Pay for Performance: Where's The Return?

BACKGROUND: Pay for performance (P4P) initiatives are designed to foster and reward improvement in health care delivery. These programs promote “value-based health care” by rewarding quality care that is characterized by a reduced amount of disproportionate spending. OBJECTIVE: To review the intent and design of P4P initiatives as well as the design and results of P4P programs in current practice. SUMMARY: Three key principles are fundamental to building a value-based health care system: measurement, transparency, and accountability. There are several levers currently driving P4P, each influencing the movement in its own way. Among these are employers, federal agencies such as the Centers for Medicare & Medicaid Services and the Department of Health and Human Services, health plans, providers, accreditors, and Congress. One key player in the P4P movement, the National Committee for Quality Assurance (NCQA), is a private, independent nonprofit health care quality oversight organization that measures and reports on health care quality and unites diverse groups around a common goal: improving health care quality. NCQA, has demonstrated several successful provider-level measurement initiatives connected to P4P programs, notable among them Bridges to Excellence programs in several markets, physician recognition programs, the Integrated Healthcare Association’s P4P initiative in California, the National Forum on Performance Benchmarking of Physician Offices and Organizations, and health plan accreditation. CONCLUSIONS: The initial data from developmental P4P programs across the nation have indicated that both financial and nonfinancial incentives motivate significant change in health care delivery, but the return on investment of these initiatives is not yet known.

AMA PRA Category 1 credits for physicians will be awarded for successful completion of this continuing education program. For faculty disclosures, please see pages S6, S10, S14, and S19. For accreditation information, please see page S20.
The articles published in this supplement represent the opinions of the authors and do not reflect the official policy or views of the Academy of Managed Care Pharmacy, the authors' institutions, or Merck & Co., Inc. unless so specified. The authors have disclosed if any unlabeled use of products is mentioned in their articles. Before prescribing any medicine, clinicians should consult primary references and full prescribing information.

Target Audience
Physicians and pharmacists practicing in a managed care environment

Learning Objectives
Upon completion of this program, participants will be able to 1. explain the benefits and risks of pay for performance (P4P) programs, 2. cite the early results of P4P programs to future performance-based measurement development, 3. differentiate the effects of P4P results versus disease management activities, 4. describe how new therapies in diabetes can be aligned with P4P objectives to improve outcomes, and 5. discuss how to align performance-based incentives with best practices using a clinical case-study approach.
T he pay for performance (P4P) movement in managed care is based on the premise of fostering and rewarding improvement in health care. Most P4P programs are constructed to promote "value-based health care" by rewarding the types of care that tend to reduce the need for the disproportionate spending on care of patients with severe disease or high risk for disease. "Value-based" health care ultimately promotes a more effective distribution of funds and efforts across the continuum of health through measurement, transparency, and accountability. P4P programs should be rooted in guidelines established from clinical evidence and developed in collaboration with purchasers, policymakers, plans, and the providers themselves.

ss NCQA and Pay for Performance
The National Committee for Quality Assurance (NCQA) is a private, independent nonprofit health care quality oversight organization. The organization measures and reports on health care quality and unites diverse groups around a common goal: improving health care quality. NCQA' s mission is to improve the quality of health care, while the organization' s vision is to transform health care through measurement, transparency, and accountability.
NCQA has demonstrated several successful provider-level measurement initiatives connected to P4P programs; notable among them are Bridges to Excellence programs in numerous markets, physician recognition programs, the Integrated Healthcare Association (IHA) P4P initiative in California, the National Forum on Performance Benchmarking of Physician Offices and Organizations, and health plan accreditation.
NCQA physician recognition programs serve as an excellent case study for the P4P movement. They demonstrate the value of measuring structure, process, and outcomes, and rewarding high performance appropriately.
Initial results from such P4P programs strongly suggest that measurement coupled with rewards leads to improvement in health care quality. Furthermore, physicians appear to be the most crucial group in the engagement and implementation of P4P initiatives, with a focus on actionable performance feedback. While the success demonstrated by these programs is encouraging for the managed care community, only time will tell if the overall movement will, in fact, be fiscally sound, as payers' return on investment (ROI) of P4P programs becomes clearer.
ss Improving Health Care Quality With P4P In 2001, the Institute of Medicine (IOM) recognized a void in the current state of health care quality and released a landmark report, Crossing the Quality Chasm: A New Health System for the 21st Century. 1 In this report, the IOM outlined several suggested initiatives for improving health care quality: creating an infrastructure to support evidence-based practice, facilitating the use of information technology (IT), preparing the workforce to better serve patients in a world of expanding knowledge and rapid change, and aligning payment incentives to encourage and support quality improvement. This alignment of payment incentives characterizes P4P and its role in the improvement of health care quality.
Central to P4P program design is the concept of "value-based health care." In the current state of health care, a disproportionate amount of resources and efforts are being spent on patients who are suffering from or who are at high risk for severe disease. This disparity in spending gives short shrift to efforts to promote wellness. In an environment where 46 million Americans do not have access to health insurance, maximizing health derived from limited health care resources is a moral and public health imperative. Value-based health care seeks to create a more uniform continuum of spending across all patient groups, keep healthy people healthy, and help people with early-stage chronic disease from progressing to more serious, complex conditions ( Figure 1).

ss Measurement in P4P
Three key principles are fundamental to building a value-based health care system: measurement, transparency, and accountability. These principles are presented in deliberate order. Measurement is the primary concept in improving the quality of health care; without measurement, there is no baseline level of quality against which to gauge improvement. Transparency serves 2 purposes: first, it ensures that quality data is translated into measures and reports that consumers and purchasers can understand and use to make informed decisions. Second, valid measurement is crucial to assigning accountability: those who deliver health care cannot be held accountable for improvement if the quality of care itself is not accurately measured.
Metrics employed in P4P initiatives should be evidence-based and reflect a consensus of key stakeholders. P4P metrics come in 2 forms: measures and guidelines. Measures indicate desired clinical outcomes. For instance, a measure might assess whether a physician' s patients with cardiac conditions control their lowdensity lipoprotein cholesterol (LDL-C) to 100 mg/dL or less. On the other hand, guidelines monitor evidence-based "best-practices" that lead to that desired results; a related guideline would ask whether the same doctor ordered annual cholesterol tests for his cardiac patients. Measures and guidelines alike are useful as measurement tools in P4P programs.
A useful P4P measure is relevant, sound, and feasible: measures that lack these qualities are a waste of time and resources. Relevant measures will be designed to have a significant impact on health outcomes; advance cost-effective, evidence-based methods of care; have strategic significance to practitioners and health plans; and will assess aspects of care that can be improved. Measures must also be sound; that is, specific enough to reliably provide accurate assessments of care. Only trustworthy, evidence-based measures will be embraced by health plans, providers, consumers, and purchasers. Feasible P4P measures are practical for use in the "real world" of health care: they are characterized by precise specifications, a reasonable cost of measurement, and assured confidentiality. Feasible P4P measures also promote accountability by being open to a third-party audit, such as by a health plan or other P4P program sponsor.
The sorts of P4P programs that are possible are driven by 2 factors: the level where measurement takes place-the plan, hospital, or physician office -and the availability of relevant data. Take, for instance, the measurement of pharmacy claims data. Over the short term, limited initiatives at the plan or physician level may arise from pharmacy claims data taken in isolation. However, a more comprehensive pharmacy P4P program would require that those same pharmacy claims be linked to diagnoses or other data that may not be widely collected at the present time.
On the plan level, measurement has led to improved health care, specifically for certain Health Plan Employer Data and Information Set (HEDIS) measures developed by NCQA. In the 5 years from 1999 to 2004, NCQA documented an average increase of more than 52% for HEDIS effectiveness-of-care measures for chicken pox vaccination, hypertension, LDL-C control, LDL-C control among patients with diabetes, and asthma ( Figure 2). 2 Similar improvement has also been observed on the physician level, where measurement has also led to advances in clinical quality. For example, performance among applicants to the Diabetes Physician Recognition Program (DPRP) showed substantial improvement in key measures, such as glycosylated hemoglobin (A1C) control, blood pressure control, and lipid control. 3 ss Many Voices in the P4P Debate Entities ranging from individual physicians to federal agencies have waded into the P4P fray, each influencing the movement in its own way. Employers, health plans, accreditors, and Congress have also weighed in. The Centers for Medicare & Medicaid Services (CMS) and the Department of Health and Human Services (DHHS) influence providers through demonstration projects and payment updates. Employers in several markets have banded together to sponsor programs that provide financial incentives to high-performing physicians and practices. Likewise, health plans support provider-performance measurement and recognition programs; while providers and consumers alike benefit from transparent ratings. Accreditors further influence the P4P movement by creating new evaluation tools for measurement and reporting.
Federal agencies impact P4P on the physician level primarily through CMS' s Physician Voluntary Reporting Program (PVRP). This program uses several different methods for collecting data from physicians' offices, including claims for Current Procedural Terminology (CPT) Category II codes and G-codes. NCQA is assisting in the development of a validation methodology for the program through confidential feedback reports. Other Medicare demonstration projects such as the Physician Group Practice Demonstration, which involves the coordination of Part A and Part B services, also exist. DHHS also provides grants to build electronic health systems, including $18.6 million for 12 regions to link doctor offices, clinics, and hospital networks using open data standards by the end of 2006. 4 The impact of health plans and employers on P4P can be seen in collaborative efforts such as the IHA and Bridges to Excellence programs in which NCQA plays a key role, as well as in the Massachusetts Health Quality Partners and Minnesota Community Partnership. Conversely, single-plan initiatives exist on this level. Notable among them are those developed by the Excellus/Rochester IPA and BlueCross/BlueShield of Michigan; the latter program demonstrated improved cardiac care and a 45% reduction in infection rates. 5 New NCQA voluntary standards for health plans further influence these efforts through the promotion of standardized measurement and reporting of physician and hospital performance results. The new Physician and Hospital Quality (PHQ) standards have found robust support in the market: the program has earned 37 employer and consumer endorsements and 49 plans are early adopters of PHQ. 6 Plans that voluntarily participate are required to use standardized measures, provide transparency about measurement, share measurement with those being measured, collaborate with other plans contracting with the same providers, and use results for reporting and for other quality improvement (QI) activities.
ss California P4P-IHA One of these programs-IHA' s P4P initiative in California-has demonstrated both longevity and success in the P4P arena. IHA comprises 7 health plans with 6.2 million commercial managed care organization (MCO) enrollees and 225 capitated medical groups that provide for multiple health plans. 7 There are 35,000 physicians in IHA, and Kaiser Permanente joined the association at the end of 2006. 7 In this landmark collaboration, health plans and medical groups agree on measures, and plans combine their data. Clinical data is administrative only in IHA, and patient experience data surveyed at the group level. IT adoption is a performance standard employed by IHA that is evaluated by NCQA; NCQA also serves as the data aggregator. Based on the results of this monitoring, health plans make individual decisions on rewards with P4P recommendations.
The clinical measures employed by IHA have evolved over the past 3 years, from 6 measures in 2003 to 10 measures in 2005: childhood immunizations, cervical cancer screening, breast cancer screening, asthma management, A1C screening, A1C control, LDL-C screening among patients who had a cardiac event, LDL-C control <130 mg/dL for patients who had a cardiac event or were diagnosed with diabetes, chlamydia screening, and appropriate treatment for children with upper respiratory infection. 8 Looking at the third-year results-that is, results reflective of care delivered in 2005-significant improvement was demonstrated in 3 categories: breast cancer screening (64% of physician groups improved in year 3), cervical cancer screening (61% improved), and diabetes: A1C screening and control (54% improved screening).
A particularly notable result arising from the third-year data was an average improvement of 40% among those physician groups who achieved a full IT score. 8 This suggests that adoption of IT facilitates gains in quality. The relationship between IT adoption and clinical quality is most prominently displayed among groups scoring between 0% and 20% of the IT score. Groups who more fully integrated technology into their delivery systems tended to post higher clinical quality scores. 8 NCQA/American Diabetes Association (ADA) Diabetes, including 2,600 physicians; and American Heart Association (AHA)/ American Stroke Association (ASA)/NCQA Heart/Stroke, including 400 physicians. Measures included in these programs include structure, process, and outcomes of excellent care management. More than 3,800 physicians are recognized nationally through these programs, and they are rewarded by many health plans and Bridges to Excellence employers. 6 To earn recognition, physicians must achieve certain standards of care across their entire panel of eligible patients. In the DPRP, for example, the total weight of all the scored measures is 100; physicians must achieve 75% to receive recognition. 9 Between 2003 and 2005, there was a 33% increase in the number of DPRP-recognized physicians. In Bridges to Excellence Diabetes Care Link areas, the increase was particularly striking-the number of recognized physicians increased 450% over the same time frame. 6 ss Conclusions The initial data from fledgling P4P programs across the board has indicated that while financial incentives do indeed motivate significant change, nonfinancial support also promotes quality improvement. The engagement of physicians is critical to the P4P movement, and public reporting heightens physician awareness. Experience has shown that actionable feedback on performance is a prerequisite to gains in clinical quality. Data integrity increases trust in the results. 10 NCQA' s data on the first wave of P4P programs has indicated that measurement provides physicians with a new perspective on their practice and that practices change their processes and delivery systems in order to meet program standards. National standards appear to be just as difficult to achieve for small and large practices, but having national measures helps reward programs get started. Physicians (i.e., generalists and some specialists) appreciate consistent requirements for measures. Clinical data continues to be difficult to obtain, as chart abstraction, especially in the absence of widespread use of electronic medical records, requires a significant commitment of time and resources.
P4P is not a silver bullet to cure all ills; it is a useful tool to align payment incentives. How a P4P initiative is conducted is just as important as whether it is conducted-it must be based on widely recognized, evidence-based measures and developed in collaboration with the providers it proposes to measure. Measurement and payment functions must be kept separate, and payers are right to be skeptical of additive payments.
The bottom line is that measurement plus rewards equals improvement in health care quality. While initial P4P programs have demonstrated success in improving health care quality, the financial implications of P4P are not yet known since the ROI data pertaining to such programs is incomplete at this time.

DISCLOSURES
This article is based on a presentation given by the author at a symposium, "Pay for Performance: Where' s the Return?" held October 4, 2006, at the Academy of Managed Care Pharmacy' s 2006 Educational Conference in Chicago, Illinois. The symposium was supported by an educational grant from Merck & Co., Inc. The author discloses that she has received an honorarium from Merck & Co., Inc. for participation in the symposium and this supplement. She discloses no potential bias or conflict of interest relating to this article. D isease management (DM) and pay for performance (P4P) are 2 recently introduced intervention movements in managed care, designed to improve the collaboration of efforts within the health care system and the resulting patient outcomes. More specifically, DM is a system of coordinated health care interventions and communications for populations with conditions in which patient self-care efforts are significant. 1 P4P is typically more variable in its definition, becoming whatever the local providers and plans decide it should be; however, P4P always involves an incentive component in which payment is defined based on meeting specific, previously agreed-upon process or outcomes targets.
Partners HealthCare System (PHS) is a large integrated health care organization in eastern Massachusetts that has been a leader in the P4P movement. One group in the PHS network, Brigham and Women' s Physicians Organization (BWPO), provides health care services to 42,000 patients through its primary care physicians (PCPs). P4P programs exist within BWPO for inpatient admissions, diabetes, and radiology, among others.
The inpatient admissions P4P initiative from BWPO seeks to reduce costs by preventing both unnecessary admissions and admissions to more costly institutions within the network, using a scoring system and through disease management. The P4P program for diabetes aims to improve care by promoting the glycosylated hemoglobin (A1C) and low-density lipoprotein cholesterol (LDL-C) screening as well as eye exams for patients with diabetes. The radiology measure employs a computerized ordering system to monitor outpatient testing and promotes peer-to-peer consultation to prevent rising costs through unnecessary tests. Collectively, BWPO' s P4P programs aim to improve the quality and efficiency of care within the organization.
While it is too early to determine if DM or P4P is having a greater effect on care quality or cost-effectiveness, several measures can be taken in the meantime to continue to encourage the movement of these programs in the right direction. Understanding the problems, encouraging informed decision making, focusing on incentives, and creating partnerships with providers and DM vendors are all key actionable items that will improve the chance of success for both managed care quality initiatives.

ss Overview of DM and P4P
Although both are still in their fledgling stages, DM and P4P have shown significant promise in increasing the quality and costeffectiveness of health care in the managed care markets. Both movements are based on the premise of promoting collaboration between all parties in managed care, and both aim to improve patient outcomes while lowering total costs in the long term.
A system of coordinated health care interventions and communications for populations with conditions in which patient Merging P4P and Disease Management: How Do You Know Which One Is Working?
Allen L. Smith, MD, MS ABSTRACT BACKGROUND: An intervention movement in managed care, disease management (DM), is a system of coordinated health care interventions and communication for populations with conditions in which patient self-care efforts are significant. Another managed care intervention movement, pay for performance (P4P), involves an incentive component in which payment is defined based on meeting specific, previously agreed-upon process or outcomes targets.
OBJECTIVE: To explore the various characteristics of DM and P4P interventions, determine how they differ, and explore the differences in results of programs in current practice.
SUMMARY: In DM, regular ongoing evaluation of clinical, humanistic, and economic outcomes plays a crucial role in reducing costs and improving quality of care. The goal of improving overall patient health in DM is also accomplished by supporting the physician or practitioner/patient relationship and plan of care. P4P initiatives vary more according to the needs and preferences of local providers and plans than do DM initiatives. While DM programs can be implemented without necessarily changing how providers deliver health care, P4P requires new programs and/or systems within the provider sector to improve patient care quality and/or efficiency. P4P initiatives also typically involve the upside or downside risk by physicians/hospitals. Partners HealthCare, based in Boston, features P4P initiatives for inpatient admissions, diabetes, and radiology that have all been met with success.
CONCLUSIONS: Both DM and P4P initiatives have been successful in managed care. However, in terms of determining whether DM or P4P initiatives are more effective in improving the quality and efficiency of health care delivery, it is simply too early to tell at this time.
self-care efforts are significant, DM emphasizes the prevention of disease as a primary objective. 1 This is accomplished using evidence-based practice guidelines and patient empowerment strategies. In DM, regular ongoing evaluation of clinical, humanistic, and economic outcomes plays a crucial role in reducing costs and improving quality of care. 1 The goal of improving overall patient health in DM is also accomplished by supporting the physician or practitioner/patient relationship and plan of care. 1 P4P initiatives vary more according to the needs and preferences of local providers and plans. Furthermore, while DM programs can be implemented without necessarily changing how providers deliver health care, P4P requires new programs and/or systems within the provider sector to improve patient care quality and/ or efficiency. P4P initiatives also typically involve the upside or downside risk by physicians/hospitals. One characteristic universal to all P4P programs is the concept of granting incentives to providers based on meeting specific, previously agreed-upon process or outcomes targets.

ss P4P at Partners HealthCare System
Based in Boston, PHS is an integrated health system founded by Brigham and Women' s Hospital (BWH) and Massachusetts General Hospital in 1994. 2 PHS includes academic medical centers, community hospitals, specialty hospitals, community health centers, a physician network, home health and long-term care services, and other health-related entities. 2 Within BWPO in PHS, P4P initiatives exist in the areas of inpatient admissions, diabetes, radiology, pharmacy, and electronic medical records. The programs operate by withholding 10% of physician/ hospital fees and returning those fees based on whether quality and efficiency targets are achieved. The inpatient admissions, diabetes, and radiology initiatives are subsequently discussed further.

ss Inpatient Admissions P4P Program
Results of rounding on 50 general medicine admissions at BWH revealed that ~50% of all admitted patients had a chronic illness and a prodrome that precipitated the admission. Furthermore, patients were not always accessing care as early as possible, and many of these admissions may have been prevented with earlier treatment. In order to reduce these costly hospital admissions within network, BWPO designed a medical management program. This program includes the 42,000 lives covered by BWPO PCPs and involves $4.4 million of annual risk, three fourths of which is at stake for BWH and one fourth of which is at stake for BWPO. The program features a numerical rating system that assigns values to different types of hospital admissions and nonadmissions to assess efficiency: One admission to an academic medical center is assigned 1.4, one admission to a community hospital is assigned 1.0, and one observation to any hospital is assigned 0.0. Because academic medical centers are associated with higher costs and are therefore assigned a higher value, the program discourages unnecessary admission to these high-cost centers.
This proactive approach is a good method to address the challenge of unnecessary admissions. Root causes for unnecessary admissions should be understood, but the focus must always remain on the patient. The program must not attempt to reduce all admissions, only those admissions that are preventable. Also, in an attempt to reduce costs, community hospitals and medical observation should be made readily available to the patient when it is clinically appropriate. It is important to maintain a collaborative approach to the problem and thereby prevent duplication of efforts, which only leads to further expenditures.
Preventable admissions for patients with chronic illness can be reduced by improving patient self-management and ensuring that patients see their PCPs earlier in their illness prodromes. 3,4 Improving patient self-management and adherence to medications and therapies can be achieved through patient education (i.e., face-to-face education during hospital stays as well as education administered via follow-up outbound phone calls) and by increasing enrollment and participation in preexisting disease management programs. In order to ensure that patients see their PCPs earlier, patient knowledge about acute symptoms should be strengthened and the perceived or actual barriers to physician access should be removed.
At BWH, a "Plan and Promise" strategy was implemented to promote adherence and develop an illness plan by identifying symptom progression early, presenting patients with a customized strategy to follow, and addressing barriers to access. In order to promote the success of this "plan," the organization "promised" to triage outbound calls from patients and insure timely access to clinicians.
The incorporation of these measures into BWPO' s P4P program has been met with success that can be documented both on an individual level and on a plan level. One patient, a 50-year-old woman with recurrent serious secondary colon infections (Clostridium difficile), had experienced 6 admissions over 3 months. The patient waited before each admission to seek care, leading to the exacerbation of her illness to the point that hospitalization was necessary. Through BWPO' s program, patient education administered by a case manager led the patient to seek care earlier in the course of her subsequent infection, and it was treated before hospitalization was necessary.
Another patient positively affected by BWPO' s program was a 47-year-old male with a history of stroke, diabetes, cardiomyopathy, and atrial fibrillation. The patient was prescribed 17 medications and had multiple emergency department visits and admissions. Further complicating the case, the patient could not afford his medication copays despite working 2 jobs. This patient was referred to a social worker through BWH and received financial assistance for his medication, which ultimately led to improved control over his comorbid disease. He has not been admitted to the hospital since these interventions commenced.
These individual success stories are demonstrative of the type of processes that translate into financial gains for BWH and BWPO, with an improvement in BWPO inpatient withhold return from 66% to 88% between 2005 and 2006 year to date. The financial implications of this are significant because the improved projected 2006 return represents $1 million for BWH and BWPO combined.
ss Diabetes P4P Program BWPO' s diabetes P4P initiative measures and monitors screening rates (A1C, LDL-C, and eye exam) to improve the quality of care among its more than 1,300 patients with diabetes. With $1.1 million at risk for BWPO, the program' s goal is to exceed the 90th percentile in each Health Plan Employer Data and Information Set (HEDIS)-determined measure in each screening category.
The program employs a "centrally guided, locally led" strategy in which a central diabetes patient outreach coordinator tracks data and performance progress while PCP practice staff and PCPs work directly with the patients. Furthermore, electronic records and claims are used to track patient screening data while a central office sends reminder letters with prefilled lab slips to patients. In this hybrid approach to diabetes care, an outreach coordinator provides monthly "out-of-compliance" reports to practice staff, who call patients to schedule appointments. In addition, phlebotomists make home visits, and incentives and celebratory rewards are given to PCPs, ophthalmologists, and staff.
BWPO' s diabetes P4P program demonstrated consistent improvement over a 3-year span in terms of screening rates for A1C, LDL-C, and eye exams. Between the years of 2002 and 2005, screening rates for A1C improved from 67% to 93%, screening rates for LDL-C improved from 78% to 94%, and eye exam rates improved from 54% to 67% (Figure 1), all of which met the P4P targets.
Upon reviewing the characteristics of BWPO' s diabetes P4P program, it may appear that the initiative could actually be considered DM. The program does, in fact, share many characteristics with a typical DM program, including the support of physicians in executing a plan of care, the use of similar record or claims data, and the importance of patient communication. However, BWPO' s diabetes P4P program differs from DM programs in that it relies on a direct relationship with PCP practices, whereas DM programs have limited interaction with PCP practices. Furthermore, the diabetes P4P program incorporates the use of incentives for PCPs, while the DM program does not.
ss Radiology P4P Program The radiology P4P program at BWPO is based on the concept of evidence-based medicine for imaging. In order to improve patient safety, quality, and efficiency of care, the radiologist serves as a physician consultant, both before and after testing, thus fostering appropriate use of radiology services. The program includes the 42,000 patients seen by BWPO PCPs and involves $2.2 million at risk, 80% of which belongs to BWH and 20% of which belongs to BWPO.
The radiology measure looks at outpatient testing rates per 1,000 patients for magnetic resonance imaging (MRI), computed tomography (CT) scan, and nuclear cardiology. A computerized physician order entry (CPOE) system is in place to aid in the ordering of imaging services, and the CPOE system is integrated into the electronic medical records system to prevent duplicate or unnecessary testing. Decision support and peer-topeer consultation also reduce the likelihood of unnecessary testing, and physician profiling helps program leaders target specific physicians for intervention.
The radiology P4P program appears to be demonstrating success, with more and more physicians embracing the CPOE system. While only a fraction of physicians used the system in October 2002, by August 2004, the number of patients tested via CPOE was approaching 75% of total patients with a radiologic exam ( Figure 2). This program shares many characteristics with prior authorization programs being utilized by other health plans to reduce unnecessary radiology testing; however, a major difference with the BWPO approach is that it is integrated into the electronic ordering system and involves local peer-to-peer interaction rather than communicating with personnel from a third-party vendor.
ss Conclusions BWPO' s P4P programs aim to improve the quality and efficiency of care within the organization through local efforts and central guidance. While these programs share several characteristics with DM programs, they all differ in their methodology.
In terms of determining which of the 2 types of programs are the most effective in improving the quality and efficiency of health care delivery, it is simply too early to tell at this time. In the meantime, several measures can be taken to ensure that both the DM and P4P initiatives are successful in achieving their goals. Understanding the problem being targeted is imperative, as is emphasizing disciplined decision making before initiating a program and excellent implementation efforts. A focus on incentives and partnership between providers and DM vendors, where possible, may further increase the likelihood of program success. Finally, assumptions should be questioned periodically and external review welcomed to ensure progression of these initiatives in the right direction. The DM and P4P movements are likely to grow in the future, parallel to the growth in the demand for improved quality and efficiency of care. The federal government may further the development of these movements by pushing for P4P and possibly DM for Medicare and other public sector programs, but if they do so, it will most likely be on a very tight budget. Stakeholders should be cognizant of excessive hype and the "creative" use of statistics to justify various DM and P4P initiatives, as this will increase confusion about what really works.
The improvement of health care quality is a complex problem with many facets and, as such, there will be no panaceas. Comprehensive efforts featuring multiple components and coordinated efforts, however, have shown promise. Technology, such as electronic medical records, will assist in these efforts, provided there is sufficient leadership, systems, and incentives to effect measurable change. Privacy concerns will also remain an issue and should be addressed to prevent the misuse of protected information. Although cooperative, multistakeholder solutions are challenging to develop and maintain, they will likely be met with success in the improvement of health care quality once in place.   1 In addition, this multiorgan disease is a growing health problem for future generations, with a reported 7% to 45% increase in children with type 2 diabetes. 2 In an effort to control this epidemic of diabetes, pay for performance (P4P) initiatives, such as those based on the National Committee for Quality Assurance' s (NCQA) Health Plan Employer Data and Information Set (HEDIS) diabetes care measure, have emerged. New diabetes treatment strategies, including these P4P initiatives, have placed an increasing emphasis on tight glycemic control. Still, despite these efforts, fewer than 2% of American adults with diabetes receive optimal quality of care, and national averages of blood glucose levels are well above desired target goals. 3,4 Diabetes is a complex condition, often accompanied by significant comorbidities. This complexity, coupled with the tendency for patients to be diagnosed late in course of illness, are just 2 possible reasons for this alarming trend and the current state of diabetes management. An examination of a case study of a 62-year-old male with type 2 diabetes and multiple comorbidities will be used to illustrate the impact of these factors on diabetes management.
The case study highlights the role of pharmacological agents with different mechanisms of action, in addition to nutrition and lifestyle changes, and emphasizes the importance of patient education and empowerment in the treatment of diabetes. All of these factors can contribute to improvements in P4P scores.
A focus on postprandial hyperglycemia, in addition to fasting blood glucose excursions, may serve to improve the current state of diabetes care. Adding such recently introduced agents as the incretin hormones, glucagon-like protein-1 (GLP-1) analogs, and dipeptidyl peptidase-IV (DPP-IV) inhibitors to the existing treatment plans may prove more beneficial in improving glycemic control than traditional agents. These therapies target different processes in the course of diabetes, exerting their effects on postprandial glucose excursions. The use of individually tailored therapy, employing the most effective pharmacologic agents possible in conjunction with nutrition and lifestyle changes, remains the best option. Best practices such as these can be enhanced through P4P initiatives to further the effectiveness of diabetes therapy in managed care.

ss Incidence and Prevalence of Diabetes
Diabetes is one of the most prevalent diseases in the United States, with one person in America being diagnosed every 25 seconds. 1 A total of 20.8 million people in the United States have diabetes, including 10.3 million adults over the age of 60 years, and more Clinical Case Study: Achieving Long-term Control of Insulin Resistance Susan A. Cornell, BS, PharmD, CDE, CDM ABSTRACT BACKGROUND: A total of 20.8 million people in the United States have diabetes, including 10.3 million adults over the age of 60 years, and more than 6 million people remain undiagnosed. Although diabetes is widely recognized as a prevalent and serious disease in managed care, current care is suboptimal, with less than 2% of American adults with diabetes receiving optimal quality of care.
OBJECTIVE: To review the various treatment interventions available in diabetes care, including the use of pay for performance (P4P) initiatives.
SUMMARY: In an effort to improve the current state of diabetes care, the NCQA's Health Plan Employer Data and Information Set (HEDIS) diabetes measure was developed as a means to better promote monitoring of various clinical markers in patients with this disease. This measure has been employed in P4P initiatives across the country by granting incentives to providers who have a prespecified proportion of their patients with diabetes meeting the measure. Likewise, to improve outcomes, many experimental and recently approved treatment options for diabetes target different processes in the course of the disease.
CONCLUSIONS: An effective program for the management of a diabetes patient population must be multidisciplinary, coordinating the efforts of many different levels of health care providers. Furthermore, components commonly incorporated in P4P initiatives, such as patient self-management education, provider contact, and the use of the American Diabetes Association standards of care measures for screening and lab levels, are imperative to the success of a diabetes management program. than 6 million people remain undiagnosed. 1 Mortality rates from diabetes and its related conditions, such as cardiovascular disease, highlight the seriousness of the this disease and contribute to it being the sixth leading cause of death listed on U.S. death certificates in 2002. 6 Furthermore, the problem of diabetes is growing, as approximately 35 million people in the United States have prediabetes. 1 This epidemic is not confined to adults or the elderly-reports indicate a 7% to 45% increase in children with type 2 diabetes. 2

ss Current Trends in Diabetes Management
Although diabetes is widely recognized as a prevalent and serious disease in managed care, current care is suboptimal. In fact, fewer than 2% of American adults with diabetes are receiving optimal quality of care. 3 Tight glycemic control, a fasting blood glucose goal of <110 mg/dL and a 2-hour postprandial goal of <140 mg/dL, (Table 1) 5 is a key component of diabetes management, since the maintenance of blood glucose within a narrow target range can minimize or prevent the microvascular and macrovascular complications associated with the disease. While this concept is universally accepted, the national averages of blood glucose levels remain well above desired target goals, indicating the need for improved management and care. Recent estimates of the U.S. averages were >200 mg/dL for fasting plasma glucose (FPG), >300 mg/dL for postprandial blood glucose (PPG), and greater than 9.5% for glycosylated hemoglobin (A1C). 4 In an effort to improve the current state of diabetes care, the NCQA' s HEDIS diabetes measure was developed as a means to better promote monitoring of various clinical markers in patients with this disease. The HEDIS diabetes measure evaluates the percentage of plan members with diabetes (aged 18-75 years) with A1C screening; A1C controlled to <9.0% (which will change to <7.0% in 2007); low-density lipoprotein cholesterol (LDL-C) screening; LDL-C controlled to <100 mg/dL; eye examination; and microalbuminuria screening. 8 This measure has been employed in P4P initiatives across the country by granting incentives to providers who have a prespecified proportion of their patients with diabetes meeting the measure.
Over the past decade, providers in the United States have shifted the treatment paradigm toward a strategy that incorporates the use of more pharmacologic agents and fewer dietary or lifestyle interventions or insulin therapy alone. The use of pharmacotherapy alone in treatment of diabetes increased from 45% to 53% in type 2 patients, while the use of dietary interventions and insulin alone decreased from 24% and 27% to 16% and 20%, respectively, in patients with type 2 diabetes (Figure 1). 9 These changes in diabetes pharmacotherapy do not appear to have resulted in improved levels of glycemic control. Indeed, the proportion of patients achieving the American Diabetes Association' s (ADA' s) recommended A1C level of <7.0% as measured in the National Health and Nutrition Examination Survey (NHANES) III and IV actually decreased from 44.5% to 35.8% between the years 1994 and 2000, indicating the need to continue to improve education and self-management efforts. 9 It is possible that increasing reliance solely on pharmacological measures, especially monotherapy, resulted in an overall worsening of the condition in patients, as demonstrated by the trend toward less-effective glucose control. Examination of a case study of a 62-year-old male with type 2 diabetes for 4 years helps to demonstrate the challenges inherent in the treatment of patients struggling with the disease. It is important to remember that the ADA estimates that by the time a patient with type 2 diabetes is finally diagnosed, they have actually had the disease for approximately 9 years. 10 Therefore, from the time of diagnosis, treatment plans need to be aggressive in order to optimally manage this disease, since most patients already have complications resulting from this delay in diagnosis.

Nationally Accepted Targets for Glycemic Control
The patient' s laboratory and physical findings in this study are outlined in Table 2. As with most diabetes patients, this particular patient has concurrent cardiovascular disease, specifically hypertension and dyslipidemia. To assist in the management of these conditions, the patient had been prescribed glyburide 5 mg two times daily, metformin 850 mg three times daily, an ACE inhibitor, a diuretic, and a low-dose statin.
Illustrating the complexity of treating such a high-maintenance disease, the patient acknowledged having a self-monitoring blood glucose (SMBG) meter but reported having "no time" to check his blood glucose levels. In addition, the patient admitted to a sedentary lifestyle, which led to his gradual increase in weight since diagnosis. He also described poor nutritional habits, such as eating fast food more than 3 times a week, thereby adding to poor glucose control and a need for a more aggressive treatment plan.
Diabetes self-management education (DSME) is the cornerstone of optimal management of this disease. As such, patient selfmanagement education and quality physician contact commonly incorporated in P4P initiatives may benefit this patient. Motivational and behavioral change strategies that support lifestyle improvements would likely lead to better glucose control, enhanced quality of life, and an increased likelihood of treatment success. In addition, at the time of diagnosis, the patient did not receive a dilated eye exam, a monofilament foot exam, or screening for kidney disease, as mandated in the ADA standards of care. This demonstrates the need and value of P4P and other initiatives that can increase the likelihood of better managing patients with diabetes.
To improve the patient' s overall health, treatment goals include reducing his A1C to <7% (preferably <6.5%), and reducing LDL-C to <100 mg/dL. In addition, the patient is scheduled for regular dilated eye exams, monofilament foot exams, and screening for kidney disease to ensure consistency in monitoring and care. A self-management diabetes education program was also implemented for this patient to better manage his diabetes by increasing his SMBG meter use to twice daily, improving his nutritional habits, and increasing his physical activity. These common P4P-related activities ultimately served as the base for the patient' s diabetes management.
Although lifestyle played an important role in the progression of this patient' s disease, his diabetes remained uncontrolled despite pharmacological treatment with glyburide and metformin.
Until the past decade, few pharmacologic options existed that focused on treating the cause of diabetes as opposed to the symptoms. Additionally, diabetes is a multifaceted disease requiring different agents that work on different organs. Recent pharmacological advancements have introduced new agents that may play an important role in improving the glucose control of this patient as well as that of others. In order to determine whether a switch in therapy is necessary, the current regimen must be evaluated to determine the drugs' efficacy in reducing the patient' s A1C to <7%. If the current pharmacologic agents (glyburide, metformin) at their current dose are determined to have suboptimal effectiveness, action must be taken either in considering dose adjustment or adding or switching agents to reduce and maintain A1C levels. In terms of switching agents, alternative conventional therapies for this particular patient include alpha glucosidase inhibitors, glitinides, or thiazolidinediones (TZDs), while emerging therapies include GLP-1 analogs or DPP-IV inhibitors.

ss Emerging Therapies in Diabetes Management
Many experimental and recently approved treatment options for diabetes target different processes in the course of the disease. In particular, a focus on improvement in the regulation of the gastrointestinal motility and the postprandial hyperglycemia that accompany the disease may serve to improve the current state of diabetes care. Postprandial hyperglycemia is an important predictor of cardiovascular disease and needs to be accurately monitored to properly adjust dietary intake, oral medications, and insulin. 11, 12 Postprandial blood glucose levels are required to accurately adjust bolus insulin doses. 13 In addition, the postprandial glucose values tend to correlate better with A1C than fasting glucose, especially at values near the 7% range. 14 A recent study by Monnier et al. suggests that approximately 70% of the glucose that contributes to an A1C of 7% comes from the after-meal or postprandial glucose level. 15 Emerging therapies in diabetes care include the GLP-1 analogs and DPP-IV inhibitors, which each exert their effect on different portions of the same pathway in the mechanism of the disease. The GLP-1 analogs, or incretin mimetics, act by mimicking the action of GLP-1, an incretin hormone secreted in the L-cells of the intestinal mucosa that is released in response to food and binds to the pancreatic β-cell to regulate insulin secretion, slow gastric emptying, and inhibit gluconeogenesis. Agents from this category include pramlintide and exenatide, which are approved for use and are administered as a subcutaneous injection prior to major meals, and liraglutide, which is in development and not available for use. 16 While GLP-1 analogs affect the processes involved in diabetes by mimicking the action of GLP-1 and enhancing natural GLP-1derived processes, DPP-IV inhibitors act by preventing the action of DPP-IV, the key enzyme involved in the degradation of GLP-1. 17 This results in an increase in (1) an endogenous GLP-1 in response to a meal, (2) suppression of glucagon release, and (3) improvement in peripheral glucose utilization. The DPP-IV inhibitors enhance the long-term effects of GLP-1, such as increasing insulin synthesis and promoting the preservation, neogenesis, and restoration of β-cells (Figure 2). DPP-IV inhibitors are oral agents and include sitagliptin, which was recently approved for use in the United States, and vildagliptin and saxagliptin, which are in development. 18 ss

Conclusions: Best Practices in Diabetes Management
An effective program for the management of a diabetes patient population must be multidisciplinary, coordinating the efforts of a primary care provider, endocrinologist, diabetes educator, dietitian, pharmacist, and support staff. Components commonly incorporated in P4P initiatives, such as patient self-management education, provider contact, and the use of the ADA standards of care measures for screening and lab levels, are imperative to the success of the program. As an example, organizations such as Kaiser Permanente; Health Partners in Minneapolis, Minnesota; and the Family Medicine Center at the University of Oklahoma Health Science Center, Oklahoma City, have designed and implemented multidisciplinary diabetes management programs that have been met with success to date.
Diabetes management must focus on establishing and maintaining glycemic control, reducing the impact of comorbid conditions, and minimizing or preventing long-term complications of diabetes. Treatment should be multifaceted and include aggressive lifestyle modification as well as safe and efficacious pharmacologic agents. The HEDIS measure for comprehensive diabetes management serves as an excellent benchmark for ideal care. Whether part of a P4P program or not, achievement of the HEDIS measure for comprehensive diabetes management can favorably improve outcomes.

DISCLOSURES
This article is based on a presentation given by the author at a symposium, "Pay for Performance: Where' s the Return?" held October 4, 2006, at the Academy of Managed Care Pharmacy' s 2006 Educational Conference in Chicago, Illinois. The symposium was supported by an educational grant from Merck & Co., Inc. The author discloses that she has received an honorarium from Merck & Co., Inc. for participation in the symposium and this supplement. She discloses no potential bias or conflict of interest relating to this article. B usiness health costs are currently on the rise. These costs are being shifted to employers, and managed care stakeholders are left looking for a means to repair the disconnect between rapidly rising expenditures without correspondingly rapid improvements in care. 1 Pay for performance (P4P) initiatives are one such means in which clinical evidence is being used by managed care organizations (MCOs) to drive financial incentives and align physicians' and MCO goals, thereby improving delivery of care. This daunting task is ultimately achieved by 1 or more of the following: (1) reducing medical errors, clinical variation, and acute treatment episodes; (2) publishing quality health cost data; (3) linking provider bonuses to improved performance; and (4) implementing new information technology (IT), improving efficiency in care delivery.
P4P programs employ metrics to incentivize improved care. These metrics typically fall under the categories of clinical, patient satisfaction, efficiency, and IT. Several types of incentives may be offered, including bonuses, adjustable fee schedules, withholds, and nonfinancial incentives. While several challenges exist in creating an effective P4P initiative, such as data-related constraints, market consolidation, and difficulties in creating a collaborative relationship between payers and providers, successful programs have emerged early throughout the United States.
SelectHealth in Salt Lake City, Utah, is an example of one organization that has developed and implemented a P4P program, which has been met with initial success. SelectHealth' s program employs quality improvement measures as well as efficiency measures that, on the pharmacy side, are focused on encouraging generic and formulary drug utilization. The P4P program at SelectHealth has been successful in improving these pharmacy measures, with generic drug utilization improving from 42% to 56% and formulary drug utilization improving from 87% to 88% in family practice between the years 2002 and 2006.
Participation in the P4P movement has been on the rise over the past 4 years, with an increasing number of programs every year. While P4P programs have demonstrated significant improvements in health care delivery, these initiatives typically come at a similarly significant cost to MCOs. Ultimately, only improved outcomes will justify the long-term costs of P4P programs.

ss Need for P4P Initiatives
Per capita health care expenditure increases were relatively consistent until 1963, at which point a rapid rise in spending occurred. 2 The introduction of Medicare caused a paradigm shift. Prior to that time, third-party payment was made through a few large industrial corporations. Third-party payers were the vast minority.
Following the advent of Medicare, third-party payers became relatively commonplace, introducing the notion in health care that Pharmacy Management Approach: How Do We Align All the Incentives? Jeffrey D. Dunn, PharmD, MBA ABSTRACT BACKGROUND: While health care costs continue to rise and shift toward employers, a parallel improvement in health care quality has not been evident. As a means to repair this apparent disconnect, pay for performance (P4P) initiatives are being implemented across the country.
OBJECTIVE: To explore the need for P4P in the current state of health care delivery and review the design, components, and results of P4P programs.
SUMMARY: In P4P, clinical evidence is used by managed care organizations (MCOs) to drive financial incentives and align physicians and MCO goals, thereby improving delivery of care. At the center of all P4P programs are specific metrics, employed to measure the quality of care by which incentives are provided. These metrics fall into 4 main categories: clinical, patient satisfaction, efficiency, and technology. After metrics are employed and a provider is determined to be deserving of an incentive according to the P4P program in place, several different options exist and vary by plan in terms of incentive type. Primarily, these types of incentives include bonuses, adjustable fee schedules, and withholds. SelectHealth, a nonprofit health insurance company serving members in Utah and Idaho, implemented a primary care incentive program in 2002 for several different conditions and for pharmacy utilization that has been successful to date. those individuals consuming services were not directly responsible for paying for the services. This transformation had a significant impact on the demand for health products and services. Widespread third-party payment also changed the way physicians thought about prescribing care. Likewise, the growth of third-party payment changed the way that technology vendors thought about development; it became increasingly likely that firms could generate return on investment (ROI) for the development of innovative medical devices, drugs, etc. The expansion of health care financing and investment combined to cause an increase in health care costs that has continued to the present day.
However, while health care costs continue to rise, government payment increases are consistently falling short of professional and facility costs, thereby shifting these costs to employers. 3 In 1980, 28.3% of personal health expenditures were paid by private health insurance; by 2000, national health expenditures per capita had risen to approximately $4,500, 34.6% of which was paid by the private sector. 2 The result has meant an increasing proportion of employer profit is being allocated to employee health costs. Furthermore, a significant disparity currently exists between these rising health care expenditures without a parallel improvement in health care quality. 1 In 2005, the National Committee for Quality Assurance (NCQA) estimated that thousands of avoidable deaths and billions of dollars in avoidable costs were caused by unexplained variations in care in the previous year (see Table). 4 Although the quality of health care has increased over past levels, these data indicate that there is room for improvement.
Considering this disconnect between increasing expenditures and unsatisfactory care, managed care stakeholders need to find more effective ways to allocate funding and improve the overall quality of health care. P4P initiatives provide one such solution in that they are based on clinical evidence and use incentives to improve the delivery of health care.

ss Overview of P4P
The concept of P4P is based on the notion that MCOs and providers must think collectively when it comes to the delivery of care. In these collaborative efforts, incentives motivate providers to apply a best-practice approach to disease treatment, maintenance, and prevention. P4P initiatives seek to reduce errors, clinical variation, and acute treatment episodes; publish quality health cost data; link bonuses to improvement; improve efficiency in care delivery; and implement new IT. 5 P4P programs share several common features, all of which are driven by evidence-based medicine and the concept that provider incentives are often misaligned for delivering quality health care. P4P programs routinely measure clinical outcomes, compliance with clinical process standards, and cost management. 5 Mechanisms for continuous improvement are also commonly featured in MCO P4P initiative structures, along with specific incentives for the increased use of IT. 5 Often, the incorporation of a consumer satisfaction measure is included, as are mechanisms to shift market share based on quality performance. 5 At the center of all P4P programs are specific metrics employed to measure the incentives earned. These metrics fall into 4 main categories: clinical, patient satisfaction, efficiency, and technology. Clinical metrics are based on specific evidence-based measures in health care such as childhood immunizations, breast cancer screening, asthma control medications, and glycosylated hemoglobin (AIC) levels and screening for diabetes. 6 Patient satisfaction metrics assess the level of care patients report receiving, such as access to or communication with physicians. 6 Efficiency metrics look toward performance, such as generic substitution rates, utilization of services (episodes of care), and practice patterns with apparent financial implications. 6 Finally, technology metrics recognize the value of employing technology to improve patient registries, emergency medical services, lab systems, or the use of electronic medical claims submission. 6 Metrics in all 4 categories come from a variety of sources, both internal and external to the MCOs employing them. External sources of P4P metrics include widely recognized national health organizations such as NCQA (e.g., the Health Plan Employer Data and Information Set [HEDIS], Bridges to Excellence, etc.), National Quality Forum, American Medical Association (AMA) Consortium, or the Centers for Medicare & Medicaid Services (CMS). Internal sources of P4P metrics include those clinical and efficiency measures developed by the organizations themselves.
MCO plans offer several different types of incentives when a provider' s performance metric attains thresholds as specified in the MCO' s P4P program. Primarily, the incentives can be categorized as bonuses, adjustable fee schedules, and withholds. Bonuses are the most popular type of incentive and are usually paid out annually. Bonuses are also easy to administer since they can be stratified for different measures. 5 Adjustable fee schedules tend to correspond to specialist incentive programs and are therefore more difficult to administer. These incentives are also typically Quality Gaps Remain* implemented retroactively. Withholds are the least popular approach for granting incentives and are often used by plans that do not have a proactively budgeted incentive pool. 5 While these types of incentives are all financial in nature, nonfinancial incentives also exist and may be awarded in a P4P program. Nonfinancial incentives include the simplification of administrative tasks and public reporting or honor rolls. 6 ss Challenges and Promise in P4P The design and implementation of an effective P4P initiative must take into consideration several challenges that may hinder the success of the program. Considering the wealth of accurate information necessary in a P4P system, data-related constraints are of primary concern. Specifically, the timeliness of data, accuracy of data, availability of pharmacy and lab data, requirement to use chart data, exchange of data with other physicians, and small or limited amounts of data are all issues that must be addressed in comprehensive P4P programs. 6 IT support systems must be standardized, easy to access, provide real-time data, and produce robust and meaningful data. 7 Other challenges arising in P4P programs include the degree of market consolidation required and the collaborative relationship between payers and providers necessary for short-and long-term program success. 7 Furthermore, the structure of a P4P initiative must be designed to overcome provider resistance, skepticism, and legal challenges. 7 Considering these apparent challenges to the development and implementation of an effective P4P program, MCOs can employ several measures to increase the program' s chance of success. Collaboration of payers and providers in the P4P development stages allows for multidimensional performance initiatives with long-term benefits for each party. 7 In that manner, these collaborations foster a sound organizational infrastructure, cooperative culture, and enhanced professional resources vital to the long-term success of P4P programs. 7 Clinicians should be involved in program design and implementation, and the use of evidence-based guidelines is imperative. 7 Finally, rigorous follow-up processes should be established with a substantial commitment to IT support for monitoring and maintaining the program' s success. 7 Despite the apparent challenges in P4P, the movement is thriving in managed health care, with more programs arising each year (see Figure). Currently, one third of all health plans feature some form of P4P initiative, totaling 115 operational programs and encompassing 53 million patient lives. 8 A wide variety of P4P methodologies are in use although, to date, no universal approach to P4P has emerged. 7 Still, similarities exist among current P4P initiatives, with bonuses being the most commonly employed incentive by health maintenance organizations, and primary care physicians being the stakeholders most commonly receiving incentives. 6 ss SelectHealth Primary Care Incentive Program One example of a P4P initiative that has been successful thus far is that of SelectHealth, a nonprofit health insurance company serving members in Utah and Idaho and an integrated subsidiary of Intermountain Healthcare. Since June 1, 2002, SelectHealth has used quality improvement and efficiency measures to evaluate qualifying providers for the provision of P4P incentives.
Among the P4P quality measures implemented at SelectHealth in 2002 were those for asthma care; otitis media; childhood immunizations; depression care; diabetes care; and preventative care, which covers both breast cancer screening and smoking cessation.
The diabetes care quality measures include the number of patients with a diagnosis of diabetes who had an A1C test in the past twelve months and the number of patients with a diagnosis of diabetes who had an LDL-C test in the past 24 months. Performance prior to the P4P implementation for both of the measures was 76%. Subsequently, the 50% and 100% benchmarks for the 2 measures were set at 78% and 83%, respectively.
Pharmacy efficiency measures in place at SelectHealth include generic and formulary prescribing rates. A third measure, which changes yearly, was incorporated later in the program: in 2005, this measure evaluated antidepressant use and, in 2006, this measure evaluated statin use. Monitoring in the program is provided by pharmacy reports, which include the average cost per prescription, number of prescriptions per patient, and cost per  prescription for individual providers, provider peer group, and health plan as a whole. The reports specify drug types as nonformulary brand name, formulary brand name, and formulary generic. Cost for these 3 categories of drugs is also reported by individual drug.
Pharmacy support in the SelectHealth P4P program is provided by way of special reports listing patients using nonformulary medications and corresponding chart stickers. These reports are run and distributed by request, with a commitment to a 10-day turnaround time. SelectHealth pharmacists and/or provider relations representatives provide ongoing education to and tracking and monitoring of clinics in support of the P4P program. The Primary Care Guidance Committee (consisting of multidisciplinary health care professionals) is updated quarterly regarding the status of prescribing among providers.
Providers participating in the SelectHealth P4P program are granted specific benefits, which further provide incentives to physicians and aid in program success. For example, participating physicians are the first to receive information on continuing medical education opportunities. Likewise, these physicians receive pharmacy help-desk support and quarterly distributed pharmacy reports to help manage patients. Participating physicians are also the first to hear about formulary changes and receive a pharmacy packet with a pocket information card, poster, and other support material.
Physicians qualify for SelectHealth P4P participation and incentives on a group level, not by individual physician. For example, for a physician group that consists of 5 physicians, all 5 will receive the efficiency incentive (provided quality measures are met) if the aggregate data for their group meets program targets. Physicians receive incentives based on their groups' targeted generic and formulary prescribing rates and quality measures.
In 2006, the payout breakdown in internal medicine/family practice was 25% for the formulary target, 50% for the generic target, and 25% for the generic statin target. Measurement for this latter metric was only evaluated for the last 6 months of the year when 2 statins were available generically. The 2006 payout breakdown in pediatrics was 25% for the formulary target and 75% for the generic target. The statin measure was not included in pediatrics because it is not applicable. Ideas for a third measure to replace inapplicable measures in this group are being sought on an ongoing basis. Improvement in reaching targets for generic and formulary drug utilization has been observed over the lifetime of the SelectHealth P4P program (2002)(2003)(2004)(2005)(2006) in the family practice, internal medicine, and pediatric specialties, demonstrating the effectiveness of the initiative.

ss Conclusions
Continually rising health care expenditures and inadequate government payment increases for professional and facility services are shifting costs to employers. This, coupled with a lack of parallel improvement in care, demonstrates the need for a system in which funding can be more effectively used to ensure quality health care delivery. 1 P4P initiatives seek to improve the delivery of services in managed care by using evidence-based medicine to drive financial incentives. Clinical, efficiency, patient satisfaction, and IT metrics are commonly employed in these initiatives. Successfully implemented P4P programs are capable of enhancing performance on specific metrics through program design.
While P4P programs are becoming increasingly common in managed care, data on whether these initiatives improve outcomes and manage costs is still limited. 9 Although many current programs have demonstrated short-term success, long-term outcomes are still needed to justify the cost of P4P initiatives in improving the delivery of quality health care.

DISCLOSURES
This article is based on a presentation given by the author at a symposium, "Pay for Performance: Where' s the Return?" held October 4, 2006, at the Academy of Managed Care Pharmacy' s 2006 Educational Conference in Chicago, Illinois. The symposium was supported by an educational grant from Merck & Co., Inc. The author discloses that he has received an honorarium from Merck & Co., Inc. for participation in the symposium and this supplement. He discloses no potential bias or conflict of interest relating to this article.

PROGRAM EVALUATION
Pay for Performance: Where's the Return?
The Postgraduate Institute for Medicine (PIM) respects and appreciates your opinions. To assist us in evaluating the effectiveness of this activity and to make recommendations for future educational offerings, please take a few minutes to complete this evaluation form. You must complete this evaluation form online to receive acknowledgment of participation for this activity.
Please answer the following questions by selecting the appropriate number: 5 = Outstanding 4 = Good 3 = Satisfactory 2 = Fair 1 = Poor Extent to Which Program Activities Met the Identified Objectives Upon completion of this activity, participants should be better able to • explain the benefits and risks of pay for performance (P4P) programs 5 4 3 2 1 • cite the early results of P4P programs to future performance-based measurement development 5 4 3 2 1 • differentiate the effects of P4P results versus disease management activities 5 4 3 2 1 • describe how new therapies in diabetes can be aligned with P4P objectives to improve outcomes 5 4 3 2 1 • discuss how to align performance-based incentives with best practices using a clinical case-study approach 5 4 3 2 1 Overall Effectiveness of the Activity • was timely and will influence how I practice 5 4 3 2 1 • will assist me in improving patient care 5 4 3 2 1 • fulfilled my educational needs 5 4 3 2 1 • avoided commercial bias or influence 5 4 3 2 1 Impact of the Activity The information presented: (select all that apply) u Reinforced my current practice/treatment habits u Will improve my practice/patient outcomes u Provided new ideas or information I expect to use u Enhanced my current knowledge base Will the information presented cause you to make any changes in your practice? u Yes u No If yes, please describe any change(s) you plan to make in your practice as a result of this conference: Future Activities • Do you feel future activities on this subject matter are necessary and/or important to your practice? Ì Yes Ì No • Please list any other topics that would be of interest to you for future educational activities:

Follow-up
As part of our ongoing continuous quality-improvement effort, we conduct postactivity follow-up surveys to assess the impact of our educational interventions on professional practice. Please indicate your willingness to participate in such a survey: u Yes, I would be interested in participating in a follow-up survey u No, I'm not interested in participating in a follow-up survey Additional comments about this activity: For Physicians Only I certify my actual time spent to complete this educational activity to be: ____________ u I participated in the entire activity and claim 1.5 AMA credits. u I participated in only part of the activity and claim _____ credits.