This is the accessible text file for GAO report number GAO-07-947 entitled 'Defense Health Care: Under TRICARE Children's Hospitals Paid More Than Other Hospitals After Accounting for Patient Complexity' which was released on July 31, 2007. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to Congressional Committees: United States Government Accountability Office: GAO: July 2007: Defense Health Care: Under TRICARE, Children's Hospitals Paid More Than Other Hospitals After Accounting for Patient Complexity: GAO-07-947: GAO Highlights: Highlights of GAO-07-947, a report to congressional committees Why GAO Did This Study: Under the Department of Defense’s (DOD) TRICARE health program, hospitals that treat primarily children—designated by DOD as children’s hospitals—are paid differently from other types of civilian hospitals through a children’s hospital differential payment. Representatives of children’s hospitals state that payments for children’s hospital services do not fully recognize the higher complexity of children’s hospital patients. Acknowledging concerns over payments for children’s hospital services, the National Defense Authorization Act for Fiscal Year 2006 directed GAO to study DOD’s current system of payments to children’s hospitals. This report examines (1) the effect of the differential on TRICARE’s base payments to children’s hospitals, (2) differences in diagnosis and complexity between TRICARE pediatric patients at children’s hospitals and those at other hospitals, (3) the extent to which TRICARE payment differences across hospitals reflect differences in patient complexity, and (4) recent trends in TRICARE pediatric patients’ use of children’s hospital services. To do this, GAO analyzed pertinent TRICARE claims data for fiscal years 2003 through 2006 and interviewed relevant DOD officials and representatives of children’s hospitals. What GAO Found: In fiscal year 2007, TRICARE’s base payments, a key component of the program’s hospital payment formula, were 61 percent higher for facilities that TRICARE defines as children’s hospitals than for other hospital types. Base payments to children’s hospitals have been substantially higher than base payments to other hospitals since 1989. However, the relative difference in base payments has decreased over time, and will continue to decrease, as the children’s hospital differential is not adjusted for inflation. From fiscal year 2003 through fiscal year 2006, excluding newborns, the types of diagnoses for TRICARE pediatric patients at children’s hospitals were similar to those treated at medical centers, hospitals that also provide specialized pediatric services. TRICARE pediatric patients at children’s hospitals had a similar level of complexity to those at medical centers and were substantially more complex than those at community hospitals, facilities that focus on more routine children’s care. GAO measured the complexity of patients using a tool that classifies hospital stays into a more refined set of groups than TRICARE’s system. Indirect measures of complexity, such as the length of a hospital stay, also showed similarities between TRICARE pediatric patients at children’s hospitals and those at medical centers. GAO found that after adjusting for differences in patient complexity, TRICARE payments to children’s hospitals were substantially greater per admission than TRICARE payments to medical centers and community hospitals. Specifically, holding patient complexity constant, children’s hospitals were paid 22 percent more than medical centers and 53 percent more than community hospitals. The number of TRICARE pediatric admissions at children’s hospitals increased from 5,027 in fiscal year 2003 to 7,083 in fiscal year 2006. The percentage of TRICARE pediatric admissions in civilian hospitals that occurred at children’s hospitals also increased during this time period. The increase in the use of children’s hospital services is consistent with statements from representatives of children’s hospitals, who said that their hospitals are committed to accepting and caring for TRICARE patients. GAO’s findings show TRICARE’s hospital payment system functioning largely as DOD expected, as the difference in base payments to children’s hospitals and other hospitals was designed to endure but diminish over time. GAO has no data on other factors that might support payment differences, however, GAO’s findings suggest that further increasing payments to children’s hospitals is not supported on the basis of patient complexity. In commenting on a draft of this report, DOD agreed with GAO’s findings and concluding observations. [Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-07-947]. To view the full product, including the scope and methodology, click on the link above. For more information, contact Laurie Ekstrand at (202) 512-7114 or ekstrandl@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: TRICARE's Base Payments to Children's Hospitals Substantially Higher Than Base Payments to Other Hospitals, Though Relative Difference in Payments Decreases over Time: TRICARE Pediatric Patients at Children's Hospitals Similar to Those at Medical Centers in Terms of Diagnoses and Complexity: After Adjusting for Patient Complexity, Children's Hospitals Were Paid More per TRICARE Pediatric Admission Than Other Hospitals: Rising Number of TRICARE Admissions at Children's Hospitals Suggests No Decline in Access: Concluding Observations: Agency and Professional Association Comments and Our Evaluation: Appendix I: Scope and Methodology: Appendix II: Comments from the Department of Defense: Appendix III: GAO Contact and Staff Acknowledgments: Tables: Table 1: Average TRICARE Patient Complexity by Hospital Type, Fiscal Year 2003 through Fiscal Year 2006: Table 2: Indirect Measures of TRICARE Pediatric Patient Complexity by Hospital Type, Fiscal Year 2003 through Fiscal Year 2006: Table 3: Average TRICARE Payment per Pediatric Admission, Patient Complexity, and Payment Adjusted for Complexity, by Hospital Type, Fiscal Year 2003 through Fiscal Year 2006: Table 4: Average Complexity per TRICARE Pediatric Admission by Hospital Type, Using Both TRICARE DRGs and APR-DRGs, Fiscal Year 2003 through Fiscal Year 2006: Table 5: Change in TRICARE Base Payments to Children's Hospitals Compared to the Change in Hospital Inflation, Fiscal Year 1992 through Fiscal Year 2006: Figures: Figure 1: Civilian Hospital Types Treating TRICARE Pediatric Patients in Fiscal Year 2006, by Number of Admissions and Number of Hospitals: Figure 2: Location of Children's Hospitals That Admitted TRICARE Pediatric Patients in Fiscal Year 2006: Figure 3: TRICARE Base Payments to Children's Hospitals Compared to TRICARE Base Payments to Other Hospitals, Fiscal Year 1989 through Fiscal Year 2007: Figure 4: Relative Difference between TRICARE's Base Payment to Children's Hospitals and Other Hospitals, in Percentages, Fiscal Year 1989 through Fiscal Year 2007: Figure 5: Most Common TRICARE Children's Hospital Admissions by Major Diagnostic Category (MDC) and Hospital Type, Excluding Newborns, Fiscal Year 2003 through Fiscal Year 2006: Figure 6: TRICARE Pediatric Admissions to Children's Hospitals, Fiscal Year 2003 through Fiscal Year 2006: Figure 7: Percentage of TRICARE Pediatric Civilian Hospital Admissions Occurring in Children's Hospitals, Fiscal Year 2003 through Fiscal Year 2006: Abbreviations: APR-DRG: All Patient Refined Diagnosis-Related Group: ASA: adjusted standardized amount: CMS: Centers for Medicare & Medicaid Services: DOD: Department of Defense: DRG: diagnosis-related group: MCSC: managed care support contractor: MDC: major diagnostic category: MTF: military treatment facility: NACH: National Association of Children's Hospitals: TMA: TRICARE Management Activity: United States Government Accountability Office: Washington, DC 20548: July 31, 2007: Congressional Committees: Of the more than 9 million individuals who were eligible for TRICARE-- the health program that is managed by the Department of Defense (DOD)- -at the end of fiscal year 2006, about 2 million were children. Under TRICARE, beneficiaries can receive care from military treatment facilities (MTF), which are owned and operated by DOD, or from civilian providers. Of the $10 billion that TRICARE paid for civilian health care services in fiscal year 2006, TRICARE spent a small fraction-- approximately $430 million--on inpatient hospital services for children. Of this amount, about $114 million--26 percent--went to children's hospitals, defined by TRICARE as hospitals in which the majority of patients are under the age of 18.