United States General Accounting Office GAO Report to the Chairman and Ranking Minority Member, Subcommittee on Military Personnel, Committee on National Security, House of Representatives August 1997 DEFENSE HEALTH CARE TRICARE Resource Sharing Program Failing to Achieve Expected Savings GAO/HEHS-97-130 United States GAO General Accounting Office Washington, D.C. 20548 Health, Education, and Human Services Division B-276961 August 22, 1997 The Honorable Stephen E. Buyer Chairman The Honorable Gene Taylor Ranking Minority Member Subcommittee on Military Personnel Committee on National Security House of Representatives The Department of Defense’s (DOD) nationwide managed care program, called TRICARE, is intended to improve the military community’s access to health care while maintaining quality and controlling costs. TRICARE represents a significant effort to reform DOD’s $15 billion per year health system. DOD’s approach to this reform involves a unique partnership between military and civilian health care entities that will include seven multistate managed care support contracts together estimated to cost about $17 billion over 5 years. In that partnership arrangement, resource sharing is an important cost-saving feature. To share resources, the contractor supplements the capacity of a military hospital or clinic by providing civilian personnel, equipment, or supplies. DOD has estimated that resource sharing could save about $700 million1 over 5 years for the contracts under way during our review. Because of your Subcommittee’s continuing interest in DOD’s use of support contracts to help deliver health care and, specifically, to control costs, we reviewed the cost-saving feature of resource sharing. In particular, we focused on (1) whether resource sharing savings are meeting DOD’s projections and thus helping control TRICARE costs, (2) what problems DOD might be encountering in pursuing resource sharing, and (3) actions and alternatives pursued by DOD to overcome those problems. Also, we considered the implications of resource sharing within the broader context of TRICARE’s overall cost-effectiveness. In doing our work, we held discussions and examined records at DOD headquarters in Washington, D.C.; TRICARE regional offices; military hospitals and clinics; and contractor offices. We focused on resource sharing experiences under the first four contracts, involving seven TRICARE regions that had been active long enough for us to draw conclusions about the progress of resource sharing. We also examined 1 This does not include expected savings from the most recently awarded contract, which began operating April 1, 1997, or from the remaining two contracts yet to be awarded. Page 1 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 TRICARE policies and plans DOD has devised or is considering that will affect resource sharing’s future. Finally, we reviewed data, which DOD provided in response to our findings, suggesting that TRICARE savings other than from resource sharing were occurring. Appendix I further describes our work’s scope and methodology. Appendix III shows the contract start dates and related TRICARE regions. Thus far, DOD and the contractors have made agreements likely to save Results in Brief about 5 percent of DOD’s overall resource sharing savings goal.2 At that rate, after 9 to 24 months of operations, about $36 million of the expected $700 million in savings will be realized. New agreements are being considered, but neither DOD nor the contractors are confident that pending agreements will be reached or that further cost savings can be attained. Because resulting TRICARE contract costs may be greater than anticipated, both parties may experience related financial losses. Problems impeding progress on resource sharing agreements and the related savings have included lack of clear program policies and priorities, uncertainty about cost effects on military hospitals, lack of financial rewards for the hospitals entering into such agreements, and changes in military hospital capacities after contractors developed bids. In response, DOD has revised policies, improved training and analytical tools, and taken other steps to promote resource sharing under the contracts, but to date, these efforts have not been sufficient to bring needed results. For the last two contracts, soon to be awarded, DOD is applying a revised financing approach that includes resource sharing but at a reduced level. The new approach allocates more funds to the military hospitals and less to the contractors, enabling the hospitals to directly acquire and use outside resources rather than use resource sharing with the contractor. But how the military hospitals, other sources, and contractors would interact under the new approach is still being defined and has not been tested. Therefore, it is not possible to say whether it will work. Moreover, resource sharing problems will not be automatically eliminated and may be exacerbated when used in combination with revised financing. For the future, DOD plans even broader changes intended to simplify military hospital budgeting and support contract operations. DOD has concluded that the current approach, including resource sharing, does not 2 Resource sharing also includes support contractor conversions of previous agreements DOD hospitals had with civilian providers. But these agreements existed before TRICARE, such that savings associated with them were not part of DOD’s new projected TRICARE resource sharing savings. Page 2 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 provide adequate accountability and incentives for delivering cost-effective care. While the military hospitals and contractors could still use resource sharing, it no longer would be the basis for projecting major savings and lowering bids at the contract’s outset. But, as DOD goes about the budgeting and contracting changes, unless they are specifically addressed, the kinds of policy, priority, and procedural problems that plagued resource sharing may continue to impair any new initiatives’ cost-savings efforts. In addition, such TRICARE efforts may be made even more challenging by leadership changes in health care management now under way in DOD. DOD officials acknowledged their resource sharing savings problems but told us that lower than expected contract award prices have led to over $2 billion in unexpected, offsetting savings. They said that the contract amounts have underrun cost projections, made as early as 1993, of what CHAMPUS3 costs would have been during the contract periods. While TRICARE’s overall cost-effectiveness was beyond our review’s scope, there are reasons to question the currency and analytical completeness of DOD’s preliminary savings claims. Thus, we support DOD’s current plans4 to undertake a detailed analysis, based on more up-to-date cost data and estimates, of TRICARE’s overall cost-effectiveness. DOD’s primary medical mission is to maintain the health of 1.6 million Background active duty service personnel5 and provide health care during military operations. Also, as an employer, DOD offers health care to 6.6 million other military-related beneficiaries, including dependents of active duty personnel and military retirees and their dependents. Most care is provided in about 115 hospitals and 470 clinics—referred to as military treatment facilities, or MTFs—worldwide, operated by the Army, Navy, and Air Force. DOD’s direct care system is supplemented by care paid for by DOD but provided in civilian facilities. In fiscal year 1997, DOD expects to spend about $12 billion providing care directly and about $3.5 billion for care in civilian facilities. 3 The Civilian Health and Medical Program of the Uniformed Services, a DOD-administered insurance-like program that has traditionally supplemented DOD’s direct care system. 4 Also, as required by the National Defense Authorization Act for Fiscal Year 1996, DOD has contracted for an independent study to, among other things, review TRICARE’s cost-effectiveness. Under the contract, the first results report is due by January 31, 1998. 5 Also includes members of the Coast Guard and the Commissioned Corps of the Public Health Service and the National Oceanic and Atmospheric Administration, who are also eligible for military health care. Page 3 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 In response to increasing health care costs and uneven access to care, in the late 1980s, DOD initiated, under congressional authority, a series of demonstration programs to evaluate alternative health care delivery approaches. On the basis of this experience, DOD designed TRICARE as its managed health care program. The TRICARE program uses regional managed care support contracts to augment its MTFs’ capacities by having contractors perform some managed care functions, including arranging civilian sector care. Altogether, seven managed care support contracts will be awarded covering 11 TRICARE regions (see app. II). To coordinate MTF and contractor services and monitor care delivery, each region is headed by a joint-service administrative organization called a “lead agent.” Thus far, DOD has awarded five contracts to three health care companies covering eight TRICARE regions. The contracts are competitively awarded and fixed price, although the price is subject to specified adjustments for changes in beneficiary population, MTF workload, and other factors beyond the contractor’s control. DOD officials believe that care provided to its patients at military facilities is less expensive than such patients’ care at civilian facilities. Resource