Payment for Quality in End-Stage Renal Disease

The Centers for Medicare and Medicaid Services (CMS) End-Stage Renal Disease (ESRD) Program has served as a model for health care policy innovation because of several unique features: the program provides dialysis and kidney transplantation services where the alternative to renal replacement therapy is death; the program has a circumscribed, easily tracked population of patients; and the cost associated with the program has afforded administrators and providers opportunity for innovation and improvement in care delivery over the past three decades (1). Additionally, the wealth of detailed information available from unique databases such as the United States Renal Data System that have been designed to follow the ESRD population have facilitated assessment of the effect of evidence-based guidelines that have shaped health care delivery and policy. The recent publication of the Final Physician Fee Schedule Rule by CMS, the release of the Medicare Payment Advisory Commission (MedPAC) report on payment for out-patient dialysis, and the Medicare Modernization Act (MMA) of 2003 have focused attention on the goal of linking payments to quality care in the ESRD setting. The leadership of the American Society of Nephrology and the National Kidney Foundation recently convened a working group of experts to examine whether the ESRD system of payment can be redesigned to encourage quality-based care delivery. The deliberations of this group (Appendix) helped formulate this article. Quality Improvement Efforts and Patient Outcomes in the ESRD Program The ESRD Program has long been associated with quality assurance and continuous quality improvement efforts that have taken place at the individual dialysis facility and provider level (2). Notably, the Forum of ESRD Networks are not-for-profit contractors with CMS that are charged with oversight of outpatient dialysis and transplantation facilities, whose central activity is the development and use of quality measures and quality improvement processes. The central activity of the 18 networks is the development and use of quality measures and quality improvement processes. In 1992, Medicare introduced the Health Care Quality Improvement Program (HCQIP), with the goal of a systems focus on consistent improvements in patient outcomes (3). Included in the ESRD HCQIP was the National/Network ESRD Core Indicators Project, CMS’s first nationwide population-based study designed to assess and identify opportunities to improve the care of ESRD patients. The ESRD Core Indicators Project has subsequently been merged into CMS’s ESRD Clinical Performance Measures Project on the basis of the National Kidney Foundation’s Dialysis Outcome Quality Initiative (DOQI) Clinical Practice Guidelines. Sixteen ESRD Clinical Performance Measures (CPM) (five for hemodialysis [HD] adequacy, three for peritoneal dialysis [PD] adequacy, four for anemia management, and four for vascular access) were developed. The ESRD CPM are collected annually on a national random sample of adult hemodialysis patients and peritoneal dialysis patients. These efforts have led to consistent and widespread improvement in ESRD care. For example, in each of the years from 1993 to 2003, measures of dialysis adequacy as well as the treatment of anemia improved nationwide (Figure 1). The consistent improvement in these clinical performance measures serve as a model of quality improvement for other aspects of the health care industry. CMS has also sponsored a collaboration entitled “The National Vascular Access Improvement Initiative” with the Institute for Health Care Improvement and the ESRD Networks to improve the quality of vascular access for hemodialysis patients. The Forum of ESRD Networks, Renal Physician’s Association, and the National Patient Safety Foundation have completed the initial phases of an ESRD Patient Safety Initiative, designed to identify action agendas to improve patient safety in dialysis care. Thus, there have been major efforts and several tangible successes in improving overall quality in the ESRD program. However, there remains wide variation in meeting these metrics, and there have also been only minimal improvements over a decade in risk-adjusted mortality for dialysis patients (4). Thus, there are opportunities for continued quality improvement.Figure 1.: (A) Percentage of adult in-center hemodialysis patients with mean delivered calculated, single-session single pool (sp)Kt/V ≥1.2. (B) Percentage of adult in-center hemodialysis patients with mean hemoglobin ≥11 g/dl. Adapted from: Centers for Medicare & Medicaid Services. 2003 Annual Report, End-Stage Renal Disease Clinical Performance Measures Project. Department of Health and Human Services, Centers for Medicare & Medicaid Services, Center for Beneficiary Choices, Baltimore, Maryland, December 2003.Defining and Measuring Quality Care The Institute of Medicine defines health care quality as “the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge” (5). Donabedian (6) first proposed a framework of assessing quality of medical care consisting of structure, process, and outcome. In this framework, structure consists of the stable elements that form the basis of the health care system; process consists of the technical and interpersonal components of what is done within the structure; and outcome consists of what happens to the health of the patient and the end results of health care practices or interventions. In Donabedian’s framework, structure influences process and both structure and process influence outcomes (Figure 2). Providing financial incentives for the desired outcomes themselves