By Christopher Guadagnino, Ph.D.
A practice that violates the federal anti-kickback statute, the Stark physician self-referral law and the civil monetary penalties provision of the Social Security Act is gaining legitimacy as a catalyst for physician-hospital collaboration and clinical quality improvement. As the federal government expands its menu of pay-for-performance quality and efficiency experiments to address health care cost inflation and avert a Medicare Trust Fund crisis, it has recently added two demonstration projects that explore a business model known as gainsharing.
A consortium of 12 New Jersey hospitals, as well as some Pa. hospitals and health systems, have applied to participate in this new opportunity and hope to prove that the gainsharing concept can play an important role in health care delivery redesign.
While government and commercial insurers have launched a variety of programs that financially reward physicians and hospitals for providing more efficient and higher quality care, hospitals are still prevented by law from doing so under gainsharing arrangements, in which hospitals provide physicians monetary incentives to modify their practice patterns in ways that save the hospital money, paid for by a share of the hospital’s savings.
Analysts point out that considerable waste in the health care delivery system comes from a lack of economic integration between hospital and physician reimbursement. Under the prospective DRG reimbursement structure, hospitals are at risk for the costs of care they provide to patients, while physicians are typically reimbursed separately for their professional services and often do not have the incentive to be concerned about hospital costs. Physicians’ choice of procedures, supplies and devices, particularly in specialties such as cardiac and spine surgery, drives a large chunk of inpatient care costs and – according to gainsharing’s rationale – if physicians are given the incentive to learn about those costs and to be mindful of efficiency when making their clinical choices, quality care can be delivered at significantly lower cost.
After reviewing a number of gainsharing requests from hospitals, the Department of Health and Human Services’ Office of Inspector General (OIG) concluded in a 1999 advisory, however, that gainsharing arrangements pose a high risk of abuse and warned that the law prohibits hospitals from paying physicians to reduce or limit clinical services to Medicare or Medicaid beneficiaries – for fear that such incentives may limit physicians’ ability to exercise independent professional judgment in the best interest of patients, that gainsharing incentives could induce physicians to refer patients to the hospital with which they have the most lucrative arrangement, and that the arrangement might also be used to disguise kickbacks.
In recent years, as the imperative of value-driven medicine has become mainstream in health care, gainsharing appears to be moving beyond these legal and regulatory stigmas and, although it is still proscribed by law, it is poised to expand well beyond the narrowly defined contexts in which it has been practiced thus far.
Pa. and New Jersey hospital proposals for Medicare’s gainsharing demonstration projects reveal an approach to gainsharing that far transcends the “first generation” model in which hospitals share financial savings with physicians who cut out waste largely attributable to variation in the ways they select and use medical devices and supplies during a narrow range of surgical procedures. Under the new proposals, hospitals seek to share financial savings with any physician caring for Medicare patients who reduces costs – by reducing hospital readmissions, length of stay and avoidable emergency room utilization – and who improves quality by adhering to best practice guidelines.
Gainsharing’s Track Record
Current law is a reflection of competing policy issues surrounding gainsharing: on the one hand, the disinterested physician is in the best position to determine the best treatment for patients, and economic incentives can lead to bad outcomes; on the other, there is ubiquitous waste and inefficiency built into a system that has costs borne by hospitals and utilization decisions made by physicians who are insulated from those costs, says Karl Thallner, Jr., Esq., a partner with ReedSmith LLP who heads the health care law practice of the firm’s Philadelphia office.
From a policy perspective, there is much to be said for aligning incentives to improve efficiencies, particularly as hospitals are battling significant marketing efforts of medical device manufacturers – which Thallner says are smart enough to know they need to market the latest technologies to physicians. The right balance must be found between the competing policy objectives, and recent developments represent a “toe in the water,” says Thallner.
Following the chill it placed on gainsharing with its 1999 advisory, OIG issued advisories in 2001 and in 2005 that sparked speculation about a renaissance of the concept. In those advisories OIG reiterated that gainsharing is illegal, but agreed not to prosecute a handful of hospitals that proposed narrowly focused gainsharing programs with multiple safeguards built in to prevent erosion of care and improper influence on physicians’ clinical decision-making and referral patterns. The agency warned, however, that its advisories were specific to those hospitals and that other institutions remained vulnerable to prosecution if they engaged in gainsharing activities.