sharing arrangements are designed to permit DOD and the contractor to share contractor-provided personnel, equipment, supplies, and other items in an effort to maximize savings. To identify resource sharing opportunities, contractors analyze such data as historical health care costs, workload, and care use, and visit military facilities. They then project the expected savings from providing care in military facilities rather than in potentially more expensive civilian settings. The contract is designed so that the contractor’s expected savings over the contract’s life from the resource sharing are deducted from the contractor’s final offer when bidding on a contract. The contract price thus reflects such anticipated savings through shared resources. The contract also is subject to a risk-sharing arrangement under which the government and the contractor share responsibility for health costs that overrun the contract price. Contractors are at risk for their bid amount of health care profit plus up to 1 percent of the bid health care price. Beyond that, the contractor and the government share in losses until an amount prepledged by the contractor, called “contractor equity,” is depleted. At that time the government becomes fully responsible for any further losses. Page 4 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 Thus, DOD’s initially realized savings in the form of a lower contract price could be reduced or lost if actual health care expenses are higher than anticipated. Accordingly, DOD encourages MTFs to help the contractor achieve projected resource sharing volume and savings. Resource sharing savings, along with expected savings from other sources, such as negotiated provider discounts; better health care utilization management; and better claims management, including collections from other health insurance plans, contribute to government and contractors’ overall financial gains. The combined expected savings from resource sharing and other sources are important as offsets to the increased costs of managing care under TRICARE. Also, statutorily, TRICARE costs cannot be greater than the health costs DOD otherwise would have incurred under CHAMPUS and the direct care system in the program’s absence (National Defense Authorization Acts for Fiscal Years 1994 and 1996, P.L. 103-160 and P.L. 104-106, 10 U.S.C. 1073 note). We reported last year that lack of resource sharing progress was one area that could impair efforts to contain related TRICARE costs and achieve savings. We reported that resource sharing was a complex and difficult process and that the process’ details were not well developed or understood, including uncertainty about how resource sharing agreements may affect contract price adjustments.6 DOD and the contractors are not attaining major new savings through Resource Sharing resource sharing agreements, and the potential for new agreements and Savings Falling Short further savings appears limited. On the basis of progress to date and of Projections discussions with DOD and contractor officials, achieving overall projected resource sharing savings appears highly unlikely. For the contracts under way, DOD projected saving about $700 million, including $116 million through the current operating years. The contractors’ projections were similar. But by March 1997, after 9- to 24-month contract operating periods, new resource sharing agreements represented only about 5 percent of the savings needed to achieve DOD’s projected savings. In addition to the new agreements, contractors have also converted into resource sharing agreements previously existing agreements that MTFs had with civilian providers before TRICARE became operational. At one MTF, for example, on the day the support contract 6 Defense Health Care: Managed Care Plan Progressing, but Cost and Performance Issues Remain (GAO/HEHS-96-128, June 14, 1996). Page 5 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 became operational, seven existing agreements were converted to resource sharing agreements. But savings associated with those converted agreements do not represent new TRICARE savings and thus were not part of DOD’s new projected savings. Support contractors and DOD are aware of the lack of progress in resource sharing. One contractor’s representative told us that achievements so far are just previous agreement conversions and that a more aggressive approach toward new agreements is needed. Another said lack of progress in negotiating new agreements remains their greatest TRICARE contract concern. DOD officials expressed mixed views ranging from optimism that resource sharing momentum will build to the belief that the approach simply will not work as envisioned. At this time, the potential for further resource sharing savings appears limited. In March 1997, the contractors had about 170 new resource sharing possibilities in some stage of cost and workload data gathering or analysis, or in some way being considered as potential agreements. For example, one region had 39 resource sharing possibilities under development, covering an array of services such as cardiology, radiology, and internal medicine. Officials told us, however, that considerable analysis was needed before potential savings could be reliably estimated and that some of the proposals likely would not prove cost-effective. Meanwhile, additional proposals are being added and existing ones deleted as the proposal and evaluation processes continue. But, as previously indicated, savings to date show that not enough is being done to reach DOD’s projected resource sharing savings levels. In addition to agreements already implemented or under consideration, by Problems March 1997, over 260 other resource sharing proposals had been either Encountered in rejected or otherwise not further pursued. Our analysis indicated that Establishing various impediments exist to resource sharing, including lack of clear policies, program complexity, lack of MTF incentives, and military Agreements downsizing. Issued in December 1994, DOD resource sharing guidance stated that MTFs had an obligation to help contractors reach the bid amount of resource sharing savings. But the guidance also instructed MTFs to look for other, possibly more cost-effective ways to increase MTF resource use, such as by reallocating existing resources, referring patients to other MTFs, or directly contracting with civilian care providers other than the support contractor. Page 6 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 When some MTFs pursued such alternatives, one contractor objected, stating its belief that resource sharing was the first alternative for increasing MTF use. In November 1996, DOD issued new guidance stating that resource sharing was the first alternative and that MTF commanders should make good faith efforts to work with contractors to execute such agreements. MTF and contractor officials cited the resource sharing approach’s complexity as another factor limiting progress. The agreements require considerable financial analysis to assess their cost-effectiveness potential (see app. IV). Also, the agreements involve intricate issues of how much credit contractors should receive for adding to the MTFs’ workload and how that credited workload will affect the contract price. MTF officials told us they did not understand all of the agreements’ financial implications, largely because they did not control or understand all the data and analyses used. They were concerned that workload shifts between MTFs and contractors, and ensuing bid price adjustments, would enable contractors to gain at MTFs’ workload and budgetary expense. At two MTFs, for example, proposed gastroenterology assistance agreements, projected to save over $400,000, were rejected because of unresolvable MTF concerns about possible effects on the overall contract price. Both contractor and MTF officials expressed concerns—and resulting hesitance to enter into agreements—about the reliability of data used to analyze agreements’ potential cost-effectiveness. According to contractors, for example, several MTFs supplied inappropriate data, such as personnel salaries and hospital maintenance, that hampered their analyses of the proposals’ likely costs and other effects. At one of the MTFs, eight proposed agreements were rejected because of data problems. Tied to the complexity and data problems, a lack of incentive to enter into agreements because MTFs do not share in resulting savings was also cited by MTF officials. DOD and the Services have not established a savings return policy for MTFs that have resource sharing agreements. Instead, after consideration, the Services decided that any such savings are to be retained at the Service level for reallocation as needed within the system. Still another MTF resource sharing disincentive is that the agreements can actually increase facility costs. For example, an agreement to provide an anesthesiologist, so the MTF can do more surgeries, will in turn result in related radiology, laboratory, and pharmacy costs. Unless contractors compensate MTFs for such costs, MTFs’ overall costs may increase. While Page 7 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 the contracts provide for such contractor compensatory payments, a July 1996 DOD policy clarification was issued to help facilitate such payments. A remaining challenge has been MTF and contractor negotiations on what costs to apply to individual agreements. Both DOD and the contractors cited military downsizing, including at the MTFs, as another limiting factor. Resource sharing opportunities identified