would seem to be the simplest and most straightforward strategy for implementation; however, there is often limited understanding of the many determinants of desired patient outcomes and attaching reimbursement to a global measure may not direct providers to the specific activities that can achieve a desirable outcome. By contrast, there are often more extensive data as to specific performance measures that are associated with improved patient outcomes including in ESRD patients. It is important to recognize, however, that improving structure/process components that lead to improved intermediate outcomes might not necessarily improve overall outcomes. As an example, in a cross-sectional study, an increased frequency of physician visits to dialysis patients improved compliance with attendance at dialysis sessions, the latter being associated with reduced mortality in other analyses. Nevertheless, increased physician visits was not itself associated with reduced mortality (7).Figure 2.: Framework for assessing the quality of medical care. Reprinted with permission from Donabedian A: Evaluating the quality of medical care. Milbank Mem Fund Q 44: 166–206, 1966.Current Payment System to Physicians for Dialysis Services In the November 7, 2003, Final Physician Fee Schedule Rule, CMS announced its final decision on the new Monthly Capitation Payment (MCP, the per patient, per month payment that Medicare provides to the nephrologist for ESRD services) that would be effective January 1, 2004. CMS has implemented a system that uses temporary “G” codes that establish different payments for the services represented by the MCP on the basis of the number of nephrologist-dialysis patient interactions in the month. Four nephrologist-dialysis patient interactions would receive an increased payment, but one, two, or three visits/mo payments for the MCP services would be reduced. On the basis of the published payment in the final rule, the approximate adult MCP payment levels are: $287 for four visits or more; $239 for two to three visits; and $191 for one visit. In the final rule, the degree of payment variability has been compressed compared with the August 15, 2003, notice of proposed rulemaking, but continues to adhere to the graduated payment system. For home dialysis, the MCP will be reimbursed at $237 for a full month of home dialysis. Two sets of home dialysis codes have been established, one for the full month and one for less than a full month. Current Payment System to Facilities for Dialysis Services Medicare pays a prospective payment (the composite rate) for each dialysis treatment provided in dialysis facilities (in-center) or in patients’ homes. The average composite rate was about $130 per dialysis treatment and this accounted for 59% of total Medicare payments to facilities in 2001. In addition, facilities receive separate payments for services such as injectable drugs, and these payments averaged about $80 per dialysis treatment in 2001 and constituted 41% of total Medicare payments to facilities. Medicare spending for outpatient dialysis services increased 9% between 1996 to 2001, largely driven by growth in the ESRD population and increased use of new technologies and drugs. The MMA of 2003 recommends the following important changes in the payment for ESRD: (1) a case mix–adjusted payment system for certain services; (2) a report to the Congress on the design and features of a bundled prospective payment for dialysis services; (3) the conduct of a demonstration study of a bundled payment system; (4) other changes including an update of the composite rate by 1.6 percent in 2005 and restoration of the exemption to the composite rate for pediatric facilities, and (5) changes in payment for separately billable drugs. Case Mix–Adjusted Payments Beginning on January 1, 2005, the Secretary of Health and Human Services is required to enact a basic case mix–adjusted payment system for dialysis services. The current composite rate is adjusted for differences in labor costs by means of two dated hospital wage indices. MedPAC has recommended that the Secretary develop a wage index based on market wage rates for occupations typically used in furnishing dialysis. The key is that the Medicare Modernization Act (MMA) requires that the case mix–adjusted payment system result in the same aggregate amount of expenditures for such services as would have been made in 2005, 2006, and 2007 if payments were not case mix adjusted. The new composite rate payment will include the “spread” between the current average wholesale prices and the acquisition cost of injectable drugs and biologicals, while the acquisition cost of these injectable drugs and biologicals will be paid outside the composite rate. Bundled Payment System The Secretary is required to submit to the Congress by October 1, 2005, a report on broadening the outpatient dialysis payment system to include injectable drugs, laboratory tests, and other items currently excluded from it. Demonstration Project Beginning on January 1, 2006, the Secretary is required to conduct a 3-yr demonstration to test a broader payment bundle that includes injectable drugs and clinical laboratory tests that are currently excluded from it. Payment for Quality in Health Care Efforts have begun to redesign the system of reimbursement to provide financial incentives for improved patient/population outcomes (8,9). Physicians are generally familiar with financial incentives designed to achieve “quality” health care and are more likely to respond to such incentives than to financial incentives designed to control the use of services (9). Designing a reimbursement system that provides financial incentives for desired health outcomes requires consensus as to what constitutes desirable patient outcomes, and what the components of health care delivery are that lead to these desirable outcomes (so-called quality health care). Achieving this consensus for a given disease or syndrome requires long-term, extensive research as to the characteristics of the population with the disease and longitudinal outcome data. The lack of this consensus has limited the use of financial incentives for quality on a national level in the United States. The Institute of Medicine has recently called on government payers to increase payments to health care providers for the delivery of high-quality care (5). Several private payers have already begun a transition to financial incentives for improved adherence to standards of care, improved patient outcomes, and patient satisfaction (9). This is a important departure from incentives in an earlier era that were primarily targeted at controlling utilization. The National Health System in the United Kingdom has recently adopted a wide-ranging initiative to increase payment for performance (10,11). However, in the United States, there has not yet been an attempt on a national level for government payers to increase payments to health care providers on the basis of a measurement of higher-quality care. Thus, a plan to institute payment for quality in the ESRD program would be a unique and groundbreaking effort. Feasibility of Payment for Quality in Dialysis Care Currently, dialysis conditions of coverage are mainly structural requirements designed to ensure the capacity of facilities to safely furnish quality health care. Medicare can exclude dialysis facilities not meeting its conditions of coverage, although the program has rarely done so. The environment is opportune for the introduction of quality incentives for outpatient dialysis reimbursement for several reasons: Evidence-Based Clinical Practice Guidelines Can Be Used to Develop Payment Incentives. The publication of the NKF K/DOQI guidelines for hemodialysis, peritoneal dialysis, vascular access, and anemia in 1997 set the stage for quality metrics. These guidelines have been updated and the scope widely expanded to include nutrition, bone and mineral metabolism, lipid disorders, BP, and cardiovascular disease. Although all clinical practice guidelines are not necessarily suitable to be converted to performance measures, the existing guidelines can be used as the initial template for determining clinical importance, ability to be measured, and ability to be acted upon. Current Data Collection Vehicles Can Allow Timely Collection and Analysis of Information on Processes and Outcomes of Care. CMS already has in place a mechanism to collect dialysis adequacy information from the claims submitted by providers. CMS recently implemented the Standardized Information Management System, a national information infrastructure that electronically links all 18 ESRD Networks with the agency. A direct linkage between dialysis units and the Networks is also in a testing phase. Although the compatibility of multiple information systems may require further investigation, the potential exists for necessary linked data to be available. Case Mix Adjusters Are Being Developed. An important safeguard in developing and implementing financial incentives is to ensure that providers do not “cherry-pick” patients (refuse to care for patients who are sicker or who require more complex care). Current risk adjusters only explain 20% to 25% of the variability, but the new Medicare + Choice adjusters of about 70 diseases may be more helpful, although this would need to be prospectively verified. A Culture of Payment for Quality Is Emerging in Dialysis Facilities. At least two national dialysis providers link Medical Director’s payment to participation in multidisciplinary quality improvement initiatives as well as clinical outcomes in the dialysis facility. One major dialysis provider sets companywide quality targets for a given year and divides its portfolio of facilities into performance quartiles at the end of the year. Medical Directors of facilities in the top three quartiles receive a 15%, 10%, or 5% bonus on their Medical Director’s payments, respectively, whereas the lowest quartile receives remedial assistance. Another major dialysis provider does not provide but has a 25% that is if quality are for Quality in ESRD Care of Incentives. dialysis facilities and be to the same set of performance because delivery of quality care to ESRD patients requires a and continuous collaboration between these two This be because provider and facility payments are both Medicare Nevertheless, quality may be more by one than the A case in is dialysis units would have control of the of but can this if in the care of the patient incentives to improve care may to financial differences that can currently between and for example, with vascular access or with the use of peritoneal dialysis for Quality Quality or Quality The as to whether incentives be based on improving care from current national or further MedPAC that a and will providers on the basis of providers at both of the quality to achieve financial incentives payments only to improvements from current levels providers who achieve improvement but at the of the quality In contrast, a single encourage all providers to However, the level of a might providers at the end from to and might not high-quality providers to improve outcomes. the of risk both the risk of providers their patients. would have an to patients with outcome if the payment were linked to In contrast, providers would have an to patients with