[Footnote 1] Children's hospitals accounted for 7.8 percent of TRICARE's pediatric admissions.[Footnote 2] Inpatient care for children admitted as TRICARE patients can be provided in a variety of civilian hospital settings. Since children's hospitals treat primarily children, they are generally freestanding, meaning that they are not part of a larger hospital that focuses on adult care.[Footnote 3] In contrast, some medical centers that treat adults have a designated pediatric inpatient unit.[Footnote 4] Both children's hospitals and medical centers specialize in treating children with certain rare and complex conditions, performing procedures such as pediatric heart surgeries.[Footnote 5] Community hospitals, on the other hand, typically provide children with more routine services, such as newborn care.[Footnote 6] Like many other hospitals that participate in TRICARE, children's hospitals are generally paid under a prospective payment system. In a prospective payment system, hospitals receive a fixed, predetermined amount per hospital stay.[Footnote 7] Payment is based on the patient's diagnosis and procedures performed during the hospital stay. Stays are classified into diagnosis-related groups (DRG) based on the information that hospitals submit on their claims. Each DRG is assigned a weight, which is a measure of the resources typically required to treat patients whose hospital stays are classified in that DRG, with higher weights reflecting greater use of resources. Because the most resource- intensive cases can be considered the most complex cases, the DRG weight is also called a measure of complexity. Under the prospective payment system, to determine the amount a hospital is to be paid for a single stay the DRG weight is multiplied by the base payment.[Footnote 8] For all hospitals other than children's hospitals, the base payment equals the adjusted standardized amount (ASA), which is TRICARE's annual estimate of the average cost per hospital stay. For children's hospitals, the base payment equals the ASA plus an add-on payment known as the children's hospital differential.[Footnote 9] Children's hospitals have received the differential since their payment began under DOD's prospective payment system on April 1, 1989. Previously, children's hospitals were paid based on their charges--the amount they billed for their services--and DOD viewed payments based on charges as excessive. The purpose of the differential was to recognize that children's hospitals typically had higher charges than other hospitals for the same services and to prevent any reduction in payments to children's hospitals as a result of the transition from a charge-based system to a prospective payment system. Medical centers and community hospitals that treat children admitted as TRICARE patients do not receive the children's hospital differential. Representatives of the National Association of Children's Hospitals (NACH) have stated that TRICARE's prospective payment system does not adequately compensate children's hospitals.[Footnote 10] In particular, these representatives contend that TRICARE pays children's hospitals at rates that are below their costs of care. In addition, NACH representatives state that the children treated at children's hospitals typically have more complex conditions than children at other types of hospitals. Recognizing concerns over TRICARE's payments to children's hospitals, the National Defense Authorization Act for Fiscal Year 2006 directed us to study the effectiveness of the current system of differential payments to children's hospitals under TRICARE.[Footnote 11] Specifically, as discussed with the committees of jurisdiction, this report examines (1) the effect of the differential on TRICARE's base payments to children's hospitals, (2) differences in diagnosis and complexity between TRICARE pediatric patients at children's hospitals and those at other hospitals, (3) the extent to which TRICARE payment differences across hospitals reflect differences in patient complexity, and (4) recent trends in TRICARE pediatric patients' use of children's hospital services. To examine the effect of the differential on children's hospital base payments, we analyzed TRICARE data on base payments from fiscal year 1989 to fiscal year 2007. To compare the diagnoses and complexity of patients at children's hospitals with patients at other hospital types, we used information from TRICARE claims data for all pediatric inpatient admissions to civilian hospitals in the United States for fiscal year 2003 through fiscal year 2006 and a tool to measure patient complexity that was developed by a health information company with input from NACH.[Footnote 12] This tool classifies hospital stays into a more refined set of diagnostic groups than TRICARE's DRG system. We used the same claims data and classification tool to determine the extent to which differences between TRICARE's payments to children's hospitals and TRICARE's payments to other hospitals reflect differences in patient complexity. To identify recent trends in TRICARE pediatric patients' use of children's hospital services, we also analyzed TRICARE pediatric inpatient claims data from fiscal year 2003 through fiscal year 2006. In addition, we interviewed DOD officials on hospital payment policy and representatives of children's hospitals to learn their perspective on the effect of TRICARE's payment policies on TRICARE beneficiaries' access to children's hospital services. We did not attempt to calculate the costs of admissions at children's hospitals because we determined that sufficiently reliable data on children's hospital costs for TRICARE admissions were not available.[Footnote 13] We found that some data fields in the TRICARE claims data were not sufficiently reliable, and we therefore did not use these fields in our analyses. We determined the remaining TRICARE claims data to be sufficiently reliable for the purposes of this report. (See app. I for a detailed explanation of our scope and methodology.) We conducted our work from July 2006 through June 2007 in accordance with generally accepted government auditing standards. Results in Brief: In fiscal year 2007, TRICARE's base payments, a key component of TRICARE's hospital payment formula, were 61 percent higher for facilities that TRICARE defines as children's hospitals than for other hospital types. Base payments to children's hospitals have been substantially higher than base payments to other hospitals since 1989. However, the relative difference in base payments has decreased over time. The relative difference in base payments will continue to decrease, as the children's hospital differential is not adjusted for inflation. From fiscal year 2003 through fiscal year 2006, excluding newborns, the types of diagnoses for TRICARE pediatric patients at children's hospitals were similar to those treated at medical centers. TRICARE pediatric patients at children's hospitals had a similar level of complexity to those at medical centers and were substantially more complex than those at community hospitals. We measured the complexity of patients using a tool that classifies hospital stays into a more refined set of groups than TRICARE's system. Indirect measures of complexity, such as the length of a hospital stay, also showed similarities between TRICARE pediatric patients at children's hospitals and those at medical centers. We found that after we adjusted for differences in patient complexity, TRICARE payments to children's hospitals were substantially greater per admission than TRICARE payments to medical centers and community hospitals. Specifically, holding patient complexity constant, children's hospitals were paid 22 percent more than medical centers and 53 percent more than community hospitals. The number of TRICARE pediatric admissions at children's hospitals increased from 5,027 in fiscal year 2003 to 7,083 in fiscal year 2006. The percentage of TRICARE pediatric admissions in civilian hospitals that occurred at children's hospitals also increased during this time period. The increase in the use of children's hospital services is consistent with statements from representatives of children's hospitals, who said that their hospitals are committed to accepting and caring for TRICARE patients. Our findings show TRICARE's hospital payment system functioning largely as DOD expected, as the difference in base payments to children's hospitals and other hospitals was designed to endure but diminish over time. We have no data on other factors that might support payment differences, however, our findings suggest that further increasing payments to children's hospitals is not supported on the basis of patient complexity. In its comments on a draft of this report, DOD stated that it agreed with our findings and concluding observations. NACH agreed with our findings that TRICARE pediatric patients at children's hospitals were clinically similar to TRICARE pediatric patients at medical centers, and that TRICARE pays children's hospitals more than other hospitals, after accounting for patient complexity. Background: Children's hospitals constitute a small fraction of civilian hospitals providing inpatient services to TRICARE pediatric patients. Children's hospitals have been paid the children's hospital differential since 1989, when they were incorporated under DOD's prospective payment system. The Number and Location of TRICARE Admissions at Children's Hospitals: In fiscal year 2006, there were 7,083 TRICARE pediatric admissions to children's hospitals (see fig. 1). A similar number of admissions, 6,416, occurred in medical centers. In contrast, 77,866 pediatric admissions took place in community hospitals.[Footnote 14] The number of community hospitals that treated TRICARE pediatric patients was substantially higher than the number of children's hospitals or medical centers that treated TRICARE pediatric patients in fiscal year 2006. Specifically, 3,441 community hospitals treated TRICARE pediatric patients compared with 67 children's hospitals and 62 medical centers. Figure 1: Civilian Hospital Types Treating TRICARE Pediatric Patients in Fiscal Year 2006, by Number of Admissions and Number of Hospitals: [See PDF for image] Source: GAO analysis of TRICARE claims data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. In this report, a medical center refers to a teaching hospital that includes a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital." A community hospital is any hospital that was not a children's hospital or medical center under our definitions. [End of figure] TRICARE admissions to children's hospitals were concentrated in a subset of these hospitals in fiscal year 2006. Of the 67 children's hospitals that treated TRICARE pediatric patients, 14 accounted for more than half of the TRICARE children's hospital admissions, and 30 children's hospitals accounted for 84 percent. Children's hospitals that treated TRICARE pediatric patients in fiscal year 2006 are spread throughout the United States (see fig. 2). Similarly, children's hospitals that had more than 200 TRICARE admissions were located in areas that were diverse geographically. States that were home to these high-volume children's hospitals include California, Virginia, Pennsylvania, Texas, Washington, and Alabama. Figure 2: Location of Children's Hospitals That Admitted TRICARE Pediatric Patients in Fiscal Year 2006: [See PDF for image] Source: GAO analysis of TRICARE claims data and MapInfor (map). Note: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. [End of figure] The Establishment of the Children's Hospital Differential: DOD began paying hospitals under its prospective payment system in October 1987, although children's hospitals and certain other types of hospitals were initially exempted.[Footnote 15] TRICARE's prospective payment system was modeled on Medicare's prospective payment system. DOD is required by law to follow Medicare's rules with regard to payment to providers to the extent practicable.[Footnote 16] In 1988, after discussions with children's hospital representatives, DOD proposed including children's hospitals under the prospective payment system and recommended paying those hospitals the children's hospital differential. In December 1988, DOD issued a final rule placing children's hospitals under the prospective payment system and establishing the differential.[Footnote 17] DOD began paying children's hospitals under the prospective payment system on April 1, 1989. DOD established the differential with the goal of ensuring that payments to children's hospitals were not reduced as a result of the transition from the previous charge-based payment system to the prospective payment system as well as to recognize that children's hospitals typically charged more than other hospitals for the same services. The value of the differential is based on a calculation made by DOD that sought to ensure revenue neutrality to children's hospitals. The regulation that established the differential states that it is not to be updated for inflation, and it has not been.[Footnote 18] As of 2007, the value of the differential was set at $2,635.41, and it has changed only twice since 1989.[Footnote 19] When DOD first proposed adopting the children's hospital differential, it expressed concern about the prospective payment system's ability to account for the complexity of children's hospital patients.[Footnote 20] DOD noted that children's hospitals could be particularly susceptible to issues in measuring complexity since children's hospitals often treat complex cases. Like other prospective payment systems, DOD's system does not capture every difference in complexity. For example, patients whose hospital stays are classified into DRG 98, pediatric cases of bronchitis and asthma, may vary in levels of complexity: one patient may have a severe case of bronchitis, while another patient may have a mild case.[Footnote 21] However, all hospital stays in DRG 98 receive the same DRG weight and therefore are paid the same rate.[Footnote 22] As a result, a hospital that consistently treats patients with severe cases of bronchitis will be paid no more for those admissions than a hospital that consistently treats patients with less severe cases of bronchitis, even though the hospital would likely incur higher costs for treating the more severe cases.[Footnote 23] However, it is also expected that at most hospitals, these differences in complexity will "balance out." In other words, a hospital may treat some patients who have severe cases of bronchitis, but the hospital will also treat some patients who have mild cases of bronchitis, so that overall the hospital will treat children who are at the average complexity of the DRG. TRICARE's Base Payments to Children's Hospitals Substantially Higher Than Base Payments to Other Hospitals, Though Relative Difference in Payments Decreases over Time: TRICARE's base payments to children's hospitals have been substantially higher than base payments to other hospital groups, although the relative difference in base payments has declined over time. For fiscal year 2007, TRICARE's base payments to children's hospitals were set 61 percent higher than base payments to all other hospitals. However, the relative difference between TRICARE's base payments to children's hospitals and base payments to other hospitals has decreased, and it will continue to decrease over time. Due to the Differential, Children's Hospitals Have Received Substantially Higher Base Payments from TRICARE: As a result of the children's hospital differential, children's hospitals have received substantially higher base payments than other hospitals under TRICARE's prospective payment system--61 percent higher in fiscal year 2007. Base payments to children's hospitals have been substantially higher than base payments to other hospitals since the children's hospital differential was established in 1989 (see fig. 3). So long as the TRICARE prospective payment system continues to include a children's hospital differential, base payments to children's hospitals will always be higher than base payments to other hospitals. Figure 3: TRICARE Base Payments to Children's Hospitals Compared to TRICARE Base Payments to Other Hospitals, Fiscal Year 1989 through Fiscal Year 2007: [See PDF for image] Source: GAO analysis of TRICARE payment data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. Other hospitals include medical centers, which in this report are teaching hospitals that include a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital," and community hospitals, which in this report are hospitals that were not children's hospitals or medical centers under our definitions. This figure reflects the two changes to the children's hospital differential. The first change occurred in fiscal year 1992, when the children's hospital differential for hospitals in large urban areas and hospitals in other areas was adjusted (the adjustment for the differential for children's hospitals in large urban areas was so slight that it is difficult to discern from the figure). In fiscal year 2005, the children's hospital differential was increased for hospitals located in areas other than large urban areas. In this figure, that change is reflected in the data for fiscal year 2007, which is the first year shown after the fiscal year 2005 change. [End of figure] Relative Difference in TRICARE's Base Payments between Children's Hospitals and Other Hospitals Has Decreased over Time: Although TRICARE's base payment to children's hospitals remains higher than the base payment to other hospitals, the relative difference between the two base payments has decreased, as the ASA has been adjusted for inflation and the children's hospital differential has not. In fiscal year 1989, the base payment to children's hospitals in large urban areas was 92 percent greater than the base payment to other hospitals in those areas (see fig. 4). Eighteen years later, the relative difference in base payments has been reduced. By fiscal year 2007, TRICARE's base payment to children's hospitals in large urban areas exceeded TRICARE's base payment to other hospitals in large urban areas by 61 percent. The relative difference in base payments between children's hospitals and other hospitals in areas other than large urban areas has also decreased. In fiscal year 1989, the base payment to children hospitals in other areas was 79 percent greater than the base payment to other hospitals in those areas. In fiscal year 2007, the base payment to children's hospitals in other areas exceeded the base payment to other hospitals in those areas by 61 percent.[Footnote 24] Figure 4: Relative Difference between TRICARE's Base Payment to Children's Hospitals and Other Hospitals, in Percentages, Fiscal Year 1989 through Fiscal Year 2007: [See PDF for image] Source: GAO analysis of TRICARE payment data. Notes: Percentages represent the amount by which children's hospital base payments are higher than base payments to other hospitals--for example, 92 means that children's hospital base payments were 92 percent higher than base payments to other hospitals. Projections are based on the assumption that the ASA continues to increase at an annual rate of 2.4 percent. A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. Other hospitals include medical centers, which in this report are teaching hospitals that include a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital," and community hospitals, which in this report are hospitals that were not children's hospitals or medical centers under our definitions. [End of figure] The relative difference in base payments will continue to decline so long as the ASA is increased to account for inflation and the children's hospital differential is not. Since 1989, the ASA for hospitals in large urban areas has increased at an average annual rate of 2.4 percent. If that rate continues, the base payment to children's hospitals will be 45 percent higher than the base payment to other hospitals in 2020. The relative difference will never disappear entirely, however, as long as children's hospitals continue to receive the children's hospital differential. TRICARE Pediatric Patients at Children's Hospitals Similar to Those at Medical Centers in Terms of Diagnoses and Complexity: From fiscal year 2003 through fiscal year 2006, children's hospitals treated TRICARE pediatric patients for the same types of diagnoses as medical centers, with the exception of newborns, which more often received care at medical centers than at children's hospitals. TRICARE patients at children's hospitals were similar in complexity levels to TRICARE pediatric patients treated at medical centers. In contrast, TRICARE patients at children's hospitals were more than three times as complex as those at community hospitals. Children's Hospitals and Medical Centers Treated TRICARE Pediatric Patients for Similar Types of Diagnoses: Children's hospitals and medical centers treated TRICARE pediatric patients for similar types of diagnoses from fiscal year 2003 through fiscal year 2006, although children's hospitals were less likely to treat newborns. Once newborns are excluded, the pattern of diagnoses at children's hospitals was very similar to the pattern of diagnoses at medical centers (see fig. 5). Newborns accounted for about 10 percent of TRICARE pediatric patients at children's hospitals, 35 percent of TRICARE pediatric patients at medical centers, and 73 percent of TRICARE pediatric patients at community hospitals. Figure 5: Most Common TRICARE Children's Hospital Admissions by Major Diagnostic Category (MDC) and Hospital Type, Excluding Newborns, Fiscal Year 2003 through Fiscal Year 2006: [See PDF for image] Source: GAO analysis of TRICARE claims data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. In this report, a medical center refers to a teaching hospital that includes a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital." [End of figure] For patients at both children's hospitals and medical centers, the three most common major diagnostic categories were related to the respiratory system, nervous system, and digestive system. Common diagnoses related to these systems include asthma, seizure and headache, and appendicitis, respectively. Compared to medical centers, children's hospitals were slightly more likely to treat children with circulatory system disorders, such as hypertension and heart failure. TRICARE Pediatric Patients at Children's Hospitals Were Similar to Those at Medical Centers Based on Measures of Complexity: We found that from fiscal year 2003 through fiscal year 2006, the average complexity of TRICARE pediatric patients at children's hospitals was about 10 percent higher than the average complexity of TRICARE pediatric patients at medical centers. For the same time period, the average complexity of pediatric patients at children's hospitals was more than three times as high as the average complexity of pediatric patients at community hospitals. In conducting this analysis, we used a tool that measures the complexity of diagnostic groups; a score of 1.0 serves as a reference point for relative complexity.[Footnote 25] Using this reference, we found that the average patient complexity of pediatric admissions at children's hospitals was 1.62, while at medical centers the score was 1.47 (see table 1). In contrast, the average pediatric patient complexity at community hospitals was .52. The relatively low level of complexity of patients at community hospitals is driven by the large percentage of normal newborns, babies that do not have any complications and therefore have a low level of complexity.[Footnote 26] Table 1: Average TRICARE Patient Complexity by Hospital Type, Fiscal Year 2003 through Fiscal Year 2006: Hospital type: Children's hospitals; Average patient complexity[A]: 1.62. Hospital type: Medical centers; Average patient complexity[A]: 1.47. Hospital type: Community hospitals; Average patient complexity[A]: .52. Source: GAO analysis of TRICARE claims data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. In this report, a medical center refers to a teaching hospital that includes a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital." A community hospital is any hospital that was not a children's hospital or medical center under our definitions. [A] The average patient complexity is often called the case mix index. [End of table] Indirect measures of complexity--length of hospital stay, hospital transfers, and in-hospital deaths--show comparable differences. From fiscal year 2003 through fiscal year 2006, length of hospital stay for pediatric admissions at children's hospitals and medical centers averaged about 6 days; transfers from another hospital were somewhat more frequent at children's hospitals than at medical centers; and frequency of pediatric admissions ending in death was about 1 percent in both settings (see table 2). In contrast, stays at community hospitals averaged 3.5 days and percentages of transfers and in- hospital deaths at community hospitals were substantially lower, at about 3 percent and less than 1 percent, respectively. Table 2: Indirect Measures of TRICARE Pediatric Patient Complexity by Hospital Type, Fiscal Year 2003 through Fiscal Year 2006: Hospital type: Children's hospitals; Average length of stay per admission (in days): 6.1; Percentage of admissions that are transfers from other hospitals: 8.8; Percentage of admissions ending in death: 1.1. Hospital type: Medical centers; Average length of stay per admission (in days): 6.2; Percentage of admissions that are transfers from other hospitals: 5.5; Percentage of admissions ending in death: 1.4. Hospital type: Community hospitals; Average length of stay per admission (in days): 3.5; Percentage of admissions that are transfers from other hospitals: 3.3; Percentage of admissions ending in death: 0.4. Source: GAO analysis of TRICARE claims data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. In this report, a medical center refers to a teaching hospital that includes a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital." A community hospital is any hospital that was not a children's hospital or medical center under our definitions. [End of table] After Adjusting for Patient Complexity, Children's Hospitals Were Paid More per TRICARE Pediatric Admission Than Other Hospitals: After comparing pediatric patients at children's hospitals to patients at other hospital types, we examined hospitals' payments per admission, adjusting for patient complexity. Using claims data from fiscal year 2003 through fiscal year 2006, we found that after adjusting for patient complexity, children's hospitals were paid substantially more per admission than both medical centers and community hospitals (see table 3). Table 3: Average TRICARE Payment per Pediatric Admission, Patient Complexity, and Payment Adjusted for Complexity, by Hospital Type, Fiscal Year 2003 through Fiscal Year 2006: Hospital type: Children's hospitals; Average payment per admission: $16,367; Average patient complexity: 1.62; Average payment adjusted for complexity: $10,089. Hospital type: Medical centers; Average payment per admission: $12,131; Average patient complexity: 1.47; Average payment adjusted for complexity: $8,275. Hospital type: Community hospitals; Average payment per admission: $3,401; Average patient complexity: .52; Average payment adjusted for complexity: $6,596. Source: GAO analysis of TRICARE claims data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. In this report, a medical center refers to a teaching hospital that includes a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital." A community hospital is any hospital that was not a children's hospital or medical center under our definitions. The average payment per admission has been adjusted upwards for medical centers and community hospitals to account for payments for capital and direct medical education expenses. Average payment adjusted for complexity equals average payment per admission divided by average patient complexity. However, due to rounding, the calculations do not work out perfectly. [End of table] We adjusted for patient complexity for the three hospital types by dividing the average payment per pediatric admission by the average patient complexity. For example, across the 4-year period, TRICARE payments to children's hospitals--adjusted for the average patient complexity--averaged $10,089 per patient, based on an average payment of $16,367 per admission and an average complexity of 1.62. This average complexity-adjusted payment to children's hospitals was 22 percent higher than the equivalent amount paid to medical centers, which was $8,275. TRICARE payments to children's hospitals were 53 percent higher than those made to community hospitals for pediatric patients, which were $6,596 after adjusting for patient complexity. Rising Number of TRICARE Admissions at Children's Hospitals Suggests No Decline in Access: From fiscal year 2003 through fiscal year 2006, TRICARE pediatric admissions at children's hospitals rose steadily, suggesting that access to children's hospital services has not decreased in recent years. Specifically, the total number of TRICARE pediatric admissions rose from 5,027 admissions in fiscal year 2003 to 7,083 admissions in fiscal year 2006 (see fig. 6). This change represents an increase of 41 percent for the time period. Figure 6: TRICARE Pediatric Admissions to Children's Hospitals, Fiscal Year 2003 through Fiscal Year 2006: [See PDF for image] Source: GAO analysis of TRICARE claims data. Note: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. [End of figure] The proportion of TRICARE pediatric civilian hospital admissions that occurred in children's hospitals also increased in recent years. In fiscal year 2006, children's hospitals accounted for 7.8 percent of all TRICARE pediatric admissions to civilian hospitals, up from 6.2 percent in fiscal year 2003 (see fig. 7). Figure 7: Percentage of TRICARE Pediatric Civilian Hospital Admissions Occurring in Children's Hospitals, Fiscal Year 2003 through Fiscal Year 2006: [See PDF for image] Source: GAO analysis of TRICARE claims data. Note: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. [End of figure] The increase in the use of children's hospital services is consistent with statements made by representatives of children's hospitals about their policy toward TRICARE patients.[Footnote 27] These representatives stated that children's hospitals are committed to treating all children, including TRICARE patients, because of their legal obligations as nonprofit hospitals as well as their mission to serve all patients.[Footnote 28] These statements, coupled with recent trends in utilization, suggest that TRICARE pediatric patients' access to children's hospitals has not declined in recent years. Concluding Observations: The current children's hospital payment system is functioning largely as DOD expected. In establishing a policy of inflation updates to the ASA, but no inflation updates to the children's hospital differential, DOD set up a system in which the difference between children's hospital base payments and base payments to other hospitals would endure, but would be reduced gradually over time. This reduction has taken place as planned. Given the lack of reliable data, we cannot know the cost to children's hospitals of treating TRICARE beneficiaries and thus cannot know how their costs compared to payment amounts. Although greater patient complexity has been cited as a rationale for larger payments to children's hospitals, our analysis shows that patient complexity for children's hospital admissions was roughly comparable to those at medical centers. While we have only limited indicators of the extent to which TRICARE pediatric patients have access to children's hospitals, we did not find data that would support concerns about access problems. Agency and Professional Association Comments and Our Evaluation: We obtained written comments on a draft of this report from DOD, which are reprinted in appendix II. DOD concurred with our findings and conclusions and said that the report was technically accurate. We also obtained oral comments from representatives of NACH. They agreed with our finding that TRICARE pays children's hospitals more than other hospitals, after accounting for patient complexity, and agreed with our finding that TRICARE pediatric patients at children's hospitals were clinically similar to TRICARE pediatric patients at medical centers. Despite this similarity, NACH said the two types of hospitals have important differences--most notably that medical centers are typically larger institutions than children's hospitals and therefore can achieve greater economies of scale. Given this difference, NACH officials noted the importance of examining whether TRICARE's payments met children's hospital costs. However, as we state in the report, this analysis was beyond the scope of our work--as agreed to with the committees of jurisdiction--because sufficiently reliable data on children's hospital costs were not available. NACH officials raised a concern related to our analysis of the percentage difference in complexity-adjusted payments to children's hospitals and other hospital types. Specifically, they suggested that the percentage difference between complexity-adjusted payments at children's hospitals and other hospital types would change if outlier claims--claims with unusually high charges given their DRGs--were analyzed separately. We could not perform this analysis because the TRICARE claims data base could not be used to reliably identify all claims that were cost outliers. Noting that utilization of children's hospital services is an imperfect measure of access, NACH officials suggested that the increase in the use of children's hospital services could have resulted from community hospitals providing fewer specialty pediatric services. NACH officials also said that our findings could have resulted from increases in the number of children enrolled in TRICARE. However, as noted in our report, the percentage of all TRICARE pediatric admissions that occurred in children's hospitals also increased, supporting our finding that access to children's hospitals does not appear to have declined. Additionally, we received technical comments from NACH, which we incorporated as appropriate. We are sending copies of this report to the Secretary of Defense, and other interested parties. We will also provide copies to others on request. In addition, the report is available at no charge on GAO's Web site at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-7114 or ekstrandl@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix III. Signed by: Laurie Ekstrand: Director, Health Care: List of Committees: The Honorable Carl Levin: Chairman: The Honorable John McCain: Ranking Member: Committee on Armed Services: United States Senate: The Honorable Daniel K. Inouye: Chairman: The Honorable Ted Stevens: Ranking Member: Subcommittee on Defense: Committee on Appropriations: United States Senate: The Honorable Ike Skelton: Chairman: The Honorable Duncan Hunter: Ranking Member: Committee on Armed Services: House of Representatives: The Honorable John P. Murtha: Chairman: The Honorable C.W. Bill Young: Ranking Member: Subcommittee on Defense: Committee on Appropriations: House of Representatives: [End of section] Appendix I: Scope and Methodology: To analyze the change in TRICARE base payments over time, we obtained data on the adjusted standardized amount (ASA) and the children's hospital differential from the TRICARE Management Activity (TMA), the office that manages TRICARE. Using these data, we calculated the base payment to children's hospitals and to other hospitals for each year since 1989. Most of the remainder of our analysis was based on claims data we obtained from TMA. The data include TRICARE claims from U.S. civilian hospitals from fiscal year 2003 through fiscal year 2006 for all patients under the age at 18 at the time of admission. To analyze the claims data, we divided providers into three separate categories: children's hospitals, medical centers, and community hospitals. We identified children's hospitals as those designated as such by TRICARE. A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. We identified hospitals as medical centers if they contained a pediatric inpatient unit that was designated as a "children's hospital within a hospital" by the National Association of Children's Hospitals (NACH). We classified all other hospitals as community hospitals. In analyzing diagnoses and complexity, we examined a subset of claims. Our analysis was of TRICARE's prospective payment system, and therefore we aimed to exclude all claims that were paid outside the prospective payment system. We excluded claims from hospitals that are exempt from TRICARE's prospective payment system. This group of providers includes psychiatric hospitals, rehabilitation hospitals, sole community hospitals, and all institutions in Maryland (hospitals in Maryland are exempt from TRICARE's prospective payment system).[Footnote 29] We excluded claims that had an indicator stating that they were paid according to an alternative payment system, such as a per diem payment system. We excluded claims that were paid by a health insurance program other than TRICARE, since these claims can be paid according to the payment rules of the other payer, with TRICARE as the secondary payer. We excluded all claims related to bone marrow transplants, cystic fibrosis, or care for children with HIV, since those claims are excluded from TRICARE's prospective payment system in cases for which the patient is a child. As a result of these exclusions, our universe of claims was reduced from 348,225 claims to 265,857 claims. We included claims that were paid under a discount rate agreement. These claims accounted for about half of the claims in our analysis. The discounted claims can be paid as a percentage discount off the prospective payment rate, or they can be paid under an alternate payment methodology. We included these claims even though some of these claims may not have been paid under the prospective payment system. We concluded that regardless of whether these claims were paid under a prospective payment system, the terms of the discount rate agreement were based on the fact that the hospital was eligible to be paid under a prospective payment system. Measuring Complexity of Admissions: To account for the complexity of admissions, we obtained the All Patient Refined Diagnosis-Related Group (APR-DRG) grouper program from 3M Health Information Systems (3M). The APR-DRG grouper program was developed by 3M with input from NACH, which offered its expertise on classifying pediatric admissions. The APR-DRG grouper program divides claims into groups, known as APR-DRGs. We applied the APR-DRG program to the subset of claims that we analyzed. (We also excluded claims that the APR-DRG grouper program could not categorize). We also obtained a file of APR-DRG weights from 3M, and we merged this file with our claims data based on the APR-DRG assigned to each claim. We used this APR-DRG weight as our refined measure of complexity. Like TRICARE's DRG grouper program, the APR-DRG grouper program assigns claims to a diagnosis group (called an APR-DRG in the case of the APR- DRG grouper) based on diagnostic, procedural, and demographic information on the claim. However, the APR-DRG grouper divides claims into a greater number of categories than the TRICARE DRG grouper program. The APR-DRG grouper program divides claims into 1,258 categories; in comparison, the TRICARE DRG grouper program divides claims into 553 categories. Since the APR-DRG grouper divides claims into more groups that are more clinically homogeneous, there is less variation in complexity within those groups. For example, TRICARE's DRG grouper program would classify a severe case of pediatric asthma into the same category as a mild case of pediatric asthma, so long as the patient did not require ventilator support. The APR-DRG grouper program, on the other hand, would place these two cases into separate categories and therefore assign them different weights. As a result, the APR-DRG grouper program produces a more refined measure of complexity, as compared to the TRICARE DRG grouper program. The average complexity of children's hospital claims varied depending on which grouper program was used to measure complexity. The average complexity of TRICARE pediatric admissions to children's hospitals was 1.62 using the APR-DRG grouper, 4 percent higher than the average complexity of TRICARE pediatric admissions to children's hospitals when the TRICARE DRG grouper was used to measure complexity (see table 4). The average complexity of TRICARE pediatric admissions at medical centers was approximately the same, regardless of which grouper program was used to measure complexity. In contrast, the average complexity of TRICARE pediatric admissions at community hospitals was lower when the APR-DRG grouper was used to measure complexity than when the TRICARE DRG grouper was used to measure complexity. Table 4: Average Complexity per TRICARE Pediatric Admission by Hospital Type, Using Both TRICARE DRGs and APR-DRGs, Fiscal Year 2003 through Fiscal Year 2006: Hospital type: Children's hospitals; Average complexity per admission: Using TRICARE's DRGs: 1.56; Average complexity per admission: Using APR-DRGs: 1.62; Percentage difference between APR-DRG complexity and TRICARE DRG complexity: 4. Hospital type: Medical centers; Average complexity per admission: Using TRICARE's DRGs: 1.47; Average complexity per admission: Using APR-DRGs: 1.47; Percentage difference between APR-DRG complexity and TRICARE DRG complexity: 0. Hospital type: Community hospitals; Average complexity per admission: Using TRICARE's DRGs: .56; Average complexity per admission: Using APR- DRGs: .52; Percentage difference between APR-DRG complexity and TRICARE DRG complexity: -8. Source: GAO analysis of TRICARE claims data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. In this report, a medical center refers to a teaching hospital that includes a pediatric inpatient unit that is designated by NACH as a "children's hospital within a hospital." A community hospital is any hospital that was not a children's hospital or medical center under our definitions. [End of table] In comparing payments to complexity, we adjusted for complexity by dividing the payment for the claim by the APR-DRG weight. For claims that occurred at medical centers and community hospitals, the payment on the claim was increased by a percentage adjustment. We applied this percentage adjustment to account for payments that medical centers and community hospitals receive for their direct medical education and capital expenses, payments that children's hospitals do not receive. We calculated a percentage adjustment of 8.7 percent for community hospitals and 9.6 percent for medical centers based on data on capital and direct medical education payments provided by TMA. Measuring Inflation: To assess the level of hospital inflation, we analyzed data from the Centers for Medicare & Medicaid Services (CMS) on the agency's Inpatient Prospective Payment System Hospital 2002 Input Price Index and compared it to TRICARE base payments to children's hospitals. Since 1992, the percentage increase in TRICARE base payments to children's hospitals has been less than the percentage increase in hospital costs. From fiscal years 1992 through 2006, hospital inflation has increased an average of 3.2 percent annually (see table 5). In contrast, base payments to children's hospitals in large urban areas have increased by 1.2 percent annually, while base payments to children's hospitals in other areas have increased by 2.1 percent annually. Table 5: Change in TRICARE Base Payments to Children's Hospitals Compared to the Change in Hospital Inflation, Fiscal Year 1992 through Fiscal Year 2006: Measure: Hospital inflation; Percentage increase, FY 1992 through FY 2006: Average annual increase: 3.2; Percentage increase, FY 1992 through FY 2006: Cumulative increase: 56.0. Measure: Base payment to children's hospitals in large urban areas; Percentage increase, FY 1992 through FY 2006: Average annual increase: 1.2; Percentage increase, FY 1992 through FY 2006: Cumulative increase: 18.5. Measure: Base payment to children's hospitals in other areas; Percentage increase, FY 1992 through FY 2006: Average annual increase: 2.1; Percentage increase, FY 1992 through FY 2006: Cumulative increase: 34.5. Source: GAO analysis of TRICARE payment data and CMS hospital inflation data. Notes: A children's hospital under TRICARE is one in which at least 50 percent of a hospital's patients are children. Hospital inflation is measured by the CMS Inpatient Prospective Payment System Hospital 2002 Input Price Index. [End of table] TRICARE base payments to children's hospitals in other areas increased at a faster rate than TRICARE base payments to children's hospitals in large urban areas for two primary reasons. In fiscal year 2003, the ASA for children's hospitals in other areas was increased to match the higher ASA for children's hospitals in large urban areas. In addition, in fiscal year 2005 the differential for children's hospitals in other areas was increased to the level of the higher differential for children's hospitals in large urban areas. As a result of these two changes, TRICARE base payments to children's hospitals in other areas increased by 24 percent from fiscal year 2003 through fiscal year 2005. We conducted our work from July 2006 through June 2007 in accordance with generally accepted government auditing standards. [End of section] Appendix II: Comments from the Department of Defense: Health Affairs: The Assistant Secretary Of Defense: 1200 Defense Pentagon: Washington, DC 20301-1200: Jul 16 2007: Ms. Laurie Ekstrand: Director, Health Care: U.S. Government Accountability Office: 441 G Street, N. W. Washington, DC 20548: Dear Ms. Ekstrand: This is the Department of Defense (DoD) response to the General Accountability Office (GAO) draft report, GAO-07-947, "Defense Health Care: Under TRICARE, Children's Hospitals Paid More than Other Hospitals After Accounting for Patient Complexity," dated June 27, 2007 (GAO Code 290560). Thank you for the opportunity to review and provide comments on the Draft Report. We have reviewed the report for technical accuracy and agree with all of the findings. In addition, I concur with the Draft Report's conclusions. DoD is pleased that the GAO found that TRICARE's hospital payment system for children's hospitals is functioning largely as expected. My points of contact are Ms. Reta Michak (Functional) at (303) 676-3440 and Mr. Gunther Zimmerman (Audit Liaison) at (703) 681-3492. Sincerely, Signed by: S. Ward Casscells, MD: [End of section] Appendix III: GAO Contact and Staff Acknowledgments: GAO Contact: Laurie Ekstrand, (202) 512-7114 or ekstrandl@gao.gov: Acknowledgments: In addition to the contact above, Phyllis Thorburn, Assistant Director; Alexander Dworkowitz; Hannah Fein; Jenny Grover; Darryl Joyce; Richard Lipinski; and Dae Park made key contributions to this report. FOOTNOTES [1] In this report, a children's hospital refers to a hospital that TRICARE identifies as a children's hospital. At least 50 percent of a hospital's patients must be children in order for TRICARE to identify the hospital as a children's hospital. In addition, TRICARE officials use information from the American Hospital Association to confirm that a hospital is a children's hospital. [2] In this report, we define pediatric admissions as admissions of children under age 18. [3] Examples of children's hospitals are Children's Hospital of Philadelphia and Children's Hospital of The King's Daughters in Norfolk, Virginia. [4] Examples of medical centers with a designated pediatric inpatient unit are the University of Michigan Health System, which includes C.S. Mott Children's Hospital, and the University of California Los Angeles Medical Center, which includes Mattel Children's Hospital. [5] In this report, we define medical centers as hospitals that contain a pediatric inpatient unit that is designated by the National Association of Children's Hospitals as a "children's hospital within a hospital." Some hospitals commonly described as medical centers are included in this category--if these hospitals do not contain a specialized pediatric unit and do not meet the criteria of a children's hospital, they are categorized as community hospitals. See app. I for more information. [6] In this report, we define any hospital that is not a children's hospital or a medical center as a community hospital. This group of community hospitals includes some hospitals that may be referred to as medical centers elsewhere. [7] Prospective payment systems are designed to give hospitals incentives to contain costs in that hospitals are allowed to retain any funds not spent on care. [8] Hospitals also have their payment adjusted based on the area wage level and for their indirect medical education expenses, which are calculated based on the ratio of medical residents to hospital beds. In addition, all hospitals except children's hospitals can receive separate payments for their capital and direct medical educational expenses. [9] The term base payment is not used by TRICARE in the same context. In this report, our definition of base payment is the amount that is multiplied by the DRG weight to determine actual payment, before adjustments for the area wage level and indirect medical education expenses are applied. Our use of the term reflects language used by the Centers for Medicare & Medicaid Services. [10] Members of this organization include more than 120 hospitals that focus on treating children. [11] See Pub. L. No. 109-163, § 734, 119 Stat. 3136, 3353-55; S. Rep. No. 109-69, at 337 (2006). [12] Our unit of analysis was a hospital admission. Multiple admissions of the same patient during our period of analysis would be counted separately. [13] Prior to the release of this report, DOD and NACH were planning to produce a reliable measurement of children's hospital costs of treating TRICARE beneficiaries. [14] An additional 61,438 pediatric admissions occurred in U.S.-based MTFs in fiscal year 2006. [15] TRICARE does not pay all hospitals under a prospective payment system. Rehabilitation hospitals and psychiatric hospitals, among others, are by regulation exempt from the system. In addition, TRICARE maintains networks of providers, and hospitals can join that network and negotiate a discount rate agreement with managed care support contractors (MCSC), organizations that manage provider networks on behalf of TRICARE. This discount can take the form of a discount off the prospective payment rate. Alternatively, MCSCs can negotiate to pay hospitals under a different methodology, such as a per diem rate. Claims that were paid at a discount were included in our analysis. For more information, see app. I. [16] See 10 U.S.C. § 1079(j)(2). [17] See 53 Fed. Reg. 50515-20 (Dec. 16, 1988). This final rule was consistent with the Department of Defense Appropriations Act for Fiscal Year 1989, Pub. L. No. 100-463, § 8091, 102 Stat. 2270, 2270-33 to 2270- 34 (1988). [18] See 32 C.F.R. § 199.14(a)(1)(iii)(E)(4)(v) (2006). [19] From April 1, 1989, to April 1, 1992, children's hospitals that had a high volume of TRICARE admissions (defined as 50 or more TRICARE admissions per year) received a hospital-specific children's hospital differential, and the remaining hospitals were assigned one of two national differentials: one for children's hospitals in large urban areas and another, lower differential for children's hospitals in other areas. (A hospital was considered to be located in a large urban area if was located in a metropolitan statistical area, as defined by the Office of Management and Budget, that had a population of more than 1 million, or in a New England County Metropolitan Area with a population of more than 970,000.) On April 1, 1992, DOD stopped paying high-volume hospitals a hospital-specific differential and recalculated the values of the national differentials to include data from the high-volume children's hospitals. This was the first change in the differential. The second change occurred at the beginning of fiscal year 2005, when the value of the differential for children's hospitals in other areas was increased to the value of the differential for children's hospitals in large urban areas. [20] See 53 Fed. Reg. 20576, 20579-80 (June 3, 1988). [21] Children who are suffering from asthma or bronchitis and need to be placed on a ventilator are typically classified into DRGs other than DRG 98. [22] This assumes the admission is not classified as an outlier. [23] This assumes the two hospitals have the same wage adjustment and are paid the same amount for their indirect medical education expenses. [24] The decline in the relative difference in base payments between children's hospitals and other hospitals in other areas was mitigated by the 2005 increase in the children's hospital differential for children's hospitals in other areas. [25] We measured complexity using the All Patient Refined Diagnosis- Related Group (APR-DRG) grouper program, which is a more refined measure of complexity than that used by TRICARE. The company that developed the APR-DRG grouper refers to patient complexity as severity of illness. For more information, see app. I. [26] From fiscal year 2003 through fiscal year 2006, the average pediatric patient complexity, excluding normal newborns, was 1.65 at children's hospitals, 1.73 at medical centers, and .96 at community hospitals. [27] Any hospital that participates in Medicare is legally required to accept TRICARE patients, and many children's hospitals accept Medicare patients. See 42 U.S.C. § 1395cc(a)(1)(J). However, hospitals are not required to join TRICARE's network of providers. TRICARE beneficiaries who need a referral to see an out-of-network provider could face restrictions in accessing children's hospital services if many children's hospitals declined to join TRICARE's network. [28] Children's hospital representatives did express concern about the level of TRICARE payments affecting their ability to maintain readily available services and noted that this could have a negative impact on patient waiting times. [29] Maryland hospitals are also exempted from Medicare's prospective payment system. GAO's Mission: The Government Accountability Office, the audit, evaluation and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's Web site (www.gao.gov). Each weekday, GAO posts newly released reports, testimony, and correspondence on its Web site. To have GAO e-mail you a list of newly posted products every afternoon, go to www.gao.gov and select "Subscribe to Updates." Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office 441 G Street NW, Room LM Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov Automated answering system: (800) 424-5454 or (202) 512-7470: Congressional Relations: Gloria Jarmon, Managing Director, JarmonG@gao.gov (202) 512-4400 U.S. Government Accountability Office, 441 G Street NW, Room 7125 Washington, D.C. 20548: Public Affairs: Paul Anderson, Managing Director, AndersonP1@gao.gov (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, D.C. 20548: