I 1 United.States General Accounting Office Report to the Ranking Minority G;AO Member, Committee on Veterans’ Affairs, U.S. Senate Aplill990 VA HEALTH CARE Better Procedures Needed to Maximize Collections From Health Insurers RESTRICTED-- Not to be released outside the General Accounting Office unless specifically approved by the Office of Congressional Relations. GAO/HID90-64 Human Resources Division B-237001 April 6,199O The Honorable Frank H. Murkowski Ranking Minority Member Committee on Veterans’ Affairs United States Senate Dear Senator Murkowski: In response to your request, this report addresses the Department of Veterans Affairs’ collections from health insurers for the cost of care provided to certain insured veterans. Basically, we found that collections from insurers greatly exceeded the cost of collection. But we estimated that the Department had collected from insurers only about one-third of the possible collections for medical services provided to insured veterans. Unless you publicly announce its contents earlier, we plan no further distribution of this report for 30 days. At that time we will send copies to the Secretary of Veterans Affairs and other interested parties. The report was prepared under the direction of David P. Baine, Director, Federal Health Care Delivery Issues. If you have any questions, you may call him on (202) 275-6207. Other major contributors are listed in appendix XI. Sincerely yours, Lawrence H. Thompson Assistant Comptroller General Executive Summ~ The Department of Veterans Affairs (VA) spends about $10 billion annu- Purpose ally providing health care to veterans, including those who have private health insurance coverage. To help reduce the federal budget deficit, in 1986, the Congress established the right of the United States to collect from health insurers the cost of care provided to certain insured veterans. The Ranking Minority Member of the Senate Committee on Veterans’ Affairs asked GAO to determine whether VA'S collections from health insurers exceeded its collection costs. He also asked whether VA had effective procedures to (1) identify insured veterans, (2) bill insurers, and (3) collect amounts owed. VA, with 159 medical centers, operates the largest health care delivery Background system in the United States. In response to the 1986 legislation, VA directed its medical centers to bill insurers for inpatient care costs. But VA allowed medical centers to decide whether to bill insurers for outpa- tient care. To assess VA'S billing and collection procedures, GAO visited six centers and reviewed records of randomly selected insured veterans who received care during 6 months of fiscal year 1988. GAO also sent a ques- tionnaire to all VA medical centers to obtain information on their identifi- cation, billing, and collection procedures, as well as costs. $96 million of these collections was attributable to inpatient health care; the remainder, to outpatient care. GAO estimates that VA collected from insurers only about one-third of the possible collections for medical services provided to insured veterans, VA centers could increase collections significantly if they (1) employed more effective methods to identify insured veterans and bill insurers and (2) committed additional resources to collection efforts. Centers are reluc- tant to make such a commitment because (1) all amounts collected must, by law, be returned to the U.S. Treasury and (2) additional collection costs centers incur are paid from their existing medical care budgets. Page2 GAO/IiRBSO-64 VA Cdlection~ From Insurers Executive!%unmmy Principal Findings GAO Estimates VA’s GAO estimates that VA could have collected an additional $223 million from insurers in fiscal year 1988, about $180 million for inpatient care Collections Potential and about $43 million for outpatient care. Data limitations introduce considerable uncertainty into the $223 mil- lion estimate, but GAO believes that additional collections could have been at least $55 million and perhaps as much as $392 million. (See ch. 2.) Insured Veterans Not In four of the six centers visited, GAO found veterans admitted for inpa- tient care who had private insurance policies not identified by VA. These Always Identified veterans made up 15 percent of the total identified by both VA and GAO during the periods GAO sampled. VA could have billed insurers about $318,000 for these veterans’ care. Center officials offered several expla- nations for these missed billings, including (1) center admissions staffs’ failing to ask the right questions or (2) veterans being confused or afraid that disclosures about insurance would jeopardize their admittance to the centers. The centers could have identified these veterans if their processes had included steps to verify the information the veterans provided, such as reviewing administrative and medical records or contacting employers. Nationwide, about two-thirds of VA'S medical centers reported that they did not verify veterans’ statements about insurance coverage. (See ch. 3.) Inpatient Care Costs Not Even when centers identified insured veterans, they did not always bill insurers, Five medical centers GAO visited did not bill $1.8 million of the Always Billed $7 million in inpatient care costs during the first 6 months of fiscal year 1988; GAO did not assess inpatient billings at one center visited. Billings were missed because the centers relied on manual systems that were flawed. For example, billing staffs were not notified when insured veter- ans were discharged, and internal controls were inadequate to detect the missed billings. (See ch. 4.) Page3 Executive Sumnuuy Outpatient Care Costs Three medical centers GAO visited did not bill for any outpatient care, Rarely Billed and the other three billed for such care only under certain conditions. GAO estimates the six centers did not bill insurers for about $1.6 million during the first 6 months of fiscal year 1988; an estimated $200,000 was billed. Nationwide, 66 centers reported they did not bill insurers for out- patient care, and 46 billed them for less than half of the outpatient care provided in fiscal year 1988. (See ch. 6.) Medical Centers Did Not Of the 159 centers, 122 reported that adequate resources were not avail- able to collect all health care costs from insurers in fiscal year 1988. Commit Resources Needed Officials at the centers GAO visited said that they would not commit additional resources to the recovery effort until they received additional resources from headquarters or were allowed to keep a portion of the collections. (See ch. 7.) GAO recommends that the Secretary of Veterans Affairs ensure that all Recommendations medical centers to bill for outpatient care. In addition, GAO recommends that the Secretary ensure that each medical center has l effective procedures for (1) identifying all veterans with insurance and (2) billing insurers for all inpatient and outpatient care provided to insured veterans and l sufficient resources to fully implement identification, billing, and collec- tion procedures. (See ch. 7.) Given the substantial benefit to the government possible through maxi- Matter for mizing insurance collections, the Congress should ensure VA has ade- Consideration by the quate resources to fully implement its recovery efforts. If adequate Congress resources are not available through the budget process, the Congress should consider amending 38 U.S.C. 629 to allow VA to keep a portion of the amounts collected from insurers to pay its collection costs. (See ch. 7.) VA agreed with GAO'S recommendations and pointed out a number of Agency Comments actions it was taking to increase collections from health insurers, includ- ing establishing a recovery task force for medical care costs. The task force would be charged with developing a strategy to improve identifi- cation, billing, and collection procedures as quickly as possible. Page4 GAO/HRDSO-64VAcOllections FromInsurers JZxecutlveSummary VA recognized that additional funding is needed to properly carry out the billing and collection effort; it proposed legislation in its fiscal year 1991 budget request to provide that funding (see app. X). Page 6 GAO/XiRD-90-64VA Collectiona From Insurers Contents Executive Summary 2 Chapter 1 10 Introduction VA’s Health Care Delivery System 10 VA’s Authority to Recover Health Care Costs 10 VA’s Insurance Recovery Process 11 Objectives, Scope, and Methodology 12 Chapter 2 16 VA Ha Not Mawed VA’s Collections Exceeded Costs 16 GAO Estimates Additional Collection Potential 18 Its Insurance Policy Decisions and Ineffective Procedures Limited VA’s 18 Collections Collections Chapter 3 19 Need to Improve Medical Centers Visited Not Identifying All Insured 19 Veterans Identification Other Medical Centers Using Special Reviews 21 Procedures at the Medical Centers Chapter 4 23 Centers Need Better Medical Centers Missing Substantial Billing Opportunities 23 Internal Controls Lacking to Prevent Missed Billings 25 Billing Procedures to Bills Not Sent Promptly 26 Recover the Costs of Inpatient Care Chapter 5 29 VA Can Substantially Billing for Outpatient Care Optional 29 Opportunities Exist for Substantial Collections 31 Increase Recoveries of Existing Procedures Hamper Outpatient Billing 32 Outpatient Care Costs Page 6 GAO/HRD9O84 VA Collections From Insurers Contents Chapter 6 Medical Centers Did Medical Center Collection Rates Vary Widely Many Reductions in Payments Due to Policy Limitations Not Collect All Collections Reduced Because Insurers’ Requirements Not Amounts Billed to Met Some Insurers Deny Payment Because Medicare Not 38 Insurers Billed First Chapter 7 41 Stronger Management VA Headquarters Guidance Inadequate 41 Medical Centers Did Not Commit Staff Needed 41 Actions Needed to Conclusions 42 Maximize Insurance Recommendations to the Secretary of Veterans Affairs 43 Matter for Consideration by the Congress 44 Collections Agency Comments 44 Appendixes Appendix I: VA’s Daily Rates Used to Prepare Bills to 46 Health Insurers Appendix II: Total Insurance Collections and Collections 47 Per Average Daily Occupied Bed by Medical Center (Fiscal Year 1988) Appendix III: When Medical Centers Began Billing 52 Insurers Appendix IV: Estimating VA’s Potential Collections From 53 Insurers Appendix V: Methodology and Results for Our Review of 56 Samples of Insured Veterans Receiving Inpatient Care Appendix VI: Methodology and Results for Our Review of 59 Samples of Insured Veterans Receiving Outpatient Care Appendix VII: Projected Value of Amounts Collected and 62 Reasons for Less Than Full Payment for Inpatient Care Billed at Five Centers Appendix VIII: Reasons for Denial or Partial Payment of 63 Bills Appendix IX: Factors Affecting Recoveries 65 Appendix X: Comments From the Department of Veterans 66 Affairs Appendix XI: Major Contributors to This Report 69 Page 7 GAO/IUtD9064 VA collections From insurers Contents Tables Table 3.1: Insured Veterans Identified During GAO Visits 20 to Medical Centers Table III. 1: Center Responses 52 Table V. 1: Projected Value of Inpatient Care Provided to 57 Insured Veterans, October 1,1987, Through March 31, 1988, and Not Billed to Insurers Table V.2: Projected Value of Inpatient Care Provided to 57 Insured Veterans, October 1,1987, Through March 3 1, 1988, and Billed to Insurers Table V.3: Centers’ Projected Collection Rates for 58 Inpatient Care Provided, October 1, 1987, Through March 31, 1988, and Billed to Insurers Table VI. 1: Projected Value of Outpatient Care Provided 60 to Insured Veterans, October 1,1987, Through March 31, 1988, and Not Billed to Insurers Table VI.2: Projected Value of Outpatient Care Provided 60 to Insured Veterans, October 1,1987, Through March 3 1, 1988, and Billed to Insurers Table VI.3: Centers’ Collection Rates for Outpatient Care 61 Provided, October 1, 1987, Through March 31, 1988, and Billed to Insurers Table VII. 1: Projected Values Billed to Insurers for 62 Inpatient Care, October 1,1987, Through March 3 1, 1988, Collected and Not Collected From Insurers Figures Figure 1.1: Average Collection per Bed From Private 13 Health Insurers at VA Medical Centers GAO Reviewed (Fiscal Year 1988) Figure 2.1: VA’s Collections From Health Insurers by 16 Quarter (Oct. 1986-Mar. 1989) Figure 4.1: Projected Value of Inpatient Care Provided to 24 Insured Veterans (Oct. 1987-Mar. 1988) Figure 4.2: Average Time Taken to Prepare Bills 27 (Oct. 1987-Mar. 1988) Figure 5.1: Percentage of Medical Centers That Billed for 30 Outpatient Care (Fiscal Year 1988) Figure 5.2: Projected Value of Outpatient Care Provided 31 to Insured Veterans (Oct. 1987-Mar. 1988) Figure 6.1: Projected Value of Inpatient Care Billed 35 (Oct. 1987-Mar. 1988) Page 8 GAO/HRD9O44 VA Cdlection~ From Ixwrem Figure 6.2: Projected Value of Inpatient Care Billed at 36 Five Centers-Proportions Collected and Not Collected (Oct. 1987-Mar. 1988) Abbreviations GAO General Accounting Office HCFA Health Care Financing Administration HMO Health Maintenance Organizations OMB Office of Management and Budget VA Department of Veterans Affairs Page9 GAO/HID-90-64 VACollections From Incurers The Department of Veterans Affairs (VA) operates the largest health care delivery system in the United States. The Veterans’ Health-Care Amendments of 1986 established the right of the United States to recover the costs of health care provided to certain veterans covered by private health insurance. The Ranking Minority Member of the Senate Committee on Veterans’ Affairs requested that we evaluate VA’S recov- ery efforts. VA’s Health Care Most of these medical facilities are organized into 159 medical centers Delivery System throughout the United States. In fiscal year 1988, VA spent about $10 billion providing medical care to veterans, including about 1.1 mil- lion inpatient hospital stays and over 20 million outpatient visits. Veterans eligible for medical care are classified into two broad catego- ries: those with disabilities resulting from their military service and those without such disabilities. Veterans with service-connected disabili- ties are afforded the highest priority when seeking care at VA medical centers. Before 1986, in specified circumstances, VA could recover the reasonable VA’s Authority to costs of care for treatment of nonservice-connected disabilities covered Recover Health Care under a workers’ compensation law, a state law concerning no-fault costs insurance, or a state or local program of compensation for victims of personal violence. In emergency cases, VA could also charge for hospital care or medical services provided to people otherwise ineligible for VA care. VA could not recover from health insurers the cost of care provided to many privately insured veterans because most insurance policies had clauses barring reimbursement to VA facilities. In 1985, we recommended that VA be permitted to recover from health insurers the costs of care provided to privately insured veterans, except care related to service- connected disabilities.’ The Veterans’ Health-Care Amendments of 1986 (title XIX of P.L. 99-272), which became law in April 1986, authorized VA to recover the reasonable costs of care for the treatment of insured veterans who do not have service-connected disabilities (hereafter referred to as insured veterans). The cost of care provided to veterans ‘L@slation to Authorize VA Recoveries From Private Health Insurance Would Result in Substantial savings (GAO~-85-24, Feb. 26,1985X Page 10 GAO/HRD9@64VA Collections From Insmwa Chapter 1 Introduction who had service-connected disabilities could not be recovered, regard- less of whether the care was related to a nonservice-connected condi- tion.2 The law specified that all funds collected by VA be returned to the U.S. Treasury. The 1986 amendments authorized VA to receive payments from health insurers just as non-v.4 health care providers do, except that VA is prohib- ited from receiving Medicare or Medicaid payments. Insurers’ payments to VA can vary considerably depending on the type of insurance. A health insurer generally pays the covered health care costs less the patient’s coinsurance (percentage of expenses the patient must pay ) and deductible amounts. The law prohibits VA from seeking reimburse- ment for any coinsurance or deductible amounts required by health plan contracts from insured veterans. Other types of insurers, such as health maintenance organizations, would pay VA for care in limited situations, for example, a medical emergency. Still other insurers pay the policy- holders (with indemnity policies) fixed amounts per day of hospitalization. Recovering from health insurers is a three-step process-identifying VA’s Insurance veterans with health insurance, billing insurers for care, and collecting Recovery Process amounts owed. Within each medical center, the medical administration service has responsibility for identifying insured veterans and billing for their care, and the fiscal service has responsibility for collecting unpaid bills and recording collected amounts. Counterpart units at headquarters give technical assistance and general guidance. Medical centers were given latitude to develop many of their own specific procedures for identifying insured veterans, billing insurers for care, and collecting amounts owed. Each year, VA develops national daily billing rates, which cover all related costs of care, including room and board, physicians’ costs, ancil- lary services, and all interest and support costs. VA calculates the amount to bill insurers for inpatient care by multiplying the number of days of care by the daily rate. Different inpatient rates are developed for specific types of care, for example, surgery or psychiatry. For out- patient care, VA charges a fixed rate. VA fiscal year 1988 and 1989 billing rates for most types of care are listed in appendix I. *In March 1989, the Ranking Minority Member of the Senate Veterans’ Affairs Committee introduced a bill (S. 573) that would authorize VA to recover the cost of care for nonservice-connected disabilities of insured veterans who also have service-connected disabilities. This change would be in conformity with GAO’s 1985 recommendation to recover for such care. Page 11 GAO/HBD9064 VACdlectio~ Prom Insurers Chapter 1 introduction Senator Frank H. Murkowski, Ranking Minority Member of the Senate Objectives, Scope, and Committee on Veterans’ Affairs, asked us to determine whether VA'S col- Methodology lections from insurers were exceeding its costs to collect. In addition, he asked us to evaluate whether VA had implemented effective procedures to (1) identify veterans with health insurance coverage, (2) bill insurers for all care provided to these veterans, and (3) collect amounts owed. In addition, we estimated VA’S potential collections from health insurers nationwide in fiscal year 1988. To accomplish our objectives, we (1) reviewed relevant policies, proce- dures, and studies and (2) interviewed VA officials at headquarters, regional offices, and medical centers. We also visited six medical centers to assess the effectiveness of VA’S recovery process. The centers were judgmentally selected to gain a wide mix for several key factors, such as (1) size (number of beds), (2) settings (for example, rural or urban), and (3) collections per average daily occupied bed.3 Appendix II contains fis- cal year 1988 information on the factors considered for each medical center. As shown in figure 1.1, the centers were located in five of the seven VA medical regions. 3We used collections per average daily occupied bed rather than total cokctions because the average daily collection incorporated the size of the center into the assessments. Page 12 GA0/HlW90-64VA(l!d.lectiona FromInsurers -- Chapter 1 Introduction Figure 1.1: Average Collection per Bed From Private Health insurers at VA Medical Centers GAO Reviewed (Fiscal Year 1988) Seattle, WA r~ Region 6 Minneapolis, MN A Region 5 1 w3521 I I Lb- , , Region 4 L’ Albanv. NY Martinsbura. WV Region 2 - ($10,378) Bay Pines, FL Region 3 ($1,039) L Note. The average collect!on amount is the total received from pnvate health Insurers dlvlded by the average number of occupied beds. The average collection amount for all medical centers was $1,910. At each of the six medical centers, we interviewed officials in the office of the director, medical administration service, and fiscal service to dis- cuss the internal control procedures for identifying insured veterans, billing insurers for care, and collecting from insurers. We also discussed factors that aided or impeded the centers’ recovery efforts. To test the effectiveness of each center’s procedures for identifying insured veterans, we worked with staff at five of the centers to conduct a special record review for veterans (1) without service-connected disa- bilities and (2) admitted to the centers but not identified as having Page 13 GAO/HRD90-64VA Cdkctions From Insurers Chapter 1 Introduction insurance.4 The review focused on identifying key indicators of insur- ance, such as employment, income, or previous insurance coverage. If possible insurance coverage was indicated, we asked the VA reviewer to interview the veterans before they were discharged to determine whether they had insurance. The test period at the five centers ranged from 14 to 35 days, depending in part on the availability of center staff to conduct the review. To assess each center’s billing and collection procedures, we reviewed medical and administrative records for a random sample of about 30 insured veterans from each of two universes:5 9 veterans discharged from inpatient care between October 1,1987, and March 31,1988,6 and . veterans provided outpatient care between October 1, 1987, and March 31, 1988. For those patients sampled, we documented (1) whether the medical center billed for all episodes of care provided to insured veterans and sent the bills promptly, (2) how much the center collected, and (3) the reasons for collecting less than the full amount. We contacted the staff of VA’S district counsel, which served several of the medical centers, to discuss the policies and procedures for dealing with insurance cases referred to them. We used our sample results to project the value of inpatient and outpa- tient care not billed by the medical centers. For our projections, which are for the first 6 months of fiscal year 1988, we used a 95-percent con- fidence level. Because our estimates are based on samples, sampling errors are associated with them. To assess whether VA’S collections from insurers exceeded its costs to collect, we used a questionnaire for information on (1) the medical cen- ters’ estimated staffing costs and (2) the amount of collections attrib- uted to inpatient and outpatient care. We asked each of the 159 centers to complete a questionnaire because VAdoes not routinely collect these 4The sixth center, Martinsburg, was already using this special record review to identify insured veterans. 5We did not include veterans with indemnity policies or health maintenance organization coverage as insured veterans because of the limitations in reimbursement. “At the Albany Medical Center in New York, we limited our review of veterans’ records to veterans who received outpatient care. Page 14 GAO/HRD90-64 VAcOUectionsProm Insurers Chapter 1 Introduction data. All centers (1) identified by grade the staff resljonsible for billing and collecting from insurers and (2) estimated the average amount of time spent on recovery efforts for inpatient and outpatient care during a typical week in fiscal year 1988. We used the government salary rates to compute the staffing costs for this typical week and multiplied the costs for each center by 52 weeks to estimate the annual staff costs of VA med- ical centers in fiscal year 1988. The centers also estimated the percent- age of their total collections that were related to outpatient care. Using these percentages, for each center, we estimated the amount of collec- tions attributed to inpatient and outpatient care. We also used the questionnaire to collect a broad range of information on each center’s identification, billing, and collection policies and proce- dures. To gather information on industry norms and standards for bill- ing and collection procedures, we contacted the Healthcare Financial Management Association, a professional society of health care financial executives, and visited a private hospital in Seattle. To estimate VA’S potential collections from insurers nationwide in fiscal year 1988, we obtained data from VA’S 1987 Survey of Veterans.’ On the basis of survey respondents’ reported health insurance coverage and use of VA health care facilities, we estimated the total number of episodes of inpatient care and outpatient visits made by veterans with health insur- ance coverage. By applying to these estimates VA’S reported average length of inpatient stays and daily rates for inpatient and outpatient care, we estimated the total amounts potentially billable by VA for this care. We reduced the total amounts potentially billable to allow for uncollectible amounts because of insurance policy limitations, such as deductibles and copayments. Our review was carried out from December 1987 through June 1989 in accordance with generally accepted government auditing standards. 7A national survey of veterans conducted for VA by the U.S. Bureau of the Census to assessveterans’ status and well-being. Page 16 GAO/IlRD90-64 VA C~llestions Prom Insurer6 ChaI”““ 2 VA Has Not M aximized Its Insurance Collections VA'S collections from health insurers exceeded its costs in fiscal year 1988. But GAO estimates that the medical centers’ collections should have been significantly higher. Policy decisions and ineffective collec- tion procedures at the centers contributed to missed collection opportu- nities. To maximize collections, medical centers should (1) begin billing insurers for all outpatient visits and (2) develop better procedures to identify veterans with insurance and to ensure that all episodes of care are billed. The 159 medical centers spent about $8 million to collect $100 million in VA’s Collections fiscal year 1988. VA medical centers reported insurance collections of Exceeded Costs $24 million in fiscal year 1987, the first full year in which VA collected from insurers. The reported collections increased to $100 million in fis- cal year 1988, and collections started to level off during the last quarter of the fiscal year. Compared with fiscal year 1988, collections have remained relatively constant during the first 2 quarters of fiscal year 1989, as shown in figure 2.1. Figure 2.1: VA’s Collections From Health Insurers by Quarter (Oct. 1986-Mar. 1989) 40 Ddhrs in Yfflia~~ 20 10 0 Dvz. ‘86 Mar. ‘87 Jun.37 Sapt.‘87 Dac.‘87 hr.‘88 Jun. ‘80 Sopt.‘88 Doc.%B Mar. ‘89 Qartots Ending Note VA collected $302,000 In the quarter endmg Dee ‘86 Page16 GAO/IiEMW64VAColl~ona FromInsurers Chapter 2 VAHasNotM ’ * dIte IMurance collectl0M The medical centers have spent a relatively small amount of money to collect from insurers. On the basis of the information that centers sub- mitted, we estimate that during fiscal year 1988, they spent about $8 million in staffing costs to bill insurers and to collect amounts owed. Although some centers reported other costs, such as for postage, the staffing costs were the vast majority of the reported costs. Given these reported collections and costs, on a nationwide basis, VA collected about $12 for every $1 spent on medical center staff .l Most of the collections-$96 million of the $100 million-came from inpatient care. This represents a return of $16 collected for each $1 spent on medical center staff. By comparison, centers’ collections for outpatient care were an estimated $4 million because VA did not direct centers to bill insurers for such care. Most centers reported that in fiscal year 1988, they either billed insurers for outpatient care on a limited basis only or did not bill. Nonetheless, the centers collected, on average, about $2 for each $1 spent on staff costs for billing. The years when centers began billing insurers for outpatient and inpatient care is shown in appendix III. VA has conducted a pilot study at the Martinsburg Medical Center that may provide a more accurate indication of the payback potential of bill- ing for outpatient care. VA headquarters authorized three staff positions for this study,2 which began on August 22, 1988. Through March 31, 1989, Martinsburg had collected about $74,000 from insurers and spent about $18,000 on personnel costs, a return of over $4 for each $1 spent. Like most other medical centers, Martinsburg has not billed for all out- patient care provided, but the collections-to-cost ratio at Martinsburg has been increasing as the center has gained experience in billing insur- ers for such care. ‘VA’s reported collections for fiscal year 1988 could include some receipts for bills sent in fii year 1987. In addition, collections for some bills sent in foal year 1988 may be included in VA’s fiscal year 1989 reported collections. We believe that this will not materially affect the results of our analysis because the amounts involved would tend to offset each other. ‘These staff, the chief of insurance billing said, also spent time preparing other types of bills. The cost amounts shown here are the costs associated with the portion of staff time spent on billing insurers for outpatient care. Page 17 GAO/HIUMOW VA Collection Prom Insurers chapter 2 VAHasNotMaxhizdIt Inmlmnce coIlectlone On the basis of our work at six centers, we could not estimate potential GAO Estimates collections from insurers nationwide. We, therefore, estimated VA’S Additional Collection potential collections from insurers on the basis of veterans’ responses to Potential VA’S 1987 Survey of Veterans. The survey asked veterans about their insurance coverage and the extent to which they received inpatient and outpatient care at VA facilities. Using veterans’ responses to these questions, as well as average length of inpatient stays reported by VA and VA’s daily rates, we estimate that VA could have collected about $323 million from insurers in fiscal year 1988. Of this amount, about $276 million was attributable to inpatient care and $47 million to outpatient care. Given VA’S collections of $100 million, we estimate that VA could have collected an additional $223 million-$180 million for inpatient care and $43 million for outpa- tient care. Limitations in the available data cause considerable uncertainty in esti- mates of potential collections. Nevertheless, we believe VA could have increased collections by at least $55 million and perhaps as much as $392 million. Our method for estimating VA’S potential collections from insurers is discussed in appendix IV. VA did not bill insurers for millions of dollars because of (1) weaknesses Policy Decisions and in procedures at the medical centers or (2) policy decisions not to bill Ineffective Procedures insurers for all outpatient care provided. The weaknesses in procedures at the centers we visited resulted in centers not Limited VA’s Collections . identifying all veterans who had insurance coverage, . billing insurers for all episodes of inpatient care provided to insured vet- erans, and . collecting all that they should have even when the care was billed. We believe, on the basis of centers’ responses to our questionnaire, that these six centers are not unique. In response to our questionnaire, many other centers reported procedures similar to those used at the centers visited. In the following chapters, we discuss in more detail the proce- dures used. Page 18 GAO/IiRDW VA Collections From Insurers Need to Improve Identification Procedures at the Medical Centers Of five medical centers visited during our test period of 14 to 35 days, four did not identify all veterans with insurance. We reviewed available data and found that the centers had not identified many insured veter- ans during the initial admissions screening. As a result, the centers missed billing insurers for over $300,000 in care provided to these veter- ans during that period. This problem may be widespread since many other medical centers, in responding to our questionnaire, said they did not review available data after the initial admission screening to iden- tify insured veterans. During the initial admission screening, each veteran requesting medical Medical Centers care must complete an application for care.’ This application contains a Visited Not number of questions designed to gather basic data about the veteran, Identifying All Insured including whether he or she has health insurance. In identifying veter- ans with insurance, of the six medical centers we visited, five relied Veterans mainly on information that the veterans gave during the initial screen- ing process. To test the effectiveness of each center’s procedures for identifying insured veterans, we did a special review, during our site visits, of the records of veterans admitted to the five centers, but not identified as having insurance. The review focused on identifying key indicators of insurance, such as employment, income, or previous insurance coverage. When we found these indicators, we asked the VA reviewer to interview the veterans before they were discharged to determine whether they had insurance. We did not do a special review at the Martinsburg Medical Center, in West Virginia, because this center routinely did such a review. Our ran- dom sample of 32 insured veterans admitted to Martinsburg included 6 identified by the center’s special review as having insurance. If Martins- burg relied solely on the initial screening process, these veterans would have been missed. On the basis of these sample results, we estimate that the insured veterans identified through this review received an esti- mated $864,000 of the medical care billed during only the first half of fiscal year 1988. Martinsburg officials have found that this review con- tinues to be an effective way to identify insured veterans. The chief of insurance and billing, whom we asked to record the results of this ‘At some medical centers, veterans complete the applications and the medical center staff later enter the information into the center’s computer system. At other centers, medical administration service staff interview veterans and enter the information obtained directly into the computer system. Page 19 GAO/HBD9064 VAcOllections From Insurers Chapter 3 Need to Improve Identifkation procedurea at the Medical Centers review during a $-month period in fiscal year 1989, reported that she identified 27 insured veterans. These veterans received almost $225,000 of care that the center billed to insurers. The other centers we visited identified 30 insured veterans during the test period; these veterans had not been identified as having insurance during the initial screening process. They received about $3 18,000 of inpatient and outpatient care that the centers should have billed to insurers. Although the centers would probably not have collected the total amount billed because of insurance deductibles and coinsurance requirements, we believe that the results indicate significant additional collection opportunities, especially given the limited test period. Our test results are shown in table 3.1. Table 3.1: Insured Veterans Identified During GAO Visits to Medical Centers insured Additional veterans insured Value of identified veterans additional through identified veterans’ Days test existing through test medical Center conducted procedures procedures care Minneapolis 23 43 16 $121,317 Bay Pines 27 57 2 77,597 Albanv 14 18 6 76.809 West Los Angeles 35 30 6 42,656 Seattle 19 22 0 0 Total 118 170 30 $318,379 Supervisors and managers at the centers could not tell us conclusively why veterans with insurance were slipping through the initial screening process, but they offered two potential reasons. Not all clerks, said the processing unit (admitting area) supervisor at one center, ask veterans about their insurance coverage each time they are admitted to the center, even though the clerks should; if the veterans have applied for care in the past and previously indicated no insurance, the clerks do not always ask whether the veterans currently have insurance coverage. As a result, the information obtained may not be complete and accurate. In addition, center staff have suggested, veterans may not answer ques- tions about their insurance accurately, even if asked, because they (1) are confused or (2) fear that disclosure of insurance will affect their access to free health care at VA medical centers. Page 20 GAO/HlUHM4 VA CMlection~ From Insurers chapter 3 Need to Improve Identification Proceduresat the Medical Centers We believe our test understates additional collection opportunities at the centers because veterans’ applications were often incomplete. For exam- ple, at the medical center in Seattle, the employer was shown as “unknown” in 81 percent of the applications we reviewed during an 8-day period.2 The lack of employment information reduced our chances of identifying additional veterans with insurance. We also found missing data at other centers we visited. For example, at three centers, from 50 to 69 percent of the applications for our sample of veterans were missing employment information. This information is useful because many employers or former employers provide health insurance coverage, and employment information can serve as an indica- tor of insurance. At the West Los Angeles Center, the insurance coordi- nator did a special review of applications to evaluate the effectiveness of the initial screening process as a method of gathering information. In reviewing applications for a 32-day period, he found that 56 percent were missing employment information. After we did our test at the West Los Angeles Medical Center, the center successfully used the special review to identify additional insured veter- ans. The insurance coordinator reported that in December 1988, several months after the test period, 24 insured veterans were admitted to the center. The admissions staff identified 11 of these veterans. The insur- ance coordinator said that he identified the other 13 veterans by con- ducting a special review of veterans’ records and interviewing veterans, employers, and insurance companies. Other Medical Centers cial review. We contacted several of the medical centers that had Using Special Reviews reported, in their responses to our questionnaire, that they reviewed veterans’ records for indicators of insurance coverage. The medical center in Bonham, Texas, according to a center official, using this special review, identified 25 percent of all veterans with insurance. By using this review, an official at the medical center in Chilicothe, Ohio, said, the center significantly increased its collections. However, the center did not record the number of insured veterans identified using this special review. 2For 8 days during our 19-day test period, we reviewed the records to determine how many contained employer data. Page 21 GAO/HID-90-64 VA Collections From Insurers chapter 3 Need to Improve Identification Proceduresat the Medical Centers Despite the usefulness of the special review, VA had not required centers to use it, at the time of our review, and most centers were not doing so. On the basis of responses to our questionnaire and follow-up conversa- tions with medical center officials, we estimate that about two-thirds of the centers were not reviewing the records of “uninsured” veterans to check for indicators of possible insurance. Page 22 GAO/HBB9044 VA Collection From Insurers Chapter 4 Centers Need Better Billing Procedures to Recover the Costs of Inpatient Care The VA medical centers we visited were often not billing insurers for care because of ineffective procedures. During the first half of fiscal year 1988, we reviewed inpatient billing for five centers; these centers missed billing insurers for about $1.8 mil1ion.l In addition, bills, when sent, were rarely processed promptly, resulting in additional lost revenue to the government. Even when medical centers identified veterans as insured, the centers Medical Centers did not always bill insurers for all episodes of care provided. At five Missing Substantial centers, we reviewed random samples of insured veterans who were dis- Billing Opportunities charged from inpatient care during the first 6 months of fiscal year 1988. On the basis of this review, we estimate that these centers should have billed insurers for approximately $7 million of inpatient care instead of the $5.2 million we projected they billed. Our projections for each center of the billed and unbilled amounts for insured veterans are shown in figure 4.1. Appendix V contains more detailed information on our methodology. ‘There is a 95-percent chance that the true value for the population lies between $1.2 and $2.3 million. Page 23 GAO/HBDM VA Collections From Insurers Chapter 4 Centers Need Better Billing Procedures to Recover the Cost8 of Inpatient Care Figure 4.1: Projected Value of Inpatient C&e Provided to Insured Veterans (Oct. 1987.Mar 1988) 2500 Dolhnhllhouan& 2250 Zwo 1750 1500 1260 1000 750 so0 250 0 VA Ysdksl Csntws I I 1 Amount Billed I Amount Not Billed We cannot project how much the five centers would have collected had they billed insurers for these additional amounts; however, the centers could collect about $850,000 if, for the veterans in our samples, the col- lection rates for the unbilled care were the same as for the billed care. As discussed in chapter 6, we believe these collection rates and the cor- responding dollar recoveries could be increased with improved procedures. Nationwide, we estimate that medical centers could have collected another $29 million to $332 million from insurers for inpatient care in fiscal year 1988 (see ch. 2). Page 24 GAO/lHUMNM VA Cdlections FkomInsurers chapter 4 Centers Need Better Billing Proceduresto Recover the Costs of Inpatient Care Four of the five centers had problems with inpatient billing procedures, Internal Controls as indicated by the amount of missed billings we found. The centers Lacking to Prevent used varying procedures to notify billing staffs about care provided to Missed Billings insured veterans. However, three centers with high missed billings relied mainly on admissions or insurance clerks to notify billing staff when insured veterans were admitted. These centers lacked basic inter- nal controls as a backup system to help ensure that all cases were billed. Hence, when admissions or insurance clerks, for whatever reasons, did not carry out their responsibilities to notify billing clerks, care was not billed, for example: . At the West Los Angeles Medical Center, an insurance coordinator was expected to (1) verify insurance coverage for veterans identified as hav- ing insurance and (2) then send a notification to the billing staff. Because the applications and insurance information forms were often not available, the coordinator said he was unable to notify the billing staff that some insured veterans were receiving care. For example, dur- ing a 26-day period, the administrative folders or insurance forms were unavailable for 65 percent of the insured veterans admitted to the center. The center’s quality assurance staff, who previously carried out some of the insurance coordinator’s functions, agreed that the inability to locate veterans’ medical records promptly directly contributed to missed billings. At the Minneapolis Medical Center, the admitting staff were supposed to send a notification to the insurance clerk when insured veterans were admitted, and the clerk was then to notify the billing staff. For 4 veter- ans in our sample of 36, insurers were not billed using these procedures, even though the veterans’ insurance coverage was identified when they were admitted. Although we could not determine where the specific breakdown occurred for each case that was not billed, we did find that the centers could use their existing computer data more effectively to identify care that should be billed. For example, at the Seattle Medical Center, from computer information, the billing clerk periodically obtained a list of care provided to veterans that the center had identified as insured. Even though, the clerk said, this allowed her to identify care that should be billed, not all care was billed because she did not have enough time to keep up with the billing workload. Medical centers could use such lists, not only to facilitate billing, but also as an internal control to help ensure that all care was billed. When we Page 26 GAO/HRW3OM VACdlections From insurers Chapter 4 Centers Need Better Bii Proceduresto Beeover the Costa of Inpatient Care used similar lists to review records for a random sample of insured vet- erans, we found that significant amounts of inpatient care had not been billed. The Federal Claims Collection Act and implementing regulations require Bills Not Sent that agency collection efforts be prompt and aggressive. Centers should Promptly bill insurers, VA’S manual states, at the end of the first month when med- ical care was provided, which would be, at most, 30 days after the care was provided. Guidelines from the Healthcare Financial Management Association, which publishes bench marks for prompt billing and dis- seminates the data to hospitals nationwide, suggest billing within 7 days after providing care. Of the five centers we reviewed, only one billed insurers for veterans in our sample within 30 days, on average, of providing inpatient care.2 The other centers took an average of 64 to 175 days after the veteran’s dis- charge date to prepare and send bills to insurers. (See fig. 4.2.) Three of the centers-Bay Pines, Seattle, and West Los Angeles-did not send any of the bills within 30 days of discharge for the veterans we sampled who received inpatient care. 2The number of bills in our sample ranged from 12 bills sent by the West Los Angeles Medical Center to 53 sent by the Martinsburg Medical Center. Page 26 GAO/‘HRDM VACollections From Insurers Chapter 4 Centers Need Better Billing Proceduresto Eecover the Casts of Inpatient Care Figure 4.2: Average lime Taken to Prepare Billa (Oct. 1987-Mar. 1988) 200 Avorags Chys Fmm Dbthrge to Bllllng 175 l- 150 125 100 75 ---l-h 54 25 0 8 s+ VA Medical Cenlom When billing is not prompt, the government loses the use of the money during the time it is owed.3 To estimate the loss to the government due to delayed billing, we used the January 1988 U.S. Treasury Bill interest rate of 6.74 percent. We calculated the interest lost from 30 days after the date of discharge to the date the bill was sent. On the basis of this standard, we estimate that the four centers that billed, on average, more than 30 days after the veterans were discharged lost about $38,000 for fiscal year 1988 inpatient collections.4 In other words, we estimate that the government lost almost $31 per episode of inpatient care billed at these centers.5 According to center officials, one problem contributing to the delays in bill preparation was the completion of the medical discharge summary, which billing staff use to prepare the insurance bill. VA’S manual 3cOkctions may also be reduced because an insurer sometimes refuses to pay the amount owed if the time between the date of care and the date the insurer receives the bill exceeds the insurer’s accepta- ble time frame. This problem is discussed in more detail in chapter 6. 4There is a 95-percent chance that the true value of the population lies between $19,000 and 858,000. “There is a 95-percent chance that the true value of the population lies between $15 and $46. Page 27 GAO/HRD90-64 VA Cdlectiona From Insurers Chapter 4 Centers Need Better Billing Procedureato Eeeover the Costa of Inpatient Cure requires billing staff to include a copy of the discharge summary with bills to insurers. A discharge summary, prepared by a physician, records the patient’s medical condition, admission and discharge dates, and types of treatment provided. VA criteria state that these summaries should be completed within 6 working days after the patient’s dis- charge. But at five of the centers we visited, medical administration ser- vice staff sometimes had to wait from several weeks to several months, they said, for the summaries to be completed. At the Martinsburg Medical Center, which had the lowest average number of days between veterans’ discharges and preparing the bills, officials were able to minimize the number of days needed to prepare the bills, they said, by l notifying staff in departments, such as ward administration and tran- scription, which patients have insurance so that the paperwork for these veterans can be expedited and . emphasizing the importance of billing insurers, setting an informal goal of 10 days from the discharge date to the date the bill is sent. Page 28 GAO/HED9084 VA Collections From Insurers Chap tlr .5 VA Can Substantially Increase Recoveriesof Outpatient Care Costs VA headquarters did not direct medical centers to bill insurers for outpa- tient care during fiscal year 1988. In the absence of such direction, most centers reported that in fiscal year 1988, they limited the extent of their outpatient billing or did not bill insurers for outpatient care at all. We estimate that the six centers we visited did not bill insurers for about $1.6 million of outpatient care provided during the first half of fiscal year 1988. The Veterans’ Health-Care Amendments of 1986,38 U.S.C. 629(a)(l), Billing for Outpatient established the right of the United States to recover health care costs Care Optional provided to certain insured veterans. Although headquarters directed medical centers to bill insurers for inpatient care, each medical center was permitted to decide whether to bill insurers for outpatient care. Of the six centers we reviewed, three did not bill insurers for any outpa- tient visits and three billed insurers for some of the outpatient care pro- vided.’ The Martinsburg Medical Center billed insurers for some outpatient care on a limited test basis, according to the chief of the med- ical administration service at the center. We found that Martinsburg billed insurers for 2 percent of outpatient costs for insured veterans in our sample.2 The center has now established procedures, the chief of insurance billing told us, to bill insurers for all outpatient care. The Albany Medical Center billed insurers for outpatient care, according to the billing staff, when (1) the veteran had received inpatient care, (2) the veteran came to the center for outpatient care without an appoint- ment, or (3) the veteran’s income exceeded an income threshold and the veteran agreed to make a payment to VA for such care. We found that Albany billed insurers for about 22 percent of the outpatient care pro- vided to insured veterans in our sample.3 The Seattle Medical Center had procedures to bill for all outpatient care, but the center’s billing clerk did not, she said, have time to bill every case. Hence, she gave first priority to billing outpatient care provided to ‘VA defines an outpatient visit as all diagnostic and therapeutic services provided to a veteran during a single 24hour period. The appearance of a veteran at the facility solely for a prescription refill is not a visit. In fiscal year 1988, VA charged insurers $127 per outpatient visit. 2There is a 95-percent chance that the true value for the Martinsburg population lies between 0 and 5 percent. 3There is a 95-percent chance that the true value for the Albany population lies between 7 and 40 percent. Page 29 GAO/liRBW VA Collectiona From Insurers chapter5 VA Can Substantially Increase Recoveriesof Outpatient Care Cod8 insured veterans under the age of 65, which she believed had higher collection potential4 We found that Seattle billed insurers for 34 percent of the outpatient costs provided to insured veterans in our sample.6 In response to our questionnaire, most medical centers said they billed only for a limited amount of outpatient care or did not bill for any out- patient care during fiscal year 1988. For example, 29 centers said they limited outpatient billing to veterans who had also received inpatient care that had been billed to their insurers. The centers’ responses con- cerning the extent to which they billed for outpatient care in fiscal year 1988 are summarized in figure 5.1. Figure 5.1: Percentage of Medical Centers That Billed for Outpatient Care Dii Not Bill for Any Visits (Fiscal Year 1988) Billed Lees Than Half of the Tatal Visits - Billed Half or More of the Total Viila Total of 159 MedicalCenters Note: The total number of medical centers is 159. 4A veteran under the age of 65, unless disabled, would not have both Medim and privatehealth insurance. As discussed in chapter 6, medical centers have experiencedproblems collecting from some insurers when the veterans were also covered by Medicare. 5There is a 95-percent chance that the true value for the Seattle population lies between 13 and 63 percent. Page 30 GAO/HRB9O84 VA Cdlection~ From Ineurem -. chapter 5 VA Can Substantially increase Recoverie6of Outpatient Care Chstz3 We project that the six centers could have billed almost $1.8 million for Opportunities Exist outpatient care instead of the $200,000 that was billed during the first for Substantial half of fiscal year 1988 (see app. VI for more detailed information about Collections our methodology and projections).” Cur projected outpatient care costs provided to insured veterans and the projected amounts not billed at each center are shown in figure 6.2. Figure 5.2: Projected Value of Outpatient 1 Care Provided to Insured Veterans (Ott 1987.Mar 1988) DdlamInlhounnd6 400 200 --b 100 VA Ydkd Conlors Amount Billed Amount Not Billed It is difficult to estimate the amounts that the six centers could expect to collect from the insurers for this care because of the centers’ limited col- lection experience. Centers would not collect the entire amount billed because of insurance deductibles and coinsurance amounts. If the six centers had billed for all outpatient care and were able to collect at the same rate as the Albany Medical Center did for our sample of veterans 6There is a 95-percent chance that the true value for the population lies between $1.1 and $2.0 million. Page 31 GAO/HRDW64 VA CMlectiona From hsurers Chapter 5 VA Can Substantially Increase Recoveriesof Outpatient Care Costa (37 percent), these centers could collect over $575,000 of the projected amount not billed during a 6-month period. If the six centers expe- rienced the same collection rate as the Seattle Medical Center did for our sample of veterans (60 percent), these centers could collect almost $934,000 of the projected amount not billed. Appendix VI contains information on the amounts collected at the three centers that billed insurers for outpatient care. Nationwide, we estimate medical centers could have collected another $26 million to $60 million from insurers for outpatient care in fiscal year 1988 (see ch. 2). The three centers we visited that were billing for outpatient care relied Existing Procedures on labor-intensive procedures to verify and document care provided. Hamper Outpatient Basically, billing personnel at these centers were reviewing each insured Billing veteran’s medical records to (1) verify the date(s) of care, since person- nel had no other means to assure that veterans actually received care for which they were scheduled, and (2) document for the insurer the type(s) of services provided, as shown on the medical notes written by health care providers. Detailed chart reviews are time-consuming, said center personnel who did them, and ultimately limit the amount of out- patient care center personnel can bill. The elimination of labor-intensive chart reviews would give center per- sonnel more time to bill outpatient care. The private sector uses a preprinted treatment form that documents the date of care, diagnoses, tests, and treatments performed by the health care provider. If VA used such treatment forms in outpatient clinics to document the care pro- vided to insured veterans, it could eliminate the time-consuming chart reviews it now uses. At the conclusion of the insured veteran’s clinic visit, the health care professional would check the appropriate boxes on the form. Ultimately, clinic staff would return the form to the billing staff, who could use it to bill the insurer. The use of this form would not add significantly to the clinics’ workload because it (1) includes a simple check-the-box approach and (2) would only need to be completed for insured veterans. The chief of the medical administration service at the Martinsburg Medical Center thought that implementing this step would streamline the outpatient billing process. To make it easier for centers to bill outpatient care provided to insured veterans, VAheadquarters officials are spearheading another effort- Page 32 GAO/llRD90-64 VACdlectione Prom Insurers chapter 6 VACan Subetantidly Increase Recoveries3of C)lrtpfitlent Care costs developing an automated billing system -which it hopes will allow cen- ters to handle a larger volume of bills. When the automated billing sys- tem becomes operational, said the former chief of the policies and procedures division in headquarters, headquarters will expect medical centers to bill for all outpatient care. The first phase of the system, installed in October 1988, allows billing clerks to use some data already in the computer, such as the insurer’s address, to prepare a bill. The clerk must manually enter information on the dates and costs of inpa- tient or outpatient care. The accounts receivable system will help staff track bills and generate follow-up letters for overdue bills. We question whether the automated system will significantly reduce the staff time needed to prepare outpatient bills because the billing staff will continue its labor-intensive procedures to verify and document the care provided. Page 33 GAO/HRD9O84 VA Collections Prom IMWWE Medical Centers Did Not Collect All Amounts Billed to Insurers In part because of insurance policy limitations applicable to all provid- ers, the centers did not always collect the full amounts billed. However, we found a small percentage of cases in which centers did not collect as much as they should have because they did not meet specific policy pro- visions, for example, submitting bills promptly. In addition, some insur- ers denied payment for veterans covered by both Medicare and private health insurance because Medicare was not billed first. VA plans to liti- gate cases denied by these insurers to clarify VA'S legal authority to recover the cost of care in such instances. At the five centers we visited, the collection rates varied widely for our Medical Center samples of veterans. The rates for inpatient care ranged from 25 percent Collection Rates Vary to 87 percent for the veterans who received inpatient care between Widely October 1987 and March 1988.’ These rates may be somewhat conserva- tive because centers had outstanding bills that they were still trying to collect when our field work was completed. For example, some centers had referred uncollected bills to their district counsels’ offices. If the centers ultimately collect some or all of the amounts owed on outstand- ing bills, the collection rates would increase. The projected amounts billed and collected for inpatient care at the five centers are shown in figure 6.1. ‘Appendix V contains more detailed information about the collections per veteran and cokction rates. Page 34 GAO/HBD9064 VA Collections F’romInsurers Chapter 6 Medical Centers Did Not Collect All Amounts Billed tu Insurers Figure 6.1: Projected Value of Inpatient Care Billed (Oct. 1987-Mar. 1988) DollaraIn lhounnh VAMdkal Ceders I Amount Collected Amount Not cOllected To determine the reasons that the centers did not collect the full amounts billed, we reviewed the inpatient amounts billed and collected for our sample of veterans. The proportions collected and the primary reasons for reductions in payment at the centers we visited are shown in figure 6.2. Page 36 GAO/HBD9084 VACollections Prom Imsurers chapter” Medicai Centers Did Not Collect All Amounts Billed to Insurers Figure 6.2: Projected Value of Inpatient Care Billed at Five Centers-Proportions Collected and Not Collected (Oct. 1987- Policy Limitations Mar. 1988) 3% Insurers’ Requirements Not Met F Collected From lnsurets For about 13 percent of the amounts billed, we were not able to deter- mine the reasons for the reductions in payment. This is because (1) the insurers’ responses did not always state the reasons for less than full payment, (2) VA could not readily provide the insurers’ explanations of the amounts paid, or (3) the insurers had not yet responded. For cases in which we did not have an explanation for the payments from insurers and the centers collected at least 70 percent of the amounts billed, we assumed the unpaid portions were for policy limitations related to deductibles and coinsurance. More information about figure 6.2 is given in appendix VII. We also reviewed the projected amounts billed and collected to deter- mine if there were differences between the centers that would explain the differing collection rates. At the Martinsburg Medical Center, we found that insurance policy limitations accounted for the majority of the amounts not collected. In contrast, the Bay Pines Medical Center had large projected amounts denied because Medicare was not billed first. The reasons for -reductions in payments are discussed in the following sections. Page 36 GAO/ERLh9084 VACollections From Insurers Chapter 6 Medical Centers Did Not Collect All Amounts Billed to Insurers The Veterans’ Health-Care Amendments of 1986 established the right of Many Reductions in the United States to recover payments from insurers for health services Payments Due to provided to veterans to the same extent as insurers paid to non-VA Policy Limitations health care providers. VA would not collect the entire amount billed if a veteran’s insurance policy contained (1) restrictions on the types of ser- vices covered or amounts the insurer will pay or (2) requirements that the beneficiary pay a deductible or coinsurance. The five centers we visited did not collect an estimated 15 percent of the amounts billed because of policy limitations, and many other medical centers reported in our questionnaire that insurers paid less than the billed amounts for the same reason (see app. VIII). The three most fre- quently cited reasons insurers gave, as reported by centers in our ques- tionnaire, are these: l Eighty-three centers reported that insurers often paid only a portion of the amount billed because the insurer offset a portion of the charges against the veteran’s insurance deductible or coinsurance or both. Forty centers reported that they often received no payment for this reason. . Forty-eight centers reported that insurers often paid only a portion of the amount billed because a portion of the charges for services provided exceeded the limitations of the insurance policy. . Thirty-seven centers reported that insurers often paid only a portion of the amounts billed because the policy did not cover some of the services provided. Twenty-eight centers reported that they often received no payment because the policy did not cover the services provided. Sev- enty-six others reported that this sometimes occurred. These reasons do not indicate that VA was treated differently from other medical care providers. These providers, however, can bill some benefi- ciaries for any difference between the amounts billed and the insurance payments, but the amendments do not authorize VA to bill the veteran for this difference. Further, the amendments prohibit VA from collecting deductible or coinsurance amounts under health plan contracts between veterans and insurers. In these cases, VA cannot expect to collect the full amounts billed. Some insurance policies require that the medical care provider take cer- Collections Reduced tam steps in order to collect the maximum amounts allowed under the Because Insurers’ policies. We found that insurers sometimes paid only portions of the Requirements Not Met amounts owed because VA did not comply with all of the insurers’ requirements. Page 37 GAO/HRD9044 VA Collections From Insurers Chapter 6 Medical Centers Did Not Collect All Amounts Billed to Insurers Some insurers require that within 24 hours of a veteran’s admission or on the next working day after admission, the centers notify the insurers of, and obtain authorization for, admissions. If centers do not obtain such authorization, they may collect only a portion or none of the billed amounts. A VA circular issued in September 1986 required the medical centers to establish procedures for complying with insurers preadmission certifica- tion requirements. Although the centers we visited had established pro- cedures, we found that the centers did not always obtain the needed certifications from insurers, but in only a few instances had bills for vet- erans in our sample been specifically denied for this reason. A district counsel legal technician for one center, however, showed that the amounts denied by insurers can be large. She cited a case, not in our sample, of a $41,591 bill that was written off because the center did not obtain authorization, even though the center had knowledge of the insurer’s authorization requirement. Twenty-three of the medical centers reported, in response to our ques- tionnaire, that insurers often paid only a portion of the amounts billed because the centers did not obtain preadmission authorization from the insurers. Sixty other centers reported that this sometimes occurred. We also found that some collections were reduced because VA did not submit bills to insurers within insurance policy time frames. For example, the Seattle Medical Center failed to collect any of a $26,539 bill for a vet- eran in our sample. The insurer refused to pay because it received the bill more than 6 months after the last day of care. In responding to our questionnaire, 23 centers said they were sometimes denied payment because the insurers did not receive bills within insurers’ required time frames. Many veterans are covered by both Medicare and private health insur- Some Insurers Deny ance. Although some of the insurers have paid VA for care provided to Payment Because these veterans, others have denied payment because Medicare was not Medicare Not Billed billed first. We estimate, on the basis of our review of a sample of inpa- tient veteran records at five centers, that about 18 percent of the care First billed to insurers was denied because Medicare was not billed first. This was also the most frequent reason for denial of payment that centers reported in our questionnaire. One hundred centers reported that insur- ers often refused to pay any of the amounts billed because Medicare was not billed first. Twenty-six other centers reported that this reason was given sometimes. Page 38 GAO/HBD9061 VA CMkctiona From Insurers Chapter 6 Medical CentersDidNotColleetAllAmounta BilledtoInsurem Many Veterans Eligible for Many veterans treated in VA medical centers are eligible for Medicare, a Medicare Coverage federal health insurance program for people aged 65 and older and some categories of disabled people. Medicare, which is administered by the Health Care Financing Administration (HCFA), is composed of two parts. Part A, hospital insurance, covers inpatient hospital, skilled nursing facility, hospice, and home health care. Part B, supplementary medical insurance, covers many types of noninstitutional services, such as phy- sicians, clinical laboratory, x-ray, and physical therapy services. Both parts require beneficiaries to share in the cost of their care through deductibles and coinsurance. Many Veterans With Many Medicare beneficiaries, including veterans, have private insurance Medicare Also Have to cover some or all of their medical costs not paid by Medicare. This insurance can be group insurance sponsored by former employers or Private Health Insurance individual policies purchased by beneficiaries. The National Center for Health Services Research and Health Care Technology Assessment Sur- vey found that in 1987, almost 45 percent of the Medicare population aged 65 to 69 had private health insurance that was employment- related; almost 33 percent had other private health insurance. Assuming no special limitations in a veteran’s policy, it appears that VA has the legal authority to recover the cost of care from insurance designed to supplement Medicare. Section 629(a), title 38, authorizes VA, in specified circumstances, to recover the reasonable costs of care under veterans’ health-plan contracts. Subsection (i)( 1) defines a “health-plan contract” as “an insurance policy . . . or similar arrangement under which health services for individuals are provided or the expenses of such services are paid.” The definition provides no exclusions that could be relevant to a Medicare supplemental policy. The Senate Budget Com- mittee intended, it stated, that the definition “. . . be construed broadly so as to achieve broad coverage under this section with respect to the types of health plans under which recoveries and collections may be sought.” Consequently, a private insurance policy supplementing Medi- care benefits appears to fall within this definition. General counsel offi- cials in VA and HCFA also believe, they stated, that VA has the legal authority to recover health care costs from these policies. VA’s Actions to Resolve VA requested assistance from HCFA in 1988 to resolve the problem of col- letting from insurers that denied payment because Medicare had not Collection Problem been billed first. Specifically, VA asked HCFA to provide either (1) an explanation of Medicare benefits for the services provided by VA when Page 39 GAO/HRD-99.64VA Collections From Inmuer~ Chapter 6 Medical Centers Did Not Cdlect AU Amounta Billed to Insurers the insurer requires such a statement or (2) a general statement, in a form that VA could share with insurers, about Medicare payments for VA- provided services. In July 1988, HCFA responded in writing that complying with the first option would not be possible. In addition to being administratively impractical and costly, HCFA stated that it understood that 38 U.S.C. 629(i)(l)(B) precludes VA from billing Medicare, a prerequisite to issuing an explanation of Medicare benefits. However, HCFA did comply with the second option, issuing a general statement in a letter, as follows: “Because the Medicare law prohibits payment for services provided in a VA facility, the Medicare payment for any such services would always be zero, except under the limited circumstances . regarding services provided to Medicare beneficiaries not eligible for VA benefits . . . . Thus, even if VA were authorized to bill Medicare the EOMB [explanation of Medicare benefits] would state that the Medicare payment is zero.” In September 1988, VA instructed each medical center as follows: when veterans had Medicare, the center should include, with bills sent to insurers, copies of HcFA’s letter. The inclusion of HCFA’S letter has not completely overcome the collection problem. For example, in a request for advice from its district counsel, the Albany Medical Center reported that the insurers were continuing to reject the bills in spite of the HCFA denial letter. At 10 of 15 other centers we contacted, staff that used the letter reported that including it with the bills sent to the insurers had not resolved the collection problems. Since the HCFA letter has not resolved the collection problems with some insurers, VA plans to litigate cases to clarify its right to reimbursement. VA referred two cases, the VA deputy assistant general counsel said, con- cerning one insurer to the Department of Justice. This insurer denied payment for care in a VA medical center because the medical center did not bill Medicare first. VA subsequently advised us that the cases have been resolved and the insurer is now honoring VA'S claims. In addition, VA advised us, several claims against other insurers’ Medicare supple- mental policies have been referred to the Department of Justice for litigation.’ lPresumably such collection problems will tend to decrease as a result of the passage of the Omnibus Budget Reconciliation Act of 1989 (P.L. lOl-239), which, effective December 19, 1989, explicitly establishes Medicare as a secondary payer on all claims for beneficiaries also covered by an employer health plan This law undercuts the basic argument used by many insurers who reject bills for Medi- care-eligible beneficiaries. Page 40 GAO/HR.D9064 VACoUectio~ From Insurers Chl -.- IIf,!’ 7 _ Stronger Management Actions Neededto Maximize Insurmce Collections Managers at VA headquarters must make a commitment to (1) establish- ing effective recovery procedures and (2) assuring that centers have adequate resources to carry out their recovery efforts. Inadequate guid- ance hindered the ability of medical centers to establish effective proce- dures; inadequate resources caused centers not to bill insurers for all outpatient care. VA Headquarters The chief medical director gave medical centers general guidance on the Guidance Irladeq uate recovery effort along with some basic forms to use, but delegated responsibility for establishing recovery procedures to the centers. Con- sequently, each center established its own procedures for identifying veterans with insurance, billing insurers, and collecting amounts owed. In response to our questionnaire, about one-third of the centers said that guidance from headquarters was inadequate. As a result, centers missed opportunities to collect more from insurers. Medical Centers Did health insurers. VA headquarters left staffing decisions up to the discre- Not Commit Staff tion of the centers. VA’S fiscal year 1987 appropriation included 199 full- Needed time equivalent positions for two recovery efforts or, on average, about 1 position for each center.’ Information given by centers for fiscal year 1988 showed that about 370 full-time equivalent positions were working on the insurance recovery effort or, on average, about 2 full-time equiv- alent positions per center.2 However, 122 of the 159 centers responded that the number of available staff was less than adequate to effectively recover costs from insurers (see app. IX). To obtain the staffing and other resources needed to fully implement recovery from insurers, managers at five of the medical centers we vis- ited told us that they would like to keep a portion of the insurance col- lections instead of transferring them to the U.S. Treasury. One center director said he would provide the resources necessary to increase bill- ing and collections if he could get some funds to supplement his budget. Another director said that he would not devote additional staff for insurance billing purposes without receiving some benefit. He said that ‘These positions were for VA’s efforts to recover from certain veterans. who were required to make copayments toward the cost of their care, as well as from insurers. 2This estimate was calculated using the centers’ responses to our questionnaire for the hours spent on the recovery effort during a typical week in fiscal year 1988. Page 41 GAO/HRD90-64 VA Cdections From Insurers Chapter 7 Stronger Management Actions Neededto MarlmizeInaumncecollectlona he would have to absorb the staff for the insurance billing out of his current budget at a time when there were other unmet, medically related staffing needs at the center. VA headquarters is considering contracting out its health insurance bill- ing and collecting activities. Currently, VAsays, it has limited statutory authority to pay, out of amounts collected, for the billing and collection services of a contractor. Allowing VAto do this would enable VAto use its appropriated funds to provide health care to veterans. The Congress recently considered, but did not authorize, allowing VA to keep some of the funds collected from insurers to pay for its administra- tive costs. A bill (H.R. 901), passed by the House of Representatives, would require VA to develop an automated program to carry out the bill- ing and collection of amounts owed to the United States. It would also require VA to pay for the costs of administering the automated program, including the costs of any contracts for billing and collection under this program, from the amounts received through the program. Of the amounts that remain after paying for administrative costs, the bill would require VA to transfer funds as follows: to the general fund of the Treasury, an amount equal to the Congressional Budget Office’s esti- mate of collections that would have been made, in fiscal year 1OfQ without the automated program. After the administrative costs have been paid and the Treasury has received its specified amount, VA would allocate the remaining amounts to its medical care account and nursing home revolving fund. The bill would also require that VAsubmit a report, by January 1992, to the Senate and House Committees on Veter- ans’ Affairs comparing the experience of carrying out the program through the use of (1) VA employees at not more than 20 medical centers with (2) contractor services at not more than 20 medical centers. How- ever, these requirements were dropped from the bill in the Senate and not included in the version adopted by the Congress. VA medical centers had the potential to collect substantially more, per- Conclusions haps another $223 million, from private health insurers in fiscal year 1988. First, over 70 percent of the centers did not bill for outpatient care or billed for less than half of the outpatient care provided. Second, the centers struggled to establish efficient and effective billing and collec- tion procedures- over 30 percent of the potentially colhxtible costs were not billed by the six centers we visited. Third, center directors have not committed the staffing needed to do the job right-additional Page 42 Chapter 7 Stronger Management Actions Neededto Ikximhe Insurance Collections staffing is needed to handle insurance recoveries at two-thirds of the 159 medical centers, according to responses to our survey questionnaire. VA headquarters should take a more active role helping medical centers to maximize collections from private insurers. First, it should work closely with the centers to improve their procedures, especially those used to (1) identify veterans with insurance and (2) bill insurers for all episodes of inpatient and outpatient care provided. Although some improvements should save money through greater efficiencies of opera- tions, most will require medical centers to spend additional funds, a financial commitment that center directors appear reluctant to make. Sufficient resources must be dedicated to VA’S insurance recovery effort if collections are to be maximized. VA should act quickly because insur- ance policies have time limitations that could prevent VA from collecting for some care if it takes too long to bill for that care. To obtain adequate resources, VA should determine the additional resources necessary to maximize insurance collections. If these resources can not be found in its existing budget, VA should seek (1) the funds through the budget process or (2) legislative authority to pay its collection costs by using some of the funds collected from insurers. We recommend that the Secretary of Veterans Affairs ensure that all Recommendations to medical centers bill insurers for outpatient care they provide to insured the Secretary of veterans. In addition, we recommend that the Secretary ensure that Veterans Affairs each medical center has . effective procedures for (1) identifying all veterans with health insur- ance and (2) billing insurers for all episodes of inpatient and outpatient care provided to insured veterans and . sufficient resources to fully implement its identification, billing, and col- lection procedures. Matter for mizing insurance collections, the Congress should ensure that VA has the Consideration by the resources necessary to fully implement its recovery efforts. If adequate Congress resources are not available through the budget process, the Congress should consider amending 38 U.S.C. 629 to allow VA tokeep a portion of the amounts collected from insurers to defray its administrative costs. Page 43 GAO/HBDsoscl VA Cdlections From Insurera - chaptm7 Stronger Management Actions Neededto htarhh Insurance fIk&ctiona VA commented on a draft of this report on February 26, 1990 (see app. Agency Comments X). The Department agreed with our recommendations and pointed out a number of actions being taken to improve the process and increase col- lections from health insurers. Actions the Department said were being taken include l convening a task force for medical care cost recovery, charged with developing a strategy to improve identification, billing, and collection procedures as quickly as possible; . developing an administrative order, which is currently being reviewed in the Department, that would require all facilities to bill for outpatient care; l citing the insurance collection system as a material weakness in its 1989 Federal Managers’ Financial Integrity Act report; and l requesting input from field stations to ensure that debt collection proce- dures are effective after a new accounts receivable system is installed at all stations, which VA expects to be in 1990. VA also recognized that additional resources are necessary to properly carry out the billing and collection procedures; it presented such a pro- posal, it said, as part of its fiscal year 1991 budget request. After receiving VA'S comments, we reviewed VA'S fiscal year 1991 budget justification. It shows that this proposal would include establishing a third-party medical recovery account to bill and collect from third-par- ties. The fund would consist of (1) $18.6 million and 500 full-time equiv- alent staff transferred from the medical care account and (2) the appropriation of an additional $6.4 million and 300 full-time equivalent staff. All collections would be deposited into this account; amounts in excess of the billing and collection costs would be forwarded to the U.S. Treasury. VA estimates that this proposal will result in a $359 million increase in net receipts for Treasury in fiscal year 1991. Page 44 GAO/HRD90-64 VA Collections From Insurera Page 46 VA’s Daily Rates Used to Prepare Billsto I Health Insurers Fiscal year Type of care 1988 1989 Outpatient care $127 $110 Inpatient medicine: 473 483 Room, board. and nursina 266 271 Physician 91 93 Ancillary 116 119 Inpatient surgery: 611 681 Room. board. and nursina 337 375 Physician 114 128 Ancillary 160 178 Inpatient spinal cord injury: 524 539 Room. board. and nursina 306 317 Physician 62 64 Ancillary 154 158 Inpatient psychiatry: 236 232 Room. board. and nursina 166 164 Physician 32 32 Ancillary 36 36 Inpatient neurology: 393 426 Room. board. and nursina 219 237 Physician 49 53 Ancillary 125 136 Inpatient alcohol/drug: 197 213 Room. board. and nursina 119 129 Physician 37 40 Ancillary 41 44 Inpatient rehabilitation medicine: 372 374 Room. board. and nursina 215 216 Physician 44 44 Ancillary 113 114 Page 46 GAO/HRD96-64 VA Collections From Insurers Appendix II Total hsurance Collections and Collections Per Average Daily Occupied Bed by Medical Center @iscal Year 1988) Collections Per average daily Medical center location occupied Rank’ City State Total bed 1 Martinsburg WV $2,978,554 $10,378 2 White River Junction VT 890,972 7.551 3 Ene PA 628,214 7,391 4 Salt Lake City UT 1,756,649 7,083 5 Charleston SC 1,480.095 6.435 6 Iowa City IA 1,119,172 6,218 7 Saginaw MI 595,366 6,075 8 Boise ID 513,751 5,974 9 Columbra MO 1,069.550 5.243 10 Dublin GA 1,161,128 5,027 11 Fargo ND 580,724 4,921 12 Wichtta KS 528,621 4,895 13 SIOUXFalls SD 753,524 4.739 14 Marion IL 590,983 4,728 15 Madison WI 949,159 4,722 16 Fayetteville NC 1,108,635 4,718 17 Kerrville TX 866,860 4,636 18 Durham NC 1,317,622 4,544 19 San Antonio TX 1,976,689 4.288 20 Lincoln NE 472,854 4,185 21 Omaha NE 980,859 4,156 22 Fayetteville AR 441,298 4,086 23 Poplar Bluff MO 469,582 4,083 24 Altoona PA 332,684 4,057 25 Grand Island NE 297,626 3,865 26 Alexandna LA 595,916 3,845 27 Columbia SC 1.221,791 3,842 28 Richmond VA 1.868.382 3,790 29 Wilmtngton DE 567,348 3,782 30 Amarillo TX 352,057 3,745 31 Cheyenne WV 209,200 3,670 32 Albanv NY 1.304,243 3.603 33 Garnesville FL 1,233,373 3,554 34 Manchester NH 337.200 3,476 35 Louisville KY 856.932 3.469 36 Prescott AZ 354,452 3,408 (continued) Page 47 GAO/HRD-90-64VA Collections From Insurers Appendix II Total Insurance Collections and Collections Per Average Daily Occupied Bed by Medical Center (Pii Year 1988) Collections Per average daily Medical center location occupied Rank’ Citv State Total bed 37 Nashville TN $928,521 $3,389 38 Des Moines IA 393,678 3,365 39 Shreveport LA 748,864 3,358 40 Minneapolis MN 1,622.489 3.352 41 Fort Wayne IN 339,812 3.331 42 West Haven CT 1,000,300 - 3,323 43 Washrngton DC 1,359,846 3,269 44 lndranapolis IN 984,406 3.249 45 Wood WI 1,292,654 3,208 46 Lake City FL 808,704 3,196 47 Clarksburg WV 394,615 3.157 48 Mountarn Home TN 1,062,136 3,152 49 Butler PA 588,898 3,116 50 Roseburg OR 577,472 3,088 51 Providence RI 588,087 3,063 52 Fort Howard MD 517,710 3,045 53 Iron Mountain Ml 291,622 3,038 54 Leavenworth KS 803,018 3,019 55 Walla Walla WA 243,352 2,932 56 Bonham TX 175,540 2,926 57 Oklahoma Citv OK 776,262 2.897 58 Pittsburgh/ Unrv. Dr PA 1,153,162 2,883 59 Denver co 670,981 2,880 60 Hampton VA 722,011 2,735 61 Ann Arbor MI 563,301 2,695 62 Birmrngham AL 599,682 -2,642 63 East Orange NJ 1,376,306 2,597 64 Portland OR 950.014 2,561 65 Beckley WV 303,308 2,549 66 Temple TX 870,834 2,495 67 Fresno CA 311,842 - 2.417 68 Livermore CA 202,695 2,413 69 ~~ Little Rock AR 1,804,612 2,335 70 Hones IL 1,695,397 --________- ~~ 2,304 71 Syracuse NY 410.472 2,255 72 Cinclnnatl OH 433,648 2,179 (contlnuedj Page 48 GAO/HRD90-64 VA Collections From Insurers Appendix II Total Insurance Collections and Collections Per Average Daily Occupied Bed by Medical Center @iscal Year 1989) Collections Per average daily Medical center location occupied Rank’ Citv State Total bed 73 Tucson AZ $394,863 $2,170 74 Memphis TN 1,152.075 2,137 75 Seattle WA 609,045 2,100 76 B&on MA 717,416 2,092 77 Dallas TX 1,016,173 2,091 78 Augusta GA 1,611,030 2,079 79 Wilkes Barre PA 616,485 2,062 80 Atlanta GA 696,689 2,049 81 Miles City MT 124,235 2,037 82 St Cloud MN 769,342 1.938 83 Phoenrx AZ 612,031 1,895 84 Tows ME- 543,867 1,888 85 Chicago/ West Side IL 630,239 1,887 86 Ashevrlle NC 593,346 1,884 87 Albuquerque NM 581,263 1,869 88 Marlin TX 282,812 1,861 89 Grand Junction co 118,042 1,844 90 Perry Point MD 1,096,392 1,794 91 Fort Meade SD 462,888 1,794 92 Bath NY 326,128 1,782 93 Chrllrcothe OH 999,995 1,718 94 Spokane WA 172,852 1,695 95 Huntrngton WV 218,776 1,683 96 Prttsburgh/ Hrghland PA 844,311 1,682 97 Muskoqee OK 214,464 1,663 98 Brg Spring TX 226,033 1,615 99 Hot Sprinas SD 156,909 1,569 100 St LOUIS MO 889,791 1.561 101 Marion IN 686,587 1,506 102 Jackson MS 447,793 1,503 103 Loma Linda CA 355,769 1,429 104 Dayton OH 563,196 1,394 105 Knoxvrlle IA 474.430 1.375 106 Kansas City MO 383,755 1,356 107 Fort Harrison MT 142,083 1,353 (continued) Page 49 GAO/~90-64 VACollections Prom Insurers Appendix II Total Insurance Collection and CMktiona Per Average Daily Occupied Bed by Medical Center (Fi8cal Year 1988) Collections Per average dally Medical center location occupied Ranka City State Total bed 108 San Francisco CA $27 1,662 $1,352 109 Baltimore MD 209,511 1,343 110 Topeka KS 691,176 1,329 111 Sheridan WY 342,588 1,328 112 Lebanon PA 640,721 1.321 113 Danville IL 816,497 1,298 114 Lexington KY 805,392 1,272 115 San Diego CA 366,052 1,262 116 Reno NV 143,332 1.246 117 Brockton MA 838,269 1,231 118 Salem VA 726,978 1,224 119 Montaomerv ” I AL 164.365 1.209 120 Northampton MA 481,313 1,188 121 Allen Park MI 430,410 1,151 122 Newington CT 131,447 1,143 123 Long Beach CA 714,352 1,069 124 Bav Pines FL 491,587 1,039 125 Waco TX 618,055 1,030 126 Salisbury NC 703,988 1,029 127 Houston TX 716.466 1.021 128 Tampa FL 425,383 976 129 Northport NY 598,708 972 130 Tuscaloosa AL 453,294 940 131 Buffalo NY 519,709 930 132 Chicago/ Lakeside IL 222,723 928 133 Palo Alto CA 880,655 928 134 Murfreesboro TN 440,202 837 135 North Chicago IL 556,493 778 136 Tomah WI 375.728 764 137 Batavia NY 87,067 757 138 Biloxi MS 421,293 738 139 Sepulveda CA 202,665 729 140 Battle Creek Ml 411,645 644 141 Castle Point NY 110,960 638 142 Cleveland OH 361,979 624 143 New Orleans LA 181,160 596 (continued) Page 60 GAO/EIRD9044 VACMlectio~ From Insurers Appendix II Total Insurance Collections and Collections Per Average Daily Occupied Bed by Medical Center (Fiscal Year 1999) Collections Per average dally Medical center location occupied Rank0 City State Total bed 144 Martinez CA $153,107 $565 145 Miami FL 262,597 564 146 Bronx NY 211,983 558 147 American Lake WA 121,636 553 148 Tuskegee AL 323,821 546 149 Fort Lyon co 135,015 511 150 West Los Angeles CA 330,448 401 151 Canandaigua NY 239,371 401 152 Lyons NJ 330,724 398 153 San Juan PR 214,353 388 154 Brooklyn NY 146,261 266 155 Philadelphia PA 74,023 244 156 New York NY 107,441 216 157 Bedford MA 118,513 215 158 Coatesville PA 146,075 209 159 Montrose NY 84,414 117 % order of collectlons per bed Page 61 GAO/IUUHO& VA Collections From Insurers Appendix III When Medical Centers Began Billing Insurers - In our questionnaire, we asked centers when they began routinely billing insurers for inpatient and outpatient care. Responses are summarized in table III. 1. Table 111.1:Center Responses Medical centers Fiscal year Outpatient Inpatient 1986 or earlier [before Oct. 1986) 10 32 1987 (Oct. 1986~Sept. 1987) 29 119 1988 (Oct. 1987~Sept. 1988) 36 8 1989 (Oct. 1988~Sept. 1989) 6 0 Start date unknown 1 0 Not started as of Oct. 1988 77 0 Total 159 159 Page 52 GAO/MUM064 VA Cokctio~ From Inmuers Estimating VA’s Potential Collections From Insurers We used three steps to estimate VA’S potential collections from health insurers nationwide in fiscal year 1988. First, from data reported in VA’S 1987 Survey of Veterarql we estimated the total number of episodes of inpatient care or outpatient visits for insured veterans. Second, we esti- mated the total cost of this care using VA’S data. Third, we reduced the total cost of care (potential collections) to allow for health insurance policy limitations. Described below is our methodology for estimating potential collections for inpatient and outpatient care. Estimating Total Cost veterans in fiscal year 1988, we calculated the number of VA inpatient for Inpatient Care episodes, reported in VA’S 1987 Survey of Veterans, by veterans who . were without service-connected disabilities, l had been patients overnight in VA hospitals in the previous 12 months, l had health insurance plans that pay the providers directly for the ser- vices, and . were either (1) under the age of 65 or (2) the age of 65 or older and had health insurance through their employers. On the basis of the survey responses, we estimated that nationwide, about 80,000 episodes of inpatient care were provided to these insured veterans during the 12-month period covered by the survey. At the 95- percent confidence level, we calculated the resulting sampling error was 44,000 episodes.? To estimate the total number of inpatient days of care, we multiplied the number of inpatient episodes by VA’S average lengths of stay for fiscal year 1988. In the survey, the 80,000 episodes of inpatient care were placed in two categories: “surgical” or “medical,” with “examination/ diagnostic” episodes included in the medical category. None of the insured veterans reported receiving inpatient psychiatric care. We assumed that nearly all of the insured veterans would be under the age of 65. Therefore, we used VA’S average lengths of stay for veterans ‘Department of Veterans Affairs, 1987 Survey of Veterans,IM&S-M 70-89-l (Washington, D.C., July 1989). “We calculated the sampling error at the 95-percent confidence level on the basis of a formula used by the Census Bureau for the 1987 Survey of Veterans, which it did for VA. This confidence level means that the chances are about 19 out of 20 that the actual number being estimated falls within the range defined by our estimate, 80,000 episodes, plus or minus the sampling error, 44,000 episodes. Page 53 GAO/HRD90-64 VA Collections From Insurers Appendix IV Estimathg VA’sPotential Cdkctions FromInsurer under the age of 65,3 which were 10.4 days for medical care episodes and 8.6 days for surgical care episodes. To estimate VA'S total cost for this inpatient care, we multiplied the total number of inpatient days of care by VA'S reported daily rates for fiscal year 1988. These daily rates were $611 for surgical and $473 for medical care. We estimated the total cost of inpatient care for insured veterans to be $395 million, with a sampling error of plus or minus $217 million. Since our cost estimate is based on the estimated number of episodes of care, the sampling error for the costs is proportional to the sampling error for the number of episodes of care. To estimate the total number of outpatient visits for insured veterans in Estimating Total Costs fiscal year 1988, we calculated the total number of outpatient visits, for Outpatient Care reported in VA'S 1987 Survey of Veterans, by veterans who l were without service-connected disabilities, l had made visits for medical treatment to VA clinics or VA hospitals on an outpatient basis in the previous 12 months, l had insurance plans that pay the providers directly for the services, and . were either (1) under the age of 65 or (2) the age of 65 or older and had health insurance through their employers. On the basis of the survey responses, we estimated that nationwide, 180,000 insured veterans made one or more outpatient visits to VA health care facilities during the 12-month period covered by the survey. The survey asked veterans to estimate the number of VA outpatient vis- its they had made during the preceding 12 months, that is, 1 visit, 2 to 5 visits, 6 to 10 visits, 11 or more visits. To estimate the number of visits nationwide, we used the lower value of each range. This resulted in an estimate of 850,000 visits. To allow for insurance deductibles, we assumed the first 2 visits by a veteran would be used to meet the insurance deductible. Thus, we sub- tracted 37,000 visits from the total estimated visits to adjust for veter- ans who reported only 1 visit; we subtracted 284,000 visits for veterans who reported 2 or more outpatient visits. This left an estimated total of 529,000 outpatient visits for which recovery from insurers seemed possible. 3We used VA’s reported average lengths of stay for short-term discharges, which are defined as stays of 99 days or less. Page 64 GAO/HRB9O-64VACollections From Insurers Appendix IV Estimafing VA’sPotential Cdledions FromJnsurers To estimate VA’S total cost for these outpatient visits, we multiplied the number of outpatient visits by VA’S reported daily rate for fiscal year 1988, which was $127. Thus, we estimate the total cost of 529,000 out- patient visits to be $67 million. At the 95-percent confidence level, this estimate has a sampling error of plus or minus $24 million. We recognized that many health insurance policies have provisions for Estimating Potential deductibles and copayments. To allow for insurance deductibles and Collections From copayments for inpatient care, we reduced the estimated total costs by Insurers 30 percent. Thus, we assumed, on average, insurers would be liable for 70 percent of the total VA cost of providing inpatient health care. For outpatient care, as discussed above, we allowed for deductibles by reducing the number of outpatient visits. We allowed for outpatient care coinsurance by reducing the total costs for the remaining visits by 30 percent. At the 95-percent confidence level, we estimate potential collec- tions of $125 to $428 million from insurers for inpatient care in fiscal year 1988. For outpatient care, we estimate potential collections of $30 to $64 million. Page 56 GAO/HBD90-64 VA Collections From Insurers Methodology and Results for Our Review of Samplesof Insured Veterans Receiving Inpatient Care At the five medical centers whose inpatient records we reviewed, we asked center officials to give us a list of the veterans without service- connected disabilities who were (1) identified as insured and (2) dis- charged from inpatient care between October 1,1987, and March 31, 1988. From this list we randomly selected a sample of approximately 30 veterans at each medical center. We did not include a veteran in our samples if . we could not find sufficient evidence that the veteran was actually cov- ered by insurance during the 6-month period of our review, l the veteran had Health Maintenance Organization (HMO) coverage or was covered by an indemnity policy, . the veteran’s records indicated he or she had a service-connected disa- bility, or l we could not verify that the veteran received inpatient care during the period reviewed. For each veteran sampled, we reviewed the veteran’s administrative and medical records to determine (1) the value of care provided, (2) the amount of care that could have been billed but was not, (3) the amount of care that was billed to insurers, and (4) the amount that was collected. For each center visited, we projected our findings to the universe of insured veterans without service-connected disabilities who were dis- charged between October 1,1987, and March 31,1988. We also pro- jected the associated precision of our estimates for the 95-percent confidence level. The projected amounts that could have been billed to insurers but were not are summarized in table V.l. Page 56 GAO/HRD!40-64VA Cdlection~ From insurers Appendix V Methodology and Results for Our Review of Samplesof Insured Veterans Receiving Inpatient Care Table V.1: Projected Value of Inpatient Care Provided to Insured Veterans, October 1, 1987, Through March 31,1988, and Not Billed to Insurers Average Projected to medical center population amount not Range for 95-percent Range for 95-percent billed per confidence level Estimated value confidence level Medical center veteran Low High of care not billed Low High Bav Pines $3,746 $2,051 $5,441 $782,837 $428,660 $1,137,014 MartInsburg 74 12 207 14,560 2,365 40,676 Mmneapolis 1,617 219 3,015 252,317 58,227 470,447 Seattle 1,910 317 3,503 227,262 57,293 416,784 West Los Angeles 9,305 3,474 15,136 502,472 297,761 817,350 Total $2,421 $1,665’ 93,177a $1,779,448 $1,223,776’ $2,335,120’ These figures are the range for our projections, not the average or sum of the values listed for the individual medical centers. We also reviewed the records for the veterans sampled to determine the amounts that the centers billed to insurers and the amounts recovered. The amounts of care the centers billed to insurers are summarized in table V.2. Table V.2: Projected Value of Inpatient Care Provided to Insured Veterans, October 1,1987, Through March 31, 1988, and Billed to Insurers Average Projected to medical center population amount Range for 95-percent Range for 95-percent billed per confidence level Estimated value confidence level Medical center veteran Low high of care billed Low High Bay Pines $6,143 $1,494 $10,793 $1,283,943 $312,215 $2,255,671 Martmsburg 8,643 4,862 12,424 2,333,590 i ,312,aoa 3,354,372 MmneaDohs 6,136 3,885 a,357 957,203 606,045 1,308,361 Seattle 4,659 2,551 6,767 554,425 303,524 805,326 West Los Angeles 1,617 958 2,356 87,291 51,728 136,953 Total $6,456 $4.631" $8.281' $5,216,452 $3,741,6718 $6,691,233” aThese figures are the range for our prolectlons. not the average or sum of the values Ned for the rndwdual medical centers In many cases, we were unable to determine the amount that YAcould expect to collect from insurers for the care provided; this is because the medical and administrative records we reviewed did not contain ade- quate information on the veterans’ insurance coverage. The collection rates shown in table V-3 were calculated by dividing the average amount collected per veteran by the average amount billed per veteran. Page 67 GAO/IiRD-90-64VA CNlectio~ From Lnsurers Appendix V Methodology and Results for Our Review of !Sampleaof Insured Veterans Receiving Inpatient Care For various reasons, centers had not closed all cases for sampled veter- ans at the time or our review. For example, the Seattle Medical Center had referred some cases to the VA district counsel for assistance. We assumed that the centers would not collect any additional funds from open cases. If the centers did collect more money, the collection rates would increase. Table V.3: Centers’ Projected Collection Rates for Inpatient Care Provided, October 1,1987, Through March 31, 1988, and Billed to Insurers Average Projected to medical center population amount Range for 95percent Range for 9bpercent collected per confidence level Estimated collection confidence level Medical center veteran Low high rate (percent) Low’ Highb Bay Prnes $1,513 $217 $3,356 25 4 55 Martrnsburq 5,758 2,991 8,525 67 35 99 Mrnneapolts 3,376 2,043 4,709 55 33 77 Seattle 1,455 556 2,354 31 12 51 West Los Angeles 1,414 838 2,321 87 52 100 Total $3.278 $2,195 $4,358 aThis was calculated by drviding the projected amount collected per veteran at the low end of the range by the projected average amount per veteran. bThrs was calculated by dividing the projected amount collected per veteran at the high end of the range by the projected average amount per veteran. Page 58 GAO/HRD9064 VACollections From Insurers Methodology and Results for Our Review of Samplesof Insured Veterans Receiving Outpatient Care At the six medical centers whose outpatient records we reviewed, we asked center officials to give us a list of the veterans without service- connected disabilities who (1) were identified as insured and (2) received outpatient care between October 1, 1987, and March 31, 1988. From this list, we randomly selected a sample of approximately 30 vet- erans at each medical center. We did not include a veteran in our samples if . we could not find sufficient evidence that the veteran was actually cov- ered by insurance during the 6-month period of our review, l the veteran had Health Maintenance Organization (HMO) coverage or was covered by an indemnity policy, . the veteran’s records indicated he or she had a service-connected disa- bility, or l we could not verify that the veteran received outpatient care during the period reviewed. For each veteran sampled, we reviewed the veteran’s administrative and medical records to determine (1) the value of care provided, (2) the amount of care that could have been billed but was not, (3) the amount of care that was billed to insurers, and (4) the amount that was collected. We projected our findings to the universes of insured veterans without service-connected disabilities who received outpatient care between October 1, 1987, and March 31, 1988, at the centers visited. We also pro- jected the associated precision of our estimates for the 95-percent confi- dence level. The projected amounts that could have been billed to insurers but were not are summarized in table V1.1. Page 69 GAO/HID-90-64 VACollections From Insurers Appendix VI Methodology and Resulte for Our Review of Sampleaof Insured Veterans lbcelvlng Outpatient Care Table VI.l: Projected Value of Outpatient Care Provided to Insured Vetemrw, October 1,1987, Through March 31, 1988, and Not Billed to Insurer8 Average Prolected to medical center population amount not Range for 95-percent Range for 95percent billed per confidence level Estimated value confidence level Medical center veteran Low high of care not btlled Low High Albany $533 $116 $950 $468,981 $101,749 $836,213 Bav Prnes 479 270 666 245,497 138.269 352.725 Martmsburg 385 293 477 325,888 248,108 403.668 Minneapolis 506 277 735 336,468 184,055 488,881 Seattle 300 179 421 128.371 76.751 ~. 179.991 _,_- West Los Angeles 504 308 700 50,901 31,084 70,718 Total $453 s330a $578’ $1,558,106 $1,133.383’ $1.978.829’ aThese figures are the range for our projections, not the average or sum of the values llsted for the individual medtcal centers. We also reviewed the records for the veterans sampled to determine the amounts that the centers billed to insurers (see table VI.2) and the amounts recovered. Table Vl.2: Projected Value of Outpatient Care Provided to Insured Vetemnr, October 1,1987, Through March 31,1988, and Billed to Insurers Projected to medical center population Average Range for 95-percent Range for 95-percent amount billed confidence level Estimated value confidence level Medical center per veteran Low high of care billed Low High Albany $149 $39 $260 $131,355 $34,165 $228,545 Bay Pines 0 0 0 0 0 0 Martinsburg 6 1 14 5,465 508 12,187 Mmneaoolrs 0 0 0 0 0 0 Seattle 157 42 272 67,039 17,859 116,219 West Los Angeles 0 0 0 0 0 Total $59 828O $91’ $203,85: $94,7270 $312,991’ aThese figures are the range for our projections, not the average or sum of the values llsted for the indwrdual medical centers. In many cases, we were unable to determine the amount that VA could expect to collect from insurers for the care provided; this is because the medical and a.dministrative records we reviewed did not contain ade- quate information on the veterans’ insurance coverage. The collection rates shown in table VI.3 were calculated by dividing the average amount collected per veteran by the average amount billed per veteran. Page 60 GAO/HID-90-64 VA Collections Prom Insurers Appendix VI Methodology and Results for Our Review of Samplesof Insured Veterans Receiving Outpatient Care For various reasons, centers had not closed all cases for these veterans at the time or our review. We assumed that the centers would not collect any additional funds from open cases. If the centers did collect more, the collection rates would increase. Table Vl.3: Centers’ Collection Rates for Outpatient Care Provided, October 1,1987, Through March 31,1988, and Billed to Insurers Average Projected to medical center population amount Range for 95-percent Range for 95-percent collected per confidence level Estimated collection confidence level Medical center veteran Low high rate (percent) LOW Hiahb Albany ____~~-- 37 1 85 Bay Pmes ~~ 0 0 0 0 0 0 Martmsbura 1 0 2 11 0 31 Mmneapolis 0 0 0 0 0 0 Seattle 95 23 166 60 15 100 West ~.. Los Angeles 0 0 0 0 0 0 Total $28 $5 $47 aThts was calculated by dividing the projected amount collected per veteran at the low end of the range by the projected average amount per veteran. bThis was calculated by dividing the projected amount collected per veteran at the high end of the range by the projected average amount per veteran. Page 61 GAO/HRIH@64 VA Collections From Insurers AppendIs VII Projected Value of Amounts W.kcted and Reasonsfor Less Than F’ull Payment for Inpatient Care Billed at Five Centers To determine the amounts collected and the reasons for less than full collection for our five samples of insured veterans who received inpa- tient care (see app. V), we reviewed care that was billed to insurers. We projected these findings to the population of similar veterans at the five centers. The projected amounts and associated range for the g&percent confidence level are summarized in table VII.1. Table VII.1: Projected Values Billed to Insurers for Inpatient Care, October 1, Range for 9bpercent confidence level 1987, Through March 31,1988, Collected Estimated Percent and Not Collected From Insurers Amount Percent Low High (low-high) Collected $2,646,176 51 $1,787,156 $3,505,196 34-67 Not collected: Policy limitations 773,094 15 266,305 1.279.883 5-25 VA did not bill Medicare first 936,478 18 149,638 1.7828520 3-34 Insurers’ requirements not met 154,898 3 34.125 345.885 l-7 Unknown/ othera 707,683 13 346,079 1,069,287 7-21 aThese were cases (1) that were not closed at the time of our review or (2) for which the center collected less than 70 percent of the amount billed and we could not determine the reason for less than full payment. Page 62 GAO/IiRD-!I084 VACdectio~ From Insurers Appendlh 1.11I Reasonsfor Denial or Partial Payment of Bills I. Our questionnaire listed possible reasons why an insurer did not pay any of the amount billed. The centers indicated how often, during fiscal year 1988, insurers gave each of these reasons for not paying any of the amount billed: Reason for denial Very often Often Sometimes Rarely Never Insurer offset the total amount charged against the veteran’s insurance deductible 13 27 53 41 23 Policy does not cover any of the services provided 5 23 76 48 5 Bill was not received within the insurer’s reauired time frame 0 2 21 48 85 VA did not bill Medicare first 61 39 26 11 18 Policy was not in effect-veteran was not insured when care was provided 8 16 80 49 4 Care billed to an HMO was neither for emergency nor preauthorized 22 18 42 41 31 Care billed to an insurer other than an HMO was not oreauthorized 7 10 50 65 25 Services provided were not medically necessary 3 12 40 64 37 VA has no contract with the insurer 15 9 31 55 47 Insurer does not pay for care provided at a VA medical center 12 23 58 45 19 Charges for the services provided exceeded the lrmitations of the insurance policy 4 23 56 45 28 Health insurance (excluding indemnity policies) payment made directly to veteran 0 16 59 63 19 II. The questionnaire listed possible reasons why an insurer pays only a portion of the amount billed. The centers indicated how often, during fiscal year 1988, insurers gave each of these reasons for paying only a portion of the total amount billed: Reason for partial payment Very often Often Sometimes Rarely Never Insurer offset a portion of the charges agarnst the veteran’s insurance deductible and/or coinsurance 36 47 55 10 9 A portron of the charges for services provtded exceeded the limitations of the insurance poltcy 19 29 63 36 10 (contrnued) Page 63 GAO/HRD-m VACdlection~ From Insurers Appendix VIII Reasonafor Denial or Partial Payment of Bills Reason for partial payment Very often Often Sometimes Rarely Never Policy does not cover some of the services provided 8 29 08 24 8 Services billed were not preauthorized 6 17 60 58 16 Length of stay exceeded the insurer’s allowable lenqth of stav 3 9 58 61 26 Some of the services provided were not medically necessary 2 15 44 59 37 Charaes were not reasonable 0 9 31 53 64 Page 64 GAO/HRD-9M4 VA CMkctio~ Prom Insurers Appendix IS Factors Affecting Recoveries Our questionnaire asked medical centers to consider various factors that may influence how effectively the center was able to recover health care costs from insurers. For each factor, the center indicated whether it was more than, about, or less than adequate to enable the center to effec- tively recover costs from insurers. More than About Less than Factor adequate adequate adequate Number of available staff 2 35 122 Staff’s skills/exberience 28 102 29 Training you are able to provide staff 7 90 62 Computer hardware 7 70 80 Computer software 6 63 84 Guidance from VA headquarters 2 100 56 Veterans’ cooperation with your efforts 9 104 44 Insurers’ cooperation with your efforts 5 84 70 District counsel’s efforts to pursue referred cases 31 87 28 Page 66 GAO/HED96-64 VA Collections From Insurers Comments From the Department of Veterans Affairs THE SECRETARY OF VETERANS AFFAIRS WASHINGTON Hr. David P. Baine Director, Federal Health Care Delivery Issues U. S. General Accounting Office Washington, DC 20548 Dear Mr. Baine: I am pleased to respond to your January. . 2, 1990, report m CARE: Better Procedures Needed To Maxlmlze Collections From Health Insurers (GAO/BBD-90-64). Since 1986 when the Congress authorized the Department of Veterans Affairs (VA) to collect from health insurers for medical care, we have strived to fully implement this mandate. We believe the GAO report documents that the Department has made significant progress with the collection of about $100 million in Fiscal Year 1988. We also recognize that more can be done to improve the collection process. We agree with your recommendations, and actions are already underway that should improve the process and increase collections from health insurers. The actions being taken as well as other comments on your report are outlined in the enclosure. Thank you for the opportunity to comment on this report. Enclosure EJD/jev Page66 GAO/HBD~VACollectio~FromI~~rs Appendirx Comment.aFromtheDeprutment of VetemnsAfTaim Enclosure DEPARTMENT OF VETERANS AFFAIRS COMMENTSON THE JANUARY 2, 1990, GAO REPORT w . CTIONS p GAO rOCOmOnd8 that I require all mdiaal contars to bill for outpatient aara. In addition, GAO reaomands that I ensure that oath medical oantar has: l fiaative proomduraa tor (1) identifying all veterans vith in8urmae and (2) billing in8urars for all inpatient and outpatient uare provided to inwred veterans. We concur with the recommendation. The Department has convened a Medical Care Cost Recovery Task Force charged with developing a strategy to improve identification, billing, and collection procedures as quickly as possible. In addition, the Veterans Health Services and Research Administration (VRS&RA) has in the concurrence process an administrative issue that requires all facilities to incorporate billing procedures for outpatient visits. VHS6tR.A is also establishing procedures for a billing training guide. In addition, they have planned changes to the Decentralized Hospital Computer Program system that will automate additional portions of the billing process. These changes will provide additional management reports to assist in the complete and timely capturing of billing information. Finally, the VIiS&RA has scheduled a monthly national conference call for all involved services to assist with any problems they encounter. Bufficient resources to fully implement its identification, billing, and collection procedures. We concur with the recommendation and have proposed a multifaceted plan to correct the situation. The Department has listed this finding as a material weakness in the VA's 1989 Federal Manager's Financial Integrity Act (FMFIA) Report that tasks the VHS&RA to identify deficiencies and follow through with an action plan. We anticipate that the first results of this action will be completed by Uarch 1990. An internal review of the third party billing and collection activity within VA was completed in September 1989. As a result, a number of recommendations were proposed and are being considered for implementation. We plan to request input from field stations to insure that our debt collection procedures are effective at the field station level once an accounts receivable module is installed at all stations. Installation is expected this calendar year. We will enhance our accounts receivable module as permitted by our resources. Page67 GAO/HRD9M4VACoUections FromInsurers Appanihxx Cemments~mtheDeparcmentof VeteransAffair 2. We recognize that additional resources are necessary to properly perform the billing and collection efforts and have presented a proposal in the 1991 budget legislation. Moreover, if Congress can provide additional resources for this program, as suggested in the audit report, we will be able to continue to significantly improve billings and collections. The following comments are offered to specific parts of the report: Now on p 40 - At page 50 of the draft report, our referral of two cases to the Department of Justice for litigation against one insurer is discussed. This matter has since been resolved, and the insurer is now honoring VA's claims against them. Recently, however, several claims against other insurers' Medicare supplemental policies were referred to the Department of Justice for litigation. Now on p 42 - On page 52, the report discusses the need for legislation to permit VA to pay contractors out of amounts collected for billing and collection activities. Currently, there is limited statutory authority to pay for collection services out of amounts collected. Authority is lacking, however, to pay a contractor out of amounts collected for services in identifying cases and preparing bills. Now on p 21 - On page 26, the official name for the medical center in Bonham, Texas, is 'the Sam Rayburn Memorial Veterans' Center. Page68 GAO/HRDW-64VACollectionsPromI~urers pppendls XI Major contributms to This Report Human Resources Walter P. Gembacz, Assignment Manager Division, Washington, DC. Seattle Regional Office Janet E. F’risch, Evaluator-in-Charge Peter J. Bylsma, Evaluator Patricia K. Yamane, Evaluator 1 Atlanta Regional Office Mark J. Huber, Evaluator Chicago Regional Office (401973) Page 69 GAO,4JRDso&I VACollections From Insurers Requests for copies of GAO reports should be sent to: C.S. General Accounting Office Post Office Box 6015 Gaithersburg, Maryland 20877 Telephone 202-275-6241 The first five copies of each report are free. Additional copies are $2.00 each. There is a 25% discount on orders for 100 or more copies mailed to a single address. Orders must be prepaid by cash or by check or money order made out to the Superintendent of Documents.
VA Health Care: Better Procedures Needed to Maximize Collections From Health Insurers
Published by the Government Accountability Office on 1990-04-06.
Below is a raw (and likely hideous) rendition of the original report. (PDF)