outcome if payment were linked to and of Incentives. MedPAC that linking quality to payment not require by Medicare However, most dialysis providers would that a of dialysis requires including more and most would for example, increased will be required to and quality would be to with a pool for to providers quality targets because this the effect on providers who are not to quality and improvements in outside the program. A quality pool would about for dialysis facilities payments for dialysis and injectable drugs in 2001 was and for payments were in and be used for this As and providers more to being for the pool for quality increase as the number of measures used to measure quality of care. if A are as a result of improved quality of care in be to and facilities to further the pool and the quality to levels of Measures of process or outcome measures are the most and measures of quality is process and outcome measures to quality improvement efforts while that such efforts are patients quality measures be easily and to As the NKF K/DOQI clinical practice guidelines are a for available in adequacy of dialysis, vascular access, nutrition, bone and mineral metabolism, and such as patient and mortality and rates can also be However, specific for quality payments be with the of the in in Care. also to improving dialysis outcomes Outcomes are for patients who do not with their treatment Patient adherence may also be by and to be case mix adjusted. patient be in clinical As an example, performance be patients a compared with patients not being a test in for Quality in ESRD The of The of the basis for current clinical practice guidelines in the care of dialysis patients is largely from and not from clinical are a of clinical in the dialysis population that process of care is associated with an improved outcome Notably, the the most clinical to in the hemodialysis the for the of dialysis and dialysis on outcomes, data that these process would be to patient outcomes The lack of improved outcomes from clinical also the need for further clinical in the dialysis The Physician of A individual for quality ESRD care, is the that an individual nephrologist often care of than dialysis patients. At this is to the case mix with to individual physician In an study assessing the of individual physician in the care of patients with has been that a be to each to the performance measure to the level of The same to dialysis with a of may be to patients to ensure a by to care for sicker less and who have not to interventions. of Quality of The improvement that has already in anemia and dialysis adequacy that improvements in these clinical performance measures might be a (Figure 1). there can be important differences in the percent of patients a goal on the basis of the of as the of the population for a given to increase are averaged over a This is important if variability is where over of may outcomes. An of this in the dialysis population is the where a than a of is more than a the of patients whose is within with the A for a quality payment program is a for and on performance to and providers As by MedPAC and the of claims and performance CMS to has not been to information and report in a CMS would need to improve its performance in this to the Case Mix As a rule, differences in physician practice generally for less than of an outcome data are adjusted for case mix variation In the dialysis the of a number of and data of patient outcomes, and for the of patient outcomes data also that case mix may for of the differences in outcomes The A is the of patients who have been on dialysis for less than to dialysis care and the of kidney which are often of dialysis Dialysis units and for ESRD patients not be for care of dialysis. of Quality financial incentives at a number of individual of clinical quality is to improvements in the quality of care measures and over the of performance that are to financial payments will be more However, many have that there are with this This that research will in the development of a system in this and the need for information in quality The will quality-based dialysis facilities are a because are in number and have and pediatric patients to have multiple medical and many more a in payment for facilities in not meeting quality targets might lead to patients with less desirable who their and units also The Medicare ESRD program reimbursement for dialysis services to patients a and population of patients that has been with to and In addition, the United States Renal Data System has collected an extensive of data on this population the that continues to be used to ESRD patient outcomes, the determinants of these outcomes, and influence the outcomes. the ESRD population an population for which to design a reimbursement system that provides incentives for quality health care. A system for for quality dialysis care is a desirable and This system use measures of quality that can be measured, incentives between and dialysis providers to be to provide increased reimbursement for increased and be by a continuous of information to and providers to be Although many for and performance are and require and is to a quality-based program for dialysis reimbursement at this A collaboration between CMS, and dialysis providers is likely to improve the care of Medicare dialysis care, and may serve as an effective model for other components of Medicare and Dialysis Advisory Project Dialysis Advisory Dialysis Advisory Clinical Practice Advisory Clinical Practice Advisory Dialysis Advisory Care Medical Medical Care of Dialysis Advisory of The and of this the of the and not necessarily the of each working group or the that are