during the contract bidding process may no longer exist as military forces are reduced or relocated and as MTFs are closed, downsized, or converted to clinics. For example, one MTF rejected five proposals because it had subsequently reduced its operating rooms from eight to four, thus obviating the need for agreements. Resource sharing problems have prompted one contractor to request a contract price adjustment. In June 1996, near the start date of health care delivery, the contractor reported that while the other care delivery preparations had progressed well, the lack of resource sharing progress was a major problem. Projecting millions of dollars in financial losses, the contractor requested a price renegotiation. In a letter to DOD, the contractor complained about changing DOD rules on how the approach was to work, inadequate data, improper MTF incentives, insufficient MTF training in developing agreements, and postaward MTF workload and capacity changes that reduced resource sharing opportunities. DOD generally agreed that problems existed, committed to work collaboratively to resolve them, and scheduled meetings with the contractor to pursue the issues in more detail. DOD said, however, that a price renegotiation was premature at the time. As of May 1997, the contractor was still pursuing a price adjustment. DOD has acted to increase resource sharing under current contracts. For DOD’s Actions to the latest two contracts, soon to be awarded, DOD will be applying an Address Resource alternative approach, referred to as “revised financing,” that relies less on Sharing Problems resource sharing for savings but adds other challenges. For the future, DOD is planning far broader changes in MTF budgeting and support contracting, which are expected to further reduce reliance on resource sharing. Attempts to Improve DOD has worked to facilitate resource sharing through policy issuances and Resource Sharing Under provision of analytical tools. Since issuing resource sharing guidance in Current Contracts December 1994, DOD headquarters officials visited the regions to provide briefings, used a focus group to help make resource sharing easier to use, Page 8 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 developed standardized training, and attempted to promote better DOD and contractor cooperation. Also, the contractors have continued to work with the MTFs to identify and pursue resource sharing opportunities. In November 1996, DOD issued clarifying policy stating that resource sharing is to be the first alternative for recapturing private sector workload into the MTF. Lead agents and MTFs are to ensure that any other MTF actions to add or retain workload do not prevent the TRICARE support contractor from entering into cost-effective agreements and reaching their resource sharing bid amounts. In July 1996, DOD clarified its policy regarding cash payments by support contractors to MTFs for marginal costs stemming from agreements. In a related move, DOD recently made available $25 million to the Services to help pay such marginal costs, or for the MTFs to otherwise invest in agreements, and asked the Services to submit potential projects for the funds’ use. In April 1997, DOD told us that some funds had been approved for only two or three requests. To help reduce resource sharing complexities, DOD provided a financial analysis worksheet for determining whether an agreement might be cost-effective and whether the amount of recaptured workload credited to the contractor is appropriate (see app. IV). DOD later revised the worksheet to, among other things, account for different agreement types. DOD also provided an analytical model further showing the MTFs’ resource sharing’s potential financial effects. The model was introduced to the MTFs in July 1996. DOD created a resource sharing focus group after a lead agent reported in January 1996 that resource sharing was complicated and presented MTFs with disincentives. The group worked for about 6 months and recommended improvements in such areas as training, the financial analysis worksheet, and the data used to make agreements. In early 1996, DOD began developing a TRICARE Financial Management Education Program curriculum that included resource sharing and the bid price adjustment process. Program testing was completed in December and presentations have begun. In November 1996, DOD initiated a new “partnering” effort with the contractors. DOD saw a need to help MTFs and contractors work through data problems, contract ambiguities, resource constraints, and other Page 9 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 TRICARE difficulties. The partnering approach calls for a more cooperative, trusting, teamwork relationship between MTFs and support contractors, including ways to avoid disputes and to informally resolve, rather than possibly litigate, those that occur. Early actions included DOD meetings with contractors at headquarters and regional levels, contractor participation in a national TRICARE conference, and consideration of assigning representatives of lead agents and the contractors to work together at each other’s locations. The bottom-line measure of DOD’s and the contractors’ efforts is in the progress made entering new resource sharing agreements. But progress remains slow, and the prospects for additional agreements are questionable. These outcomes, along with one contractor’s request for financial relief and DOD’s recognized need to improve teamwork, indicate a need for more concerted efforts under the current contracts to reach the agreements that are pending while seeking acceptable alternatives to resource sharing. Revisions Under the Latest DOD’s revised financing approach, conceived before the first support Contracts Reduce Reliance contract began operating but applied only in the latest two, is intended to on Resource Sharing but strengthen MTF health care management. Under this approach, MTFs’ direct funding and financial responsibilities will be increased. The funding Add Complications increase will be determined by the amount of previous CHAMPUS expenditures for MTF-based TRICARE Prime7 enrollees, which DOD expects will include most MTF service areas’ beneficiaries. Thus, rather than sharing responsibility for Prime enrollees with the support contractor, the MTFs will have full funding and full responsibility for their Prime enrollees and will pay the contractor for care required from the contractor’s network. One result of this approach will be to reduce reliance on resource sharing to lower support contract costs; but it also adds new challenges and does not eliminate, and may even exacerbate, resource sharing problems. Giving the MTFs direct financial control for TRICARE Prime enrollees is aimed at providing them with clearer incentives to efficiently manage care use and to behave more like private sector HMOs. DOD saw the need for this while still arranging the earlier contracts and later viewed it as a way to relieve emerging resource sharing problems. But, under revised financing’s 7 TRICARE is designed to give beneficiaries a choice among TRICARE Prime, which is similar to a health maintenance organization (HMO); TRICARE Extra, which is similar to a preferred provider organization; and TRICARE Standard, which is the current fee-for-service-type benefit. Beneficiaries who select TRICARE Prime must enroll to receive care under this option. Page 10 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 current approach, DOD will continue sharing care costs with the contractor for beneficiaries not enrolled with the MTFs. Also, the MTFs will continue working with the new contractor toward signing resource sharing agreements. Thus, to the extent contractor reliance on resource sharing continues, the difficulties already experienced are also likely to continue. DOD believes revised financing gives MTFs added cost-saving incentives to engage in resource sharing by reducing the need for referral of their enrollees to the TRICARE support contractor. However, revised financing may add further complexity to resource sharing’s use. Because the new approach’s potential effects on resource sharing are not now known, TRICARE contract offerors must make their own assumptions and projections about such effects. Much will depend, for example, on how MTFs’ funding levels may change and the consequent alterations in their beneficiary service priorities. And the added extent of funding going to MTFs rather than to contractors will in turn depend on the MTFs’ capacities and ability to enroll beneficiaries and serve as their primary care manager—all of which have yet to be determined. Revised financing’s effects on resource sharing are uncertain and were at issue during the two affected contracts’ bidding processes. One bidder, a current TRICARE contractor, wrote to DOD to clarify what portion of the funds the MTFs and contractor respectively would control and how revised financing would affect resource sharing. In earlier discussions, the bidder told DOD the company could be creative and assume resource sharing opportunities would still exist or assume none would exist. DOD replied that the new approach’s effects on resource sharing were uncertain but that the successful bidder should work creatively with the MTFs to achieve resource sharing. DOD also amended the request for a bid proposal to provide more description and examples of how revised financing and resource sharing might be integrated. But, as with resource sharing under the current contracts, the new approach’s actual effects will not be known until it is implemented. While DOD officials in regions with contracts generally favored revised financing, they expressed concerns about poor accounting systems and lack of data on patient care costs and outcomes that MTFs will need to become effective, cost-competitive providers. Some had concerns about the general lack of MTF health care management experience and control over their staffing. MTF officials in regions about to apply revised financing have stated that they recognize their increased need for accountability, Page 11 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 adequate staffing to support their enrollees, and better information systems to support resource sharing decisions. While theoretically possible, revised financing’s potential has yet to be demonstrated. Also, while revised financing reduces reliance on resource sharing, it does not eliminate or necessarily alleviate resource sharing problems and may exacerbate such problems under the new contracts. More Contracting and For the future, DOD plans other changes to simplify TRICARE contracting Related Budgeting and MTF budgeting. The changes would incorporate revised financing and Changes Planned, With further reduce reliance on resource sharing but also would have far broader implications for current and future contracts. Adding to such Broader Implications TRICARE initiatives’ challenges are changes in DOD’s top leadership in Health Affairs. DOD is now considering alternative structures for future contracts, on the basis of our recommendations8 and those from lead agents, contractors, and others in the health care industry. The alternatives include smaller, shorter, and less prescriptive contracts, allowing contractors to rely more on their own “off-the-shelf” commercial practices. DOD has held several forums to discuss ideas and the alternative approaches’ potential advantages and disadvantages. The issues involved include effects on beneficiary choice of providers, assurance of contractor qualifications, quality of care, DOD and contractors’ risk sharing, administrative complexity, adequacy of bid competition, and DOD costs. No final decisions have been made yet. The new contract structures likely will include an approach similar to revised financing. Basically, each MTF would be funded to cover all its enrollees in TRICARE Prime, and the contractor would be funded for all other beneficiaries. Thus, each MTF and contractor would be responsible for its share of the beneficiary population’s care costs, and would reimburse each other when one provides services to the other’s beneficiaries. For example, the contractor would reimburse an MTF for caring for a nonenrollee, and one MTF would reimburse another upon referring its own enrollee for care there. One aim of the funding approach would be to eliminate reliance on resource sharing as a major source for TRICARE savings. 8 Defense Health Care: Despite TRICARE Procurement Improvements, Problems Remain (GAO/HEHS-95-142, Aug. 3, 1995). Page 12 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 In April 1997, DOD accelerated the planned change in MTF budgeting and contract financing and announced it would be effective at the start of fiscal year 1998. This means that not only will the changes apply to future contracts but also current contracts will have to be amended. DOD expects that changing the current contracts may have cost implications of unknown extent at this time for both the government and the contractors. Commenting on a February 1997 DOD policy draft, one contractor said that any change that would avoid reliance on resource sharing, bid price adjustments, and resulting MTF disincentives would be positive. The contractor added, however, that DOD needs to involve the contractors in weighing the new budgeting and financing approach’s assumptions and risks to ensure it will work; otherwise contract prices may increase to cover the unknown risks. Another contractor said that many of the details had yet to be worked out and that two remaining questions are how funding will be split between MTFs and contractors and how resource sharing will be affected. Such budgeting and contracting changes reach far beyond an expectation that they will reduce the need for resource sharing. This notwithstanding, DOD lacks a simple, stable, long-term approach to TRICARE budgeting and contracting that provides clear managed care incentives and accountability and avoids the complexities and disincentives of resource sharing. As the contractors indicated, whether the contemplated system changes succeed will depend upon how these details are worked out and how well DOD and the contractors manage the system and support each other. In addition, both the Assistant Secretary of Defense (Health Affairs) and the Principal Deputy Assistant Secretary, who have actively and forcefully led TRICARE since its beginning, have left their positions. The former Principal Deputy has taken the Assistant Secretary position in an acting capacity. The Principal Deputy position has been filled, but to date no successor to the Assistant Secretary has been nominated. These top DOD leadership changes may add to the challenge of successfully reducing reliance on resource sharing and adopting broader budgeting and contracting changes. Page 13 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 DOD officials acknowledged that resource sharing has not achieved the DOD Views Resource expected savings, but told us that lower than expected contract award Sharing as One of amounts have led to more than $2 billion in other savings. They explained Several Cost-Saving that the contract award amounts consistently have underrun DOD’s projections, required before each contract is awarded, of what CHAMPUS Features costs would be over the contracts’ lives. As an example, one region’s estimated CHAMPUS costs without the contract would have been about $2.1 billion, compared with the contract award amount of $1.8 billion; so, according to DOD, the savings would be $0.3 billion. These officials also said that overall health care data show downward MTF cost trends, further supporting managed care’s cost-saving effects—despite resource sharing’s limited showing. For example, they provided a graph showing that both direct care and CHAMPUS total costs declined steadily—by 10 percent overall—from fiscal years 1991 through 1996. While assessing TRICARE’s overall cost-effectiveness was beyond our review’s scope, there are reasons at this time to question the currency and analytical completeness of DOD’s savings claims. First, DOD’s preaward estimates of CHAMPUS costs, a key component of its savings claim, may now be outdated. The first estimate—for the Northwest Region contract—was based on cost data prior to August 1993. Over the 4 years since then, changes in such areas as benefits and allowed payments to providers would affect the results of that estimate. Second, in a separate review,9 we found that as of May 1997, the existing five contracts had been modified as many as 350 times, with the resulting potential for substantial contract cost increases attributable to TRICARE. These potential cost increases, just like the potential losses from lack of resource sharing, also would offset DOD’s projected savings. Furthermore, we recently questioned10 DOD’s cumulative 5- to 7-percent utilization management11 savings estimate in its near $15 billion to $18 billion health care budget totals for fiscal years 1998 to 2003. We reported that DOD lacked a formal methodology for developing the estimates, and we concluded overall that future health care costs likely would be greater. Lastly, DOD’s available health care cost data do not indicate whether apparent downward shifts might be due to managed care 9 We plan to report shortly on DOD’s overall management of TRICARE contract change orders. 10 Defense Health Care: Future Costs Are Likely to Be Greater Than Estimated (GAO/NSIAD-97-83BR, Feb. 21, 1997). 11 Utilization management employs such techniques as preadmission hospital certification, concurrent and retrospective reviews, and case management to determine the appropriateness, timeliness, and medical necessity of an individual’s care. Page 14 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 effectiveness or to such other factors as reductions in allowed provider payments that would have occurred in TRICARE’s absence. Thus, we support DOD’s plans to undertake a more current and complete cost analysis of MTF direct and contractor-provided care, based on recent program data, to bottom-line TRICARE’s current and future-year cost-effectiveness. At their present results levels, for the existing contracts, DOD and the Conclusions support contractor will achieve only about 5 percent of the expected $700 million in new savings, potentially causing shared financial losses and higher TRICARE costs. Progress in achieving new agreements is slow, and neither DOD nor the contractors know what resource sharing potential remains under these contracts. While DOD now seems to be moving toward a view that the approach will not work as designed, the contractors and DOD are still pursuing about 170 resource sharing possibilities in an effort to discover additional savings with which to reduce their costs. Many problems have contributed to resource sharing’s lack of success. DOD’s policies, processes, and tools for use at the local level as well as the degree of DOD and contractor collaboration have not yet been sufficient to effectively resolve the approach’s obstacles. While revised financing is feasible though unproven, its potential effects on resource sharing and on other expected savings under the latest two contracts remain to be seen. Under the new approach, resource sharing may be reduced, but its problems will remain and may become more complex as new MTF and contractor management responsibilities are introduced. DOD’s more broadly proposed MTF budgeting and support contracting changes would greatly affect future and current contracts, including further reducing resource sharing. Clearly, a simple, accountable, incentive-based approach is lacking, yet the potential effectiveness of DOD’s considered changes will largely depend on how well they are designed and implemented. As such changes further reduce resource sharing as a potential savings mechanism and as DOD looks to alternative savings sources, lessons learned from resource sharing will need to be carefully heeded and skillfully incorporated. Carrying such lessons forward may be particularly challenging as DOD changes the top leadership in Health Affairs. Page 15 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 DOD officials acknowledged that resource sharing has not, and likely will not, produce the projected savings, but contended that TRICARE’s managed care approach has produced offsetting savings in other ways. We question, however, the currency and analytical completeness of these claims and thus believe it is important that DOD proceed with its plans to reestimate TRICARE costs versus projected costs without TRICARE. We recommend that the Secretary of Defense direct the Assistant Recommendations Secretary of Defense (Health Affairs) to • determine whether any further resource sharing savings remain under the current contracts and, as appropriate, consummate promising agreements while seeking other mutually acceptable alternatives to resource sharing; • determine, to the extent the new contracts with revised financing use resource sharing, whether any such agreements are available and, as appropriate, enter promising agreements while seeking effective alternatives to resource sharing; and • incorporate, while planning for and implementing the next wave of MTF financing and contract management initiatives, such resource sharing lessons learned as the need for coherent, timely policies; clearly understood procedures; mutually beneficial incentives; and effective collaboration. DOD agreed with our recommendations and said, without elaborating, it Agency Comments had already implemented each of them. Nevertheless, while agreeing with and Our Evaluation the recommendations, DOD disagreed with the way we presented certain issues. DOD said, for example, that the report does not note the tremendous resource sharing success during the “CHAMPUS Reform Initiative” (CRI) in California and Hawaii (which preceded TRICARE) and does not note the continued success in region 9 (Southern California). Thus, DOD said the reader is led to assume that problems occurred in other regions because resource sharing was implemented on a broad scale without the requisite examination. We did not evaluate CRI resource sharing because our focus was whether resource sharing under TRICARE was producing new savings to help offset added TRICARE costs. Also, as the report notes, DOD’s reference to continued success in region 9 is basically a conversion of CRI resource sharing agreements, which do not reflect new savings under TRICARE. Page 16 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 Furthermore, DOD said the resource sharing program as currently structured was based on the best available information at the time and that the report should note that the TRICARE support contractors came to the same conclusion as DOD regarding resource sharing’s potential cost-effectiveness, even with their years of experience with managed care. But our report does not question whether DOD’s structure for TRICARE resource sharing was based on the best information available at the time. Instead, the report discusses the complex issues that arose during the implementation of resource sharing. Also, the report notes that the contractors as well initially concluded that resource sharing would be cost-effective. Also, DOD said the report treats resource sharing in isolation, as opposed to one component of a comprehensive system that has proven to be cost-effective. While we focused on resource sharing because it was expected to be a major cost-saving mechanism, we also noted that it was one of several ways in which DOD expected to achieve savings to offset TRICARE’s costs. DOD went on to state that efficiencies not achieved through resource sharing were otherwise achieved by increased MTF capability and efficiency brought about by TRICARE. As the report points out, during our review DOD presented information showing downward MTF cost trends, but these data do not show whether the trends were due to TRICARE managed care efforts or whether the costs would have declined anyway in TRICARE’s absence. DOD said managed care support (MCS) contracts have resulted in savings of $2.3 billion when compared with projected costs without the contracts. It said that we acknowledged this savings estimate but that our placement of it in the report diluted its significance. While a detailed review of overall TRICARE savings was beyond the scope of our review, as our report states, we question that savings estimate’s currency and analytical completeness, and we support DOD’s plans to undertake more current and complete analysis of TRICARE’s cost-effectiveness. We have revised the report to discuss DOD’s overall savings estimate in a separate section. DOD took issue with the report statement that, while revised financing reduces reliance on resource sharing, it does not eliminate or necessarily alleviate resource sharing problems and may exacerbate such problems under the new contracts. DOD said revised financing, in conjunction with its planned change to enrollment-based capitated budgeting for MTFs, increases incentives for MTFs to engage in resource sharing by expanding MTF funding while reducing support contractor costs. We agree that Page 17 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 revised financing, in conjunction with enrollment-based capitation, has the potential to create more incentive for the MTFs to engage in resource sharing and may similarly provide incentive to the support contractors. Still, those approaches add their own complexities and do not automatically eliminate the difficulties experienced with resource sharing. As we said in the report, the approaches are still being defined and are yet to be tested. Nonetheless, we revised the relevant text to better recognize DOD’s views on revised financing’s potential. DOD’s comments in their entirety are included as appendix V. We also obtained comments from the three current TRICARE support contractors. All expressed general agreement with the report’s overall content and completeness of subject coverage. In its comments, one contractor also offered a minor technical comment about lack of clarity in a statement defining limits on resource sharing agreement profits, which is part of the procedural description in appendix IV. The contractor pointed out, however, that there is no misunderstanding between it and DOD as to what is intended. We made no change because the appendix was presented to illustrate DOD’s guidance as it was offered. A second contractor expressed concern about its limited progress in resource sharing and about the problems and lack of success in resource sharing elsewhere, as conveyed in our report, and expressed hope that the report would help bring about favorable resolution of the problems. While stating that the report otherwise accurately portrays the resource sharing situation, the third contractor disagreed with the report’s statement that the prospects for additional resource sharing agreements are questionable. The contractor informed us that it had recently made a presentation to DOD on resource sharing shortfalls, but it also asserted that, with the right incentives and education at the MTF commander level, resource sharing is still an extremely viable program with current savings opportunities. On the basis of our analysis of the problems and overall limited resource sharing progress, the prospects for reaching new agreements seem to us to be limited. Still, the report urges DOD to identify and pursue promising resource sharing opportunities while also seeking other mutually acceptable alternatives to resource sharing. Page 18 GAO/HEHS-97-130 TRICARE Resource Sharing B-276961 We are sending copies of this report to the Secretary of Defense and interested congressional committees, and will make copies available to others upon request. Please contact me at (202) 512-7111 or Dan Brier, Assistant Director, at (202) 512-6803 if you or your staff have any questions concerning this report. Other major contributors are Elkins Cox, Evaluator-in-Charge; Allan Richardson; Beverly Brooks-Hall; and Sylvia Jones. Stephen P. Backhus Director, Veterans’ Affairs and Military Health Care Issues Page 19 GAO/HEHS-97-130 TRICARE Resource Sharing Contents Letter 1 Appendix I 22 Scope and Methodology Appendix II 24 Regions Served by the Seven Managed Care Support Contracts Appendix III 25 TRICARE Support Contracts and Start Dates Appendix IV 26 Guidance on Developing Resource Sharing Agreements Appendix V 37 Comments From the Department of Defense Figure Figure IV.1: Resource Sharing Development Flowchart 31 Page 20 GAO/HEHS-97-130 TRICARE Resource Sharing Contents Abbreviations ADD active duty dependent CHAMPUS Civilian Health and Medical Program of the Uniformed Services CPT current procedural terminology CRI CHAMPUS Reform Initiative DCP data collection period DOD Department of Defense DRG diagnosis-related group HCF Health Care Finder HMO health maintenance organization MCS managed care support MEPRS Medical Expense and Performance Reporting System MHSS Military Health Services System MTF military treatment facility NADD nonactive duty dependent NAS Non-Availability Statement RSA resource sharing agreement Page 21 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix I Scope and Methodology To assess the Department of Defense’s (DOD) experiences with resource sharing, we visited 5 (of the 7) regions where TRICARE support contractors had begun delivering health care and 11 military treatment facilities (MTF) within those regions. We also met with the two civilian TRICARE contractors that were providing health care support to the MTFs. A third contractor began providing health care on April 1, 1997, in two other regions (since combined into one region), but because of the newness of the operations, we met with this contractor briefly but did not include it in our detailed assessment of resource sharing progress and problems. Two other contracts, covering the remaining three regions, were still pending at the time of our review. We reviewed DOD and contractor projections of resource sharing costs and savings, TRICARE policies and guidance, and various efforts by DOD to promote the overall resource sharing effort. This included discussions with officials of the Office of the Assistant Secretary of Defense for Health Affairs, DOD’s TRICARE cost consultant, and contractor officials. At the contractors’ offices, we reviewed individual resource sharing project files to analyze the progress being made and determine the specific reasons why some potential agreements were not being implemented. The project files consisted of both agreements existing before TRICARE, referred to as “partnerships,” and new resource sharing agreements. Many of the partnership agreements were converted to resource sharing agreements as TRICARE became operational. To assess progress in achieving new savings under TRICARE, we identified the expected savings from the new agreements and compared the result to DOD’s overall projected TRICARE savings. We discussed information, training, and other needs with DOD officials at DOD’s Washington, D.C., headquarters and at regional and MTF levels, focusing on the factors that affected progress in resource sharing. Especially at the MTF level, we discussed officials’ understanding of, and amount of confidence in, the financial aspects of resource sharing agreements, including effects on the MTF workload and bid price adjustment. Through discussions with DOD and contractor officials and examination of records, we reviewed their experiences with planning and establishing resource sharing agreements, including the problems they encountered. We also discussed with DOD and contractor officials alternatives DOD has undertaken for the current contracts as well as policies and plans DOD has devised or is considering that will affect the future of resource sharing. Page 22 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix I Scope and Methodology At the completion of our work, we briefly reviewed DOD-provided data suggesting that TRICARE savings other than from resource sharing were occurring that more than offset the resource sharing savings shortfalls we had found. Determining TRICARE’s overall cost-effectiveness was beyond the scope of our review. Nonetheless, upon reviewing the data, we asked follow-up questions of DOD, obtained status information on DOD’s planned and under way internal and contracted studies aimed in whole or in part at determining TRICARE’s cost-effectiveness, and reviewed pertinent information from our other work in process and our issued reports. We conducted our review between June 1996 and May 1997 in accordance with generally accepted government auditing standards. Page 23 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix II Regions Served by the Seven Managed Care Support Contracts Page 24 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix III TRICARE Support Contracts and Start Dates Health care delivery Region Lead agent Contract status start date Northwest Madigan Awarded to Foundation March 1995 Health Federal Services, Inc., September 1994, for $475 million Southwest Wilford Hall Awarded to Foundation November Health Federal Services, 1995 Inc., April 1995, for $1.8 billion Southern California, San Diego, Awarded to Foundation April 1996 Golden Gate, and David Grant, Health Federal Services, Hawaii-Pacific Tripler Inc., September 1995, for $2.5 billion Southeast and Eisenhower, Awarded to Humana July 1996 Gulf South Keesler Military Healthcare Services, November 1995, for $3.8 billion Central Evans Awarded to Triwest April 1997 Healthcare Alliance, Inc., July 1996, for $2.3 billion Northeast National Capital Award expected by May 1998 September 1997 Mid-Atlantic and Portsmouth, Award expected by May 1998 Heartland Wright-Patterson September 1997 Page 25 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements To further explain the resource sharing agreement development process, the following information was condensed from selected guidance offered by lead agents. The guidance includes preparation of proposals, a chart showing the flow of agreement development (fig. IV.1), and application of a financial analysis worksheet. The Resource Sharing Program is a mechanism for providing contracted Guidance on civilian health care personnel, equipment, and/or supplies to enhance the Preparation of capabilities of MTFs to provide necessary inpatient and outpatient care to Proposals beneficiaries of the Civilian Health and Medical Program of the Uniformed Services (CHAMPUS). Resource sharing is a cooperative activity between the contractor, the lead agent, and the MTF commander. A variety of information sources and databases may be examined in looking for and evaluating resource sharing opportunities that may subsequently be developed into resource sharing proposals and agreements. Analysis of CHAMPUS Analysis of CHAMPUS utilization and cost data may identify diagnoses, Utilization and Cost Data procedures, or specialty health care, which account for significant numbers of patient encounters or high costs. A variety of reports may be useful in this regard. CHAMPUS Cost and Utilization Reports. These reports are generated by the Office of the Civilian Health and Medical Program of the Uniformed Services from health care service record data to show CHAMPUS costs and utilization, by type of health care service, for each catchment area. Those services showing high costs and/or utilization may be excellent candidates for resource sharing considerations. Non-Availability Statement (NAS) Reports. NASs authorize beneficiaries to seek certain care in civilian facilities when the MTF cannot provide the care. These reports show the numbers and types of NASs generated by each MTF. Those health care services showing large numbers of NASs being issued over time may be excellent candidates for resource sharing consideration. Health Care Finder (HCF) Referral Reports. These reports show the numbers and types of referrals of CHAMPUS-eligible beneficiaries to both MTF and civilian health care providers. High numbers of referrals to civilian Page 26 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements providers for specific health care services may indicate resource sharing opportunities. CHAMPUS Ad Hoc Claims Reports. CHAMPUS historical data may be obtained from claims data files. These data can be tailored to provide greater detail for the types of services being provided under CHAMPUS. Information from CHAMPUS Cost and Utilization Reports, NAS Reports, and HCF Referral Reports may indicate those health care specialties that warrant more detailed examination to identify potential resource sharing opportunities. MTF Capability Reports. These reports are developed by HCFs and indicate MTF capabilities. They are used by the HCFs to guide referrals into and out of MTFs. They may also provide insight into potential resource sharing opportunities. Composite Health Care System Professional Activity Study Reports. These reports may be used to identify gaps in MTF services or high referral patterns from the MTF to outside health care providers. These gaps and referral patterns may indicate additional opportunities for resource sharing. Network Provider Directory. This directory provides the numbers and types of health care providers by location. Gaps and shortages in the civilian provider network may be identified that may indicate resource sharing opportunities. Analysis of MTF MTF capabilities, staffing, workload, and backlog—both current and Capabilities projected—should be identified and evaluated to determine potential opportunities for resource sharing. MTF capabilities may be assessed using the following reports: MTF Capability Reports. (See prior description of reports.) MTF Staffing Reports. These reports are developed by MTFs and show the numbers and types of personnel assigned to, and employed by, the MTF. Careful review of staffing reports over time may indicate staffing trends, which may provide insight into both current and future resource sharing opportunities. A baseline report of regionwide staffing, by MTF, was compiled from computer tapes provided by the government to the contractor for fiscal year 1993. Page 27 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements MTF Operations Study. This report shows the historical number of health care services provided by MTFs for both inpatient and outpatient services. This report is derived from data compiled on a computer tape provided to the contractor by the government for fiscal year 1993. This information can be used to identify both current and future opportunities for resource sharing. Potential Resource Sharing Opportunities List. This list, developed during site visits at each MTF, provides resource sharing opportunities that had been identified by the MTF, after examining the demand for services and identifying shortfalls in meeting those demands. The Written Resource Once a resource sharing opportunity has been identified, the MTF Sharing Proposal completes a written request for consideration of the potential resource sharing agreement (RSA). The proposal is to show the project title, requesting MTF, point of contact, and desired start date. The expected accomplishment is to be described. For example, “This project is intended to expand Family Practice services within the hospital. This MTF currently averages 200 ambulatory care visits a month, and the implementation of this project should increase the monthly visits by an additional 200 visits. This should decrease the number of NASs issued and the concomitant CHAMPUS visits and costs.” The proposal is to include the estimated resources required, including personnel, equipment, and supplies, along with the following: Direct Workload. Provide the number of outpatient visits and/or inpatient admissions, by type of CHAMPUS beneficiary (active duty dependent [ADD], or nonactive duty dependent [NADD]) that the project is expected to provide per year. Note that the NADD category includes retirees, family members of retirees, survivors of deceased service members, and others. If possible, provide a detailed breakdown of workload numbers by current procedural terminology (CPT) or diagnosis-related group (DRG). If possible, provide the estimated cost to the MTF for each CPT and DRG code. Ancillary Workload. Provide the anticipated additional ancillary workload that the project will develop for the MTF, by type of CHAMPUS beneficiary (ADD or NADD), per year. If possible, provide a detailed breakdown of ancillary workload numbers by CPT or DRG. If possible, also provide the estimated cost to the MTF for each CPT and DRG code. Page 28 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements MTF Cost/Expense Data. Provide specific Medical Expense and Performance Reporting System (MEPRS) cost elements for the clinical function of the project. If possible, provide a detailed breakdown of MEPRS cost elements by CPT or DRG. CHAMPUS Workload Data. Provide CHAMPUS workload, within the catchment area, currently being accomplished for the clinical function of the project. If possible, provide a detailed breakdown of CHAMPUS workload and cost data by CPT or DRG. Signature and Date. Provide signature of the MTF commander, or his agent, and the date the document was signed. Example of a Resource Project Title. Internal Medicine Augmentation and Support. Sharing Proposal Purpose. The MTF had three internists assigned in fiscal year 1994, two in fiscal year 1995, and will decrease to one by June 1996. MTF workload has shown a concomitant decrease in the average number of outpatient visits, admissions, and occupied bed days. The number of NASs and visits to civilian providers under CHAMPUS has risen to absorb the demand for internal medicine services in the face of decreasing supply within the MTF. This proposed RSA, if approved, would expand the internal medicine services within the MTF and should increase the number of monthly outpatient visits by approximately 900 per month and the number of inpatient admissions by 37 per month. These increases should avoid a shift of approximately 425 outpatient visits per month to CHAMPUS with the loss of a military provider. They should also add an additional 475 outpatient visits per month to the MTF workload. Recognizing that approximately 44 percent of our CHAMPUS beneficiaries are ADDs and that 56 percent are NADDs, and using the appropriate volume trade-off factors, it should also reduce the number of visits that had previously been paid for through CHAMPUS by approximately 212 visits per month. Resources Required. To implement the proposed RSA, additional providers and support personnel will be required. Also, a financial offset for increased costs in ancillary services and supply costs will be necessary. Facility space and equipment are adequate to support the additional workload. Page 29 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements Personnel. Internist (board certified or eligible), Nurse (Licensed Vocational Nurse), with attached example of position description. Equipment. None. Supplies. No direct supplies, but, based on fiscal year 1995 MEPRS data, reimbursement for the costs of ancillary services and supplies for outpatient visits above that achieved during the data collection period, fiscal year 1995 (10,188 outpatient visits per year). Estimated at up to 5,700 visits. (For outpatient visits, example shows costs per procedure and per visit for pharmacy, laboratory, radiology, medical supplies, and other supplies.) Also, reimbursement for the cost of ancillary services and supplies for inpatient admissions above that achieved during the data collection period, fiscal year 1995 (404 admissions per year). Estimated at up to 226 admissions. (Example shows ancillary service and supply costs—based upon fiscal year 1995 MEPRS data—per procedure and per admission for same categories as for outpatient admissions.) MTF Workload Data. (Example shows internal medicine direct workload, based on fiscal year 1995 MEPRS data, in terms of outpatient visits and inpatient admissions. It shows also the internal medicine ancillary workload, based on fiscal year 1995 MEPRS data, in terms of pharmacy prescriptions, laboratory procedures, and radiology films per year for outpatient visits and inpatient admissions.) MTF Cost/Expense Data. (Example refers to attachments for MEPRS data for outpatient and inpatient care, based on fiscal year 1995 MEPRS data.) CHAMPUS Workload Data. (Example refers to attachment for CHAMPUS claims data for this catchment area based on claims data from September 1994 through August 1995.) Page 30 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements Figure IV.1: Resource Sharing Development Flowchart Source: TRICARE Desert States Financial Guide (Jan. 1997). Using the Financial Analysis Worksheet Purpose of Worksheet The standardized Internal Resource Sharing Financial Analysis Worksheet is structured to take into account three different types of proposed agreements: (1) the recapture of new workload, (2) the conversion of a partnership agreement, and (3) the replacement of a lost provider. Page 31 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements For all of these different situations, the resource sharing worksheet is designed to help the MTF answer two questions: (1) Is the proposed agreement projected to be cost-effective and (2) is the proposed contractor workload credit appropriate? An agreement is deemed cost-effective from the Military Health Services System (MHSS) perspective if the MHSS cost for the agreement (the sum of the MTF’s marginal expenditures and the contractor’s expenditures for the proposed RSA) is less than the government’s share of projected CHAMPUS savings. Assuming the cost-effectiveness test is satisfied, there are two additional criteria for evaluating whether the contractor’s workload credit is appropriate. First, the contractor credit shall not exceed the full credit (that is, 100 percent credit) that would be counted under the Guidelines for Resource Sharing Workload Reporting. Second, a prospective profit rate limit applies to RSAs for which the savings exceed those assumed in the contractor’s best and final offer. For these agreements, the contractor’s projected profit rate on resource sharing expenditures (as calculated by the worksheet) should not exceed the contractor’s overall proposed health care profit rate (on a prospective basis). For example, if a contractor proposed a 5-percent profit rate for health care costs, then the projected contractor profit on resource sharing expenditures exceeding the up-front bid price assumptions should also not exceed 5 percent. A prospective profit limit also applies to an RSA that converts an inpatient partnership agreement that existed in the data collection period (DCP) and for which CHAMPUS admissions were not counted in the DCP data. (In this case, workload credit should be negotiated as necessary to produce a projected contractor net gain approximately equal to zero, since otherwise the contractor would receive an upward price adjustment for additional NASs simply for maintaining the same workload done in the DCP under the partnership agreement.) If both of the previous questions cannot be answered “yes” for the proposed RSA, then the MTF should either renegotiate some of the terms of the proposed agreement (for example, the contractor’s workload credit) or consider other alternatives to the proposed agreement (for example, the task order resource support option). In addition to answering both previous questions for resource sharing in isolation, the resource sharing worksheet is designed to project the cost Page 32 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements impact of implementing the agreement under task order resource support rather than resource sharing, including a summary comparison of cost-effectiveness under the two options. Similarly, the worksheet shows the relative financial impact on the managed care support (MCS) contractor of resource sharing versus resource support. (Details on resource support analysis are excluded from this condensed version of the guidance.) Accrual of Savings Under the MCS contracts, resource sharing savings can accrue to the government in three ways, each of which is addressed in the worksheet. First, for those resource sharing savings investments assumed as part of the contractor’s best and final offer proposal, the contractor’s bid price includes a cost-per-eligible trend factor for resource sharing savings (that is, claims avoidance). Net of the contractor’s expected expenditures on resource sharing, this creates a lower up-front bid price (claims avoidance - resource sharing expenditures = net savings). These net savings are calculated in section I of the worksheet on an average basis (that is, using the contractor’s best and final offer assumption about the average savings to cost ratio for resource sharing). Second, if partial contractor workload credit is negotiated, the government will realize savings in the bid price adjustment for MTF utilization (the “O” factor). This can result in a more favorable bid price adjustment for the government. These savings are calculated in section II of the worksheet. Third, the government will also realize 0, 80, 90, or 100 percent of any residual savings in the risk-sharing corridor, depending on which tier of the risk-sharing corridor applies to the bid price adjustment for the option period. (The contract’s risk-sharing provisions are specified in detail in section G-5 and in appendix C in the Bid Price Adjustment Procedures Manual.) This will result in the government sharing any risk-sharing savings realized by the contractor. These savings are calculated in section IV of the worksheet. Required Completion of MTF commanders or their designated representatives are required to the Financial Analysis complete the standardized Resource Sharing Financial Analysis Worksheet Worksheet in negotiating each proposed RSA, in addition to any other analyses prepared by the contractor or the MTF (as specified in section G-5g(2) of the contract). Page 33 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements In completing the resource sharing worksheet, users should not be lulled into a false sense of security by focusing on numerical results rather than on underlying assumptions. The accuracy of assumptions such as the number of admissions and/or visits to be recaptured, the MTF’s marginal costs in recapturing these units, and the costs avoided in CHAMPUS are crucial to the accuracy of the spreadsheet’s projections. If estimates are too optimistic, even though the spreadsheet may project net gains for the government, in reality the government may experience net losses. Of course, overly pessimistic estimates can lead the government to miss out on cost-effective opportunities. The MTF Inputs Page To use the Financial Analysis Worksheet, the MTF must enter the boxed values on the “MTF Inputs” page. These include (1) the type of RSA, (2) whether the agreement converts an inpatient partnership agreement that previously existed, (3) the option period (year) covered by the proposed agreement, (4) the number of outpatient visits or inpatient admissions enabled by the agreement, (5) the expected government risk-sharing responsibility percentage, (6) the estimated volume trade-off factor used to estimate CHAMPUS avoidance savings, (7) the estimated average government cost per unit for admissions and/or outpatient visits avoided in CHAMPUS for care covered by the agreement, (8) the expected contractor expenditure under the agreement, (9) the projected MTF marginal expenditures, (10) the contractor resource sharing workload credit assumed in the analysis, (11) the sum of the projected resource sharing expenditures for those agreements approved for the lead agent region as a whole, and (12) the expected MTF payment for the contractor’s costs and the MTF’s marginal costs if the resource is acquired under task order resource support rather than resource sharing. As part of the negotiation of the RSA, the MTF commander and the contractor must agree on each estimate or assumption entered on the “MTF Inputs” page before the worksheet is finalized. The remaining sections of the Financial Analysis Worksheet do not require the MTF to enter any data or assumptions. Depending on the results shown on the “summary” page for resource sharing, however, it may be appropriate to revise some of the MTF inputs (for example, the contractor workload credit) on an iterative basis. Resource Sharing The “Summary—Resource Sharing” page lists the key results for the Summary Page proposed agreement under resource sharing. This summary shows Page 34 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements (1) whether the proposed contractor workload credit is appropriate, (2) whether government gains exceed government expenditures, (3) the projected contractor net gain under the RSA, (4) the projected government net gain, and (5) whether the proposed agreement reduces the contractor’s actual costs even if the contractor’s net gain is negative due to the average savings assumed up front in the contractor’s best and final offer. (Because the contractor reduced its best and final offer bid price based on an assumption about average savings for each RSA, some actual agreements are expected to produce savings that are smaller than this assumed average, but are still positive. This perspective is particularly relevant for conversion of partnership agreements, since the contractor is not likely to achieve new savings simply for continuing previous partnership agreements under the same terms as RSAs. The net contractor gain after taking account of average up-front savings from the best and final offer is likely to be negative, yet converting a cost-effective partnership agreement allows the contractor to avoid an increase in CHAMPUS claims costs that would otherwise result.) If the “Summary—Resource Sharing” page shows that the contractor workload credit is not appropriate and/or government gains do not exceed government expenditures, then one option for the MTF is to adjust the proposed contractor workload credit on an iterative basis until the proposed agreement satisfies both requirements. It may also be appropriate for the MTF to renegotiate other terms of the proposed agreement (for example, the level of resources to be provided by the contractor). If it is not possible to determine a workload credit percentage that results in a “yes” response to both questions, given all of the other input assumptions agreed upon by the MTF commander and the contractor, then the proposed RSA should not be approved (unless the lead agent determines that the proposed agreement still warrants approval due to compelling circumstances). Resource Sharing Page The resource sharing worksheet page has five sections. Section I estimates the net resource sharing savings under this agreement that would already be reflected in the contractor’s proposed bid price, based on the average-savings-to-cost ratio used to develop the resource sharing savings trend factor in the contractor’s best and final offer. Section II estimates the effect of the RSA, including the contractor’s workload credit, on the MTF utilization adjustment in the bid price adjustment formula (that is, the “O” factor adjustment). Page 35 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix IV Guidance on Developing Resource Sharing Agreements Section III estimates the actual savings (that is, cost avoidance) in CHAMPUS health care costs as a result of the RSA. Section IV estimates the residual gain in CHAMPUS (that is, the difference between the adjusted bid price for health care costs and the actual health care costs) under the proposed RSA. The section also estimates the government and contractor portions of these gains, since the gains would be subject to risk sharing between the government and contractors. Section V provides the two necessary results of this analysis (for an assessment of resource sharing in isolation). First, is the contractor credit for resource sharing workload assumed in the analysis appropriate? Second, does the analysis indicate that the proposed RSA would be cost-effective for the government from the MHSS perspective? Page 36 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix V Comments From the Department of Defense Page 37 GAO/HEHS-97-130 TRICARE Resource Sharing Appendix V Comments From the Department of Defense Now on p. 12. (101493) Page 38 GAO/HEHS-97-130 TRICARE Resource Sharing Ordering Information The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Orders by mail: U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD 20884-6015 or visit: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (301) 258-4066, or TDD (301) 413-0006. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. For information on how to access GAO reports on the INTERNET, send an e-mail message with "info" in the body to: email@example.com or visit GAO’s World Wide Web Home Page at: http://www.gao.gov PRINTED ON RECYCLED PAPER United States Bulk Rate General Accounting Office Postage & Fees Paid Washington, D.C. 20548-0001 GAO Permit No. G100 Official Business Penalty for Private Use $300 Address Correction Requested
Defense Health Care: TRICARE Resource Sharing Program Failing to Achieve Expected Savings
Published by the Government Accountability Office on 1997-08-22.
Below is a raw (and likely hideous) rendition of the original report. (PDF)