Although a well-established body of medical literature points to unexplained variation in physician practice patterns – and costs – that lead to comparable outcomes, hospitals remain prohibited by law from providing economic incentives to physicians for incorporating cost efficiency into their practice, or to adopt best practice protocols to reduce variation and attendant waste, says Joane Goodroe, president of Georgia-based Goodroe Healthcare Solutions, LLC.
Goodroe says the concept of gainsharing gives physicians with expertise the opportunity and incentive to innovate ways to reengineer patient care processes to decrease costs while maintaining, and even increasing, quality. Unlike the typical model of managed care, it is the physician who drives the decisions related to cost reduction, she notes.
Goodroe has developed a methodology that won OIG clearance for seven different gainsharing relationships between hospitals and their physicians, which focus on cardiovascular surgery and cardiology – including interventional cardiology, electrophysiology and peripheral procedures – while Goodroe says the methodology should be applicable to other procedures such as orthopedics and spinal procedures.
Goodroe’s methodology focuses on processes of care around the use of “physician preference items” – use by physicians of specific products and brands of products which becomes inculcated during training as well as by the support and service physicians receive from medical device and product vendors. The idea, she says, is to reduce variation in practice patterns by having physicians collaborate with hospitals to look for opportunities to improve care and eliminate processes that do not bring value.
Physician preference items have never been priced by true market pressures, Goodroe says, because the physician has been the vendor’s key customer while rarely if ever having been involved with price – which is handled by the hospital. Once physicians become knowledgeable and concerned about pricing, they can develop strategies to contain costs and reengineer the care of patients, while the medical profession can incorporate clinical studies with detailed cost analysis, Goodroe maintains.
These activities involve ongoing data analysis among physicians, as well as physician involvement in negotiations with product vendors, which Goodroe believes is a considerable amount of work that physicians would not do if they were not compensated for their efforts.
Goodroe’s OIG-approved gainsharing model encompasses four categories of process redesign and cost reduction:
· Product substitution, including items such as wrist splits, drapes, arm boards, sutures and catheters. The full array of product choice is available and physicians are free to make all product decisions, while less expensive products that achieve the same quality outcomes are identified.
· Product standardization, including key devices such as heart valves, stents, pacemakers and internal defibrillators. The names, clinical data and price of each product are specifically examined; physicians decide what products need to be available for performing procedures; and they make case-by-case determinations whether a standardized device is clinically appropriate.
· Perform/use as needed, including clinically significant items such as pharmaceuticals, vascular closure devices, gelfoam, and vancomycin paste. There is no economic incentive for physicians not to use such items, but physicians determine how and when to use them – for example, certain pharmaceuticals that control bleeding may not be necessary if there is not a high risk for bleeding during a procedure.
· Open as needed, including certain types of surgical kits and the cell-saver unit – items that are made immediately available in procedure rooms but are not opened prior to the procedure.
Goodroe says her gainsharing model has built in important safeguards to pass OIG muster, including:
· Delineating specific, identifiable cost-lowering activities before patient care changes are made.
· Disclosing the program to patients.
· Preserving physicians’ access to the items they used before and allowing them to make the final product and process decisions.
· Establishing a floor for product substitution, based on objective historical and clinical measures, beyond which no savings would accrue to physicians.
· Establishing an aggregate monetary cap on savings that physicians are allowed to share.
· Distributing savings to physicians on a per capita basis to mitigate any incentive for an individual surgeon to generate disproportionate cost savings.
· Preventing physicians from being paid for the same savings they achieved in the past by making that savings level the new baseline for future gainsharing involving those activities.
· Maintaining a physician practice’s historical volume and patient severity mix to prevent physicians from profiting by cherry-picking patients or changing referral patterns.
· Maintaining the acuity level of a hospital’s patient mix to prevent patient steering.
· Appointing a third-party administrator to ensure program and safeguard compliance, and to monitor effects on patient care using national quality benchmarks.
Hospitals participating in the model have seen procedure costs reduced by 10 percent, and have achieved annual savings ranging from $600,000 to $4 million in a year, says Goodroe.
Beyond cost reduction, gainsharing drives quality improvement, says Goodroe: “If every physician has a different approach to performing the same procedure, that means they’re not sharing their knowledge.” Reducing wasteful variation in the way surgeons perform a given procedure can also reduce nursing staff burnout from having to learn different procedures and processes for different surgeons, says Goodroe. Other improvements in care quality she has seen include more efficient operating room use, patients requiring shorter time under anesthesia, and needing less blood transfusion time outside the operating room.
PinnacleHealth System in Harrisburg, Pa., one of the entities that secured OIG sanction for a gainsharing agreement using the Goodroe model, has achieved clinical care improvement and a level of collaboration with physicians that transcend the monetary savings accrued from the arrangement, according to President and CEO Roger Longenderfer, M.D.
In the three years Pinnacle has run the gainsharing program – involving Harrisburg Hospital and three cardiology group practices representing about 50 physicians – the organization saved about $2.5 million by focusing on cost and best practice utilization of cardiac devices such as bare metal and drug eluting stents, pacemakers and implantable defibrillators. The cardiology groups decide how to distribute their shared annual savings to their physicians, while an individual physician might receive between $10,000 to $30,000 in incentive payments, depending on their level of involvement with the activities, says Longenderfer.
Pinnacle has seen its strongest economic return in the first couple of years, while some of its subsequent cost savings goals have not been met, after the “low hanging fruit” – such as better contract pricing – had been picked, says Longenderfer. “Your baseline keeps changing, and you must look for new and innovative changes. But, technologies keep changing too, and it’s an ongoing and constantly challenging process to control the inflation of costs,” says Longenderfer.
Pinnacle reapplies to OIG each year to continue the program. “I have always felt that gainsharing was a little about economics and a lot about physician-hospital quality improvement – sharing data. That has always been the underlying motive for us,” says Longenderfer, noting that cardiac cath lab complication rates continue to decrease since the program began.
The program gave the hospital and physicians a forum to discuss clinical variation and strive for best practice consensus – a dialogue that has fostered two clinical quality improvement initiatives, Longenderfer adds: one focused on achieving consistent diabetic measurement and management across practices, and the other addressing a more consistent use of anticoagulent. While such discussions can and do occur among physicians anyway, Longenderfer has seen them “reach a plateau,” while the shared responsibility, knowledge of costs, and financial incentives brought by the gainsharing arrangement provided “a catalyst to kick us into a new gear,” he says.
Gainsharing Demonstration Projects Emerge
The gainsharing concept has gained important allies in the past few years, and appears poised to expand well beyond the narrowly defined contexts in which it has been practiced thus far.
Shortly after OIG issued its 2005 gainsharing advisories, the Medicare Payment Advisory Commission (MedPAC) recommended that Congress grant the Department of Health and Human Services (HHS) the authority to allow gainsharing arrangements between physicians and hospitals, as long as such arrangements maintained quality of care and did not affect physician referrals. In its report to Congress this March, MedPAC reiterated that call, saying that gainsharing – which it described as shared accountability arrangements in which hospitals and physicians agree to share savings from reengineering clinical care in the hospital – has the potential to foster better communication and coordination among hospitals and physicians to improve efficiency, eliminate redundancy and reduce costs. Such efforts, said MedPAC, could include negotiating with manufacturers to obtain greater discounts, scheduling operating rooms more efficiently and complying with clinical protocols that improve efficiency and quality.
Gainsharing has won Congressional endorsement as well. The Deficit Reduction Act of 2005 requires HHS to establish a three-year DRA 5007 Medicare Hospital Gainsharing Demonstration to test and evaluate gainsharing methodologies and arrangements between hospitals and physicians. Authorization for a second gainsharing project – the Physician Hospital Collaboration Demonstration (PHCD) – stemmed from Section 646 of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003, which spawned a number of Medicare quality demonstration programs.
While the two gainsharing demonstration projects share much of the same conditions of participation – including many of the safeguards gleaned from the Goodroe model and subsequent OIG advisories – the DRA 5007 project is open to six hospital sites nationwide and will focus on short-term impacts of gainsharing programs for inpatient services, while the PHCD is open to 72 hospitals nationwide, involves tracking patients long-term for follow-up in quality improvements and cost reductions, and will include participation by consortia of health care groups – up to a maximum of 12 hospitals per consortium. The application deadlines have passed and the Centers for Medicare & Medicaid Services (CMS) says it plans to announce the awards in the coming months, according to a CMS spokesperson.
In contrast to OIG-sanctioned gainsharing projects which focus on a narrow set of surgical procedures, CMS left its gainsharing projects open to any patient care involving Medicare beneficiaries, and program proposals demonstrate the manner in which the gainsharing apparatus is capable of moving beyond efficient medical device usage, seeking to derive savings from improved care, engineered entirely by physician decision-making. Incentive payments to individual physicians, as per CMS guidelines, are limited to 25 percent of what Medicare Part B would normally pay physicians for the services in question.
PinnacleHealth System, northeastern Pa.’s Geisinger Health System, and a consortium of 12 New Jersey hospitals have all applied for a CMS gainsharing demo project.
Pinnacle’s proposal applies to general internists and seeks to find ways to provide better case management of patients with diabetes, COPD and pneumonia – both in physician offices and in the hospital, says Longenderfer. Pinnacle hopes a collaboration with physicians will lead to fewer hospital readmissions, shorter lengths of stay, and better selection of antibiotics and other drugs, he adds.
Geisinger Health System has applied for approval of a Medicare gainsharing project that will expand a successful cardiac surgery quality improvement program to eleven high-volume inpatient surgical and interventional procedures, including three types of angioplasty, carotid artery stenting, heart valve replacement, heart bypass, endarterectomy, hysterectomy, spinal fusion, hip replacement and leg amputation. Payment to physicians will be based on their adherence to nationally-recognized best practice standards for each procedure, including objective justification of each case to be performed as well as processes that have been proven to reduce adverse outcomes, according to Mark Selna, M.D., Geisinger Health System’s Associate Chief Medical Officer, Division of Clinical Effectiveness.
Savings from the program will be generated from preventing readmissions, reducing the number of patients hitting outlier payment thresholds and – as complications rates are reduced – decreasing utilization of ER services, diagnostic testing and outpatient surgery/consultations, says Selna. Any adult patient who has any of the specified high-volume surgical procedures performed at one of Geisinger’s three medical centers will be included in the quality improvement activities of the project, while the project will focus on tracking guideline compliance and readmission data for Medicare beneficiaries, notes Selna.
The proportion of these Medicare patients readmitted within 30 days (excluding trauma cases and defibrillator implant cases) as compared to the Pa. state average proportion of such readmissions will determine the expenditures saved by the Medicare Trust Fund. A semi-annual reconciliation process will determine the amount of Medicare’s savings to be reimbursed to Geisinger in the form of a Part A risk-based supplemental payment, while Geisinger would owe money to Medicare if the readmission rate were to increase above the state average proportion. Selna notes that about ten percent of the full episode of care cost for these procedures is related to inpatient readmissions – offering ample opportunity to accrue and share savings with caregivers.
Any physician rendering care to Medicare patients undergoing the defined procedures at a Geisinger hospital will be eligible to participate and, to assure that all savings are achieved largely as a result of higher quality care, a physician can receive bonus payment only if there is objective data showing that his or her patients received exemplary care, namely, meeting all of the program’s best practice guidelines defined for that procedure, says Selna. The model also rewards collaboration of all care team members – including surgeons, anesthesiologists, nurse anesthetists, and primary care physicians (for their preadmission and post-discharge activities), he adds. The pool of incentive payment dollars will be distributed to care providers at the practice or department level, where individual allocation decisions will be made.
Program design will be modeled after Geisinger’s ProvenCare Heart Congestive Heart Failure (CHF) Readmission Prevention Program for patients who undergo coronary artery bypass surgery, which applies over 45 separate best practices known to give each patient the maximum opportunity for a successful outcome – including identifying patients with CHF at the time of admission and managing their care using a combination of nurse-executed protocols. Selna notes that the program has improved mortality, dramatically reduced complications and reduced costs and readmissions over the last year at Geisinger Medical Center and Geisinger Wyoming Valley Medical Center. Geisinger’s gainsharing demonstration project will do the same for infection, hypercoagulation and cardiac ischemia, says Selna.
While Geisinger has demonstrated the success of such a program involving its employed cardiac surgeons without gainsharing incentives, Selna believes the introduction of quality-based payment incentives will serve to focus all clinicians on the objective, particularly lower-paid ancillary staff – such as nurses and physician assistants – who do much of the routine standardization work. If the health system continues to outperform its competitors, says Selna, it can grow its peripheral market share and attract higher volume for several surgical procedures, which would benefit physicians and hospital alike.
Essential to the infrastructure of the project is an electronic health record, which enables pre-admission risk assessment and documentation flowsheet, generating a customized care plan, according to Selna. During each hospitalization, each patient is monitored for each process component, which for surgery cases (unless clinically contraindicated) includes the surgical site infection (SSI) prevention bundle processes (appropriate removal of hair, appropriate antibiotic used, antibiotic given within 60 minutes prior to incision, antibiotic discontinued with 24 hours after surgery – 48 hrs for cardiac surgery, normothermia maintained, supplemental oxygen given, post-op glucose levels followed); the perioperative cardioprotection processes (beta-blocker maintained during perioperative period, beta-blocker initiated for vascular surgery patients); and medication reconciliation upon admission, during each intra-hospital transfer, and on discharge. Following discharge, the care team case management staff will confirm that the care plan is working, that all therapies are being pursued and that there are no signs of any complications.
Activities related to every patient will be measured and serve as the basis of feedback for continuous process improvement, which Selna says has been instrumental in Geisinger achieving near 100 percent compliance for the 45-plus process measures in its heart program.
The New Jersey Hospital Association (NJHA) had applied for and won approval in 2003 for an eight-hospital gainsharing demonstration project in New Jersey, which ran briefly in 2004 but was halted when four non-participating hospitals, recognizing the merits of the project, sued CMS to expand the eight hospital limit and allow them the opportunity to participate as well, according to Sean Hopkins, NJHA’s Senior Vice President of Health Economics.
NJHA and a consortium of 12 hospitals have recently submitted an application to CMS to run a similar gainsharing demonstration project in New Jersey and is awaiting review by CMS. Under the proposal, each participating hospital will establish an internal oversight committee composed of at least 50 percent physician membership to establish a dialogue seeking ways to improve quality and efficiency of care, says Hopkins.
Hospital selection criteria for the project included diversity of geographic location, teaching and non-teaching, system-affiliated and stand-alone, and sufficient Medicare discharge volume, while more than half of the state’s 81 acute care hospitals expressed interest in participating, says Hopkins. The selected participants are: Overlook Hospital, AtlantiCare Regional Medical Center, Holy Name Hospital, Hunterdon Medical Center, Jersey Shore Medical Center, JFK Medical Center in Edison, The Valley Hospital, Robert Wood Johnson University Hospital in New Brunswick, Monmouth Medical Center, St. Francis Medical Center in Trenton, Southern Ocean County Hospital, and Somerset Medical Center.
The oversight committees will recommend structural and clinical practice changes that could result in shorter inpatient stays, fewer marginal but costly diagnostic tests being performed, reductions in pharmacy expenses, a more efficient use of the operating room, avoidance of duplicative services and improvement in discharge planning, says Hopkins.
The project is open to any physician who provides services to Medicare beneficiaries at participating hospitals, while physician participation is voluntary and there are no penalties for not achieving the benchmarks, according to Hopkins. The incentive payments to individual physicians, as per CMS guidelines, are limited to 25 percent of the amount that is normally paid under Medicare Part B for cases in the gainsharing demonstration.
Examples of issues the physician-hospital committees might identify, Hopkins notes, include:
· Adding more patient transporter services to eliminate bottlenecks in the emergency room and delay in moving patients to their inpatient beds, which adds to a patient’s length of stay.
· Having physicians do their hospital rounds earlier in the day so that they may be able to write discharge orders before 12 noon, avoiding an unnecessary extra inpatient day; or having extra rounds on Fridays to avoid unnecessary weekend inpatient stays.
· Scheduling surgeons with certain operating styles – i.e., “fast cutters” or “slow cutters” – in different operating rooms to avoid losing OR time later in the day because operations ran longer than expected in the morning.
· Getting lab results sooner, allowing physicians to come to the patient’s bedside sooner and determine that they’re ready for the next level of care, or possibly discharge.
· Sharing best practice norms among physicians and hospitals to improve care quality.
· Implementing “traditional” gainsharing activities such as open and perform as needed, and product standardization and substitution.
Under the project, a base year will be used to set performance thresholds for physician incentive payments. The target thresholds will be severity-adjusted DRG performance standards for Medicare cost consumption and inpatient length-of-stay, based on the average for the top 25 percent of performing physicians in New Jersey within a specific specialty. Performance will be assessed annually, while hospitals will have the latitude of doing their evaluation on a semi-annual or even quarterly basis, says Hopkins.
NJHA will have a global steering committee that will serve as a forum to share best practices across physicians at the 12 participating hospitals, from both a clinical and nonclinical perspective, to help them work toward the target norm, says Hopkins.
The project contains another quality improvement component, Hopkins adds: to achieve quantifiable reductions in preventable errors, infections or adverse outcomes, all consortium members must also agree: to participate in the Institute for Healthcare Improvement’s 100,000 Lives Campaign, to join all NJHA-sponsored quality collaboratives – which include initiatives for ICUs, pressure ulcers, rapid response teams, and anti-microbial infections – and to submit process and outcome quality indicators to CMS.