Home Health Pay-for-Performance (January 2008–December 2009, Medicare)

Voluntary home health agencies in Illinois (N = 127), Connecticut (N = 49), Massachusetts (N = 50), Alabama (N = 54), Georgia (N = 54), Tennessee (N = 88), California (N = 134), with 50% of agencies in each state randomized into treatment; patient population:

all fee-for-service Medicare episodes

1 hospitalization measure

1 emergency care measure

5 health status measures

Incentive determination: baseline and non-participating peer comparison;

incentive pool: bonus contingent on Medicare savings from quality improvement

Home Health Value-Based Purchasing (January 2016–December 2022, Medicare)

Mandatory for Medicare-certified home health agencies in Massachusetts, Maryland, North Carolina, Florida, Washington, Arizona, Iowa, Nebraska, and Tennessee

1 hospitalization measure

1 emergency care measure

1 discharge setting measure

6 health status measures

3 care process measures

5 patient experience measures

3 pay-for-reporting measures

Incentive determination: baseline, threshold and participating peer comparison; incentive pool: withheld from usual payment, redistributed from low to high performers as penalties and bonuses; incentive size: ±3–8% of total Medicare reimbursement

Note. * Applicable measures vary depending on patient populations served.

Among the handful of studies available, most focus on determining whether long-term care providers responded to the incentives in the program. As is the case with the empirical literature on pay-for-performance in other sectors, researchers find mixed and modest effects on targeted measures.

Medicare’s Nursing Home Value-Based Purchasing and Home Health Pay-for-Performance were two similarly structured demonstration projects that did not find meaningful behavioral responses (Grabowski et al., 2017; Hittle et al., 2012). The demonstrations used a variety of measures and rewarded providers for both improvement and achievement. To maintain budget neutrality, they both incorporated a stipulation requiring Medicare cost savings among treatment providers before any pay-for-performance bonuses would be paid (White et al., 2009; L&M Policy Research and Harvard Medical School, 2013). Qualitative evidence from surveys and focus groups for the Medicare Nursing Home Value-Based Purchasing demonstration and the Home Health Pay-for-Performance also did not suggest strong and consistent behavioral changes among providers that were indicative of an improvement response. Medicaid pay-for-performance programs had similar results. Werner, Konetzka, and Polsky (2013) looked for evidence of improvement in quality among incentivized measures in Medicaid nursing home pay-for-performance programs across eight states implemented between 2001 and 2009. Across all three studies, results did not show improved outcomes following pay-for-performance (Grabowski et al., 2017; Hittle et al., 2012; Werner et al., 2013).

Similar lackluster findings were found for a Japanese pay-for-performance program on adult day-care services in Shiga prefecture that spanned 2012–2014 (Iizuka et al., 2017). The program rewarded long-term care providers a bonus for day-care centers that achieved high performance or improvements in patient physical and cognitive function. Using a difference-in-differences approach, the authors found weak evidence that the program affected patient outcomes. Further, the authors observed indications of patient selection, likely attributed to a lack of risk adjustment in the program’s scoring methods. The study highlighted the important role of the pay-for-performance design.

Part of the reasons for a lack of demonstrated quality improvement response may be due to a high level of uncertainty present in some of the programs, another consequence of how programs are designed. As is the case across pay-for-performance systems generally, incentives are muted when the relationship between the individual’s action and the reward becomes more ambiguous (Conrad, 2015). In Medicare’s Nursing Home Value-Based Purchasing and the Home Health Pay-for-Performance demonstrations, high-performing agencies were rewarded only when the quality improvement efforts from the collective treatment group achieved savings to Medicare (White et al., 2009; L&M Policy Research and Harvard Medical School, 2013). In other words, the improvement efforts by providers would not necessarily reap rewards. In fact, Medicare savings were not achieved for 6 out of 9 state-years for nursing homes and 4 out of 14 state-years for home health agencies. While the demonstrations led to positive reward pools for some providers, dependence on other providers in each peer group increased the uncertainty associated with rewards and likely contributed to small responses.

Another challenge for predicting behavior change for long-term care providers is the lack of observable information (to the analyst) on marginal costs for improvements. While the marginal benefits may be clear from the incentive structures, the marginal costs may vary by provider type. For example, Werner, Skira, and Konetzka (2016) examined three nursing home pay-for-performance programs in Colorado, Georgia, and Oklahoma from 2006 to 2009. They examined the effects of varying incentive levels, proxied by the distance between baseline performance to a performance threshold used to qualify for rewards. Theory predicts that conditional on marginal cost to improvement, rewarding providers for meeting performance thresholds means that providers with expected performance far below the threshold have low marginal incentives to improve while the marginal incentives are maximized for providers directly below the threshold (Mullen et al., 2010). Contrary to their hypothesis, the authors found that improvements were primarily driven by the lowest baseline performing nursing homes, which had low marginal incentives to improve. Their study suggests that the marginal costs for improvement may have played a role and diminished the net benefits faced by providers.

Finally, multitasking remains an important aspect of pay-for-performance programs in long-term care that could lead to unexpected responses. One source of potentially competing incentives stems from outside programs that coincide with other pay-for-performance programs. By 2009, consumer-facing report cards to address information asymmetry for long-term care providers had already been implemented for several years. Therefore, by the time that Medicare’s Nursing Home Value-Based Purchasing demonstrations began, it was somewhat unsurprising that nursing homes did not change their quality improvement behaviors in response to the demonstration (Grabowski et al., 2017).

Another source of multitasking stems from multiple measured dimensions. Konetzka et al. (2018) provide the only article that directly assesses behavior response in the presence of multitasking. They found that dimensions of quality associated with higher returns tend to be associated with greater improvements, sometimes at the cost of dimensions with lower marginal returns. They measured marginal returns in terms of the weight that a measure received in performance scoring and whether a measure was linked to pay-for-performance eligibility criteria. The authors’ results were consistent with multitasking. As outlined in recent theoretical literature, multitasking can lead to ambiguous effects on measure improvement when multiple dimensions are rewarded because quality inputs depend on the marginal revenues (or cost functions) of the unrewarded dimensions and competing rewarded dimensions (Sherry, 2016). Given that the authors were unable to observe the complete set of rewarded measures in their analysis, it is unsurprising that not all rewarded measures, including those with higher weights, were associated with improvements.

Programs for Future Research

Although many pay-for-performance programs are too new to have any published research findings on their effectiveness, pay-for-performance programs will play an increasingly important role in long-term care in the United States. The Skilled Nursing Value-Based Purchasing program is a permanent Medicare program and mandatory for all nursing homes (Centers for Medicare & Medicaid Services [CMS], 2016). Medicare will also expand the Home Health Value-Based Purchasing program to be permanent for all home health agencies by 2022 (Centers for Medicare & Medicaid Services [CMS], 2015). Several state Medicaid programs have also adopted or expanded pay-for-performance programs (Libersky et al., 2017). A handful of these state programs are mandatory for targeted providers, such as Indiana’s VBP Initiative, Minnesota’s Integrated Care System Partnership and Value-Based Reimbursement, Ohio’s Nursing Home Quality Incentive System, and Maryland’s Pay for Performance program. Some have expanded the types of patients affected to include those who are community dwelling, such as Tennessee’s Quality Improvement in Long Term Services and Supports (TN Division of TennCare, n.d.), and others have increased the number of patients directly affected by pay-for-performance, such as the Arizona Health Care Cost Containment System (Taylor et al., 2017).

There are several aspects about new Medicare programs in particular that are conducive for future research. First, in the new Home Health Value-Based Purchasing program, participation is mandatory for all Medicare certified home health agencies within nine states; states were selected by CMS to be representative of the various geographic areas of the United States, which may mitigate any selection bias and can allow for better estimation of treatment effects (Centers for Medicare & Medicaid Services [CMS], 2015). Second, the incentive sizes more than double over the years of the Home Health Value-Based Purchasing program, potentially introducing meaningful variation in incentive sizes. Third, unlike other programs, the Skilled Nursing Value-Based Purchasing program rewards only one measure, which may decrease multitasking among rewarded dimensions that is common in most pay-for-performance programs. Fourth, these two programs both incorporate hospital readmissions that overlap with other settings of care, including hospitals, and could lead to other indirect effects worthy of exploration. For instance, it is unclear whether the interplay of incentives across sectors is beneficial to patients or whether they may lead to unintended consequences, such as provider consolidation.

Newer literature on pay-for-performance in long-term care has only begun to explore the effects of specific design features on incentives and behavior, and existing literature feature little discussion of indirect effects of pay-for-performance on long-term care. Programs such as the Home Health Value-Based Purchasing and Skilled Nursing Value-Based Purchasing offer opportunities to explore these strands of research further.

Future Directions for Performance-Based Payment

Pay-for-performance is rapidly expanding across health care. The implementation of national programs with substantial financial payments at stake indicates the seriousness of policy makers in tackling quality and spending issues in health care. As with any incentive program geared toward behavior change, finding the optimal structure is difficult. Identifying inefficiencies, finding the right measures, understanding competing incentives, and calibrating the right incentive amounts are not trivial undertakings. While modest and inconsistent effect estimates have led many to conclude that pay-for-performance does not work or is not worth the risks (Berwick, 1995), this article offers a more optimistic view. A review of the previous findings points to a need for research to move beyond impact assessments and into understanding the mechanisms of behavior change in pay-for-performance.

For research, much value would be added by gaining a better understanding of how various quality dimensions are produced. This in turn can help illuminate commonalities in production of quality, for example, which can guide more impactful incentive targeting. Additionally, pay-for-performance research can focus on identifying channels for incentive failures. For example, which types of uncertainty are most problematic for providers? What types of tasks are most prone for multitasking? Where are the market failures most suited for pay-for-performance? How should unique traits of each sector affect the effectiveness of pay-for-performance?

This article also offers two areas for policy consideration. Specifically, policy makers, in designing pay-for-performance programs, should be more explicit about the inefficiencies that they seek to address. Clear problem identification can help streamline the policy response and present a framework to understand the results of the policy. Additionally, policy makers should design pilots to test various arrangements of pay-for-performance to isolate the effects of individual design traits. While these pilots may be costly in the beginning, they provide important sources of information that are likely to pay off in the future.

As a whole, both research and policy makers should increase attention to the indirect effects of pay-for-performance on long-term care. In the coming decades, demand for long-term care will surge (Johnson, Toohey, & Wiener, 2007). The way to think about the effects of these programs will need to be broadened beyond the direct effects if society is to understand the implications for pay-for-performance as an incentive tool.

Further Reading

Berwick, D. M. (1995). The toxicity of pay for performance. Quality Management in Health Care, 4(1), 27–33.Find this resource:

Cashin, C., Chi, Y.-L., Smith, P., Borowitz, M., & Thomson, S. (Eds.). (2014). Paying for performance in health care: Implications for health system performance and accountability. New York, NY: Open University Press.Find this resource:

Conrad, D. A. (2015). The theory of value-based payment incentives and their application to health care. Health Services Research, 50(December), 2057–2089.Find this resource:

Eggleston, K. (2005). Multitasking and mixed systems for provider payment. Journal of Health Economics, 24(1), 211–223.Find this resource:

Institute of Medicine. (2007). Rewarding provider performance. Washington, DC: National Academies Press.Find this resource:

Konetzka, R. T., Skira, M. M., & Werner, R. M. (2018). Incentive design and quality improvements: Evidence from state Medicaid nursing home pay-for-performance programs. American Journal of Health Economics, 4(1), 105–130.Find this resource:

Libersky, J., Stone, J., Smith, L., Verdier, J., & Lipson, D. (2017). Value-based payment in nursing facilities: Options and lessons for states and managed care plans. Hamilton, NJ: Integrated Care Resource Center.Find this resource:

Norton, E. C. (2018). Long-term care and pay-for-performance programs. Review of Development Economics, 22(3), 1005–1021.Find this resource:

Sherry, T. B. (2016). A note on the comparative statics of pay-for-performance in health care. Health Economics, 25(5), 637–644.Find this resource:

Werner, R. M., Konetzka, R. T., & Liang, K. (2010). State adoption of nursing home pay-for-performance. Medical Care Research and Review, 67(3), 364–377.Find this resource:

References

Arrow, K. J. (1963). Uncertainty and the welfare economics of medical care. American Economic Review, 53(5), 941–973.Find this resource:

Berwick, D. M. (1995). The toxicity of pay for performance. Quality Management in Health Care, 4(1), 27–33.Find this resource:

Briesacher, B. A., Field, T. S., Baril, J., & Gurwitz, J. H. (2009). Pay-for-performance in nursing homes. Health Care Financing Review, 30(3), 1–13.Find this resource:

Casalino, L. P., Elster, A., Eisenberg, A., Lewis, E., Montgomery, J., & Ramos, D. (2007). Will pay-for-performance and quality reporting affect health care disparities? Health Affairs, 26(3), w405–414.Find this resource:

Cashin, C., Chi, Y.-L., Smith, P., Borowitz, M., & Thomson, S. (Eds.). (2014). Paying for performance in health care: Implications for health system performance and accountability. New York, NY: Open University Press.Find this resource:

Cebul, R. D., Rebitzer, J. B., Taylor, L. J., & Votruba, M. E. (2008). Organizational fragmentation and care quality in the U.S. healthcare system. Journal of Economic Perspectives, 22(4), 93–113..Find this resource:

Centers for Medicare & Medicaid Services (CMS). (2011). Medicare program; hospital inpatient value-based purchasing program. Final rule. Federal Register, 76(88), 26490–26547.Find this resource:

Centers for Medicare & Medicaid Services (CMS). (2015). Medicare and Medicaid programs; CY 2016 home health prospective payment system rate update; home health value-based purchasing model; and home health quality reporting requirements. Final rule. Federal Register, 80(214), 68623–68719.Find this resource:

Centers for Medicare & Medicaid Services (CMS). (2016). Medicare program; Prospective payment system and consolidated billing for skilled nursing facilities for FY 2017, SNF value-based purchasing program, SNF quality reporting program, and SNF payment models research. Final rule. Federal Register, 81(151), 51970–52053.Find this resource:

Centers for Medicare & Medicaid Services (CMS). (2017). Medicare program; Prospective payment system and consolidated billing for skilled nursing facilities for FY 2018, SNF value-based purchasing program, SNF quality reporting program, survey team composition, and correction of the performance period for the NHSN HCP influenza vaccination immunization reporting measure in the ESRD QIP for PY 2020. Federal Register, 82(149), 36530–36634.Find this resource:

Chandra, A., Finkelstein, A., Sacarny, A., & Syverson, C. (2016). Health care exceptionalism? Performance and allocation in the US health care sector. American Economic Review, 106(8), 2110–2144.Find this resource:

Chi, Y.-L., & Hewlett, E. (2014). Brazil: São Paulo: Social organization in health. In C. Cashin, Y.-L. Chi, P. Smith, M. Borowitz, & S. Thomson (Eds.), Paying for performance in health care: Implications for health system performance and accountability. New York, NY: Open University Press.Find this resource:

Christianson, J. B., Leatherman, S., & Sutherland, K. (2008). Lessons from evaluations of purchaser pay-for-performance programs. Medical Care Research and Review, 65(6 Suppl.), 5S–35S.Find this resource:

Coleman, E. A., Mahoney, E., & Parry, C. (2005). Assessing the quality of preparation for posthospital care from the patient’s perspective: The care transitions measure. Medical Care, 43(3), 246–255.Find this resource:

Coleman, E. A., Parry, C., Chalmers, C., & Min, S.-J. (2006). The care transitions intervention: Results of a randomized controlled trial. Archives of Internal Medicine, 166(17), 1822–1828.Find this resource:

Colorado Department of Health Care Policy & Financing. (2018). 2019 nursing facility pay for performance application. Denver, CO: Colorado Department of Health Care Policy & Financing.Find this resource:

Conrad, D. A. (2015). The theory of value-based payment incentives and their application to health care. Health Services Research, 50(December), 2057–2089.Find this resource:

Cromwell, J., Trisolini, M., Pope, G., Mitchell, J., & Greenwald, L. (Eds.). (2011). Pay for performance in health care: Methods and approaches. Research Triangle Park, NC: RTI Press.Find this resource:

Doran, T., Maurer, K. A., & Ryan, A. M. (2017). Impact of provider incentives on quality and value of health care. Annual Review of Public Health, 38(1), 449–465.Find this resource:

Eggleston, K. (2005). Multitasking and mixed systems for provider payment. Journal of Health Economics, 24(1), 211–223.Find this resource:

Ellenbecker, C. H., Samia, L., Cushman, M. J., & Alster, K. (2008). Patient safety and quality in home health care. In R. G. Hughes (Ed.), Patient safety and quality: An evidence-based handbook for nurses (Chap. 13). Rockville, MD: Agency for Healthcare Research and Quality.Find this resource:

Ellis, R. P. (1998). Creaming, skimping and dumping: Provider competition on the intensive and extensive margins. Journal of Health Economics, 17(5), 537–555.Find this resource:

Forgia, G. M. L., & Couttolenc, B. F. (2008). Hospital performance in Brazil: The search for excellence. Washington, DC: World Bank.Find this resource:

Friedson, A. I., Horrace, W. C., & Marier, A. F. (2016). So many hospitals, so little information: How hospital value-based purchasing is a game of chance. Center for Policy Research, The Maxwell School Working Paper Series, Paper No. 192. Syracuse, NY: Syracuse University.Find this resource:

Gilman, M., Adams, E. K., Hockenberry, J. M., Wilson, I. B., Milstein, A. S., & Becker, E. R. (2014). California safety-net hospitals likely to be penalized by ACA value, readmission, and meaningful-use programs. Health Affairs, 33(8), 1314–1322.Find this resource:

Gilman, M., Hockenberry, J. M., Adams, E. K., Milstein, A. S., Wilson, I. B., & Becker, E. R. (2015). The financial effect of value-based purchasing and the hospital readmissions reduction program on safety-net hospitals in 2014. Annals of Internal Medicine, 163(6), 427.Find this resource:

Gneezy, U., Meier, S., & Rey-Biel, P. (2011). When and why incentives (don’t) work to modify behavior. Journal of Economic Perspectives, 25(4), 191–210.Find this resource:

Golden, B. R., & Sloan, F. A. (2008). Physician pay for performance: Alternative perspectives. In F. A. Sloan & H. Kasper (Eds.)., Incentives and choice in health care (pp. 289–300). Cambridge, MA: MIT Press.Find this resource:

Grabowski, D. C., Stevenson, D. G., Caudry, D. J., O’Malley, A., Green, L. H., Doherty, J. A., & Frank, R. G. (2017). The impact of nursing home pay-for-performance on quality and Medicare spending: Results from the nursing home value-based purchasing demonstration. Health Services Research, 52(4), 1387–1408.Find this resource:

Grabowski, D. C., & Town, R. J. (2011). Does information matter? Competition, quality, and the impact of nursing home report cards. Health Services Research, 46(6 pt.1), 1698–1719.Find this resource:

Greenhalgh, T., Howick, J., & Maskrey, N. (2014). Evidence based medicine: A movement in crisis?” BMJ, 348, g3725.Find this resource:

Harris-Kojetin, L., Sengupta, M., Park-Lee, E., Valverde, R., Caffrey, C., Rome, V., & Lendon, J. (2016). Long-term care providers and services users in the United States: Data from the National Study of Long-Term Care Providers, 2013–2014. Vital & Health Statistics. Series 3, Analytical and Epidemiological Studies, 38(x–xii), 1–105.Find this resource:

Hirth, R. A., Banaszak-Holl, J. C., & McCarthy, J. F. (2000). Nursing home-to-nursing home transfers: Prevalence, time pattern, and resident correlates. Medical Care, 38(6), 660–669.Find this resource:

Hittle, D. F., Nuccio, E. J., & Richard, A. A. (2012). Evaluation of the home health pay-for-performance demonstration: Final report. Baltimore, MD: CMS.gov.Find this resource:

Holmstrom, B., & Milgrom, P. (1991). Multitask principal-agent analyses: Incentive contracts, asset ownership, and job design [Special issue]. Journal of Law, Economics, and Organization, 7, 24–52.Find this resource:

Iizuka, T., Noguchi, H., & Sugawara, S. (2017). Pay-for-performance and selective referral in long-term care. SSRN Electronic Journal.Find this resource:

Institute of Medicine. (2007). Rewarding provider performance. Washington, DC: National Academies Press.Find this resource:

Johnson, R. W., Toohey, D., & Wiener, J. M. (2007). Meeting the long-term care needs of the baby boomers: How changing families will affect paid helpers and institutions. The Retirement Project Discussion Paper Series, Paper 07-04. Washington, DC: The Urban Institute.Find this resource:

Joynt, K. E., De Lew, N., Sheingold, S. H., Conway, P. H., Goodrich, K., & Epstein, A. M. (2017). Should Medicare value-based purchasing take social risk into account? New England Journal of Medicine, 376(6), 510–513.Find this resource:

Jung, J. K., Wu, B., Kim, H., & Polsky, D. (2016). The effect of publicized quality information on home health agency choice. Medical Care Research and Review, 73(6), 703–723.Find this resource:

Kane, R. L. (2011). Finding the right level of posthospital care: “We didn’t realize there was any other option for him.” JAMA, 305(3), 284–293.Find this resource:

Kansas State University Center on Aging. (2018). PEAK 2.0 2018–2019 Program Information. Manhattan, KS: Kansas State University.Find this resource:

Konetzka, R. T., Skira, M. M., & Werner, R. M. (2018). Incentive design and quality improvements: Evidence from state Medicaid nursing home pay-for-performance programs. American Journal of Health Economics, 4(1), 105–130.Find this resource:

Konetzka, R. T., Stuart, E. A., & Werner, R. M. (2018). The effect of integration of hospitals and post-acute care providers on Medicare payment and patient outcomes. Journal of Health Economics, 61, 244–258.Find this resource:

Kripalani, S., Theobald, C. N., Anctil, B., & Vasilevskis, E. E. (2014). Reducing hospital readmission rates: Current strategies and future directions. Annual Review of Medicine, 65(1), 471–485.Find this resource:

L&M Policy Research, and Harvard Medical School. (2013). Evaluation of the nursing home value-based purchasing demonstration. Year 3 and Final Evaluation Report. Washington, DC: L&M Policy Research.Find this resource:

Libersky, J., Stone, J., Smith, L., Verdier, J., & Lipson, D. (2017). Value-based payment in nursing facilities: Options and lessons for states and managed care plans. Hamilton, NJ: Integrated Care Resource Center.Find this resource:

Lipscomb University. (2014). Technical assistance report to Bureau of TennCare on the quality improvement in long term services and supports (QuILTSS) initiative. Nashville, TN: Lipscomb University.Find this resource:

Maryland Department of Health. (n.d.). Nursing facility pay for performance. Maryland.Gov.Find this resource:

McGuire, T. G. (2008). Physician fees and behavior: Implications for structuring a fee schedule. In F. A. Sloan & H. Kasper (Eds.), Incentives and choice in health care (p. 266). Cambridge, MA: MIT Press.Find this resource:

Mesteig, M., Helbostad, J. L., Sletvold, O., Røsstad, T., & Saltvedt, I. (2010). Unwanted incidents during transition of geriatric patients from hospital to home: A prospective observational study. BMC Health Services Research, 10(1), 1.Find this resource:

Mor, V., Intrator, O., Feng, Z., & Grabowski, D. C. (2010). The revolving door of rehospitalization from skilled nursing facilities. Health Affairs, 29(1), 57–64.Find this resource:

Mor, V., Rahman, M., & McHugh, J. (2016). Accountability of hospitals for Medicare beneficiaries’ postacute care discharge disposition. JAMA Internal Medicine, 176(1), 119.Find this resource:

Mullen, K. J., Frank, R. G., & Rosenthal, M. B. (2010). Can you get what you pay for? Pay-for-performance and the quality of healthcare providers. Rand Journal of Economics, 41(1), 64–91.Find this resource:

Naylor, M. D., Brooten, D., Campbell, R., Jacobsen, B. S., Mezey, M. D., Pauly, M. V., & Schwartz, J. S. (1999). Comprehensive discharge planning and home follow-up of hospitalized elders: A randomized clinical trial. JAMA, 281(7), 613–620.Find this resource:

Newhouse, J. P. (1970). Toward a theory of nonprofit institutions: An economic model of a hospital. American Economic Review, 60(1), 64–74.Find this resource:

Norton, E. C. (1992). Incentive regulation of nursing homes. Journal of Health Economics, 11(2), 105–128.Find this resource:

Norton, E. C. (2018). Long-term care and pay-for-performance programs. Review of Development Economics, 22(3), 1005–1021.Find this resource:

Norton, E. C., Li, J., Das, A., & Chen, L. M. (2018). Moneyball in Medicare. Journal of Health Economics, 61, 259–273.Find this resource:

O’Connor, M., & Davitt, J. K. (2012). The outcome and assessment information set (OASIS): A review of validity and reliability. Home Health Care Services Quarterly, 31(4), 267–301.Find this resource:

Olsen, R. M., Hellzén, O., & Enmarker, I. (2013). Nurses’ information exchange during older patient transfer: Prevalence and associations with patient and transfer characteristics. International Journal of Integrated Care, 13(1).Find this resource:

Pauly, M. V. (1968). The economics of moral hazard: Comment. American Economic Review, 58(3), 531–537.Find this resource:

Perraillon, M. C., Konetzka, R. T., He, D., & Werner, R. M. (2017). Consumer response to composite ratings of nursing home quality. American Journal of Health Economics, December, 1–36.Find this resource:

Pesis-Katz, I., Phelps, C. E., Temkin-Greener, H., Spector, W. D., Veazie, P., & Mukamel, D. P. (2013). Making difficult decisions: The role of quality of care in choosing a nursing home. American Journal of Public Health, 103(5), e31–37.Find this resource:

Poey, J. L., Hermer, L., Cornelison, L., Kaup, M. L., Drake, P., Stone, R. I., & Doll, G. (2017). Does person-centered care improve residents’ satisfaction with nursing home quality? Journal of the American Medical Directors Association, 18(11), 974–979.Find this resource:

Prendergast, C. (1999). The provision of incentives in firms. Journal of Economic Literature, 37(1), 7–63.Find this resource:

Public Consulting Group. (2017). Nursing facilities pay for performance application review: Data report. Denver, CO: Colorado Department of Health Care Policy and Financing.Find this resource:

Rahman, A. N., & Applebaum, R. A. (2009). The nursing home minimum data set assessment instrument: Manifest functions and unintended consequences: Past, present, and future. The Gerontologist, 49(6), 727–735.Find this resource:

Rahman, M., Grabowski, D. C., Mor, V., & Norton, E. C. (2016). Is a skilled nursing facility’s rehospitalization rate a valid quality measure? Health Services Research, 51(6), 2158–2175.Find this resource:

Richardson, S. S. (2012). Integrating pay-for-performance into health care payment systems.Find this resource:

Robinson, J. C. (2001). Theory and practice in the design of physician payment incentives. The Milbank Quarterly, 79(2), 149–177.Find this resource:

Roland, M. (2004). Linking physicians’ pay to the quality of care: A major experiment in the United Kingdom. New England Journal of Medicine, 351(14), 1448–1454.Find this resource:

Rosenthal, M. B. (2006). What is the empirical basis for paying for quality in health care? Medical Care Research and Review, 63(2), 135–157.Find this resource:

Rosenthal, M. B., & Dudley, R. A. (2007). Pay-for-performance: Will the latest payment trend improve care? JAMA, 297(7), 740–744.Find this resource:

Ryan, A. M., Krinsky, S., Kontopantelis, E., & Doran, T. (2016). Long-term evidence for the effect of pay-for-performance in primary care on mortality in the UK: A population study. The Lancet, 388(10041), 268–274.Find this resource:

Shamus, E., Fabrizi, S., & Hogan, J. (2018). A qualitative study of professional issues in home health therapy. Home Health Care Management & Practice, 30(1), 9–15.Find this resource:

Sherry, T. B. (2016). A note on the comparative statics of pay-for-performance in health care. Health Economics, 25(5), 637–644.Find this resource:

Shleifer, A. (1985). A theory of yardstick competition. RAND Journal of Economics, 16(3), 319.Find this resource:

Taylor, E., Dyer, M. B., & Bailit, M. (2017). State strategies: Value-based payment for Medicaid populations with complex care needs. Report. Princeton, NJ: Robert Wood Johnson Foundation.Find this resource:

Texas Health and Human Services. (n.d.). Quality incentive payment program for nursing homes. Austin, TX: Texas Health and Human Services.Find this resource:

Tian, W. (2006). An all-payer view of hospital discharge to postacute care, 2013. Statistical brief No. 205. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville, MD: Agency for Healthcare Research and Quality (US).Find this resource:

TN Division of TennCare. (n.d.). Tennessee health care innovation initiative: Long-term services and supports. Nashville, TN: TennCare.Find this resource:

Van Cleave, J. H., Egleston, B. L., Abbott, K. M., Hirschman, K. B., Rao, A., & Naylor, M. D. (2016). Multiple chronic conditions and hospitalizations among recipients of long-term services and supports. Nursing Research, 65(6), 425–434.Find this resource:

Werner, R. M., Konetzka, R. T., & Liang, K. (2010). State adoption of nursing home pay-for-performance. Medical Care Research and Review, 67(3), 364–377.Find this resource:

Werner, R. M., Konetzka, R. M., & Polsky, D. (2013). The effect of pay-for-performance in nursing homes: Evidence from state Medicaid programs. Health Services Research, 48(4), 1393–1414.Find this resource:

Werner, R. M., Konetzka, R. T., & Polsky, D. (2016). Changes in consumer demand following public reporting of summary quality ratings: An evaluation in nursing homes. Health Services Research, 51(June), 1291–1309.Find this resource:

Werner, R. M., Skira, M., & Konetzka, R. T. (2016). An evaluation of performance thresholds in nursing home pay-for-performance. Health Services Research, 51(6), 2282–2304.Find this resource:

Werner, R. M., Norton, E. C., Konetzka, R. T., & Polsky, D. (2012). Do consumers respond to publicly reported quality information? Evidence from nursing homes. Journal of Health Economics, 31(1), 50–61.Find this resource:

White, A., Hurd, D., Moore, T., Kramer, A., Hittle, D., & Fish, R. (2009). Nursing home value-based purchasing demonstration: Design refinements. Cambridge, MA: Abt Associates.Find this resource:

Woolhandler, S., & Himmelstein, D. U. (2015). Collateral damage: Pay-for-performance initiatives and safety-net hospitals. Annals of Internal Medicine, 163(6), 473.Find this resource:

World Bank. (2006). Brazil: Enhancing performance in Brazil’s health sector: Lessons from innovations in the state of São Paulo and the city of Curitiba.” Washington, DC: World Bank.Find this resource: