The purpose of a healthcare system should be to improve the health and functioning of its population;¹ therefore, it makes sense that health care quality has been described by the Institute of Medicine (IOM) as “the degree to which health care services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge.”² The standard framework for understanding how quality of care can be measured was proposed by Donabedian³ more than 30 years ago (Figure 1). Within this framework, care structures influence care processes, which in turn affect health outcomes. Foundational to the framework is the definition of high-quality medical care as that which is expected to achieve the best balance of health benefits and risks; or medical care that best improves health or prevents health decline. It is important to note that in both definitions, quality is defined not by absolute health outcomes, but rather in terms of the probability of achieving maximal health benefit as a result of the medical care provided. Within the context of these definitions, the IOM has defined six aims for the US healthcare system, which are often used to guide the development and implementation of quality programs.⁴ Specifically, health care should be safe, effective, patient-centered, timely, efficient, and equitable (Table 1).

The goal of pay-for-performance (P4P) programs is to promote performance improvement and in some cases to also decrease costs. Development of these programs was galvanized by reports demonstrating that health care delivery in the United States can be unsafe and inefficient. In 1999, the IOM published To Err Is Human, which estimated that medical errors due to misuse, overuse, or underuse contributed to approximately 100,000 deaths per year.⁵ This ranked medical errors as the fifth leading cause of death in the United States after heart disease, cancer, stroke, and chronic lower respiratory disease.⁶ In more recent reports, the United
States has been shown to have significantly worse health outcomes than other developed nations.⁷ Consistent with these outcomes, as many as 55% of Americans do not receive appropriate evidence-based care from their clinicians.⁸ Some reimbursement models may compete against quality improvement goals; for example, capitation may encourage underuse of effective methodologies and fee-for-service (FFS) may encourage overuse of ineffective treatments.⁴

Measuring care quality is important in orthopaedics, for which perioperative and overall complication rates for high-risk surgeries can be as high as 19% and 55%, respectively.⁹ Significant variation in orthopaedic outcomes exists across geographic areas.¹⁰,¹¹ Likewise, there is variation in the performance of care processes. For example, the Dartmouth Atlas demonstrates significant geographic variation in hospital discharges for total hip arthroplasty, total knee arthroplasty, and back surgery across the United States.¹² Likewise, the length of time for which patients are prescribed narcotics after surgery can vary by up to 25%.¹³ Although some variation can be explained by inconclusive evidence about the efficacy of care
processes generally, or for certain populations, there is also variation when the evidence is clear. Approximately 50% of patients with low back pain receive care in accordance with imaging guidelines and best practices,¹⁴ and certain procedures such as knee arthroscopy for osteoarthritis continue to be overutilized despite evidence supporting nonsurgical management.¹⁵

In 2001, IOM published Crossing the Quality Chasm, a 10-year strategy to address the six aims of health care improvement⁴ (Table 1). Inherent to the strategy was a need for system redesign to ensure that: (1) clinical decision making is aligned with a patient’s individual values and the local care delivery level; (2) the most up-to-date science is used in care delivery, and guaranteeing efficient and coordinated care at the healthcare organization level; (3) effective care is delivered across services and settings; and (4) financial, regulatory, and health policy mechanisms would advance the improvement aims.¹⁶

For decades, mechanisms to incentivize quality have been used in attempts to improve the quality of patient care. The first movement came from developing quality metrics of outcomes and evidence-based processes to allow for clinician and hospital-level measurement and reporting. The goal of measurement and reporting was to raise awareness of performance, determine baseline performance metrics, and create targets for improvement. However, measurement and reporting alone produced only modest effects and, although necessary, were not sufficient to improve quality.¹⁷

Financial incentives were introduced as another lever to align clinicians with payers and policymakers to improve the quality of care. P4P programs work by tying financial bonuses and/or penalties to quality measures that are, ideally, meaningful to clinicians and payers or policymakers. These measures are typically based on performing evidence-based processes or achieving an outcome of interest such as decreased complication rates, readmissions, or mortality. One of the benefits of P4P programs is that they offer a way to track measurement and payments. The economic theory of tying financial incentives to clinical care is the hope that clinicians will focus more resources on quality improvement to achieve those bonuses or avoid penalties. By focusing on quality improvement, the care that patients receive will improve and payers will reduce the amount of health care expenditures related to poor-quality care. In the words of a previous administrator for the Centers for Medicare & Medicaid Services (CMS), “You get what you pay for. And we ought to be paying for better quality.”¹⁸

The effects of P4P programs on improving quality of care have been mixed. Program design, the size of the incentives and/or penalties, the type of care affected, and the validity of the measures used all can affect program results.

In an international systematic review, P4P was shown to improve compliance with care processes in the short term but did not show substantial effect beyond 3 years.¹⁹ Greater improvement was observed when baseline performance was poor.
The observed effect on patient outcomes was mixed, partially because of the limited availability of rigorous studies.¹⁹

Many factors may limit the measurable effect of P4P programs. At the most basic level, a paucity of meaningful, validated, and reliable quality measures relevant to a particular program limits how well a program can be assessed. Although improving patient outcomes is the ultimate goal of any quality program, using outcomes to assess the efficacy of a P4P program requires risk adjustment using variables that may not be collected routinely in clinical practice. Additionally, most P4P programs run over a 1-year period, which may not be long enough to achieve optimal outcomes for certain diseases or procedures. Process measures are more easily measured and actionable by physicians, especially over a short time period. However, utilization of measures that physicians do not view as a meaningful representation of the quality of the care they provide can affect not only collection, but also performance on those measures. Additionally, because some measures require volume thresholds, many physicians may not qualify for many metrics.