This is the accessible text file for GAO report number GAO-13-364 entitled 'Defense Health Care: TRICARE Multiyear Surveys Indicate Problems with Access to Care for Nonenrolled Beneficiaries' which was released on April 2, 2013. 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. United States Government Accountability Office: GAO: Report to Congressional Committees: April 2013: Defense Health Care: TRICARE Multiyear Surveys Indicate Problems with Access to Care for Nonenrolled Beneficiaries: GAO-13-364: GAO Highlights: Highlights of GAO-13-364, a report to congressional committees. Why GAO Did This Study: DOD provides health and mental health care through its TRICARE program. TRICARE offers three basic options. Beneficiaries who choose TRICARE Prime, an option that uses civilian provider networks, must enroll. Beneficiaries who do not enroll in this option may obtain care from non-network providers under TRICARE Standard or from network providers under TRICARE Extra. In addition, qualified National Guard and Reserve servicemembers may purchase TRICARE Reserve Select, a plan whose care options are similar to those of TRICARE Standard and TRICARE Extra. GAO refers to servicemembers who use TRICARE Standard, TRICARE Extra, or TRICARE Reserve Select as nonenrolled beneficiaries. The National Defense Authorization Act for Fiscal Year 2008 directed DOD to conduct annual surveys over fiscal years 2008 through 2011 of both beneficiaries and civilian providers to determine the adequacy of access to health and mental health care providers for nonenrolled beneficiaries. It also directed GAO to review these surveys. This report addresses (1) what the results of the 4-year beneficiary surveys indicate about the adequacy of access to care for nonenrolled beneficiaries; (2) what the results of the 4-year civilian provider surveys indicate about civilian providers’ awareness and acceptance of TRICARE, and (3) what the collective results of the surveys indicate about access to care by geographic area. To do so, GAO interviewed DOD officials, obtained relevant documentation, and analyzed the data for both surveys over the 4-year period. What GAO Found: In its analysis of the 2008-2011 beneficiary survey data, GAO found that nearly one in three nonenrolled beneficiaries experienced problems finding a civilian provider who would accept TRICARE and that nonenrolled beneficiaries’ access to civilian primary care and specialty care providers differed by type of location. Specifically, a higher percentage of nonenrolled beneficiaries in Prime Service Areas (PSA), which are areas with civilian provider networks, experienced problems finding a civilian primary care or specialty care provider compared to those in non–Prime Service Areas (non-PSA), which do not have civilian provider networks. GAO found that the top reasons reported by nonenrolled beneficiaries for why they experienced access problems—-regardless of type of provider—-were that the providers were either not accepting TRICARE payments or new TRICARE patients. Additionally, GAO’s comparison of the Department of Defense’s (DOD) beneficiary survey data to related data from a Department of Health and Human Services survey showed that nonenrolled beneficiaries’ satisfaction ratings for primary and specialty care providers were consistently lower than those of Medicare fee-for-service beneficiaries. GAO’s analysis of the 2008-2011 civilian provider survey data found that about 6 in 10 civilian providers were accepting new TRICARE patients and the most-cited reason for not accepting new TRICARE patients was that the civilian providers were not aware of the TRICARE program. Civilian physicians’ acceptance of TRICARE has also decreased over time. Specifically, when compared to DOD’s 2005-2007 civilian physician survey results, civilian physicians’ acceptance of new TRICARE patients has decreased. This was also true whether they were accepting any new patients or new Medicare patients. Civilian providers’ awareness and acceptance of TRICARE also differed by provider type, as fewer civilian mental health care providers were aware of TRICARE or accepting new TRICARE patients than other types of providers. For example, only an estimated 39 percent of civilian mental health care providers were accepting new TRICARE patients, compared to an estimated 67 percent of civilian primary care providers and an estimated 77 percent of civilian specialty care providers. The analysis also showed that civilian providers’ awareness and acceptance of TRICARE differ by location type, as civilian providers in PSAs were less aware of TRICARE and less likely to accept new TRICARE patients than those in non-PSAs. GAO’s analysis of the collective results of the beneficiary and civilian provider survey results indicates specific geographic areas, including areas in Texas and California, where nonenrolled beneficiaries have experienced considerable access problems. In each of these areas, although almost all civilian providers were accepting new patients, less than half were accepting new TRICARE patients. In most of these areas, civilian providers most often cited reimbursement concerns as the reasons why they were not accepting any new TRICARE patients. In commenting on a draft of this report, DOD concurred with GAO’s overall findings. View [hyperlink, http://www.gao.gov/products/GAO-13-364]. For more information, contact Debra A. Draper at (202) 512-7114 or draperd@gao.gov. [End of section] Contents: Letter: Background: Nearly One in Three Nonenrolled Beneficiaries Experienced Problems Accessing Care, and They Rated Their Satisfaction with Care Generally Lower than Medicare Fee-for-Service Beneficiaries: Civilian Providers' Acceptance of New TRICARE Patients Has Decreased over Time; Mental Health Providers Report Lower Awareness and Acceptance than Other Provider Types: Collective Results of TMA's Beneficiary and Civilian Provider Surveys Indicate Specific Geographic Areas Where Nonenrolled Beneficiaries Have Experienced Access Problems: Agency Comments and Our Evaluation: Appendix I: TRICARE Management Activity's Methodology for the 2008- 2011 Beneficiary and Civilian Provider Surveys: Appendix II: Beneficiary Survey Instrument: Appendix III: Survey Instruments for Civilian Physicians and Nonphysician Mental Health Care Providers: Appendix IV: Areas Where Nonenrolled Beneficiaries Had Problems Accessing Civilian Providers, 2008-2011: Appendix V: Civilian Provider Acceptance of Any New TRICARE Patients in Hospital Service Areas Surveyed in Fiscal Year 2011: Appendix VI: Comments from the Department of Defense: Appendix VII: GAO Contact and Staff Acknowledgments: Tables: Table 1: Summary of TRICARE Options: Table 2: Comparison of TRICARE Network and Non-network Civilian Providers: Table 3: Requirements for Annual Beneficiary and Provider Surveys Contained in the National Defense Authorization Act for Fiscal Year 2008 (NDAA 2008): Table 4: Areas Where the Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Provider Was at Least the National Estimate and Where the Percentage of Civilian Providers Who Were Accepting Any New TRICARE Patients Was at or below the National Estimate, 2008-2011: Table 5: Civilian Providers' Estimated Percentage of Acceptance of New Patients and New TRICARE Patients, by Problem Area, 2008-2011: Table 6: Areas Where the Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Primary Care Provider Was at Least the National Estimate and Where the Percentage of Civilian Primary Care Providers Who Were Accepting Any New TRICARE Patients Was at or below the National Estimate, 2008-2011: Table 7: Civilian Primary Care Providers' Estimated Percentage of Acceptance of New Patients and New TRICARE Patients, by Problem Area, 2008-2011: Table 8: Number of Beneficiary Surveys Mailed, Returned, and Complete and Eligible, by Fiscal Year: Table 9: Number of Civilian Physician and Nonphysician Mental Health Provider Surveys Mailed, Returned, and Complete and Eligible, by Fiscal Year: Table 10: Prime Service Areas (PSA), Non-Prime Service Areas (non- PSA), and Hospital Service Areas (HSA) with 31 Percent or More of Nonenrolled Beneficiaries Experiencing Problems Finding Any Type of Provider, and the Willingness of Civilian Providers in the Corresponding Areas to Accept New TRICARE Patients, 2008-2011: Table 11: Prime Service Areas (PSA), Non-Prime Service Areas (non- PSA), and Hospital Service Areas (HSA) with 25 Percent or More of Nonenrolled Beneficiaries Experiencing Problems Finding a Civilian Primary Care Provider, and the Willingness of Civilian Primary Care Providers in the Corresponding Areas to Accept New TRICARE Patients, 2008-2011: Table 12: Prime Service Areas (PSA), Non-Prime Service Areas (non- PSA), and Hospital Service Areas (HSA) with 25 Percent or More of Nonenrolled Beneficiaries Experiencing Problems Finding Civilian Specialist Providers, and the Willingness of Civilian Specialist Providers in the Corresponding Areas to Accept New TRICARE Patients, 2008-2011: Table 13: Hospital Service Areas (HSA) Surveyed in 2011, and the Estimated Percentage of Civilian Providers Who Were Accepting Any New TRICARE Patients: Figures: Figure 1: Percentage of Outpatient Claims Paid for TRICARE Prime, TRICARE Standard, TRICARE Extra, and TRICARE Reserve Select for Fiscal Years 2008 through 2011: Figure 2: Types of Nonenrolled TRICARE Beneficiaries: Figure 3: Geographic Location of Nonenrolled TRICARE Beneficiaries: Figure 4: Prime Service Areas (PSA) and Non-Prime Service Areas (non- PSA) Surveyed for TRICARE Management Activity's 4-Year Beneficiary and Provider Surveys, 2008-2011: Figure 5: Hospital Service Areas (HSA) Surveyed for TRICARE Management Activity's (TMA) 4-Year Beneficiary and Provider Surveys, 2008-2011: Figure 6: Estimated Percentages of Nonenrolled TRICARE Beneficiaries Who Experienced Access Problems, by Civilian Provider Type, in Prime Service Areas (PSA) and Non-Prime Service Areas (non-PSA), 2008-2011: Figure 7: Estimated Percentages of Nonenrolled TRICARE Beneficiaries Who Experienced Access Problems, by Civilian Provider Type, in Hospital Service Areas (HSA) and Prime Service Areas (PSA)/non-PSAs Outside of Surveyed HSAs, 2008-2011: Figure 8: Top Five Reasons Reported by Nonenrolled Beneficiaries Who Experienced Problems Accessing Civilian Primary, Specialty, or Mental Health Care, 2008-2011: Figure 9: Nonenrolled TRICARE Beneficiaries' Estimated Satisfaction Ratings Compared to Those of Commercially Insured, Medicaid, and Medicare Fee-For-Service Beneficiaries, 2008-2011: Figure 10: Top Seven Categories of Reasons for Not Accepting New TRICARE Patients Reported by Civilian Providers That Were Not Accepting Any New TRICARE Patients, 2008-2011: Figure 11: Civilian Physicians' Awareness and Acceptance of TRICARE over Time in TRICARE Management Activity's (TMA) Surveys: Figure 12: Civilian Providers' Awareness and Acceptance of TRICARE, by Type of Provider, 2008-2011: Figure 13: Top Categories of Reasons Reported by Civilian Providers for Not Accepting New TRICARE Patients, by Provider Type, 2008-2011: Figure 14: Civilian Providers' Awareness and Acceptance of TRICARE in Prime Service Areas (PSA) and non-Prime Service Areas (non-PSA), 2008- 2011: Figure 15: Civilian Providers' Awareness and Acceptance of TRICARE, by Hospital Service Areas (HSA) and Prime Service Areas/non-Prime Service Areas (PSA/non-PSA) outside of the Surveyed HSAs, 2008-2011: Figure 16: Estimated Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Primary, Specialty, or Mental Health Care Provider, 2008-2011: Figure 17: Estimated Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Primary Care Provider, 2008- 2011: Figure 18: Estimated Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Specialty Care Provider, 2008- 2011: Abbreviations: BRAC: Base Realignment and Closure: CAHPS: Consumer Assessment of Healthcare Providers and Systems: DEERS: Defense Enrollment Eligibility Reporting System: DOD: Department of Defense: HHS: Department of Health and Human Services: HSA: Hospital Service Area: NDAA: National Defense Authorization Act: NDAA 2008: National Defense Authorization Act for Fiscal Year 2008: non-PS: Anon-Prime Service Area: OMB: Office of Management and Budget: PSA: Prime Service Area: TMA: TRICARE Management Activity: TRS: TRICARE Reserve Select: [End of section] United States Government Accountability Office: Washington, DC 20548: April 2, 2013: The Honorable Carl Levin: Chairman: The Honorable James Inhofe: Ranking Member: Committee on Armed Services: United States Senate: The Honorable Howard "Buck" McKeon: Chairman: The Honorable Adam Smith: Ranking Member: Committee on Armed Services: House of Representatives: In fiscal year 2012, the Department of Defense (DOD) offered health care services, including mental health care services, to about 9.7 million eligible beneficiaries in the United States and abroad through TRICARE, DOD's regionally structured health care program.[Footnote 1] Under TRICARE, beneficiaries may obtain care either from military hospitals and clinics, referred to as military treatment facilities, or from civilian providers.[Footnote 2] DOD's TRICARE Management Activity (TMA), which oversees the program, uses managed care support contractors[Footnote 3] to develop networks of civilian providers--referred to as network providers--to serve all TRICARE beneficiaries in geographic areas called Prime Service Areas (PSA).[Footnote 4] The contractors use estimates of the number of TRICARE users, among other factors, to develop provider networks and ensure adequate access to care for beneficiaries. Although some network providers may be located outside of PSAs, contractors are not required to develop networks in these areas (which we refer to as non- PSAs). The number and type of civilian providers available to serve TRICARE beneficiaries can vary depending on a beneficiary's location and choice of coverage among TRICARE's three basic plans--TRICARE Prime, TRICARE Standard, and TRICARE Extra.[Footnote 5] Beneficiaries who use TRICARE Prime, a managed care option, must enroll and can obtain care through military treatment facilities or TRICARE's civilian provider network. Beneficiaries do not need to enroll to receive care under TRICARE Standard, a fee-for-service option, or TRICARE Extra, a preferred provider organization option; they can choose to receive care through TRICARE Standard when they are seeing non-network civilian providers and through TRICARE Extra when they are seeing network civilian providers.[Footnote 6] We use the term "nonenrolled beneficiaries" for beneficiaries who are not enrolled in TRICARE Prime and who use the TRICARE Standard or Extra options, or TRICARE Reserve Select (TRS).[Footnote 7] Since TRICARE's inception in 1995, nonenrolled beneficiaries in some locations have complained about difficulties finding civilian providers who will accept them as patients. In response to these concerns, the National Defense Authorization Act (NDAA) for Fiscal Year 2004 directed DOD to monitor access to care for nonenrolled TRICARE beneficiaries through a survey of civilian providers.[Footnote 8] The act also directed GAO to review the processes, procedures, and analyses used by DOD to determine the adequacy of the number of network and non-network civilian providers and the actions DOD has taken to ensure access to care for beneficiaries who were not enrolled in TRICARE Prime. In December 2006, we reported that TMA and contractor officials used various methods to evaluate access to care, including the survey of civilian providers, and according to those officials, their methods indicated that access was generally sufficient for nonenrolled beneficiaries.[Footnote 9] Nonetheless, concerns about the ability of TRICARE beneficiaries, particularly nonenrolled beneficiaries, to access health care and mental health care continued. In light of these continued concerns about access to civilian providers, the NDAA for Fiscal Year 2008 (NDAA 2008) directed DOD to conduct annual surveys over 4 years of both beneficiaries and civilian providers to determine the adequacy of access to health care and mental health care providers for nonenrolled beneficiaries.[Footnote 10] It also directed GAO to review these surveys along with other factors such as DOD's outreach, marketing, and education efforts, and provider reimbursement issues. We have issued several reports that address the topics covered in this mandate, including a March 2010 report on the methodology and results of the first year of DOD's 4-year beneficiary and provider surveys. [Footnote 11] In our initial review of the surveys, we reported that a higher percentage of nonenrolled beneficiaries in the surveyed PSAs experienced problems accessing care from civilian primary care providers than those in the surveyed non-PSAs. However, we could not reach any generalizable conclusions about the civilian provider survey because it had not generated sufficient survey responses during the first year. The NDAA for Fiscal Year 2012 extended DOD's annual beneficiary and provider surveys for another 4 years, from fiscal years 2012 through 2015.[Footnote 12] As of early January 2013, TMA had mailed the 2012 beneficiary and civilian provider survey instruments. This report addresses DOD's beneficiary and civilian provider surveys for the first 4-year survey period, covering fiscal years 2008 through 2011. Specifically, it addresses (1) what the results of the 4-year beneficiary surveys indicate about the adequacy of access to care for nonenrolled beneficiaries, (2) what the results of the 4-year civilian provider surveys indicate about civilian providers' awareness and acceptance of TRICARE, and (3) what the collective results of the 4- year beneficiary and civilian provider surveys indicate about access to care for nonenrolled beneficiaries by geographic area. To determine what the results of the 4-year beneficiary surveys indicate about the adequacy of access to care for nonenrolled beneficiaries, we obtained and analyzed survey data on access to civilian primary,[Footnote 13] specialty,[Footnote 14] and mental health care[Footnote 15] from TMA's TRICARE Standard Surveys of Beneficiaries for 2008 through 2011. For the purposes of our analysis, we analyzed survey results for those nonenrolled beneficiaries who reported using TRICARE Standard, TRICARE Extra, or TRS the most in the last year. Because the overall response rate for the 4 years was about 38 percent,[Footnote 16] we verified that TMA's survey results were representative of the areas surveyed by reviewing TMA's nonresponse analyses and interviewing TMA officials.[Footnote 17] We also obtained and analyzed the Department of Health and Human Services' (HHS) Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey data for the same 2008-2011 period in order to compare nonenrolled TRICARE beneficiaries' satisfaction with their health care, health plan, and primary and specialty care providers to that of Medicare fee- for-service, Medicaid, and commercially insured beneficiaries. [Footnote 18] We assessed the reliability of these data by obtaining information from knowledgeable officials and reviewing related documentation, and we determined that TMA's 4-year beneficiary survey data and HHS's CAHPS data were sufficiently reliable for our purposes. To determine what the results of the 4-year civilian provider surveys indicate about civilian providers' awareness and acceptance of TRICARE, we obtained and analyzed the survey data from TMA's TRICARE Standard Surveys of Providers for 2008 through 2011. Because the overall response rate was about 42 percent,[Footnote 19] we verified that TMA's civilian provider survey results were representative of the areas surveyed by reviewing TMA's nonresponse analyses and interviewing TMA officials.[Footnote 20] We compared the civilian provider survey results to those of a national survey of physicians conducted in 2008 by the Center for Studying Health System Change to compare civilian providers' acceptance of any new TRICARE patients to providers' acceptance of any new Medicare (fee-for-service or managed care), Medicaid, and commercially insured beneficiaries.[Footnote 21] We also compared the results of TMA's 4-year civilian provider survey to those of TMA's 2005-2007 civilian physician survey to identify any changes in physicians' awareness and acceptance over time.[Footnote 22] We assessed the reliability of these data by speaking with knowledgeable officials and reviewing related documentation, and we determined that these data were sufficiently reliable for our purposes. To determine what the results of the collective analysis of the 4-year beneficiary and civilian provider surveys indicate about access to care for nonenrolled beneficiaries, we compared the results of our analyses of the 4-year beneficiary and provider survey data by specific geographic regions where possible, in order to identify areas with both high percentages of nonenrolled beneficiaries who experienced problems finding civilian providers and low percentages of civilian providers who were accepting new TRICARE patients. Specifically, we identified areas where the estimated percentage of nonenrolled beneficiaries that experienced problems finding a civilian provider was either at or above the national estimate for nonenrolled beneficiaries, using the 95 percent confidence limits. For these geographic areas, we then looked at civilian provider acceptance of new TRICARE patients and identified areas where the percentage of civilian providers that were accepting any new TRICARE patients was at or below the national estimate, using the 95 percent confidence limits. Our analyses have some limitations. In our analyses of TMA's beneficiary and provider surveys we report survey results for some individual areas, but we were unable to compare survey results among all of the individual geographic areas surveyed because of low numbers of respondents in some areas. Similarly, in our analysis of TMA's beneficiary survey we were unable to identify specific geographic areas in which nonenrolled beneficiaries experienced problems finding mental health care providers because of the low numbers of respondents who indicated that they needed mental health care. We conducted this performance audit from June 2012 through February 2013 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Background: Under TRICARE, beneficiaries have choices among various benefit options and may obtain care from either military treatment facilities or civilian providers. When nonenrolled beneficiaries receive care from civilian providers, they have the option of seeing either network or non-network providers. The NDAA 2008 directed DOD to conduct surveys of beneficiaries and civilian providers to assess nonenrolled beneficiaries' access to care. TRICARE's Benefit Options: TRICARE provides benefits through several basic options for its non- Medicare-eligible beneficiary population. These options vary by enrollment requirements, choices in civilian and military treatment facility providers, and the amount beneficiaries must contribute toward the cost of their care. Table 1 provides a summary of some of these benefit options. Table 1: Summary of TRICARE Options: TRICARE option: TRICARE Prime; Description: This managed care option requires enrollment, and all active duty servicemembers are required to use this option, while other TRICARE beneficiaries have a choice. TRICARE Prime enrollees receive most of their care from providers at military treatment facilities and also may receive care from network civilian providers. This option has the lowest out-of-pocket costs for beneficiaries. TRICARE option: TRICARE Standard and TRICARE Extra; Description: TRICARE beneficiaries who choose not to enroll in TRICARE Prime may obtain health care from non-network providers (under TRICARE Standard) or network civilian providers (under TRICARE Extra). The TRICARE Standard option is designed to provide beneficiaries with maximum flexibility in selecting providers, but beneficiaries who obtain care from a network provider, through TRICARE Extra, pay lower copayments than they would under the TRICARE Standard option. TRICARE Standard and Extra beneficiaries also may receive care from military treatment facilities, though they have a lower priority for receiving care than do TRICARE Prime beneficiaries. TRICARE option: TRICARE Reserve Select (TRS); Description: TRS is a premium-based health plan that certain National Guard and Reserve members may purchase.[A] TRS beneficiaries may obtain health care from either non-network or network providers, similarly to beneficiaries using TRICARE Standard or Extra, respectively, and will pay lower copayments for using network providers. Source: GAO summary of DOD TRICARE documentation. [End of table] [A] To be eligible for TRS, the beneficiary must be a member of the Selected Reserve of the Ready Reserve, and not eligible for or enrolled in the Federal Employees Health Benefits program, either under their own eligibility or through a family member who is enrolled in a family plan. Claims data from fiscal years 2008 to 2011 show that the percentages of the number of outpatient claims paid for TRICARE Prime and TRS have gradually increased, while the percentage of claims paid for TRICARE Standard has declined. (See figure 1.) The percentage of claims paid for TRICARE Extra has remained steady over the same period. Figure 1: Percentage of Outpatient Claims Paid for TRICARE Prime, TRICARE Standard, TRICARE Extra, and TRICARE Reserve Select for Fiscal Years 2008 through 2011: [Refer to PDF for image: vertical bar graph] Percentage of TRICARE claims paid: FY 2008[A]: TRICARE Prime: 68%; TRICARE Standard: 13%; TRICARE Extra: 17%; TRICARE Reserve Select: 1%. FY 2009[A]: TRICARE Prime: 69%; TRICARE Standard: 12%; TRICARE Extra: 17%; TRICARE Reserve Select: 2%. FY 2010[A]: TRICARE Prime: 69%; TRICARE Standard: 10%; TRICARE Extra: 17%; TRICARE Reserve Select: 3%. FY 2011[A]: TRICARE Prime: 70%; TRICARE Standard: 9%; TRICARE Extra: 17%; TRICARE Reserve Select: 4%. Source: GAO analysis of TMA data. Notes: All percentages may not add up to 100 percent because of rounding. Claims were for outpatient services provided in an office or other setting outside of an institution. Claims for services rendered at hospitals, military treatment facilities, and other institutions were excluded. TRICARE for Life and TRICARE Young Adult claims were excluded, as well as claims for medical supplies and from chiropractors and pharmacies. [A] Percentages calculated on the basis of total number of outpatient claims: 23,995,179 claims in fiscal year 2008; 26,950,329 claims in fiscal year 2009; 29,857,355 claims in fiscal year 2010; and 32,012,220 claims in fiscal year 2011. [End of figure] Starting on September 30, 2013, the number of PSAs will be reduced, and as a result, the TRICARE Prime option will be available to fewer beneficiaries. The targeted PSAs are those that are not in close proximity to existing MTFs or BRAC locations and will predominantly affect retirees and their dependents. According to a TMA official, this change is expected to affect about 171,000 retirees and dependents (37,000 in the North region, 36,000 in the West region, and 98,000 in the South region), with an estimated savings to DOD of $45 million to $56 million annually.[Footnote 23] Composition of TRICARE's Nonenrolled Beneficiary Population: In fiscal year 2011, TMA identified about 2 million nonenrolled beneficiaries (approximately one-fourth of the total eligible TRICARE population), who fell into three main categories: (1) retirees and their dependents or survivors, (2) active duty dependents, and (3) National Guard and Reserve servicemembers and their dependents.[Footnote 24] (See figure 2.) Figure 2: Types of Nonenrolled TRICARE Beneficiaries: [Refer to PDF for image: pie-chart] Retirees and their dependents or survivors: 69%; Active duty dependents: 17%; National Guard and Reserve servicemembers and their dependents[B]: 13%; Other[A]: 1%. Source: GAO analysis of TMA data. Notes: Nonenrolled beneficiaries are beneficiaries not enrolled in TRICARE Prime who are eligible for TRICARE Standard or Extra, as well as TRICARE Reserve Select enrollees. Data are for nonenrolled beneficiaries as of December 31, 2010. [A] Other nonenrolled beneficiaries include family members of deceased servicemembers and secretarial designees. [B] Limited to deactivated National Guard or Reserve servicemembers and their dependents. Dependents of activated National Guard or Reserve servicemembers are included in the "active duty dependents" category. [End of figure] Most of these nonenrolled beneficiaries lived in PSAs--areas where TRICARE managed care support contractors have developed provider networks. (See figure 3.) Figure 3: Geographic Location of Nonenrolled TRICARE Beneficiaries: [Refer to PDF for image: pie-chart] Prime Service Area: 81%; Non-Prime Service Area: 19%. Source: GAO analysis of TMA data. Note: Nonenrolled beneficiaries are beneficiaries not enrolled in TRICARE Prime who are eligible for TRICARE Standard or Extra, as well as TRICARE Reserve Select enrollees. Data are for nonenrolled beneficiaries as of December 31, 2010. [End of figure] TRICARE Network and Non-network Civilian Providers: In order for network and non-network civilian providers to be authorized to provide care and be reimbursed under TRICARE, they must meet the licensing and certification requirements of TRICARE regulations and practices for their area of health care. Individual TRICARE-authorized civilian providers can include health care providers, such as primary care physicians and specialists, as well as mental health care providers, including clinical psychologists. Table 2 provides a comparison of network and non-network civilian providers. Table 2: Comparison of TRICARE Network and Non-network Civilian Providers: Network Civilian Providers: * Are TRICARE-authorized providers who enter contractual agreements with the TRICARE regional managed care support contractors in their areas to provide health care and mental health care to TRICARE beneficiaries; Non-network Civilian Providers: * Are TRICARE-authorized providers who do not have contractual agreements with regional managed care support contractors to provide care to TRICARE beneficiaries. Network Civilian Providers: * Have agreed to accept TRICARE reimbursement rates. By law, TRICARE maximum allowable reimbursement rates generally must mirror Medicare rates, but network providers may agree to accept lower reimbursements as a condition of network membership; Non-network Civilian Providers: * May choose to accept the TRICARE reimbursement rate as payment in full for their services, or may charge up to 15 percent more than the TRICARE reimbursement rate for their services on a case-by-case basis (with the difference paid by the beneficiary). Network Civilian Providers: * Are not obligated to accept all TRICARE beneficiaries seeking care. For example, a network civilian provider may decline to accept TRICARE beneficiaries as patients because the provider's practice does not have sufficient capacity; Non-network Civilian Providers: * May accept TRICARE beneficiaries as patients on a case-by-case basis. Network Civilian Providers: * Have agreed to meet TRICARE Management Activity's access to care standards for TRICARE Prime enrollees. For example, these providers are required to offer urgent care appointments within 24 hours; Non-network Civilian Providers: * Are not required to meet TRICARE's access to care standards. Source: GAO. [End of table] DOD's Implementation of the NDAA 2008 Beneficiary and Civilian Provider Survey Requirements: The NDAA 2008 directed DOD to conduct surveys of beneficiaries and civilian providers in at least 20 PSAs and 20 non-PSAs in each of 4 fiscal years, 2008 through 2011.[Footnote 25] figure 4 shows the 80 PSAs and 80 non-PSAs surveyed over the 4-year period of 2008 through 2011. Figure 4: Prime Service Areas (PSA) and Non-Prime Service Areas (non- PSA) Surveyed for TRICARE Management Activity's 4-Year Beneficiary and Provider Surveys, 2008-2011: [Refer to PDF for image: illustrated U.S. map] For TRICARE North, South, and West regions, the map depicts the following: PSA; Non-PSA. Source: GAO analysis of TMA data (data); MapInfo (map). [End of figure] The NDAA 2008 also required DOD to consult with representatives of TRICARE beneficiaries and health care and mental health care providers to identify locations where nonenrolled beneficiaries have experienced significant access-to-care problems--which TMA uses Hospital Service Areas (HSA) to define--and to survey health care and mental health care providers in these areas.[Footnote 26] figure 5 shows the 71 HSAs identified as problem areas by providers and beneficiary groups. [Footnote 27] (See appendix I for TMA's methodology in implementing the beneficiary and civilian provider surveys.) Figure 5: Hospital Service Areas (HSA) Surveyed for TRICARE Management Activity's (TMA) 4-Year Beneficiary and Provider Surveys, 2008-2011: [Refer to PDF for image: illustrated U.S. map] For TRICARE North, South, and West regions, the map depicts the following: HSA. Source: GAO analysis of TMA data (data); MapInfo (map). Note: For the 4-year provider surveys, TMA surveyed a total of 71 HSAs from 2008 to 2011, shown above. Fifty-five of these 71 HSAs were also surveyed for the beneficiary survey from 2008 to 2010, but according to TMA officials, no HSAs were surveyed for the 2011 beneficiary survey because of funding issues. [End of figure] The NDAA 2008 also required that specific types of information be requested in the surveys. For example, the beneficiary survey must include questions to determine whether nonenrolled beneficiaries have difficulties finding a provider who will accept TRICARE, and the civilian provider survey must include questions to determine whether civilian providers are aware of TRICARE. (See appendices II and III for the 2011 beneficiary and civilian provider survey instruments, respectively.) Table 3 lists the NDAA 2008 requirements for DOD's beneficiary and civilian provider surveys. Table 3: Requirements for Annual Beneficiary and Provider Surveys Contained in the National Defense Authorization Act for Fiscal Year 2008 (NDAA 2008): Requirement: Survey goals; Requirement description: 1. Determine the number of health care providers in TRICARE Prime Service Areas (PSA) that are accepting new patients under TRICARE Standard and Extra. Requirement description: 2. Determine the number of health care providers in TRICARE non-Prime Service Areas (non-PSA) that are accepting patients under TRICARE Standard and Extra. Requirement description: 3. Determine the availability of mental health care providers in TRICARE PSAs and TRICARE non-PSAs. Requirement: Survey area selection; Requirement description: 4. Survey beneficiaries and providers in at least 20 TRICARE PSAs in each fiscal year to determine the availability of health care providers accepting new patients under TRICARE Standard and Extra. Requirement description: 5. Survey beneficiaries and providers in 20 non-PSAs in which significant numbers of beneficiaries who are members of the Selected Reserve reside, to determine the availability of health care providers accepting new patients under TRICARE Standard and Extra. Requirement description: 6. Survey beneficiaries and providers in at least 40 total PSAs and non-PSAs to determine the availability of mental health care providers. Requirement description: 7. In prioritizing areas to be surveyed, give a high priority to surveying beneficiaries and providers located in geographic areas with high concentrations of members of the Selected Reserve. Requirement description: 8. In prioritizing areas to be surveyed, consult with representatives of TRICARE beneficiaries and health care and mental health care providers to identify locations where nonenrolled beneficiaries are experiencing significant levels of access-to-care problems under TRICARE Standard or Extra and give a high priority to surveying health care and mental health care providers in these locations. Requirement: Beneficiary survey content; Requirement description: 9. Include questions in beneficiary surveys seeking information to determine whether they have difficulties in finding health care and mental health care providers willing to provide services under TRICARE Standard or Extra. Requirement: Provider survey content; Requirement description: 10. Include questions in provider surveys to determine the following: * Whether the provider is aware of the TRICARE program; * What percentage of the provider's current patient population uses any form of TRICARE; * Whether the provider accepts patients for whom payment is made under the Medicare program for health care and mental health care services; * If the provider accepts Medicare patients, whether the provider would accept new Medicare patients. Requirement: Benchmarks; Requirement description: 11. Establish benchmarks to determine the adequacy of the availability of health care and mental health care providers to beneficiaries eligible for TRICARE. Source: GAO analysis of legislation. Note: Data are based on review of the NDAA 2008 § 711(a). [End of table] We previously reported that TMA generally addressed the requirements outlined in the NDAA 2008 during the implementation of its 2008 beneficiary and provider surveys, but because of methodological considerations TMA used a different--but acceptable--approach for its selection of survey areas.[Footnote 28] We also found that TMA's methodology for both of the surveys was consistent with the Office of Management and Budget (OMB) standards for statistical surveys that we reviewed. Since then, TMA has made several minor revisions to the surveys' methodologies for 2009 through 2011, but none of these changes are inconsistent with the NDAA 2008 requirements. Nearly One in Three Nonenrolled Beneficiaries Experienced Problems Accessing Care, and They Rated Their Satisfaction with Care Generally Lower than Medicare Fee-for-Service Beneficiaries: Nearly One in Three Nonenrolled Beneficiaries Experienced Problems Finding Civilian Providers Who Would Accept TRICARE; Those in PSAs Experienced More Problems Finding Primary and Specialty Care than Those in Non-PSAs: Overall, during 2008-2011, an estimated one in three nonenrolled beneficiaries (about 31 percent) experienced problems finding any type of civilian provider--primary, specialty, or mental health care provider--who would accept TRICARE. Specifically: * an estimated 25 percent of nonenrolled beneficiaries experienced problems finding a civilian primary care provider; * an estimated 25 percent of nonenrolled beneficiaries experienced problems finding a civilian specialty care provider; and: * an estimated 28 percent experienced problems accessing a civilian mental health care provider.[Footnote 29] Overall, access to civilian primary care and specialty care providers differed for nonenrolled beneficiaries located in PSAs compared to those in non-PSAs. Specifically, we found that more nonenrolled beneficiaries in PSAs experienced problems finding civilian primary care and specialty care providers compared to those in non-PSAs. (See figure 6.) However, access to civilian mental health care providers did not differ for nonenrolled beneficiaries in PSAs and non-PSAs. Figure 6: Estimated Percentages of Nonenrolled TRICARE Beneficiaries Who Experienced Access Problems, by Civilian Provider Type, in Prime Service Areas (PSA) and Non-Prime Service Areas (non-PSA), 2008-2011: [Refer to PDF for image: vertical bar graph] Estimated percentage of nonenrolled beneficiaries: Civilian provider type: Primary care[A]; PSAs: 27%[D]; 95% confidence interval for estimates: 26-28%; Non-PSAs: 19%; 95% confidence interval for estimates: 18-20%. Civilian provider type: Specialty care[B]; PSAs: 26%[D]; 95% confidence interval for estimates: 25-27%; Non-PSAs: 20%; 95% confidence interval for estimates: 19-21%. Civilian provider type: Mental health care[C]; PSAs: 27%; 95% confidence interval for estimates: 24-30%; Non-PSAs: 28%; 95% confidence interval for estimates: 25-31%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. [A] Respondents answered "a big problem" or "a small problem" to the question that asked: In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE? Answer choices were "A big problem," "A small problem," or "Not a problem." [B] Respondents answered "a big problem" or "a small problem" to the question that asked: "In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE? Answer choices were "A big problem," "A small problem," or "Not a problem." [C] Respondents answered "a big problem" or "a small problem" to the question that asked: Based on the following: In the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan? Answer choices were "A big problem," "A small problem," or "Not a problem." [D] Within provider type, the difference in estimates between PSAs and non-PSAs is significantly different at the 95 percent confidence level. [End of figure] TMA also surveyed beneficiaries in HSAs in response to access concerns about these specific areas. We found that more nonenrolled beneficiaries in HSAs experienced problems accessing civilian specialty care than those in the areas outside of the surveyed HSAs. [Footnote 30] (See figure 7.) However, there were no statistical differences in the estimated percentages of nonenrolled beneficiaries who experienced problems finding civilian primary or mental health care providers between the HSAs and the locations surveyed outside of these areas. Figure 7: Estimated Percentages of Nonenrolled TRICARE Beneficiaries Who Experienced Access Problems, by Civilian Provider Type, in Hospital Service Areas (HSA) and Prime Service Areas (PSA)/non-PSAs Outside of Surveyed HSAs, 2008-2011: [Refer to PDF for image: vertical bar graph] Estimated percentage of nonenrolled beneficiaries: Civilian provider type: Primary care[A]; HSAs: 27%; 95% confidence interval for estimates: 24-30%; PSAs/Non-PSAs outside of surveyed HSAs: 25%; 95% confidence interval for estimates: 24-26%. Civilian provider type: Specialty care[B]; HSAs: 29%[D]; 95% confidence interval for estimates: 26-32%; PSAs/Non-PSAs outside of surveyed HSAs: 24%; 95% confidence interval for estimates: 23-25%. Civilian provider type: Mental health care[C]; HSAs: 24%; 95% confidence interval for estimates: 18-30%; PSAs/Non-PSAs outside of surveyed HSAs: 28%; 95% confidence interval for estimates: 25-31%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. Each surveyed HSA was part of a PSA or non-PSA (depending on the location), and because HSAs were not mutually exclusive of the PSAs or non-PSAs, we did not compare the results from nonenrolled beneficiaries in HSAs to nonenrolled beneficiaries in PSAs or non- PSAs. Instead, we compared the results for the nonenrolled beneficiaries in the surveyed HSAs to those nonenrolled beneficiaries in the collective areas outside the surveyed HSAs. [A] Respondents answered "a big problem" or "a small problem" to the question that asked: In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE? Answer choices were "A big problem," "A small problem," or "Not a problem." [B] Respondents answered "a big problem" or "a small problem" to the question that asked: In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE? Answer choices were "A big problem," "A small problem," or "Not a problem." [C] Respondents answered "a big problem" or "a small problem" to the question that asked: In the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan? Answer choices were "A big problem," "A small problem," or "Not a problem." [D] The estimates of nonenrolled beneficiaries who experienced problems finding a civilian specialty care provider between HSAs and PSAs/non-PSAs outside of the surveyed HSAs are significantly different at the 95 percent confidence level. [End of figure] The top two reasons reported by nonenrolled beneficiaries--regardless of type of care--for why they believed they experienced problems accessing a provider included "doctors not accepting TRICARE payments" and "doctors not accepting new TRICARE patients." (See figure 8.) Figure 8: Top Five Reasons Reported by Nonenrolled Beneficiaries Who Experienced Problems Accessing Civilian Primary, Specialty, or Mental Health Care, 2008-2011: [Refer to PDF for image: vertical bar graph] Beneficiaries’ reasons for access problems: Doctor(s) not accepting TRICARE payments: Primary care[A]: 56%; 95% confidence interval for estimates: 54-58%; Specialty care[B]: 55%; 95% confidence interval for estimates: 52-58 Mental health care[C]: 45[D]%; 95% confidence interval for estimates: 39-51%. Doctor(s) not taking new TRICARE patients: Primary care[A]: 36%; 95% confidence interval for estimates: 34-38%; Specialty care[B]: 29%; 95% confidence interval for estimates: 27-31%; Mental health care[C]: 25%; 95% confidence interval for estimates: 19-31%. Doctor(s) not taking any new patients: Primary care[A]: 27%; 95% confidence interval for estimates: 25-29%; Specialty care[B]: 18%; 95% confidence interval for estimates: 16-20%; Mental health care[C]: 17%; 95% confidence interval for estimates: 12-22%. Travel distance too long: Primary care[A]: 23%; 95% confidence interval for estimates: 21-25%; Specialty care[B]: 26%; 95% confidence interval for estimates: 24-28%; Mental health care[C]: 24%; 95% confidence interval for estimates: 19-29%. Wait for an appointment was too long: Primary care[A]: 14%[E]; 95% confidence interval for estimates: 12-16%; Specialty care[B]: 19%; 95% confidence interval for estimates: 17-21%; Mental health care[C]: 24%; 95% confidence interval for estimates: 19-29%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. Estimated percentages are out of the total estimated number of nonenrolled beneficiaries who experienced any problems accessing civilian primary, specialty, or mental health care providers. Percentages across problem types do not add up to 100 percent because respondents were able to select more than one response, and only the top five responses for primary and specialty care are shown. In addition to the responses above, the top five responses for mental health care included "Other," with an estimated 21 percent of nonenrolled beneficiaries (plus or minus 5 percentage points) indicating "Other" as a reason for having problems finding a provider. Unless otherwise noted below, differences in estimates within each problem type are not significantly different at the 95 percent confidence level. [A] Based on the following: "What problems did you encounter in finding a personal doctor who would accept TRICARE?" [B] Based on the following: "What problems did you encounter in finding a specialist who would accept TRICARE?" [C] Based on the following: "In the last 12 months, what problems did you encounter in finding treatment or counseling?" [D] The difference in estimates between mental health care and other care types is statistically significant at the 95 percent confidence level. [E] The difference in estimates between primary care and other care types is statistically significant at the 95 percent confidence level. [End of figure] Nonenrolled Beneficiaries' Satisfaction Did Not Differ across Types of Areas, but Was Generally Lower than That of Medicare Fee-for-Service Beneficiaries: Our analysis of the 4-year survey data showed that nonenrolled beneficiaries' ratings for specific satisfaction measures were similar when compared between PSAs and non-PSAs, and between surveyed HSAs and the areas outside of the surveyed HSAs. Specifically, our analysis of beneficiaries' ratings for four measures--satisfaction with primary care providers, specialty care providers, health care, and health plan--indicated no substantial differences between area types. [Footnote 31] For example, we found that about 80 percent of nonenrolled beneficiaries in both PSAs and non-PSAs rated their primary care provider as an 8 or higher on a scale from 0 to 10. [Footnote 32] Additionally, we found that nonenrolled TRICARE beneficiaries' satisfaction ratings for several of these measures were generally lower than those of Medicare fee-for-service beneficiaries and varied compared to Medicaid and commercially insured beneficiaries during the same 4-year period,[Footnote 33] according to HHS's 2008-2011 CAHPS surveys.[Footnote 34] (See figure 9.) For example, we found that fewer nonenrolled TRICARE beneficiaries rated their primary care provider, specialty care provider, and health plan as an 8 or higher compared to Medicare fee-for service beneficiaries. Figure 9: Nonenrolled TRICARE Beneficiaries' Estimated Satisfaction Ratings Compared to Those of Commercially Insured, Medicaid, and Medicare Fee-For-Service Beneficiaries, 2008-2011: [Refer to PDF for image: 4 stacked horizontal bar graphs] Beneficiary rating of primary care provider[A]: Beneficiary type: Medicare FFS; 0-7 rating: 12%; 8-10 rating: 88%; Beneficiary type: Medicaid; 0-7 rating: 23%; 8-10 rating: 77%; Beneficiary type: Commercial; 0-7 rating: 18%; 8-10 rating: 82%; Beneficiary type: TRICARE nonenrolled; 0-7 rating: 21%; 8-10 rating: 79%. Beneficiary rating of specialty care provider[B]: Beneficiary type: Medicare FFS; 0-7 rating: 14%; 8-10 rating: 86%; Beneficiary type: Medicaid; 0-7 rating: 24%; 8-10 rating: 76%; Beneficiary type: Commercial; 0-7 rating: 19%; 8-10 rating: 81%; Beneficiary type: TRICARE nonenrolled; 0-7 rating: 20%; 8-10 rating: 80%. Beneficiary rating of health care[C]: Beneficiary type: Medicare FFS; 0-7 rating: 22%; 8-10 rating: 78%; Beneficiary type: Medicaid; 0-7 rating: 32%; 8-10 rating: 68%; Beneficiary type: Commercial; 0-7 rating: 25%; 8-10 rating: 75%; Beneficiary type: TRICARE nonenrolled; 0-7 rating: 20%; 8-10 rating: 80%. Beneficiary rating of health plan[D]: Beneficiary type: Medicare FFS; 0-7 rating: 26%; 8-10 rating: 74%; Beneficiary type: Medicaid; 0-7 rating: 29%; 8-10 rating: 71%; Beneficiary type: Commercial; 0-7 rating: 38%; 8-10 rating: 62%; Beneficiary type: TRICARE nonenrolled; 0-7 rating: 34%; 8-10 rating: 66%. Source: GAO analysis of TMA and HHS data. Note: All estimates between nonenrolled TRICARE beneficiaries and other beneficiary groups are significantly different at the 95 percent confidence level. We did not adjust the CAHPS survey data for factors that could affect the various beneficiary groups' ratings, such as age or health status. [A] TRICARE beneficiaries were asked "Using any number from 0 to 10, where 0 is the worst personal doctor or nurse possible, and 10 is the best personal doctor or nurse possible, what number would you use to rate your personal doctor or nurse?" Commercial, Medicare, and Medicaid beneficiaries were asked this question of their personal doctor only. Our analysis is limited to TRICARE nonenrolled beneficiaries who indicated that their personal doctor or nurse was a civilian. [B] TRICARE and commercial beneficiaries were asked "We want to know your rating of the specialist you saw most often in the last 12 months. Using any number from 0 to 10, where 0 is the worst specialist possible, and 10 is the best specialist possible, what number would you use to rate the specialist?" Medicare and Medicaid beneficiaries were asked the same question, but only in reference to the last 6 months. Our analysis is limited to TRICARE nonenrolled beneficiaries who indicated that they had seen a civilian specialist in the last 12 months. [C] TRICARE and commercial beneficiaries were asked "Using any number from 0 to 10, where 0 is the worst health care possible, and 10 is the best health care possible, what number would you use to rate all your health care in the last 12 months?" Medicare and Medicaid beneficiaries were asked the same question, but only in reference to the last 6 months. [D] TRICARE, commercial, Medicare, and Medicaid beneficiaries were asked "Using any number from 0 to 10, where 0 is the worst health plan possible, and 10 is the best health plan possible, what number would you use to rate your health plan?" [End of figure] Civilian Providers' Acceptance of New TRICARE Patients Has Decreased over Time; Mental Health Providers Report Lower Awareness and Acceptance than Other Provider Types: Nationwide, an estimated 82 percent of civilian providers indicated they were aware of the TRICARE program, but only an estimated 58 percent were accepting new TRICARE patients, according to our analysis of the 2008 through 2011 civilian provider survey results.[Footnote 35] When compared to a national provider survey, civilian providers' acceptance of new TRICARE patients was less than providers' acceptance of other types of beneficiaries. Specifically, a survey of physicians in 2008 by the Center for Studying Health System Change found that about 96 percent of physicians accepted new commercially insured beneficiaries, about 86 percent accepted new Medicare beneficiaries, and about 72 percent accepted new Medicaid beneficiaries.[Footnote 36] According to the TRICARE survey results, when asked the reasons for not accepting new TRICARE patients, the most-cited category by those civilian providers who were not accepting any new TRICARE patients was that the provider "was not aware of the TRICARE program/not asked/don't know about TRICARE." (See figure 10 for the top 7 categories of reasons for why civilian providers were not accepting new TRICARE patients.) Additionally, while nonenrolled beneficiaries cited that providers were not accepting TRICARE for payment as the top reason why any providers were unwilling to accept them as patients, the providers cited it as the third highest reason in addition to "don't know/no answer." Figure 10: Top Seven Categories of Reasons for Not Accepting New TRICARE Patients Reported by Civilian Providers That Were Not Accepting Any New TRICARE Patients, 2008-2011: [Refer to PDF for image: vertical bar graph] Categories of reasons for not accepting any new TRICARE patients: Estimated percentage of civilian providers: Not aware of TRICARE/not asked/don't know about TRICARE: 23%[A]; 95% confidence interval for estimates: 21-25%; Miscellaneous: 20%[A]; 95% confidence interval for estimates: 19-21%; Don’t know/no answer: 14%[B]; 95% confidence interval for estimates: 13-15%; Reimbursement: 14%[B]; 95% confidence interval for estimates: 13-15%; Insurance image problems/Issues with TRICARE in past: 10%[C]; 95% confidence interval for estimates: 9-11%; Specialty not covered: 9%[D]; 95% confidence interval for estimates: 8-10%; Not accepting patients: 8%[E]; 95% confidence interval for estimates: 7-9%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. Estimated percentages are out of the total estimated number of civilian providers who were not accepting any new TRICARE patients. Percentages across problem types do not add up to 100 percent because respondents were able to select more than one response, and only the top seven responses are shown. [A] For these two categories of reasons, the differences in estimates between them and the other categories of reasons, as well as between each other, are statistically significant at the 95 percent confidence level. [B] For these two categories of reasons, the differences in estimates between these categories of reasons and the others are statistically significant at the 95 percent confidence level. However, the differences between the two categories of reasons are not statistically significant at the 95 percent confidence level. [C] For the category "insurance image problems/issues with TRICARE in past," the differences in estimates between this category of reasons and all others (except for "specialty not covered") are statistically significant at the 95 percent confidence level. [D] For the category "specialty not covered," the differences in estimates between this category of reasons and the others (except for "insurance image problems/issues with TRICARE in past" and "not accepting patients") are statistically significant at the 95 percent confidence level. [E] For the category "not accepting patients," the differences in estimates between this category of reasons and all others (except for "specialty not covered") are statistically significant at the 95 percent confidence level. [F] The "miscellaneous" category includes reasons such as "not a provider/signed provider," and "working as locum tenens," which means that the provider substitutes for the regular provider when that regular provider is absent. [End of figure] When we compared the results of TMA's 2008-2011 civilian provider survey (excluding nonphysician mental health providers) to the results of its 2005-2007 civilian physician survey,[Footnote 37] we found that although civilian physicians' awareness has increased over time, their acceptance of new TRICARE patients has decreased over time.[Footnote 38] This was also true whether they were accepting any new patients or new Medicare patients. For example, civilian physicians' acceptance of any new TRICARE patients has decreased from about 76 percent in 2005- 2007 to an estimated 70 percent in 2008-2011.[Footnote 39] (See figure 11.) Figure 11: Civilian Physicians' Awareness and Acceptance of TRICARE over Time in TRICARE Management Activity's (TMA) Surveys: [Refer to PDF for image: horizontal bar graph] Aware of the TRICARE program[A]: Civilian physicians (TMA’s 2005-2007 civilian physician survey): 87%; 95% confidence interval for estimates: 86-88%; Civilian physicians (TMA’s 2008-2011 civilian provider survey): 91%; 95% confidence interval for estimates: 90-92%. Accepting any new TRICARE patients[B]: Civilian physicians (TMA’s 2005-2007 civilian physician survey): 76%; 95% confidence interval for estimates: 75-77%; Civilian physicians (TMA’s 2008-2011 civilian provider survey): 70%; 95% confidence interval for estimates: 69-71%. Accepting any new TRICARE patients if accepting new patients[C]: Civilian physicians (TMA’s 2005-2007 civilian physician survey): 81%; 95% confidence interval for estimates: 80-82%; Civilian physicians (TMA’s 2008-2011 civilian provider survey): 74%; 95% confidence interval for estimates: 73-75%. Accepting any new TRICARE patients if accepting new Medicare patients[D]: Civilian physicians (TMA’s 2005-2007 civilian physician survey): 87%; 95% confidence interval for estimates: 86-88%; Civilian physicians (TMA’s 2008-2011 civilian provider survey): 77%; 95% confidence interval for estimates: 76-78%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. A statistically significant difference exists between civilian physicians from the 2005-2007 surveys and those from the 2008-2011 surveys for each of the questions at the 95 percent confidence level. Civilian physicians consist of civilian primary care and specialty care physicians, including psychiatrists. [A] Respondents answered yes to the following question: "Is the provider aware of the TRICARE health care program?" [B] Respondents answered "for all claims" or on a "claim-by-claim basis" to the following question: "As of today, is the provider accepting new TRICARE Standard patients?" [C] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting any new patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents providers' indications that they were accepting new TRICARE Standard patients on either a "claim- by-claim basis" or "for all claims." [D] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting new Medicare patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents providers' indications that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [End of figure] When analyzed further by provider type, we found that civilian primary and specialty care providers had higher awareness and acceptance of TRICARE than civilian mental health care providers. (See figure 12.) Specifically, only an estimated 39 percent of civilian mental health providers were accepting new TRICARE beneficiaries, compared to an estimated 67 percent of civilian primary care providers and an estimated 77 percent of civilian specialty care providers.[Footnote 40] Figure 12: Civilian Providers' Awareness and Acceptance of TRICARE, by Type of Provider, 2008-2011: [Refer to PDF for image: horizontal bar graph] Aware of the TRICARE program[A]: Primary care providers: 92%; 95% confidence interval for estimates: 91-93%; Specialty care providers (excluding psychiatrists): 92%; 95% confidence interval for estimates: 91-93%; Mental health care providers: 68%; 95% confidence interval for estimates: 67-69%. Accepting any new TRICARE patients[B]: Primary care providers: 67%; 95% confidence interval for estimates: 66-68%; Specialty care providers (excluding psychiatrists): 77%; 95% confidence interval for estimates: 76-78%; Mental health care providers: 39%; 95% confidence interval for estimates: 38-40%. Accepting any new TRICARE patients if accepting any new patients[C]: Primary care providers: 72%; 95% confidence interval for estimates: 71-73%; Specialty care providers (excluding psychiatrists): 78%; 95% confidence interval for estimates: 77-79%; Mental health care providers: 41%; 95% confidence interval for estimates: 40-42%. Accepting any new TRICARE patients if accepting any new Medicare patients[D]: Primary care providers: 75%; 95% confidence interval for estimates: 73-77%; Specialty care providers (excluding psychiatrists): 79%; 95% confidence interval for estimates: 78-80%; Mental health care providers: 57%; 95% confidence interval for estimates: 55-59%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. With the exception of primary care providers' and specialty care providers' awareness of the TRICARE program, a statistically significant difference exists between primary care providers, specialty care providers, and mental health care providers for each question at the 95 percent confidence level. [A] Respondents answered yes to the following question: "Is the provider aware of the TRICARE health care program?" [B] Respondents answered "for all claims" or on a "claim-by-claim basis" to the following question: "As of today, is the provider accepting new TRICARE Standard patients?" [C] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting any new patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [D] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting new Medicare patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [End of figure] The categories of reasons cited for not accepting new TRICARE patients also differed by provider type. For example, civilian mental health care providers more often cited "not aware of TRICARE/not asked/don't know about TRICARE" than civilian primary or specialty care providers. Additionally, the top category of reasons cited by civilian primary care providers was that they were "not accepting patients" while the top category of reasons cited by specialty providers was "reimbursement." (See figure 13 for the top categories of reasons for civilian providers not accepting new TRICARE patients, by provider type.) Figure 13: Top Categories of Reasons Reported by Civilian Providers for Not Accepting New TRICARE Patients, by Provider Type, 2008-2011: [Refer to PDF for image: vertical bar graph] Categories of reasons for not accepting any new TRICARE patients: Not aware of TRICARE/not asked/don't know about TRICARE: Primary care providers: 6%; 95% confidence interval for estimates: 4-8%; Specialty care providers (excluding psychiatrists): 9%; 95% confidence interval for estimates: 6-12%; Mental health care providers: 30%[A]; 95% confidence interval for estimates: 28-32%. Miscellaneous[D]: Primary care providers: 15%; 95% confidence interval for estimates: 13-17%; Specialty care providers (excluding psychiatrists): 19%; 95% confidence interval for estimates: 17-21%; Mental health care providers: 21%[B]; 95% confidence interval for estimates: 19-23%. Don’t know/no answer: Primary care providers: 13%; 95% confidence interval for estimates: 11-15%; Specialty care providers (excluding psychiatrists): 15%; 95% confidence interval for estimates: 11-19%; Mental health care providers: 14%; 95% confidence interval for estimates: 12-16%. Reimbursement: Primary care providers: 19%; 95% confidence interval for estimates: 17-21%; Specialty care providers (excluding psychiatrists): 22%; 95% confidence interval for estimates: 19-25%; Mental health care providers: 11%; 95% confidence interval for estimates: 10-12%. Insurance image problems/Issues with TRICARE in past: Primary care providers: 6%; 95% confidence interval for estimates: 4-8%; Specialty care providers (excluding psychiatrists): 6%; 95% confidence interval for estimates: 4-8%; Mental health care providers: 11%[A]; 95% confidence interval for estimates: 9-13%. Specialty not covered: Primary care providers: 5%; 95% confidence interval for estimates: 4-6%; Specialty care providers (excluding psychiatrists): 10%; 95% confidence interval for estimates: 7-13%; Mental health care providers: 9%; 95% confidence interval for estimates: 8-10%. Not accepting patients: Primary care providers: 25%[C]; 95% confidence interval for estimates: 23-27%; Specialty care providers (excluding psychiatrists): 5%; 95% confidence interval for estimates: 3-7%; Mental health care providers: 4%; 95% confidence interval for estimates: 3-5%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. Estimated percentages are out of the estimated number of civilian primary care, specialty care, and mental health care providers who were not accepting any new TRICARE patients. Percentages across problem types do not add up to 100 percent because respondents were able to select more than one response, and only the top seven responses are shown (ranked by the overall categories of reasons reported by all civilian providers, regardless of area). Unless otherwise noted below, differences in estimates within each problem type are not significantly different at the 95 percent confidence level. [A] For the categories "not aware of TRICARE/not asked/don't know about TRICARE," "reimbursement," and "insurance image problems/issues with TRICARE in past," the differences in estimates between mental health care providers and other provider types are statistically significant at the 95 percent confidence level. [B] For the category "miscellaneous," the difference in estimates between mental health care providers and primary care providers is statistically significant at the 95 percent confidence level. [C] For the categories "specialty not covered," and "not accepting patients," the differences in estimates between primary care and other provider types are statistically significant at the 95 percent confidence level. [D] The "miscellaneous" category includes reasons such as "not a provider/signed provider," and "working as locum tenens," which means that the provider substitutes for the regular provider when that regular provider is absent. [End of figure] We also found that providers' awareness and acceptance of TRICARE differed by type of area. Similar to TMA's nonenrolled beneficiary survey, which showed that nonenrolled beneficiaries in PSAs generally experienced more problems finding providers than their counterparts in non-PSAs, our analysis of the 2008 through 2011 civilian provider survey indicated that civilian providers in PSAs were less aware of TRICARE and less accepting of new TRICARE patients than civilian providers in non-PSAs. Specifically, an estimated 81 percent of civilian providers in PSAs were aware of the TRICARE program, compared to an estimated 87 percent of civilian providers in non-PSAs,[Footnote 41] and an estimated 56 percent of civilian providers in PSAs were accepting any new TRICARE patients, compared to an estimated 66 percent of those providers in non-PSAs.[Footnote 42] (See figure 14.) Figure 14: Civilian Providers' Awareness and Acceptance of TRICARE in Prime Service Areas (PSA) and non-Prime Service Areas (non-PSA), 2008- 2011: [Refer to PDF for image: horizontal bar graph] Estimated percentage of civilian providers: Aware of the TRICARE program[A]: PSAs: 81%; 95% confidence interval for estimates: 80-82%; Non-PSAs: 87%; 95% confidence interval for estimates: 86-88%. Accepting any new TRICARE patients[B]: PSAs: 56%; 95% confidence interval for estimates: 55-57%; Non-PSAs: 66%; 95% confidence interval for estimates: 65-67%. Accepting any new TRICARE patients if accepting new patients[C]: PSAs: 59%; 95% confidence interval for estimates: 58-60%; Non-PSAs: 70%; 95% confidence interval for estimates: 69-71%. Accepting any new TRICARE patients if accepting new Medicare patients[D]: PSAs: 71%; 95% confidence interval for estimates: 70-72%; Non-PSAs: 79%; 95% confidence interval for estimates: 78-80%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. A statistically significant difference exists between civilian providers in PSAs and those in non-PSAs for each of the questions at the 95 percent confidence level. [A] Respondents answered yes to the following question: "Is the provider aware of the TRICARE health care program?" [B] Respondents answered "for all claims" or on a "claim-by-claim basis" to the following question: "As of today, is the provider accepting new TRICARE Standard patients?" [C] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting any new patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents providers' indications that they were accepting new TRICARE Standard patients on either a "claim- by-claim basis" or "for all claims." [D] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting new Medicare patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents providers' indications that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [End of figure] Civilian providers in HSAs were more frequently aware of TRICARE and accepting of new TRICARE beneficiaries than civilian providers in the PSAs and non-PSAs outside of these HSAs.[Footnote 43] (See figure 15.) These HSAs represented locations that were identified by beneficiary and provider groups to TMA as potentially having access problems. Figure 15: Civilian Providers' Awareness and Acceptance of TRICARE, by Hospital Service Areas (HSA) and Prime Service Areas/non-Prime Service Areas (PSA/non-PSA) outside of the Surveyed HSAs, 2008-2011: [Refer to PDF for image: horizontal bar graph] Estimated percentage of civilian providers: Aware of the TRICARE program[A]: HSAs: 89%; 95% confidence interval for estimates: 88-90%; PSAs/Non-PSAs outside of surveyed HSAs: 81%; 95% confidence interval for estimates: 80-82%. Accepting any new TRICARE patients[B]: HSAs: 60%; 95% confidence interval for estimates: 59-61%; PSAs/Non-PSAs outside of surveyed HSAs: 57%; 95% confidence interval for estimates: 56-58%. Accepting any new TRICARE patients if accepting new patients[C]: HSAs: 63%; 95% confidence interval for estimates: 62-64%; PSAs/Non-PSAs outside of surveyed HSAs: 60%; 95% confidence interval for estimates: 59-61%. Accepting any new TRICARE patients if accepting new Medicare patients[D]: HSAs: 76%; 95% confidence interval for estimates: 75-77%; PSAs/Non-PSAs outside of surveyed HSAs: 72%; 95% confidence interval for estimates: 71-73%. Source: GAO analysis of TMA data. Notes: Error bars display 95 percent confidence intervals for estimates. Each HSA is part of a PSA or non-PSA (depending on the location); and because HSAs are not mutually exclusive of the PSAs or non-PSAs, we did not compare the results from nonenrolled beneficiaries in HSAs to nonenrolled beneficiaries in PSAs or non-PSAs. We compared the results for the nonenrolled beneficiaries in the surveyed HSAs to those nonenrolled beneficiaries in the areas outside of HSAs. The difference in estimates between HSAs and PSAs/non-PSAs outside of surveyed HSAs for each question is statistically significant at the 95 percent confidence level. [A] Respondents answered yes to the following question: "Is the provider aware of the TRICARE health care program?" [B] Respondents answered "for all claims" or on a "claim-by-claim basis" to the following question: "As of today, is the provider accepting new TRICARE Standard patients?" [C] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting any new patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [D] Respondents answered yes to questions that asked the following: "As of today, is the provider accepting new Medicare patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [End of figure] Collective Results of TMA's Beneficiary and Civilian Provider Surveys Indicate Specific Geographic Areas Where Nonenrolled Beneficiaries Have Experienced Access Problems: An analysis of the collective results of the multiyear beneficiary and civilian provider surveys indicated particular geographic areas where nonenrolled beneficiaries are experiencing considerable access problems. These locations are defined as areas where (1) the percentage of nonenrolled beneficiaries who experienced difficulties finding a civilian provider was at least the national estimate and (2) the percentage of civilian providers who were accepting any new TRICARE patients was at or below the national estimate.[Footnote 44] Using these criteria, we identified a number of areas where beneficiaries were having access problems, mostly in Texas.[Footnote 45] (See appendix IV for detailed information about these areas and how they were determined.) In determining areas where nonenrolled beneficiaries were experiencing access problems to any type of civilian provider, we first identified 24 individual areas (out of the 215 individual areas surveyed by the 2008-2011 beneficiary surveys)[Footnote 46] where the estimated percentage of nonenrolled beneficiaries who experienced difficulties finding any type of civilian provider met or exceeded the national estimate (31 percent). Of these, we identified 2 PSAs where the estimated percentage of civilian providers who were accepting any new TRICARE patients was at or below the national estimate (58 percent)-- Central/Southern-Central Coastal California and Northeastern Texas. Additionally, we identified 2 HSAs that also met these criteria, one of which is contained within the Northeastern Texas PSA. Table 4 shows each of these areas with the estimated percentage of (1) nonenrolled beneficiaries who experienced problems finding any type of civilian provider and (2) civilian providers who were accepting any new TRICARE patients. Table 4: Areas Where the Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Provider Was at Least the National Estimate and Where the Percentage of Civilian Providers Who Were Accepting Any New TRICARE Patients Was at or below the National Estimate, 2008-2011: Prime Service Areas (PSA): Area name: 1. Central/Southern-Central Coastal California; Estimated percentage of beneficiaries with a problem finding any type of civilian provider (margin of error)[A]: 48% (12%); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 45% (8%). Area name: 2. Northeastern Texas; Estimated percentage of beneficiaries with a problem finding any type of civilian provider (margin of error)[A]: 47% (10%); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 53% (6%). Hospital Service Areas (HSA): Area name: 3. Austin, Texas; Estimated percentage of beneficiaries with a problem finding any type of civilian provider (margin of error)[A]: 58% (18%)[C]; Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 46% (6%). Area name: 4. Dallas/Ft. Worth, Texas[D]; Estimated percentage of beneficiaries with a problem finding any type of civilian provider (margin of error)[A]: 48% (14%); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 50% (6%). Source: GAO analysis of TMA data. Notes: The margin of error is at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the civilian provider survey. Areas in this table had an estimated 31 percent or more of nonenrolled beneficiaries who were having difficulties finding any type of civilian provider who will accept TRICARE (the national estimate, or greater) and equal to or less than an estimated 58 percent of civilian providers who were accepting new TRICARE patients (the national estimate or less). Both determinations were made using the estimates' margins of error at the 95 percent confidence level. Estimated percentages and margins of error have been rounded to the nearest whole number. [A] Estimated percentage is based on the number of nonenrolled beneficiaries who responded "a big problem" or "a small problem" to any one of the following three questions: (1) "In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE?"; (2) "In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE?"; or (3) "In the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan?" We also limited nonenrolled beneficiary responses to those who indicated their provider was a civilian provider. [B] Estimated percentage is based on the number of civilian providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [C] This estimate has a relative margin of error of 30 percent or greater. [D] The Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [End of table] For the overlapping PSA and HSA (Northeastern Texas and Dallas/Fort Worth), we found that although a high percentage of civilian providers were accepting new patients (between 95 and 97 percent), only about half of these providers were accepting any new TRICARE patients. (See table 5.) For the remaining PSA (Central/Southern-Central California) and HSA (Austin, Texas), between 92 and 98 percent of civilian providers were accepting new patients, and less than half of those providers were accepting any new TRICARE patients. Further, of the civilian providers in all of these areas who were accepting new Medicare patients, between 65 and 70 percent were also accepting any new TRICARE patients. Reimbursement was the most cited reason for providers not accepting new TRICARE patients for all of the areas except the PSA in California for which "not aware of the TRICARE program" was the most cited reason. Table 5: Civilian Providers' Estimated Percentage of Acceptance of New Patients and New TRICARE Patients, by Problem Area, 2008-2011: Prime Service Areas (PSA): Area name: 1. Central/Southern-Central Coastal California; Estimated percentage of civilian providers accepting any new TRICARE patients (margin of error)[A]: 45% (8%); Estimated percentage of civilian providers accepting any new patients (margin of error): 92% (5%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 48% (8%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: 66% (10%). Area name: 2. Northeastern Texas; Estimated percentage of civilian providers accepting any new TRICARE patients (margin of error)[A]: 53% (6%); Estimated percentage of civilian providers accepting any new patients (margin of error): 97% (2%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 55% (6%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: 70% (7%). Hospital Service Areas (HSA): Area name: 3. Austin, Texas; Estimated percentage of civilian providers accepting any new TRICARE patients (margin of error)[A]: 46% (6%); Estimated percentage of civilian providers accepting any new patients (margin of error): 98% (2%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 47% (6%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: 65% (8%). Area name: 4. Dallas/Ft. Worth, Texas[D]; Estimated percentage of civilian providers accepting any new TRICARE patients (margin of error)[A]: 50% (6%); Estimated percentage of civilian providers accepting any new patients (margin of error): 95% (3%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 53% (6%); Estimated percentage of civilian providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: 70% (7%). Source: GAO analysis of TMA data. Notes: The margin of error is at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the civilian provider survey. Areas in this table had an estimated 31 percent or more of nonenrolled beneficiaries who were having difficulties finding any type of civilian provider who will accept TRICARE (the national estimate, or greater) and equal to or less than an estimated 58 percent of civilian providers who were accepting new TRICARE patients (the national estimate or less). Both determinations were made using the estimates' margins of error at the 95 percent confidence level. Estimated percentages and margins of error have been rounded to the nearest whole number. Each surveyed HSA was part of a PSA or non-PSA (depending on the location). [A] Estimated percentage is based on the number of civilian providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [B] Estimated percentage is based on the number of civilian providers who answered yes to questions that asked the following: "As of today, is the provider accepting any new patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [C] Estimated percentage is based on the number of civilian providers who answered yes to questions that asked the following: "As of today, is the provider accepting new Medicare patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim- by-claim basis" or "for all claims." [D] The Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [End of table] When analyzing this data by type of provider (primary care, specialty, and mental health), we found four areas where the percentage of civilian primary care providers who were accepting any new TRICARE patients was at or below the national estimate, but did not find similarly low-percentage areas for civilian specialty care providers. Because of the low numbers of survey responses, we are unable to report survey results for access problems to civilian mental health care providers. Civilian Primary Care Providers: In determining areas where nonenrolled beneficiaries experienced access problems to civilian primary care providers, we first identified 21 individual areas where the estimated percentage of nonenrolled beneficiaries who experienced difficulties finding a civilian primary care provider met or exceeded the national estimate (25 percent). Of these, we identified 2 PSAs where the estimated percentage of civilian primary care providers who were accepting any new TRICARE patients was at or below the national estimate (67 percent)--Northeastern Texas and Eastern-Central Texas. We also identified 2 HSAs that met these criteria, each of which was contained in one of the PSAs we identified. Table 6 shows each of these areas with the estimated percentage of (1) nonenrolled beneficiaries who experienced problems finding a civilian primary care provider and (2) civilian primary care providers who were accepting any new TRICARE patients. Table 6: Areas Where the Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Primary Care Provider Was at Least the National Estimate and Where the Percentage of Civilian Primary Care Providers Who Were Accepting Any New TRICARE Patients Was at or below the National Estimate, 2008-2011: Prime Service Areas (PSA): Area name: 1. Northeastern Texas; Estimated percent of beneficiaries with a problem finding a civilian primary care provider (margin of error)[A]: 40% (10%); Estimated percent of civilian primary care providers accepting new TRICARE patients (margin of error)[B]: 48% (10%). Area name: 2. Eastern-Central Texas; Estimated percent of beneficiaries with a problem finding a civilian primary care provider (margin of error)[A]: 38% (12%)[C]; Estimated percent of civilian primary care providers accepting new TRICARE patients (margin of error)[B]: 53% (10%). Hospital Service Areas (HSA): Area name: 3. Austin, Texas[D]; Estimated percent of beneficiaries with a problem finding a civilian primary care provider (margin of error)[A]: 56% (18%)[C]; Estimated percent of civilian primary care providers accepting new TRICARE patients (margin of error)[B]: 42% (11%). Area name: 4. Dallas/Ft. Worth, Texas[E]; Estimated percent of beneficiaries with a problem finding a civilian primary care provider (margin of error)[A]: 40% (14%)[C]; Estimated percent of civilian primary care providers accepting new TRICARE patients (margin of error)[B]: 51% (12%). Source: GAO analysis of TMA data. Notes: The margin of error is at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the civilian provider survey. Areas in this table had an estimated 25 percent or more of nonenrolled beneficiaries who were having difficulties finding a civilian primary care provider who will accept TRICARE (the national estimate, or greater) and equal to or less than an estimated 67 percent of civilian primary care providers who were accepting new TRICARE patients (the national estimate or less). Both determinations were made using the estimates' margins of error at the 95 percent confidence level. Estimated percentages and margins of error have been rounded to the nearest whole number. [A] Estimated percentage is based on the number of beneficiaries who responded that they used TRICARE Standard, TRICARE Extra, or TRICARE Reserve Select the most in the last 12 months, and of those, the number who responded "a big problem" or "a small problem" to the question that asked "In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE?" We also limited nonenrolled beneficiary responses to those who indicated their provider was a civilian provider. [B] Estimated percentage is based on the number of civilian primary care providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [C] These estimates have relative margins of errors that are 30 percent or greater. [D] The Austin, Texas, HSA is part of the Eastern-Central Texas PSA. [E] The Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [End of table] As we similarly found in the areas where nonenrolled beneficiaries were having access problems for any type of civilian provider, we found that between 94 and 97 percent of civilian primary care providers in the Northeastern Texas PSA/Dallas/Ft. Worth HSA and the Eastern-Central Texas PSA/Austin, Texas, HSA were accepting new patients, but only around half of them were accepting new TRICARE patients.[Footnote 47] (See table 7.) Further, of the civilian primary care providers in the two PSAs who were accepting new Medicare patients, between 59 and 68 percent were accepting any new TRICARE patients.[Footnote 48] Reimbursement was the most cited reason by civilian primary care providers for not accepting any new TRICARE patients in each of these areas except for the Dallas/Ft. Worth, Texas, HSA, for which "don't know/no answer" was the most cited reason. Table 7: Civilian Primary Care Providers' Estimated Percentage of Acceptance of New Patients and New TRICARE Patients, by Problem Area, 2008-2011: Prime Service Areas (PSA): Area name: 1. Northeastern Texas; Estimated percentage of civilian primary care providers accepting any new TRICARE patients (margin of error)[A]: 48% (10%); Estimated percentage of civilian primary care providers accepting any new patients (margin of error): 95% (5%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 51% (11%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: 59% (13%). Area name: 2. Eastern-Central Texas; Estimated percentage of civilian primary care providers accepting any new TRICARE patients (margin of error)[A]: 53% (10%); Estimated percentage of civilian primary care providers accepting any new patients (margin of error): 96% (4%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 55% (10%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: 68% (15%). Hospital Service Areas (HSA): Area name: 3. Austin, Texas[D]; Estimated percentage of civilian primary care providers accepting any new TRICARE patients (margin of error)[A]: 42% (11%); Estimated percentage of civilian primary care providers accepting any new patients (margin of error): 97% (4%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 43% (11%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: [E]. Area name: 4. Dallas/Ft. Worth, Texas[F]; Estimated percentage of civilian primary care providers accepting any new patients (margin of error): 51% (12%); Estimated percentage of civilian primary care providers accepting any new patients (margin of error): 94% (6%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting any new patients (margin of error)[B]: 54% (12%); Estimated percentage of civilian primary care providers accepting any new TRICARE patients, if accepting new Medicare patients (margin of error)[C]: [E]. Source: GAO analysis of TMA data. Notes: The margin of error is at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the civilian provider survey. Areas in this table had an estimated 25 percent or more of nonenrolled beneficiaries who were having difficulties finding any type of civilian provider who will accept TRICARE (the national estimate, or greater) and equal to or less than an estimated 67 percent of civilian providers who were accepting new TRICARE patients (the national estimate or less). Both determinations were made using the estimates' margins of error at the 95 percent confidence level. Estimated percentages and margins of error have been rounded to the nearest whole number. [A] Estimated percentage is based on the number of civilian primary care providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [B] Estimated percentage is based on the number of civilian primary care providers who answered yes to questions that asked the following: "As of today, is the provider accepting any new patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [C] Estimated percentage is based on the number of civilian primary care providers who answered yes to questions that asked the following: "As of today, is the provider accepting new Medicare patients?" and "As of today, is the provider accepting new TRICARE Standard patients?" The yes response to this question represents the providers' indication that they were accepting new TRICARE Standard patients on either a "claim-by-claim basis" or "for all claims." [D] The Austin, Texas, HSA is part of the Eastern-Central Texas PSA. [E] Because the number of responses was below 50, we do not present the estimates and margins of error for these locations. [F] The Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [End of table] Civilian Specialty Care Providers: In determining areas where nonenrolled beneficiaries are experiencing access problems to civilian specialty care providers, we first identified nine individual areas where the estimated percentage of nonenrolled beneficiaries who experienced difficulties finding a civilian specialty care provider met or exceeded the national estimate (25 percent). Unlike the collective results for "any civilian provider" and "civilian primary care providers," when we examined civilian specialty care providers' responses for these areas, we did not identify any geographic areas where the estimated percentage of civilian specialty care providers who were accepting any new TRICARE patients was at or below the national estimate (77 percent) when accounting for the margins of error at the 95 percent confidence limit. For the nine areas where the estimated percentage of beneficiaries who experienced difficulties finding a civilian specialty care provider met or exceeded the national estimate, the percentage of civilian specialty care providers who were accepting new TRICARE patients ranged from 75 to 86 percent.[Footnote 49] Civilian Mental Health Care Providers: Because of the low numbers of survey responses for beneficiaries who said they needed civilian mental health care, we are unable to report correlated survey results for access problems to civilian mental health care providers.[Footnote 50] However, given the nationwide shortage of certain types of mental health providers and the survey results that only 39 percent of civilian mental health care providers were accepting new TRICARE patients, access to mental health care providers is a concern for all TRICARE beneficiaries, including those who use the TRICARE Standard and Extra options. Agency Comments and Our Evaluation: In reviewing a draft of this report, DOD concurred with our overall findings and provided technical comments, which we incorporated where appropriate. (See appendix VI.) We are sending copies of this report to the Secretary of Defense and appropriate congressional committees. The report is also available at no charge on GAO's website at [hyperlink, http://www.gao.gov]. If you or your staff members have any questions about this report, please contact me at (202) 512-7114 or draperd@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 members who made key contributions to this report are listed in appendix VII. Signed by: Debra A. Draper: Director, Health Care: [End of section] Appendix I: TRICARE Management Activity's Methodology for the 2008- 2011 Beneficiary and Civilian Provider Surveys: The National Defense Authorization Act for Fiscal Year 2008 (NDAA 2008) directed the Department of Defense (DOD) to determine the adequacy of the number of health care and mental health care providers that currently accept nonenrolled beneficiaries as patients under TRICARE, DOD's health care program. We use the term "nonenrolled beneficiaries" for beneficiaries who are not enrolled in TRICARE Prime and who use the TRICARE Standard or Extra options, or TRICARE Reserve Select (TRS).[Footnote 51] The NDAA 2008 also included specific requirements related to the number and priority of areas to be surveyed, including the populations to be surveyed each year, content for each type of survey, and the use of benchmarks. Within DOD, the TRICARE Management Activity (TMA), which oversees the TRICARE program, has the lead responsibility for designing and implementing the nonenrolled beneficiary and civilian provider surveys. The following information describes TMA's methodology, including its actions to address the requirements for each of the following: (1) survey area, (2) sample selection, (3) survey content, and (4) the establishment of benchmarks. Survey Area Selection: The NDAA 2008 specified that DOD survey beneficiaries and providers in at least 20 TRICARE Prime Service Areas (PSA),[Footnote 52] and 20 geographic areas in which TRICARE Prime is not offered--referred to as non-Prime Service Areas (non-PSA)--each fiscal year, 2008 through 2011. The NDAA 2008 also required DOD to consult with representatives of TRICARE beneficiaries and health care and mental health care providers to identify locations where nonenrolled beneficiaries have experienced significant access-to-care problems, and give a high priority to surveying health care and mental health care providers in these areas. Additionally, the NDAA 2008 required DOD to give a high priority to surveying areas in which a high concentration of Selected Reserve servicemembers live. In designing the 2008 through 2011 nonenrolled beneficiary and civilian provider surveys, TMA defined 80 PSAs and 80 non-PSAs that allowed it to survey the entire country over a 4-year period, and subsequently develop estimates of access to health care and mental health care at service area and national levels. TMA identified the 80 PSAs by collecting zip codes where TRICARE Prime was offered from officials within each of the three TRICARE Regional Offices. TMA grouped these zip codes into 80 nonoverlapping areas so that each area had roughly the same number of TRICARE-eligible beneficiaries. Because non-PSAs had not previously been defined, TMA sought to define them by grouping all zip codes not in PSAs into one large area using Hospital Referral Regions,[Footnote 53] which are groupings of Hospital Service Areas (HSA).[Footnote 54] TMA divided the large area into 80 non-PSAs so that each area had roughly the same number of TRICARE-eligible beneficiaries. To identify locations where nonenrolled beneficiaries and health care and mental health care providers have identified significant levels of access-to-care problems under TRICARE Standard and Extra, TMA spoke with groups representing beneficiaries and health care and mental health care providers, as well as officials at the TRICARE Regional Offices. These groups suggested cities and towns where access should be measured (in addition to the larger PSAs and non-PSAs), and HSAs corresponding to each city and town were then identified. On the basis of the groups' recommendations, multiple lists were created and sorted in priority order: 21 HSAs were surveyed in the 2008 surveys;[Footnote 55] 9 HSAs in the 2009 surveys; 25 HSAs in the 2010 surveys; and 16 HSAs in the 2011 civilian provider survey. This resulted in a total of 55 HSAs surveyed for the nonenrolled beneficiary survey, and 71 HSAs surveyed in the civilian provider survey (the 71 HSAs includes the same 55 HSAs surveyed for the nonenrolled beneficiary survey and an additional 16 that were selected for the 2011 fielding).[Footnote 56] Although the NDAA 2008 required DOD to give a high priority to surveying areas in which a high concentration of Selected Reserve servicemembers live, TMA officials decided to randomly select areas for the surveys in order to produce results that could be generalized to the populations in the areas surveyed and to survey the entire United States over the 4-year period--an approach we deemed acceptable in our previous report.[Footnote 57] Survey Sample Selection: Nonenrolled Beneficiary Survey Sample Selection: TMA selected its sample of beneficiaries who met its criteria for inclusion in the beneficiary survey using DOD's Defense Enrollment Eligibility Reporting System (DEERS),[Footnote 58] a database of DOD beneficiaries who may be eligible for military health benefits. TMA determined a beneficiary's eligibility to be included in the nonenrolled beneficiary survey if DEERS indicated that the individual met five criteria: 1. eligible for military health care benefits as of the date of the sample file extract; 2. age 18 years old or older; 3. not an active duty member of the military; 4. residing in one of the 20 randomly selected PSAs or 20 randomly selected non-PSAs to be surveyed that year; and: 5. not enrolled in TRICARE Prime, or is enrolled in TRS.[Footnote 59] From this database, TMA randomly sampled 1,000 beneficiaries from each PSA and non-PSA--a sample size that would achieve TMA's desired sample error rate.[Footnote 60] For the 2008, 2009, and 2010 survey fieldings, TMA used a sample size between approximately 40,000 and 50,000 beneficiaries. Because of budgetary constraints, the sample size of the 2011 nonenrolled beneficiary survey was decreased to around 34,000.[Footnote 61] Because of this reduction, the 2011 sample was further stratified by using claims data to identify beneficiaries who would likely self-report as TRICARE Standard and Extra users. [Footnote 62] After receiving the returned surveys, TMA identified the responses that it considered complete and eligible on the basis of whether the beneficiary had answered at least half of TMA's identified "key" questions. Table 8 shows the number of nonenrolled beneficiary surveys mailed, by fiscal year. Table 8: Number of Beneficiary Surveys Mailed, Returned, and Complete and Eligible, by Fiscal Year: Fiscal year: 2008; Final count mailing attributed to this year: 51,568; Complete and eligible surveys returned[A]: 20,431; Complete and eligible responses from nonenrolled beneficiaries who used TRICARE Standard, Extra or TRICARE Reserve Select[B]: 6,936. Fiscal year: 2009; Final count mailing attributed to this year: 40,996; Complete and eligible surveys returned[A]: 16,767; Complete and eligible responses from nonenrolled beneficiaries who used TRICARE Standard, Extra or TRICARE Reserve Select[B]: 5,690. Fiscal year: 2010; Final count mailing attributed to this year: 46,063; Complete and eligible surveys returned[A]: 16,793; Complete and eligible responses from nonenrolled beneficiaries who used TRICARE Standard, Extra or TRICARE Reserve Select[B]: 6,027. Fiscal year: 2011; Final count mailing attributed to this year: 38,214; Complete and eligible surveys returned[A]: 12,599; Complete and eligible responses from nonenrolled beneficiaries who used TRICARE Standard, Extra or TRICARE Reserve Select[B]: 5,397. Fiscal year: Total; Final count mailing attributed to this year: 176,841; Complete and eligible surveys returned[A]: 66,590; Complete and eligible responses from nonenrolled beneficiaries who used TRICARE Standard, Extra or TRICARE Reserve Select[B]: 24,050. Source: TMA. [A] TRICARE Management Activity (TMA) identified the responses that it considered complete and eligible based on whether the beneficiary had answered at least half of TMA's identified "key" questions. [B] Complete and eligible responses from a nonenrolled beneficiary that used TRICARE Standard, Extra, or TRICARE Reserve Select are those that were complete and eligible, and the respondent answered that he or she used TRICARE Standard or Extra or TRICARE Reserve Select in response to the following question: "Which health plan did you use for all or most of your health care in the last 12 months?" [End of table] Civilian Provider Survey Sample Selection: For each survey fielding, TMA selected the civilian provider sample within the same 20 PSAs and 20 non-PSAs that had been randomly selected for that year's nonenrolled beneficiary survey, as well as civilian providers in the HSAs identified by beneficiary and provider groups as having significant levels of access-to-care problems under TRICARE Standard and Extra. TMA used the American Medical Association Physician Masterfile to select a sample of physicians who were licensed, office-based civilian medical doctors or licensed civilian doctors of osteopathy within the specified locations who were engaged in more than 20 hours of patient care each week. The American Medical Association Physician Masterfile is a database of physicians in the U.S.--Doctors of Medicine and Doctors of Osteopathic Medicine--that includes data on all physicians who have the necessary educational and credentialing requirements. This "Masterfile" did not differentiate between TRICARE's network and non-network civilian providers, which TMA deemed acceptable to avoid any potential bias in TMA's sample selection. As such, TMA selected this file because it is widely recognized as one of the best commercially available lists of providers in the United States and contained more than 940,000 physicians along with their addresses, phone numbers, and information on practice characteristics, such as their specialty.[Footnote 63] According to TMA, the American Medical Association updates physicians' addresses monthly and other elements through a rotating census methodology involving approximately one-third of the physician population each year. Although the Masterfile is considered to contain most providers, deficiencies in coverage and inaccuracies in detail remain. Therefore, TMA attempted to update providers' addresses and phone numbers and ensure that providers were eligible for the survey by also using state licensing databases, local commercial lists, and professional society and association lists. For its 2008 and 2009 mental health care provider sample selection, TMA selected a sample of mental health care providers from two sources: the American Medical Association's Masterfile of psychiatrists, and LISTS, Inc.--a list of names with contact information assembled from state licensing boards. For the 2010 and 2011 mental health care provider sample selections, TMA also used mental health specialty areas from the National Plan and Provider Enumeration System database maintained by the Centers for Medicare & Medicaid Services, in addition to data from LISTS, Inc., and the psychiatrist data from the American Medical Association's Masterfile. According to TMA, it selected these sources for mental health care providers because they have been identified as the most comprehensive databases for these health care providers. From these data sets, TMA planned to randomly sample about 800 providers (400 each of physicians and mental health care providers) from each PSA, non-PSA, and HSA--a sample size that would achieve TMA's desired sample error rate.[Footnote 64] In those instances where there were not 800 providers in a single area, TMA selected all of the providers in that area to receive surveys. As the PSA and non-PSA regions were formed on the basis of the number of beneficiaries and not the number of civilian providers, some regions with a large number of civilian providers were sampled at relatively low rates in 2008, 2009, and 2010. To improve the precision of national estimates, in 2011 TMA selected six areas to oversample: (1) Southeastern N.Y. and Northern N.J. (New York City); (2) Los Angeles, Calif.; (3) Eastern Mass. (Boston); (4) Northeastern/Central Ohio (Cleveland); (5) Southeastern/Northern Mich. (Detroit); and (6) Northwestern/Northeastern/Central-Eastern Ill. and Southwestern Wisc. (Chicago). Therefore, in 2011, a supplemental sample of 4,800 providers was drawn for these 6 PSAs, thereby increasing the numbers of eligible providers in each area: * 1,600 providers from the two 2008 PSAs (Los Angeles, California, and Southeastern New York/Northern New Jersey); * 800 providers from the one 2009 PSA (Eastern Massachusetts); and: * 2,400 providers from the three 2010 PSAs (Northeastern/Central Ohio, Southeastern/Northern Michigan, and Northwestern/Northeastern/Central- Eastern Illinois/Southeastern Wisconsin). Upon receipt of the returned surveys, TMA identified the responses that it considered complete and eligible based on the following criteria for respondents: (1) if the provider answered "yes" to the questions that asked whether the provider offers care in an office- based location or private practice; (2) for the nonphysician mental health survey, if the provider responded he or she was one of the six TRICARE participating specialties: certified clinical social worker, certified psychiatric nurse specialist, clinical psychologist, certified marriage and family therapist, pastoral counselor, or mental health counselor; and (3) the provider had to have completed three key questions on the physician survey instrument, or three key questions on the nonphysician mental health provider survey instrument. Table 9 shows the number of civilian provider surveys mailed, by fiscal year. Table 9: Number of Civilian Physician and Nonphysician Mental Health Provider Surveys Mailed, Returned, and Complete and Eligible, by Fiscal Year: 2008 total; Final count mailing attributed to this year: 40,589; Completed surveys returned[A]: 18,557; Complete and eligible responses[B]: 11,358. Physician; Final count mailing attributed to this year: 20,193; Completed surveys returned[A]: 9,123; Complete and eligible responses[B]: 7,628. Nonphysician mental health; Final count mailing attributed to this year: 20,396; Completed surveys returned[A]: 9,434; Complete and eligible responses[B]: 3,730. 2009 total; Final count mailing attributed to this year: 52,234; Completed surveys returned[A]: 20,726; Complete and eligible responses[B]: 14,017. Physician; Final count mailing attributed to this year: 23,031; Completed surveys returned[A]: 9,243; Complete and eligible responses[B]: 8,036. Nonphysician mental health; Final count mailing attributed to this year: 29,203; Completed surveys returned[A]: 11,483; Complete and eligible responses[B]: 5,981. 2010 total; Final count mailing attributed to this year: 51,358; Completed surveys returned[A]: 22,564; Complete and eligible responses[B]: 14,822. Physician; Final count mailing attributed to this year: 25.095; Completed surveys returned[A]: 11,278; Complete and eligible responses[B]: 9,183. Nonphysician mental health; Final count mailing attributed to this year: 26,263; Completed surveys returned[A]: 11,286; Complete and eligible responses[B]: 5,639. 2011 total (supplement total)[B]; Final count mailing attributed to this year: 50,593 (4,800); Completed surveys returned[A]: 20,264 (1,649); Complete and eligible responses[B]: 13,156 (1,052). Physician[C]; Final count mailing attributed to this year: 24,498 (2,400); Completed surveys returned[A]: 10,279 (829); Complete and eligible responses[B]: 8,266 (657). Nonphysician mental health[C]; Final count mailing attributed to this year: 26,095 (2,400); Completed surveys returned[A]: 9,985 (820); Complete and eligible responses[B]: 4,890 (395). Overall total; Final count mailing attributed to this year: 194,774; Completed surveys returned[A]: 82,111; Complete and eligible responses[B]: 55,019. Source: GAO analysis of TMA data. [A] TRICARE Management Activity (TMA) considered a survey complete if the provider completed three key questions on the physician survey instrument, or three key questions on the non-physician mental health provider survey instrument that asked about the providers' location of practice and awareness and acceptance of TRICARE. [B] TMA considered a survey complete and eligible if: (1) the provider completed three key questions on the physician survey instrument, or three key questions on the non-physician mental health provider survey instrument; (2) the provider answered "yes" to the questions that asked whether the provider offers care in an office-based location or private practice; and (3) for the non-physician mental health survey, if the provider responded they were one of the six TRICARE participating specialties: certified clinical social worker, certified psychiatric nurse specialist, clinical psychologist, certified marriage and family therapist, pastoral counselor, or mental health counselor. [C] As the Prime Service Area and non-Prime Service Area regions were formed based on the number of beneficiaries and not the number of civilian providers, some regions with a large number of civilian providers were sampled at relatively low rates in 2008, 2009, and 2010. To improve the precision of national estimates, TMA selected six regions to oversample in 2011. These numbers are not included in the 2008, 2009, and 2010 counts. [End of table] Beneficiary and Provider Survey Content: Nonenrolled Beneficiary Survey Content: The NDAA 2008 required that the beneficiary survey include questions to determine whether TRICARE Standard and Extra beneficiaries have had difficulties finding physicians and mental health care providers willing to provide services under TRICARE Standard or TRICARE Extra. TMA's 2008 nonenrolled beneficiary survey included 91 questions that addressed, among other things, health care plans used; perceived access to care from a personal doctor, nurse, or specialist; the need for treatment or counseling; and ratings of health plans. TMA based some of its 2008 nonenrolled beneficiary survey questions on those included in the Department of Health and Human Services' Consumer Assessment of Healthcare Providers and Systems (CAHPS), a national survey of beneficiaries of commercial health insurance, Medicare, Medicaid, and the Children's Health Insurance Program. Over the 4 years of the nonenrolled beneficiary survey fielding, TMA added three additional questions to the original 91 questions in the 2008 nonenrolled beneficiary survey that covered topics about the beneficiaries' flu-shot history, and what they liked and disliked about TRICARE Standard and Extra. Additionally, in 2011, "TRICARE Young Adult" and "TRICARE Retired Reserve" were added to the response selections for the question that asked about the health plan the beneficiary used. (See appendix II for a copy of the 2011 beneficiary survey instrument.) When TMA began mailing the beneficiary survey, it included a combined cover letter and a questionnaire to all beneficiaries in its sample-- with the option of having beneficiaries complete the survey by mail or Internet. The cover letter provided information on the options available for completing the survey, as well as instructions for completing the survey by Internet. If the beneficiary did not respond to the mailed questionnaire, TMA mailed a second combined cover letter and questionnaire 4 weeks later encouraging the beneficiary to complete the survey. Civilian Provider Survey Content: For the civilian provider survey, the NDAA 2008 required questions to determine: (1) whether the provider is aware of TRICARE; (2) the percentage of the provider's current patient population that uses any form of TRICARE; (3) whether the provider accepts Medicare patients for health care and mental health care; and (4) if the provider accepts Medicare patients, whether the provider would accept new Medicare patients. TMA obtained clearance for its provider survey from the Office of Management and Budget (OMB) as required under the Paperwork Reduction Act.[Footnote 65] Subsequent to this review, OMB approved an 11-item questionnaire for physicians (including psychiatrists) and a 12-item questionnaire for nonphysician mental health providers. The mental health care providers' version of the survey includes an additional question about what type of mental health care the provider practiced. Beginning with the 2009 civilian provider survey, an additional follow-up question was added that asked the provider what type of practice they practiced in if the provider indicated that they were not in private practice. Although a civilian provider's indication that the provider was not in private practice still made the provider's responses ineligible for the survey, the additional information from these nonprivate practice civilian providers could be used by TMA to glean additional information about civilian providers. (See appendix III for a copy of the 2011 civilian provider survey instruments.) When TMA began mailing the provider survey, it included a combined cover letter and a questionnaire to each provider in the sample. The providers had the option of completing the survey by mail, fax, or Internet. The cover letter provided information on the options available for completing the survey, as well as instructions for completing the survey by Internet. If the provider did not respond to the mailed questionnaire, TMA mailed a second combined cover letter and questionnaire about 4 weeks later encouraging the provider to complete the survey. Survey Benchmarks: In accordance with the NDAA 2008, TMA identified benchmarks for analyzing the results of the beneficiary and civilian provider surveys. Because TMA based some of its 2008 beneficiary survey questions on those included in the CAHPS surveys, it was able to compare the results of those questions with its 2008 through 2011 beneficiary survey results. To benchmark its provider survey, TMA compared the results of its 2008 through 2011 surveys with the results of its 2005, 2006, and 2007 provider surveys. A TMA official noted that TMA was unaware of any external benchmarks that would be applicable to its surveys of providers. Analyses of Survey Results: Analysis of Nonenrolled Beneficiary Survey Results: In analyzing the results of the nonenrolled beneficiary survey, TMA representatives conducted yearly nonresponse analyses because the overall response rate for the surveys was around 38 percent.[Footnote 66] To conduct this analysis for the 2008, 2009, and 2010 survey years, TMA did the following: (1) compared key beneficiary demographic characteristics of respondents to those of nonrespondents (e.g., beneficiary gender and age) and (2) interviewed a sample of beneficiaries who did not respond to the original survey or the follow- up second mailing and compared their responses with the original survey respondents. Because of budgetary constraints during the 2011 survey year, TMA only compared key beneficiary demographic characteristics of respondents to those of the nonrespondents. The results of TMA's nonresponse analyses indicated that respondents to the nonenrolled beneficiary survey differed substantially from the surveyed population in some demographic characteristics. For example, the analyses indicated that retirees, dependents of retirees, and dependents of survivors were overrepresented in the study, and dependents of active duty servicemembers, dependents of Guard/Reserve personnel, and dependents of inactive guard personnel were underrepresented in the study. Additionally, in each of the years in which TMA representatives conducted follow-up interviews (2008-2010), they found some response differences between survey respondents. For example, each year in follow-up interviews of nonrespondents, they found these beneficiaries rated their primary care provider and health plans more favorably than beneficiaries who responded to the survey. According to TMA representatives, they used a weighting scheme to reflect the survey population proportions to correct any bias as a result of survey nonresponse. Analysis of Civilian Provider Survey Results: In analyzing the results of the provider survey, TMA conducted a nonresponse analysis because the overall response rate to the surveys was about 42 percent. To conduct this analysis for the 2008, 2009, and 2010 surveys, TMA did the following: (1) compared key provider demographic characteristics of respondents to those of nonrespondents (for example, provider type and area) and (2) interviewed a sample of physicians and mental health care providers who did not respond to the survey, follow-up second mailing, or follow-up telephone calls and compared their responses with the survey respondents. Because of budgetary constraints during the 2011 survey year, TMA only compared key provider demographic characteristics of respondents to those of the nonrespondents. The results of TMA's nonresponse analyses indicated that there are some demographic differences between respondents and those who did not respond. For example, the analyses indicated that in some years psychiatrists were underrepresented in the survey samples. Overall, however, the results were consistent among the nonresponse analyses and indicated little variation between respondents and nonrespondents. As TMA used in the weighting scheme for the nonenrolled beneficiary survey, TMA used a weighting scheme to reflect the survey population proportions to correct any bias as a result of survey nonresponse. [End of section] Appendix II: Beneficiary Survey Instrument: The National Defense Authorization Act for Fiscal Year 2008 (NDAA 2008) directed the Department of Defense (DOD) to determine the number of health care and mental health care providers that currently accept nonenrolled beneficiaries as patients under TRICARE, DOD's health care program. For the purpose of this report, we use the term "nonenrolled beneficiaries" for beneficiaries who are not enrolled in TRICARE Prime and who use the TRICARE Standard or Extra options, or TRICARE Reserve Select (TRS).[Footnote 67] Specifically, the NDAA 2008 specified that DOD conduct surveys of beneficiaries each fiscal year, 2008 through 2011. The NDAA 2008 also required that the beneficiary survey include questions seeking information from nonenrolled beneficiaries to determine whether they have had difficulties finding health care and mental health care providers willing to accept them as patients. For the 2008 fielding of the beneficiary survey, 91 questions were included in the survey instrument. Over the next 3 years of the beneficiary survey's fielding, TRICARE Management Activity (TMA) used the same 91 questions and added these additional questions: * For the 2009 survey fielding and beyond, TMA added Question #81, which asked "When did you last have a flu shot?" for a total of 92 questions in 2009; * For the 2010 survey fielding and beyond, TMA added two questions (Questions #75 and #76) that asked what the beneficiary liked and disliked about TRICARE Standard and Extra, respectively, for a total of 94 questions in 2010 and 2011. In addition, for the 2011 survey instrument, "TRICARE Young Adult" and "TRICARE Retired Reserve" were added to the response selections for Question #2, which asked "By which health plan are you currently covered?" Following is the actual survey instrument from the 2011 fielding that TMA used to obtain information from nonenrolled beneficiaries. July 21, 2011: Dear: We need your help! The Department of Defense needs your help in completing the enclosed June 2011 Health Care Survey of DoD Beneficiaries. Our mission is to provide beneficiaries with the highest quality health care that we can. To accomplish this, we need to know what we are doing right and what needs improvement. We depend on you to keep us informed. By sharing your thoughts and feelings about your health care experiences, you can help us make health care better for all beneficiaries and their families. If you have already completed the survey online, we thank you and please disregard this letter. This survey asks about your experiences and satisfaction with the health care services you have received in the past 12 months. You are one of a few military beneficiaries who have been selected for this study. You have been chosen as part of a scientific sample of health plan members. To get accurate results, we need to get answers from you and other people we ask to take part in this survey. We hope you will take the time to answer these questions. Most people find it takes only 15 minutes to answer these questions. Of course, what you have to say is private. Your answers will be part of a pool of information from others like you. What you write will be used only by this study. You may choose to fill out this survey or not If you choose not to, this will not affect the benefits you get. Your responses are important to us even if you do not receive your health care through the military. For your convenience, you can also complete the survey online by using the link and password below, If your installation's server blocks the survey site, you can complete the survey online using a civilian internet source: www.synovate.net/healthsurvey11: ID: 5100010: Password: 9999999: If you have questions about the survey, need the survey sent to your new address or do not wish to participate, please contact the Survey Processing Center. You can reach them by email at surveydodq2@synovate.net; by calling 1477-236-2390; or sending a fax to 1-800-409-7681. Please reference your ID number, 12345678, in all communication. For information about the legitimacy of the survey, please go to the TRICARE Web site at www.tricare.milihpae/home and click on the List of Approved Surveys. The DoD Report Control Symbol for this survey is RCS# DD-HA(A) 1942. Thank you for your time and assistance in this very important effort. Sincerely, Thomas V. Williams, Ph.D. Director, Health Program Analysis and Evaluation Directorate: Office of the Assistant Secretary of Defense (Health Affairs)/TRICARE Management Activity: [End of letter] Health Care Survey of DoD Beneficiaries: June 2011: According to the Privacy Act of 1974 (Public Law 93-579). the Department of Defense is required to inform you of the purposes and use of this survey. Please read it carefully. Authority: 10 U.S.C., Chapter 55; Section 706, Public Law 102484; E.O. 9397. Purpose: This survey helps health policy makers gauge beneficiary satisfaction with the current military health care system and provides valuable input from beneficiaries that will be used to improve the Military Health System. Routine Uses: None. Disclosure: Voluntary. Failure to respond will not result in any penalty to the respondent. However, maximum participation is encouraged so that data will be as complete and representative as possible. Your Privacy: Your participation in this survey effort is very important Your responses are confidential and your participation is voluntary. The number on the back of this survey is only used to let us know if you returned your survey so we don't have to send you reminders. This is your opportunity to tell officials of your opinions and experiences with the current military health care system. II is also an opportunity to provide feedback and identify areas where improvements are needed. The survey processing renter removes all identifying information before sending the results to the Department of Defense. Your information is grouped with others and no individual information is shared. Only group statistics will be compiled and reported. No information about you as an individual wilt be disclosed. Survey Instructions: Answer all the questions by checking the box to the left of your answer, You are sometimes told to skip over some questions in this survey. When this happens you will see an arrow with a note that tells you what question to answer next, like this: Yes: Go To Question 9; No: Please return the completed questionnaire in the enclosed postage-paid envelope within seven days. If the envelope is missing, please send to: Office of the Assistant Secretary of Defense (Health Affairs): TMA/HPAE: c/o Synovate Survey Processing Center: PO Box 5030: Chicago, IL 60680-4138: Survey Starts Here: As an eligible TRICARE beneficiary, please complete this survey even if you did not receive your health care from a military Wily Reese recognize that some specific questions about TRICARE benefits may not apply to you, depending on your entitlement and particular TRICARE program. This survey is about the health cam of the person whose name appears on the cover letter. The questionnaire should be completed by that person. If you are not the addressee, please give this survey to that person 1. Are you the person whose name appears on the cower letter? Yes: Go To Question 2; No 4 Please give this questionnaire to the person addressed on the cover letter. 2. By which of the following health plans are you currently covered? Mark All That Apply: Military Health Plans: TRICARE Prime (including TRICARE Prime Remote and TRICARE Overseas); TRICARE Extra or Standard (CHAMPUS); TRICARE Plus; TRICARE for Life; TRICARE Supplemental Insurance; TRICARE Reserve Select; TRICARE Retired Reserve; TRICARE Young Adult; Continued Health Care Benefit Program (CHCBP) (a COBRA-like premium- based health care program). Other Health Plans: Medicare; Federal Employees Health Benefit Program (FEHEP); Medicaid or other state health insurance; A civilIan HMO (such as Kaiser); Other civilian health insurance (such as Blue Cross); Uniformed Services Family Health Plan (USFHP); The Veterans Administration (VA); Government health insurance from a country other than the US; Not sure. 3. Which health plan did you use for at or most of your health care In the last 12 months? Mark Only One Answer: TRICARE Prime; TRICARE Extra or Standard (CHAMPUS) TRICARE Plus; TRICARE Reserve Select; TRICARE Retired Reserve; TRICARE Young Adult; Continued Health Care Benefit Program (CHCBP) (a COBRA-like premium- based health care program); Medicare (may include TRICARE for Life); Federal Employees Health Benefit Program (FEHBP); Medicaid or other state health insurance; A civilian HMO (such as Kaiser); Other civilian health insurance (such as Blue Cross); Uniformed Services Family Health Plan (USFHP); The Veterans Administration (VA); Government health insurance from a country other than the US; Not sure; Did not use any health plan in the last 12 months: Go To Question 5. 4. How many months or years In a row have you been in this heath plan? Less than 6 months; 6 up to 12 months; 12 up to 24 months; 2 up to 5 years; 5 up to 10 years; 10 or more years. 5. Many beneficiaries who are eligible for TRICARE also have the opportunity to obtain other civilian health insurance through their job or a family member's job, through COBRA, or through retirement coverage from a previous job, or from some other group. COBRA lets beneficiaries pay to keep their coverage temporarily when they leave their job. Do you have the opportunity to obtain civilian health Insurance for yourself through some civilian group? Yes; No: Go To Question 9. 6. What options do you have for obtaining civilian coverage? Mark All That Apply: Through my current employer; Through COBRA from my previous employer; Through retirement coverage from my previous employer; Through a family member's current employer; Through COBRA from a family members previous employer; Through retirement coverage from a family members previous employer; Through another organization; Through a government program; Don't know. 7. Are you now covered by a civilian health insurance policy? Yes; No: Go to Question 9. 8. Are you alone covered or are you and others in your household covered by the civilian health insurance policy? I alone am covered; I and at least one other person in my household are covered. 9. Have you used TRICARE for any health care (not including for prescription drugs) in the past 12 months? Yes: Go To Question 11; No. 10. Why haven't you used TRICARE? Mark All That Apply: I have a greater choice of doctors with my civilian plan; My personal doctor is not available to me through TRICARE; My TRICARE regular doctor is no longer available to me; My TRICARE specialist is no longer available to me; My preferred doctors do not accept TRICARE; I prefer civilian hospitals; There are no military facilities near me; I have to travel toe far to see my TRICARE doctor; I get better customer service with civilian plans; TRICARE benefits are poor compared to my civilian plan; It is easier for me to get care through my civilian plan; I do not want to pay the premium for TRICARE; I pay less for civilian care than I would for TRICARE; I have not needed health care; Another reason. For the remainder of this questionnaire, the term health plan refers to the plan you indicated in Question 3, Your Personal Doctor Or Nurse: The next questions ask about your own health care. Do not include care you got when you stayed overnight in a hospital. Do not include the times you went for dental care visits. 11. A personal doctor or nurse is the health provider who knows you best. This can be a general doctor, a specialist doctor, a nurse practitioner, or a physician assistant. Do you have one person you think of as your personal doctor or nurse? Yes; No: Go To Question 15. 12. Using any number from 0 to 10 where 0 is the worst personal doctor or nurse possible, and 10 Is the best personal doctor or nurse possible, what number would you use to rate your personal doctor or nurse? 0: Worst personal doctor or nurse possible; 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: Best personal doctor or nurse possible; I don't have a personal doctor or nurse. 13. How long does it take you to travel to your personal doctor or nurse? Less than 15 minutes; 15 to 30 minutes; 31 minutes to 60 minutes (1 hour); 61 minutes to 90 minutes; 91 minutes to 120 minutes (2 hours); More than 120 minutes (2 hours). 14. Did you have the same personal doctor or nurse before you joined this health plan? Yes: GO TO QUESTION 16. No. 15. Since you joined your health plan, how much of a problem, if any, was it to get a personal doctor or nurse you are happy with? A big problem; A small problem; Not a problem. 16. Where is your personal doctor or nurse located? Mark Only One Answer: A military facility — This includes: Military clinic, Military hospital, PRIMUS clinic, NAVCARE clinic: Go To Question 18; A civilian facility — This includes: Doctor's office, Clinic, Hospital, Civilian TRICARE contractor; Uniformed Services Family Health Plan facility (USFHP); Veterans Affairs (VA) clinic or hospital; I do not have a personal doctor or nurse. 17. In the last 12 months, did you try to find a personal doctor or nurse who was located at a military treatment facility? Yes; No: Go To Question 20. 18. How much of a problem, if any, was it to find an available personal doctor or nurse at a military treatment facility? A big problem; A small problem; Not a problem: Go To Question 20. 19. What is the biggest problem you encountered trying to find a personal doctor or nurse at a military treatment facility? Mark Only One Answer: The military facilities near me have downsized or closed; The wait for an appointment at the military treatment facilities near me is too long; The waiting rooms at the military facilities near me are crowded or uncomfortable; The staff at the military treatment facilities near are not helpful or courteous; I have had problems communicating with doctor(s) at the military treatment facilities; Another reason. 20. Is your personal doctor or nurse a civilian? Yes; No: Go To Question 25; I do not have a personal doctor or nurse: Go To Question 25. 21. The TRICARE civilian provider network is made up of the doctors, clinks, hospitals and other health care providers who are part of DoD's preferred provider pool. Is your personal doctor or nurse part of the TRICARE civilian provider network? Yes; No. 22. What is the specialty of your personal doctor or nurse? Mark Only One Answer: Family Medicine or General Practitioner; Internist; Pediatrician; OB-GYN; Geriatrician or Geriatric Nurse; Preventive Medicine; Nurse Practitioner or Physician's Assistant; Other specialty. 26. In the last 12 months, how much of a problem, if any, was It to see a specialist that you needed to see? A big problem; A small problem; Not a problem; I didn't need a specialist in the last 12 months. 27. In the last 12 months, did you see a specialist? Yes; No: Go To Question 36. 23. In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE? A big problem; A small problem; Not a problem: Go To Question 25. 24. What problems did you encounter in finding a personal doctor who would accept TRICARE? Mark All That Apply: Travel distance too long; Problems communicating with doctor; Doctor(s) not taking any new patients; Doctor(s) not taking new TRICARE patients; Doctor(s) not accepting TRICARE payments; Could not find the specialty I wanted; Did not like doctor(s); Wait for an appointment was too long; Could not find information about doctors; Other. Getting Health Care From A Specialist: When you answer the next questions, do not include dental visits. 25. Specialists are doctors like surgeons, heart doctors, allergy doctors, skin doctors, and others who specialize in one area of health care. In the last 12 months, did you or your doctor think you needed to see a specialist? Yes No: Go To Question 27. 28. We want to know your rating of the specialist you saw most often in the last 12 months. Using Rey number from 0 to 10 where 3 is the worst specialist possible, and 10 is the best specialist possible, what number would you use to rate the specialist? 0: Worst specialist possible; 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: Best specialist possible; I didn't see a specialist in the last 12 months. 29. How long does it take you to travel to the specialist you saw most in the past 12 months? Less than 15 minutes 15 to 30 minutes 31 minutes 4360 minutes (1 hour) 61 minutes to 90 minutes 91 minutes to 120 minutes (2 hours) More than 120 minutes (2 hours) 30. In the last 12 months, did you sees civilian specialist? Yes; No: Go To Question 31. 31. In the last 12 months, was the civilian specialist you saw most the same doctor as your personal doctor? Yes; No. 32. In the last 12 months, was the civilian specialist you saw most part of the TRICARE civilian provider network? Yes; No. 33. In the last 12 months, what was the specialty of the civilian specialist you saw most? Mark Only One Answer: Surgeon; Cardiologist (heart doctor); Allergist; Dermatologist (skin doctor); Rheumatologist (specialist of the joints); Endocrinologist (thyroid, hormone and diabetes specialist); Urologist (specialist of the urinary tract and male reproductive system) Oncologist (cancer specialist); Orthopedist (specialist of the bones, muscles and their connected tissues); Ear, nose and throat specialist; Obstetrician/Gynecologist; Ophthalmologist; Other. 34. In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE? A big problem; A small problem; Not a problem: Go To Question 36. 35. What problems did you encounter in finding a specialist who would accept TRICARE? Mark All That Apply: Travel distance too long; Problems communicating with doctor; Doctor(s) not taking any new patents; Doctor(s) not taking new TRICARE patients; Doctor(s) not accepting TRICARE payments; Could not find the specialty I wanted; Did not like doctor(s); Wait for an appointment was loo long; Could not find intonation about doctors; Other. Calling Doctors' Offices: 36. In the last 12 months, did you call a doctor's office or clinic during regular office hours to get help or advice for yourself? Yes; No: Go To Question 38. 37. In the last 12 months, when you called during regular office hours, how often did you gel the help or advice you needed? Never; Sometimes; Usually; Always; I didn't call for help or advice during regular office hours in the last 12 months. Your Health Care In The Last 12 Months: 38. In the last 12 months, did you have an illness, injury, or condition that needed care right away in a clinic, emergency room, or doctor's office? Yes; No: Go To Question 41. 39. In the last 12 months, when you needed care right away for an illness, injury, or condition, how often did you get care as soon as you wanted? Never; Sometimes; Usually; Always; I didn't need care right away for an illness, injury or condition in the last 12 months. 40. In the last 12 months, when you needed care right away for an illness, injury, or condition, how long did you usually have to wait between trying to get care and actually seeing a provider? Same day; 1 day; 2 days; 3 days; 4-7 days; 8-14 days; 15 days or longer; I didn't need care right away for an illness, injury or condition in the last 12 months; 41. A health provider could be a general doctor, a specialist doctor, a nurse practitioner, a physician assistant, a nurse, or anyone else you would use for health care. In the last 12 months, not counting the times you needed health care right away, did you make any appointments with a doctor or other health provider for health care? Yes No: Go To Question 44. 42. In the last 12 months, not counting times you needed health care right away, how often did you get an appointment for health care as soon as you wanted? Never; Sometimes; Usually; Always; I had no appointments in the last 12 months. 43. In the last 12 months, not counting the times you needed health care right away, how many days did you usually have to wait between making an appointment and actually seeing a provider? Same day 1 day; 2-3 days; 4-7 days; 8-14 days; 15-30 days; 31 days or longer; I had no appointments In the last 12 months. 44. In the last 12 months, how many times did you go to an emergency room to get care for yourself? None; 1; 2; 3; 4; 5 to 9; 10 or more. 45. In the last 12 months, not counting times you went to an emergency room), how many times did you go to a doctor's office or clinic to get care for yourself? None: Go To Question 58; 1; 2; 3; 4; 5-9; 10 or more. 46. In the last 12 months, did you or a doctor believe you needed any care, tests, or treatment? Yes No: Go To Question 48. 47. In the last 12 months, how much of a problem, if any, was it to get the care, tests or treatment you or a doctor believed necessary? A big problem; A small problem; Not a problem; I had no visits in the last 12 months. 48. In the last 12 months, did you need approval from your health plan for any care, tests, or treatment? Yes; No: Go To Question 50. 49. In the last 12 months, how much of a problem, if any, were delays in health care while you waited for approval from your health plan? A big problem; A small problem; Not a problem; I had no visits in the last 12 months. 50. In the last 12 months, how often were you taken to the exam room within 15 minutes of your appointment? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 51. In the last 12 months, how often did office staff at a doctor's office or clinic treat you with courtesy and respect? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 52. In the last 12 months, how often were office staff at a doctor's office or clinic as helpful as you thought they should be? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 53. In the last 12 months, how often did doctors or other health providers listen carefully to you? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 54. In the last 12 months, how often did doctors or other health providers explain things in a way you could understand? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 55. In the last 12 months, how often did doctors or other health providers show respect for what you had to say? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 56. In the last 12 months, how often did doctors or other health providers spend enough time with you? Never; Sometimes; Usually; Always; I had no visits in the last 12 months. 57. Using any number from 0 to 16, where 0 Is the worst health care possible, and 10 is the best health care possible, what number would you use to rate all your health care in the last 12 months? 0: Worst health care possible; 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: Best healthcare possible; I had no visits in the last 12 months. 58. In the last 12 months, where did you go most often for your health care? Mark Only One Answer: A military facility — This includes: Military clinic, Military hospital, PRIMUS clinic, NAVCARE clinic; A civilian facility —This includes: Doctor's office, Clinic, Hospital, Civilian TRICARE contractor; Uniformed Services Family Health Plan facility (USFHP); Veterans Affairs (VA) clinic or hospital; I went to none of the listed types of facilities in the last 12 months. Treatment Or Counseling: 59. In the last 12 months, did you need any treatment or counseling for a personal or family problem? Yes; No: Go To Question 74. 60. In the fast 12 months, what type of provider did you want to see most for this treatment or counseling? MARK ONLY ONE ANSWER: Psychologist; Psychiatrist; Psychotherapist; Social worker; Mental health counselor; Marriage or family therapist; Your personal doctor or nurse; Other; Don't know. 61. In the last 12 months, did you receive treatment or counseling for a personal or family problem? You; No: Go To Question 65. 62. In the last 12 months, did you receive this treatment or counseling from a civilian provider? Yes; No: Go To Question 64. 63. In the last 12 months, did you receive this treatment or counseling from a provider In TRICARE's civilian network? Yes; No. 64. In the last 12 months, what type of provider did you see most often for this treatment or counseling? Mark Only One Answer: Psychologist; Psychiatrist; Psychotherapist; Social worker; Mental health counselor; Marriage or family therapist; Your personal doctor or nurse; Other; Don't know. 65. In the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan? A big problem; A small problem; Not a problem: Go To Question 67. 66. In the last 12 months, what problems did you encounter In finding treatment or counseling? Mark All That Apply: Travel distance too long; Problems communicating with doctor; Doctor(s) or counselor(s) not taking new patients; Doctor(s) or counselor(s) not taking new TRICARE patients; Doctor(s) or counselor(s) not accepting TRICARE payments; Could not find the specialty I wanted; Did not like doctor(s) or counselors); Wait for an appointment was too long; Could not find information about doctors or counselors; Other. 67. In the last 12 months, did you need treatment or counseling right away? Yes; No: Go To Question 69. 68. In the last 12 months, when you needed treatment or counseling right away, how often did you see someone as soon as you wanted? Never; Sometimes; Usually; Always. 69. In the last 12 months, did you need approval for any treatment or counseling? Yes; No: Go To Question 71. 70. In the last 12 months, how much of a problem, if any, were delays in treatment or counseling while you waited for approval? A big problem; A small problem; Not a problem. 71. In the last 12 months, did you call customer service to get information or help about treatment or counseling? Yes; No: Go To Question 13. 72. In the last 12 months, how much of a problem, if any, was it to get the help you needed when you called customer service? A big problem; A small problem; Not a problem. 73. Using any number from 0 to 10 where 0 is the worst treatment or counseling possible, and 10 is the best treatment or counseling possible, what number would you use to rate all your treatment or counseling in the last 12 months? 0: Worst treatment or counseling possible; 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: Best treatment or counseling possible. I didn't receive treatment or counseling in the last 12 months Your Health Plan: The next question asks about your experience with your health plan. By your health plan, we mean the health plan you marked in Question 3. 74. Using any number from 0 to 10, where 0 is the worst health plan possible, and 10 is the best health plan possible, what number would you use to rate your health plan? 0: Worst health plan possible; 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: Best health plan possible. Even if you do not use TRICRARE Standard or Extra, we'd like to know what you like and dislike about these plans compared to civilian plans. 75. What do you like about TRICARE Standard and Extra? Mark All That Apply. I have a better choice of doctors with TRICARE than with a civilian plan; My preferred personal doctor is only available U3 me through TRICARE; I want to be time I can always use military health care; I get better customer service with TRICARE than with civilian plans; It is easier to get care through TRICARE than a civilian plan; The premium for TRICARE is lower than the premium for civilian coverage; Copays and deductibles cost less through TRICARE than a civilian plan; Civilian benefits are poor compared to TRICARE; Other; 76. What do you dislike about TRICARE Standard and Extra? Mark All That Apply: I have a better choice of doctors with a civilian plan than with TRICARE; My preferred personal doctor is not available to me through TRICARE; I worry about losing access to civilian coverage; I get better customer service with civilian plans than with TRICARE; It is easier to get care through a civilian plan than TRICARE; The premium tor TRICARE is too high ; Copays and deductibles cost more through TRICARE than a civilian plan; TRICARE benefits are poor compared to a civilian plan; Other. Preventive Care: Preventive care is medical cars you receive that is intended to maintain your good health or prevent a future medical problem. A physical or blood pressure screwing are examples of preventive care. 77. When did you last have a blood pressure reading? Less than 12 months ago; 1 to 2 years ago; More than 2 years ago. 78. Do you know if your blood pressure Is too high? Yes, it is too high; No, it is not too high; Don't know. 79. Have you ever smoked at least 100 cigarettes In your entire We? Yes; No: Go To Question 82; Don't know: Go To Question 82. 80. Do you now smoke every day, some days or not at all? Every day; Some days; Not at all: Go To Question 82; Don't know: Go To Question 82. 81. In the last 12 months, on how many visits were you advised to quit smoking by a doctor or other health provider In your plan? None 1 visit; 2 to 4 visits; 5 to 9 visits; 10 or more visits; I had no visits in the last 12 months. 82. When did you last have a cholesterol screening, that is, a test to determine the level of cholesterol in your blood? Less than 12 months ago; 1 to 2 years ago; More than 2 but less than 5 years ago; 5 or more years ago; Never had a cholesterol screening. 83. When did you last have a flu shot? Less than 12 months ago; I to 2 years ago; More than 2 years ago; Never had a flu shot. 84. Are you male or female? Male: Go To Question 88; Female. 85. When did you last have a Pap smear test? Within the last 12 months; 1 to 3 years ago; More than 3 but less than 5 years ago; 5 or more years ago; Never had a Pap smear test. 88. Are you under age 40? Yes: Go To Question 88; No. 87. When was the last time your breasts were checked by mammography? Within the last 12 months; 1 to 2 years ago; More than 2 but less than 5 years ago; 5 or more years ago; Never had a mammogram. About You: 88. In general, how would you rate your overall health now? Excellent; Very good; Good; Fair; Poor. 89. Are you limited in any way in any activities because of any Impairment or health problem? Yes; No. 90. What is the highest grade or level of school that you have completed? 8th grade or less; Some high school, but did not graduate; High school graduate or GED; Some college or 2-year degree; 4-year college graduate; More than 4-year college degree. 91. Are you of Hispanic or Latino origin or descent? Mark 'No' if not Spanish/Hispanic/Latina. No, not Spanish, Hispanic, or Latino; Yes, Mexican, Mexican American, Chicano; Yes, Puerto Rican; Yes, Cuban; Yes, other Spanish, Hispanic, or Latino. 92. What Is your race? Mark One Or More races to indicate what you consider yourself to be. White; Black or African American; American Indian or Alaska Native; Asian (e.g., Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese); Native Hawaiian or other Pacific Islander (e g., Samoan, Guamanian, or Chamorro). 93, What is your age now? 18 to 24; 25 to 34; 35 to 44; 45 to 54; 55 to 64; 65 to 74; 75 or older. 94. Which of the following income ranges is closest to your family's (2010) total income from all sources? Your best estimate would be fine. Less than $10,000; $10,000 to $24.999; $25,000 to $49,999; 950,000 to $74,999; $75,000 to $99,999; 5100,000 to $124,999; $125,000 to $149,999; $150,000 and above; Don't know. Thank You For Taking The Time To Complete The Survey! Your generous contribution will greatly aid efforts to improve the health of our military community. Return your survey in the postage-paid envelope. If the envelope is missing, please send to: Office of the Assistant Secretary of Defense (HA): TMA/HPAE: c/o Synovale Survey Processing Center: PO Box 5030: Chicago, IL 60680-4138: Questions about the survey? Email: survey-dodq2@synovate.net Toll-free phone (in the US, Puerto Rico and Canada): 1-877-236-2390, available 24 hours a day. Toll-free fax (lathe US and Canada): 1-800-409-7681. When calling or writing, please provide your name, address, and the 8- digit number above your address on the envelope. Questions about your TRICARE coverage? For additional information on TRICARE, or if you are not sure about your benefits, or it you don't have a primary care manager. contact the TRICARE Service Center in your region: North: 1-877-874-2273; South: 1-800-444-5445; West: 1-888-874-9378; Outside the US: 1-888-777-8343; The website is: www.tricare/mil/contactus. Veterans Contact the US Department of Veterans Affairs at 1-877-422- VETS, or go to www.va.gov. [End of section] Appendix III: Survey Instruments for Civilian Physicians and Nonphysician Mental Health Care Providers: The National Defense Authorization Act for Fiscal Year 2008 (NDAA 2008) directed the Department of Defense (DOD) to determine the number of health care and mental health care providers that currently accept nonenrolled beneficiaries as patients under TRICARE, DOD's health care program. For the purpose of this report, we use the term "nonenrolled beneficiaries" for beneficiaries who are not enrolled in TRICARE Prime and who use the TRICARE Standard or Extra options, or TRICARE Reserve Select (TRS).[Footnote 68] Specifically, NDAA 2008 directed DOD to survey providers each fiscal year, 2008 through 2011. The NDAA 2008 also required that the provider survey include questions seeking information to determine (1) whether the provider is aware of the TRICARE program, (2) the percentage of the provider's current patient population that uses any form of TRICARE, (3) whether the provider accepts Medicare patients, and (4) if the provider accepts Medicare patients, whether the provider would accept new Medicare patients. DOD implemented two versions of its provider survey, one for physicians, including psychiatrists, and one for nonphysician mental health providers.[Footnote 69] For the 2008 fielding of the civilian provider survey, 11 and 12 questions were included in the physician and nonphysician mental health provider survey instruments, respectively. Over the next 3 years of the civilian provider survey's fielding, TRICARE Management Activity (TMA) generally used the same questions, but made the following adjustments to the survey instruments: * Beginning with the 2009 fielding of both survey instruments and beyond, TMA adjusted Question #1 which asked the provider whether they provided health care to patients in an office-based practice (for physicians) or a private practice (for nonphysician mental health care providers) so that a "no" response would no longer instruct the provider to stop answering the survey at that point. Instead, the revision directed the provider to the newly added Question #1a that asked the provider what type of practice they were in (if they answered "no" to Question #1). * For the 2010 and 2011 fieldings of the physician survey instrument, TMA also adjusted Question #1 from "Does [the provider] provide treatment to patients through an office-based practice?" to "Does [the provider] provide treatment to patients through private practice?" Following are the actual survey instruments from the 2011 fielding that TMA used to obtain information from physicians and nonphysician mental health care providers. Office of the Assistant Secretary Of Defense: Health Affairs: Tricare Management Activity: Health Program Analysis And Evaluation Directorate: Unique ID: FOR: [Title] [Insert Provider Name] Street Address: City, State, and Zip: Dear Billing Manager for [Title] [Insert Provider Name], Hello! The physician named above has been selected to participate in a very important survey effort. In support of U.S. military men and women, Congress has directed the Department of Defense to survey civilian physicians across the U.S. to determine whether military service members and their families have access to the health care they need. A substantial amount of health care to service members and their families is delivered by private, civilian physicians like [Title] [Insert Provider Name], and we need your help. We are asking you to please answer the questions on the back of this letter on behalf of the physician above and return it within five days. There are several ways to complete this survey, which should only take five minutes of your time: * Complete the survey on the reverse side of this letter and return it via postal mail in the enclosed postage paid envelope * Complete the survey on the reverse side of this letter and fax it to 1-800-585-9446 * Complete the survey on the internet at the following URL: http://www.dodcv08com. Your unique login name: xxxxxxxx. Your unique password: xxxxxxxx. We recognize that there may be more than one provider in your office and ask that you complete the survey for the provider listed above. Since we may survey more than one provider in your office, please complete each survey for the appropriate provider named above. If you are not the appropriate person to answer these questions, please pass this on to the person in your office most familiar with the [Title] [Insert Provider Namers billing and insurance. Thank you in advance for your cooperation and help as we examine this important issue that impacts our American service men and women. If you have questions about this survey, please call Synovate between the hours of 8AM and 5PM Eastern Time at 1-800-228-6764. Sincerely yours, Thomas V. Williams, Ph.D. Director, Health Program Analysis and Evaluation Directorate: Office of the Assistant Secrete of Defense (Health Affairs) TRICARE Management Activity: Survey Questions On Reverse Side: We estimate this survey will take an average five (5) minutes to complete. including the time for reviewing instructions, getting the needed data, and completing and reviewing the survey. You may send comments regarding our estimate or any other aspect of this survey, including suggestions for reducing the completion time, to Department of Defense, Washington Headquarters Services, Executive Services Directorate, Information Management Division (OMB Number 0720-0031). The OMB number above is currently valid, and you are not required to respond, unless this number is displayed. This Official DoD survey may be confirmed at the TRICARE website http://www.tricare.mil/hpae/home, click on the List of Approved Surveys, and find "Survey of Civilian Provider Acceptance of TRICARE Standard." Privacy Act Statement: According to the Privacy Act of 1974 (Public law 93-579), the Department of Defense is required to inform you of the purposes and use of this surrey. Please read carefully. Authority: Section 711 of the National Defense Authorization Act for Fiscal Year 2008 (Public Law (P.L.) 110-181. Purpose: Mandated by Congress, this confidential Survey of civilian providers helps TRICARE health policy makers gauge civilian provider awareness and acceptance of the TRICARE Standard health care benefit option, and will provide valuable aggregated input to help improve the Military Health System. Routine Uses: Those disclosures generally permitted under 5.US.C.552a(b) of the Privacy Act. Disclosure: Providing information In this questionnaire is voluntary. There is no penalty if you choose not to respond. However, maximum participation is encouraged no that data will be as complete and representative as possible. You may notice a number on this survey: this number is used only to let us know if you returned the survey to minimize sending you reminders. Q1. Does [Title) (Insert Provider Name) provide treatment to patients through private practice? (By this we mean that the provider is working in a setting where he/she can decide or influence the decision regarding which insurance to accept.) Yes: (Go to Q2); No, does not provide treatment, or has retired). (Thank you, please return the questionnaire); No, not in private practice: (Go to Q1a). Q1a. What type of practice is [Title) (Insert Provider Name) in? (please choose one): Government: Federal, State or other municipality; School, University or other academic institution; Hospital staff; Contractor providing services exclusively to government clients; Rehab Facility, Nursing Home, or Home Health Provider; Closed Panel HMO; Other. Q2. Is [Title] [Insert Provider Name] aware of the TRICAPE health care program? Yes; No; Don't Know. Q3. As of today, Is [Title] [Insert Provider Name] a contracted member of the TRICARE network of health care providers? Yes; No; I Don't Know. Q4. As of today, is [Title] [Insert Provider Name] accepting new TRICARE Stoddard patients? No: (Go to Q5); Yes, an a claim by claim basis only: (Go to Q6); Yes, for all claims: (Go to Q6); I Don't know: (Go to Q6). Q5. If you answered "no" to Q4 above, why Is [Title) [insert Provider Name] not accepting new TRICARE Standard patients? Please list all the reasons. If you need additional space, please include a separate sheet of paper. Q6. What percentage of patients seen by (Title) [Insert Provider Name] use any form of TRICARE? If unsure, please write down your best guess. None: Dr. [Insert Last Name] has no TRICARE patients; percent use some form of TRICARE; Don't Know. Q7. Does [Tide] (Insert Provider Name) accept any Medicare patients? Yes; No; I Don't Know. Q8. As of today, is [Title] [Insert Provider Name) accepting new Medicare patients? Yes: Thank you, please return the questionnaire; No: (Go to Q9); I Don't Know: (Go to Q10). Q9. If you answered "no" to Q8 above, why Is [Title] [Insert Provider Name) not accepting new Medicare patients? Please list all reasons. If you need additional space, please include a separate sheet of paper. Q10. Does [The) [Insert Provider Name] accept any insurance plans? Yes; No. Q11. As of today is (Title) (Insert Provider Name) accepting any new patients? Yes; No; I Don't Know. Thank you for taking part in this survey. Please put this in the enclosed postage-paid envelope and return it to the Survey Processing Center or fax the survey to Synovate at 1-800-585-9446. If you have any questions about TRICARE, its specific health plans, or the benefits it provides, please visit the TRICARE web site at www.tricare.osd.mil for assistance. [End of section] Appendix IV: Areas Where Nonenrolled Beneficiaries Had Problems Accessing Civilian Providers, 2008-2011: The 2008-2011 beneficiary survey indicated individual areas where nonenrolled beneficiaries experienced problems finding "any civilian provider," civilian primary care providers, and civilian specialty care providers.[Footnote 70] We define these locations as areas where the percentage of nonenrolled beneficiaries who experienced difficulties finding a civilian provider was at the national estimate or higher. Problems Finding Any Provider: We identified 24 individual areas (out of the 215 individual areas surveyed by the 2008-2011 beneficiary surveys)[Footnote 71] where the percentage of nonenrolled beneficiaries who experienced problems finding any type of provider who would accept TRICARE met or exceeded the national estimate.[Footnote 72] We then identified 49 additional areas where the percentage of nonenrolled beneficiaries who experienced these problems was less than the national estimate. [Footnote 73] The remaining 130 areas had estimates that ranged from 18 to 50 percent, but because of their confidence intervals, were neither above nor below the 31 percent threshold.[Footnote 74] Figure 16 shows the geographic distribution of these three categories of areas. Figure 16: Estimated Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Primary, Specialty, or Mental Health Care Provider, 2008-2011: [Refer to PDF for image: illustrated U.S. map] Areas depicted on the map: 31% or more experienced problems: Fewer than 31% experienced problems; Could not determine if estimate was above or below 31%. Source: GAO analysis of TMA data (data); MapInfo (map). Notes: Nationwide, an estimated 31 percent of nonenrolled beneficiaries experienced problems finding any civilian provider (i.e., a civilian primary, specialty, or mental health care provider). We used the lower 95 percent confidence limit to identify areas for which 31 percent or more of nonenrolled beneficiaries experienced problems finding any civilian provider. We used the upper 95 percent confidence limit to identify areas for which fewer than 31 percent of nonenrolled beneficiaries experienced problems. Areas depicted in white indicate areas that did not fall into either of the above categories due to their 95 percent confidence interval. We excluded areas from our analysis with fewer than 30 respondents combined for the three survey questions that asked if beneficiaries had problems finding a personal doctor or nurse, specialist, or treatment and counseling within the last 12 months. TRICARE Management Activity did not identify additional Hospital Service Areas to survey for its 2011 beneficiary survey. [End of figure] Table 10 lists the 24 individual areas where at least 31 percent of nonenrolled beneficiaries experienced problems finding any type of provider who would accept TRICARE patients, and the area's corresponding estimated percentage of civilian providers who would accept new TRICARE patients. Table 10: Prime Service Areas (PSA), Non-Prime Service Areas (non- PSA), and Hospital Service Areas (HSA) with 31 Percent or More of Nonenrolled Beneficiaries Experiencing Problems Finding Any Type of Provider, and the Willingness of Civilian Providers in the Corresponding Areas to Accept New TRICARE Patients, 2008-2011: Area: 1. Austin, TX[C]; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 58 (18)[D]; Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 46 (6). Area: 2. Anchorage, AK[E]; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 56 (20)[D]; Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 68 (4). Area: 3. AK; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 51 (17)[D]; Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 75 (4). Area: 4. AK; Area type: non-PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 51 (15); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 70 (14). Area: 5. Central-Eastern TX; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 49 (12); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 59 (5). Area: 6. Western-Central WA; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 48 (15)[D]; Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 52 (8). Area: 7. Dallas/Ft. Worth, TX[F]; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 48 (14); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 50 (6). Area: 8. Central/Southern-Central Coastal CA; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 48 (12); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 45 (8). Area: 9. Fredericksburg, VA[G]; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 48 (11); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 74 (6). Area: 10. Columbia/Sumter, SC; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 47 (13); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 72 (6). Area: 11. Prince William Co., VA[H]; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 47 (11); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 74 (6). Area: 12. Southern-Central AZ; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 47 (11); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 59 (7). Area: 13. Northeastern TX; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 47 (10); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 53 (6). Area: 14. Central-Northern VA; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 45 (8); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 75 (4). Area: 15. Fairfax Co., VA[I]; Area type: HSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 44 (10); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 60 (5). Area: 16. Northeastern OK; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 43 (12); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 57 (6). Area: 17. Washington, D.C.; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 43 (11); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 55 (7). Area: 18. Central-Southern MD; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 43 (9); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 53 (6). Area: 19. Southern AZ PSA; Southeastern CA; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 42 (10); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 60 (5). Area: 20. Southeastern FL; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 42 (9); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 58 (6). Area: 21. Southwestern MI; Area type: non-PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 41 (11); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 66 (7). Area: 22. LA; Southwestern MS; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 41 (9); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 60 (7). Area: 23. Western-Central/Northern/Southern TX; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 41 (9); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 68 (7). Area: 24. Central-Northern/Central-Eastern FL; Area type: PSA; Estimated percentage of beneficiaries with a problem finding any type of provider (margin of error)[A]: 40 (9); Estimated percentage of civilian providers accepting new TRICARE patients (margin of error)[B]: 71 (6). Source: GAO analysis of TMA data. Notes: The margins of error are at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the civilian provider survey. To be included in this table, areas had an estimated 31 percent or more of nonenrolled beneficiaries who were having difficulties finding a provider who would accept TRICARE as payment (using the estimate's margin of error at the 95 percent confidence level). Estimated percentages and margins of error have been rounded to the nearest whole number. Each surveyed HSA was part of a PSA or non-PSA (depending on the location). [A] Estimated percentage is based on the number of nonenrolled beneficiaries who responded "a big problem" or "a small problem" to any one of the following three questions: (1) "In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE?"; (2) "In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE?"; or (3) "In the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan?" [B] Estimated percentage is based on the number of civilian providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [C] Although most of the Austin, Texas, HSA is within the Eastern- Central Texas PSA, one of its zip codes is part of the Western- Central/Northern/Southern Texas PSA. [D] These estimates have relative margins of error that are 30 percent or greater. [E] The Anchorage, Alaska, HSA is part of the Alaska PSA and the Alaska non-PSA. [F] The Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [G] The Fredericksburg, Virginia, HSA is part of the Central-Northern Virginia PSA. [H] The Prince William County, Virginia, HSA is part of the Central- Northern Virginia PSA and the Central-Southern Maryland PSA. [I] The Fairfax, Virginia, HSA is part of the Central-Southern Maryland PSA and the Washington, D.C. PSA. [End of table] Problems Finding Civilian Primary Care Providers: We identified 21 individual areas where the percentage of nonenrolled beneficiaries who experienced problems finding a civilian primary care provider who would accept TRICARE patients met or exceeded the national estimate.[Footnote 75] We then identified 50 additional areas where the percentage of nonenrolled beneficiaries who experienced these problems was less than the national estimate.[Footnote 76] The remaining 129 areas had estimates that ranged from 13 to 44 percent, but because of their confidence intervals, were neither above nor below the 25 percent threshold.[Footnote 77] Figure 17 shows the geographic distribution of these three categories of areas. Figure 17: Estimated Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Primary Care Provider, 2008- 2011: [Refer to PDF for image: illustrated U.S. map] Areas depicted on the map: 25% or more experienced problems: Fewer than 25% experienced problems; Could not determine if estimate was above or below 25%. Source: GAO analysis of TMA data (data); MapInfo (map). Notes: Nationwide, an estimated 25 percent of nonenrolled beneficiaries experienced problems finding a civilian primary care provider. We used the lower 95 percent confidence limit to identify areas for which 25 percent or more nonenrolled beneficiaries experienced problems finding a civilian primary care provider. We used the upper 95 percent confidence limit to identify areas for which fewer than 25 percent of nonenrolled beneficiaries experienced problems. Areas depicted in white indicate areas that did not fall into either of the above categories. We excluded areas from our analysis with fewer than 30 respondents to the survey question that asked: "In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE?" TRICARE Management Activity did not identify additional Hospital Service Areas to survey for its 2011 beneficiary survey. [End of figure] Table 11 lists the 21 individual areas where at least 25 percent of nonenrolled beneficiaries experienced problems finding a civilian primary care provider who would accept TRICARE patients, and the areas' corresponding estimated percentage of civilian primary care providers who would accept new TRICARE patients. Table 11: Prime Service Areas (PSA), Non-Prime Service Areas (non- PSA), and Hospital Service Areas (HSA) with 25 Percent or More of Nonenrolled Beneficiaries Experiencing Problems Finding a Civilian Primary Care Provider, and the Willingness of Civilian Primary Care Providers in the Corresponding Areas to Accept New TRICARE Patients, 2008-2011: Area: 1. Austin, TX[C]; Area type: HSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 56 (18)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 42 (11). Area: 2. Western-Central WA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 47 (16)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 60 (13). Area: 3. Prince William Co., VA[E]; Area type: HSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 44 (12); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 80 (10). Area: 4. Southern-Central AZ; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 44 (12); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 67 (13). Area: 5. Los Angeles, CA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 42 (14)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 69 (9). Area: 6. Columbia/Sumter, SC; Area type: HSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 42 (13)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 84 (8). Area: 7. Central-Eastern TX; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 41 (13)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 67 (9). Area: 8. Dallas/Ft. Worth, TX[F]; Area type: HSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 40 (14)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 51 (12). Area: 9. Northeastern TX; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 40 (10); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 48 (10). Area: 10. LA; Southwestern MS; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 39 (10); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 71 (11). Area: 11. Asheville, NC; Area type: HSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 38 (11)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 68 (11). Area: 12. Southwestern MI; Area type: non-PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 38 (11)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 79 (9). Area: 13. Central GA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 38 (12)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 77 (9). Area: 14. Eastern-Central TX; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 38 (12)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 53 (10). Area: 15. Washington, D.C.; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 38 (11); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 59 (14). Area: 16. Central/Southern-Central Coastal CA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 37 (12)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 64 (12). Area: 17. Western NY; Area type: non-PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 37 (12)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 64 (13). Area: 18. Central MS; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 35 (11)[D]; Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 88 (7). Area: 19. Central-Northern VA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 35 (8); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 82 (6). Area: 20. Central-Southern MD; Area type: PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 33 (9); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 69 (10). Area: 21. Northern/Central/Western NM; Northeastern AZ; Southwestern CO; Area type: non-PSA; Estimated percent of beneficiaries with a problem finding a primary care provider (margin of error)[A]: 33 (8); Estimated percent of primary care providers accepting new TRICARE patients (margin of error)[B]: 66 (12). Source: GAO analysis of TMA data. Notes: The margins of error are at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the provider survey. To be included in this table, areas had an estimated 25 percent or more of nonenrolled beneficiaries who were having difficulties finding a civilian primary care provider who would accept TRICARE as payment (using the estimate's margin of error at the 95 percent confidence level). Estimated percentages and margins of error have been rounded to the nearest whole number. Each surveyed HSA was part of a PSA or non-PSA (depending on the location). [A] Estimated percentage is based on the number of nonenrolled beneficiaries who responded "a big problem" or "a small problem" to the question that asked "In the last 12 months, how much of a problem was it to find a personal doctor or nurse who would accept TRICARE?" [B] Estimated percentage is based on the number of civilian primary care providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [C] The Austin, Texas, HSA is part of the Eastern-Central Texas PSA. [D] These estimates have relative margins of error that are 30 percent or greater. [E] The Prince William County, Virginia, HSA is part of the Central- Northern Virginia PSA and Central-Southern Maryland PSA. [F] The Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [End of table] Problems Finding Civilian Specialty Care Providers: We identified nine individual areas where the percentage of nonenrolled beneficiaries who experienced problems finding a civilian specialty care provider who would accept TRICARE patients met or exceeded the national estimate.[Footnote 78] We then identified 34 additional areas where the percentage of nonenrolled beneficiaries who experienced these problems was less than the national estimate.[Footnote 79] The remaining 144 areas had estimates that ranged from 14 to 47 percent, but because of their confidence intervals, were neither above nor below the 25 percent threshold.[Footnote 80] Figure 18 shows the geographic distribution of these three categories of areas. Figure 18: Estimated Percentage of Nonenrolled Beneficiaries Who Experienced Problems Finding a Civilian Specialty Care Provider, 2008- 2011: [Refer to PDF for image: illustrated U.S. map] Areas depicted on the map: 25% or more experienced problems: Fewer than 25% experienced problems; Could not determine if estimate was above or below 25%. Source: GAO analysis of TMA data (data); MapInfo (map). Notes: Nationwide, an estimated 25 percent of nonenrolled beneficiaries experienced problems finding a civilian specialty care provider. We used the lower 95 percent confidence limit to identify areas for which 25 percent or more of nonenrolled beneficiaries experienced problems finding a civilian specialty care provider. We used the upper 95 percent confidence limit to identify areas for which fewer than 25 percent of nonenrolled beneficiaries experienced problems. Areas depicted in white indicate areas that did not fall into either of the above categories. We excluded areas from our analysis with fewer than 30 respondents to the survey question that asked: "In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE?" TRICARE Management Activity did not identify additional Hospital Service Areas to survey for its 2011 beneficiary survey. [End of figure] Of the nine individual areas where at least 25 percent of nonenrolled beneficiaries experienced problems finding a civilian specialty care provider who would accept TRICARE patients, one of the areas had less than 50 civilian specialty care respondents to the civilian provider survey--TMA's threshold for reporting civilian provider survey results. Therefore, we only included eight areas in our collective analysis of access to specialty care in the beneficiary and civilian provider survey results. Table 12 lists these eight individual areas and the area's corresponding estimated percentage of civilian specialty care providers that would accept new TRICARE patients. Table 12: Prime Service Areas (PSA), Non-Prime Service Areas (non- PSA), and Hospital Service Areas (HSA) with 25 Percent or More of Nonenrolled Beneficiaries Experiencing Problems Finding Civilian Specialist Providers, and the Willingness of Civilian Specialist Providers in the Corresponding Areas to Accept New TRICARE Patients, 2008-2011: Area: 1. AK; Area type: PSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 49 (17)[C]; Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 83 (6). Area: 2. Northwestern/Central/Central-Eastern WA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 45 (17)[C]; Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 84 (8). Area: 3. Central-Eastern TX; Area type: PSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 42 (15)[C]; Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 76 (8). Area: 4. Central-Northern VA; Area type: PSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 40 (9); Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 85 (6). Area: 5. Northeastern TX; Area type: PSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 39 (13)[C]; Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 75 (7). Area: 6. Western-Central/Northern/Southern TX; Area type: PSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 39 (12)[C]; Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 79 (11). Area: 7. Prince William Co., VA[D]; Area type: HSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 50 (13); Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 86 (8). Area: 8. Fredericksburg, VA[E]; Area type: HSA; Estimated percent of beneficiaries with a problem finding a specialty care provider (margin of error)[A]: 38 (12)[C]; Estimated percent of civilian specialty care providers accepting new TRICARE patients (margin of error)[B]: 78 (9). Source: GAO analysis of TMA data. Notes: The margins of error are at the 95 percent confidence level. Areas were considered only if they had at least 30 respondents for the beneficiary survey and at least 50 respondents for the provider survey. To be included in this table, areas had an estimated 25 percent or more of nonenrolled beneficiaries who were having difficulties finding a civilian specialty care provider who would accept TRICARE as payment (using the estimate's margin of error at the 95 percent confidence level). Estimated percentages and margins of error have been rounded to the nearest whole number. Each surveyed HSA was part of a PSA or non-PSA (depending on the location). [A] Estimated percentage is based on the number of nonenrolled beneficiaries who responded "a big problem" or "a small problem" to the question that asked "In the last 12 months, how much of a problem was it to find a doctor with this specialty who would accept TRICARE?" [B] Estimated percentage is based on the number of civilian specialty care providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [C] These estimates have relative margins of error that are 30 percent or greater. [D] The Prince William County, Virginia, HSA is part of the Central- Northern Virginia PSA. [E] The Fredericksburg, Virginia, HSA is part of the Central-Northern Virginia PSA. [End of table] Problems Finding Civilian Mental Health Care Providers: Because of the low number of nonenrolled beneficiary responses to the questions about civilian mental health care,[Footnote 81] we are unable to identify specific geographic areas where nonenrolled beneficiaries have access problems to civilian mental health care providers. Of the 215 areas surveyed in the 4-year beneficiary survey, only 5 areas had 30 or more respondents--TMA's threshold for reporting beneficiary survey results--who indicated that they needed mental health care and received it from a civilian provider. Additionally, for those 5 areas that did have at least 30 nonenrolled beneficiary responses, the margins of error were between 10 and 25 percentage points. [End of section] Appendix V: Civilian Provider Acceptance of Any New TRICARE Patients in Hospital Service Areas Surveyed in Fiscal Year 2011: The TRICARE Management Activity (TMA) fielded its provider and beneficiary surveys to the same Hospital Service Areas (HSA) each year with one exception. Because of resource constraints, the 2011 fielding of the beneficiary survey did not include any HSAs. However, 16 HSAs were included in the 2011 fielding of the provider survey. Because beneficiaries were not surveyed for these HSAs, they are not included in our collective analysis of the beneficiary and civilian provider survey results. Table 13 lists the 16 HSAs that were surveyed in the 2011 civilian provider survey fielding and the estimated percentage of civilian providers who were accepting any new TRICARE patients. Table 13: Hospital Service Areas (HSA) Surveyed in 2011, and the Estimated Percentage of Civilian Providers Who Were Accepting Any New TRICARE Patients: HSA: 1. Oklahoma City, OK; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 51 (10). HSA: 2. Madison, WI; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 52 (9). HSA: 3. Athens, OH; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 52 (10). HSA: 4. Tucson, AZ; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 56 (5). HSA: 5. Tulsa, OK; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 58 (9). HSA: 6. Nashville, TN; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 65 (7). HSA: 7. Lihue/Waimea/Wailuku, HI; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 66 (7). HSA: 8. Birmingham, AL; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 67 (6). HSA: 9. Laramie, WY; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 71 (14). HSA: 10. Hopkinsville, KY; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 72 (11). HSA: 11. Tacoma, WA; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 75 (8). HSA: 12. Augusta, GA; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 80 (5). HSA: 13. Rapid City, SD; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 81 (6). HSA: 14. Columbus, GA; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 84 (6). HSA: 15. Hampton/Newport News, VA; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 85 (4). HSA: 16. Petersburg/Hopewell, VA; Estimated percent of civilian providers accepting new TRICARE patients (margin of error)[A]: 91 (6). Source: GAO analysis of TMA data. Notes: The margins of error are at the 95 percent confidence level. Areas were considered only if they had at least 50 respondents for the civilian provider survey. [A] Estimated percentage is based on the number of civilian providers who answered "for all claims" or a "claim-by-claim basis" to the question that asked "As of today, is the provider accepting new TRICARE Standard patients?" [End of table] [End of section] Appendix VI: Comments from the Department of Defense: The Assistant Secretary of Defense: Health Affairs: 1200 Defense Pentagon: Washington, DC 20301-1200: March 27, 2013: Ms. Debra A. Draper: Director, Health Care: U.S. Government Accountability Office: 441 G Street, N.W. Washington, DC 20548: Dear Ms. Draper: This is the Department of Defense's response to the Government Accountability Office (GAO) Draft Report, GA0-13-364, "Defense Health Care: TRICARE Multi-year Surveys Indicate Problems with Access to Care for Nonenrolled Beneficiaries," dated February 25, 2013, (GAO Code 291045). Thank you for the opportunity to review and comment on the draft report. Overall, I concur with the draft report's findings and conclusions. The report does not contain any recommendations, and I have no significant technical changes to offer other than what we have provided to the analysts. I thank you for your detailed review of our survey methodology and processes. My points of contact on this matter are Dr. Richard Bannick (Functional) and Mr. Gunther Zimmerman (Audit Liaison). Dr. Bannick may be reached at (703) 681-3638, and Mr. Zimmerman may be reached at (703) 681-4360. Sincerely, Signed by: Jonathan Woodson, M.D. [End of section] Appendix VII: GAO Contact and Staff Acknowledgments: GAO Contact: Debra A. Draper, (202) 512-7114 or draperd@gao.gov: Staff Acknowledgments: In addition to the contact named above, Bonnie Anderson, Assistant Director; Jennie Apter; Linda Galib; Giselle Hicks; Jeff Mayhew; Lisa Motley; Dan Ries; and Eric Wedum made key contributions to this report. [End of section] Footnotes: [1] Eligible beneficiaries include active duty personnel and their dependents, medically eligible National Guard and Reserve servicemembers and their dependents, and retirees and their dependents and survivors. Active duty personnel include Reserve component members on active duty for at least 30 days. [2] Through individual agreements between military treatment facilities and the Department of Veterans Affairs' medical centers, eligible beneficiaries may also receive certain types of care from Department of Veterans Affairs' medical centers in some locations. [3] TMA uses managed care support contractors in each of the three TRICARE regions (North, South, and West) to develop networks of civilian providers and to perform other customer-service functions, such as processing claims and assisting beneficiaries with finding providers. [4] These geographic areas are determined by the Assistant Secretary of Defense for Health Affairs and are defined by a set of five-digit zip codes, usually within an approximate 40-mile radius of a military treatment facility. In addition to developing networks of civilian providers in PSAs, the managed care support contracts also require the contractor to develop civilian provider networks at all Base Realignment and Closure (BRAC) sites, which are military installations that have been closed or realigned as a result of decisions made by the Commission on Base Realignment and Closure. [5] TRICARE offers several other plans, including TRICARE Reserve Select (TRS) for certain National Guard and Reserve servicemembers, and TRICARE Young Adult (Prime and Standard options) for servicemembers' dependents up to age 26. TRICARE also offers TRICARE for Life to TRICARE beneficiaries who are eligible for Medicare and enroll in Part B. Under the TRICARE for Life program, TRICARE processes claims after they have been adjudicated by Medicare. [6] All beneficiaries may obtain care at military treatment facilities, although priority is given to active duty personnel and then to beneficiaries enrolled in TRICARE Prime. [7] We include TRS beneficiaries in our definition of nonenrolled beneficiaries because, although they must enroll in the plan, they can receive care from network or non-network providers similarly to TRICARE Standard and Extra beneficiaries. We did not include TRICARE Young Adult-Standard Option beneficiaries in our analysis because this plan did not become available until May 2011. [8] See Pub. L. No. 108-136, § 723, 117 Stat. 1392, 1532-34 (2003), and S. Rep. No. 108-46, at 330 (2003). [9] GAO, Defense Health Care: Access to Care for Beneficiaries Who Have Not Enrolled in TRICARE's Managed Care Option, [hyperlink, http://www.gao.gov/products/GAO-07-48] (Washington, D.C.: Dec. 22, 2006). [10] See Pub. L. No. 110-181, § 711(a), 122 Stat. 3, 190-91, and S. Rep. No. 110-77, at 359-60 (2007). [11] We have previously issued three reports that address the issues covered in this mandate. See GAO, Defense Health Care: DOD Lacks Assurance That Selected Reserve Members Are Informed about TRICARE Reserve Select, [hyperlink, http://www.gao.gov/products/GAO-11-551] (Washington, D.C.: June 3, 2011); Defense Health Care: Access to Civilian Providers under TRICARE Standard and Extra, [hyperlink, http://www.gao.gov/products/GAO-11-500] (Washington, D.C.: June 2, 2011); and Defense Health Care: 2008 Access to Care Surveys Indicate Some Problems, but Beneficiary Satisfaction Is Similar to Other Health Plans, [hyperlink, http://www.gao.gov/products/GAO-10-402] (Washington, D.C.: Mar. 31, 2010). [12] See Pub. L. No. 112-81, § 721(a), 125 Stat. 1298, 1479 (2011). [13] We use the term "civilian primary care" to refer to instances where respondents indicated that their personal doctor or nurse was a civilian. [14] We use the term "civilian specialty care" to refer to instances where respondents indicated that they had seen a civilian specialist within the last year. [15] We use the term "civilian mental health care" to refer to instances where respondents indicated that they had received treatment or counseling for a personal or family problem from a civilian provider within the last year. [16] For the 4 years of surveys, TMA mailed 176,841 surveys and received 66,590 returned surveys that were complete and eligible responses. Complete and eligible responses included those TRICARE beneficiaries who answered at least half of the TMA-identified "key" questions. [17] A nonresponse analysis is used to verify that nonrespondents to the survey would not answer differently from those who did respond and that the respondents are representative of the target population, thus ensuring that the results can be generalized to the population from which the sample was chosen. TMA concluded that the results of the beneficiary survey nonresponse analyses suggested that although there were some differences in the demographic profile, they were not associated with systematic differences in satisfaction with care. TMA officials also told us that the final postsurvey weights used in their analysis accounted for the key-characteristic differences in survey respondents compared with nonrespondents identified through the nonresponse analyses. [18] HHS's CAHPS survey is a national survey of beneficiaries of commercial health insurance, Medicare, Medicaid, and the Children's Health Insurance Program. We limited our CAHPS analysis to Medicare fee-for-service, Medicaid, and commercial CAHPS surveys, and pooled the data for each from 2008 through 2011 in order to compare the results to TMA's 4-year beneficiary surveys over the same period. We did not adjust the CAHPS survey data for factors that could affect the various beneficiary groups' ratings, such as age or health status. [19] For the 4 years of surveys, TMA mailed 194,774 surveys and received 82,111 returned surveys that were complete. A survey was considered complete if the provider answered three TMA-identified "key" questions that asked about the providers' location of practice and awareness and acceptance of TRICARE. [20] From the results of the civilian provider survey nonresponse analyses, TMA concluded that although there were some demographic and response differences between respondents and nonrespondents, the differences were not large or systematic. TMA officials also told us that the final postsurvey weights used in their analysis accounted for the key-characteristic differences in survey respondents compared with nonrespondents identified through the nonresponse analyses. [21] The Center for Studying Health System Change is a nonpartisan health policy research organization that conducts research and analysis focused on the U.S. health care system to inform the thinking and decisions of policymakers in government and private industry. The 2008 Health Tracking Physician Survey covered a wide variety of physician and practice dimensions, from basic physician demographic information, practice organization, and career satisfaction, to insurance acceptance, compensation arrangements, information- technology use, and charity care provision. [22] TMA's 2005-2007 civilian physician survey was sent to physicians only and did not include nonphysician mental health providers. Therefore, when comparing to TMA's 2005-2007 civilian physician survey, we show the results of TMA's 2008-2011 civilian provider survey results for civilian physicians only, which consist of civilian primary care and specialty care physicians, including psychiatrists. [23] TMA officials estimate that the shift from TRICARE Prime to TRICARE Standard will increase the out-of-pocket costs of a retiree family of three, for example, by about $700 per year. [24] Although TMA can identify which beneficiaries have not enrolled, it does not have complete information on which beneficiaries intend to use their benefits. [25] In designing the beneficiary and civilian provider surveys, DOD defined 80 distinct PSAs and 80 distinct non-PSAs (representing the entire country), and surveyed 20 of each in fiscal years 2008 through 2011. This allowed DOD to survey the entire country over a 4-year period. At the end of the 4-year period, each year's survey results were combined and weighted to develop estimates of access to health care and mental health care at individual service area, regional, and national levels. [26] TMA identified HSAs to include in its survey sampling locations on the basis of the recommendations of groups representing TRICARE beneficiaries and civilian providers, which identified specific cities and towns in which these groups were aware of beneficiaries having problems accessing civilian TRICARE providers. HSAs, as defined by a Dartmouth College study, are collections of zip codes organized into over 3,000 geographic regions in which Medicare beneficiaries seek the majority of their care from one hospital or a collection of hospitals, and have nonoverlapping borders and contain all U.S. zip codes without gaps in coverage. The HSAs surveyed in the beneficiary and civilian provider surveys are within the 80 PSAs or 80 non-PSAs surveyed. [27] Of the 71 HSAs, all were included for the civilian provider survey, but only 55 HSAs were included for the beneficiary survey. According to TMA officials, the 16 HSAs that were included in the 2011 civilian provider survey were not included in the 2011 beneficiary survey due to funding issues. [28] We previously reported that, according to a TMA official responsible for implementing the surveys, TMA did not give a high priority to areas where higher concentrations of Selected Reserve servicemembers live because it decided to randomly select the areas to be surveyed in order to produce results that could be generalized to the populations from which the survey samples were selected. Since TMA planned to survey the entire United States over the 4-year period, its 4-year survey results would include any locations with a higher concentration of Selected Reserve servicemembers. See [hyperlink, http://www.gao.gov/products/GAO-10-402]. [29] The margins of error for the estimates of beneficiary problems finding civilian primary, specialty, and mental health care providers at the 95 percent confidence level are plus or minus 1, 1, and 3 percentage points, respectively. These estimates are not significantly different from each other at the 95 percent confidence level. [30] Each surveyed HSA was part of a PSA or non-PSA (depending on the location), and because HSAs were not mutually exclusive of the PSAs or non-PSAs, we did not compare the results from nonenrolled beneficiaries in HSAs to nonenrolled beneficiaries in PSAs or non- PSAs. Instead, we compared the results for the nonenrolled beneficiaries in the surveyed HSAs to those nonenrolled beneficiaries in the areas outside the surveyed HSAs. [31] In our comparison across location types for all of the satisfaction measures in our analysis, there was one statistical difference at the 95 percent confidence level for nonenrolled beneficiaries' 8-10 ratings of their health care in PSAs (about 79 percent) compared to those in non-PSAs (about 82 percent). Additionally, there was one statistical difference at the 95 percent confidence level for nonenrolled beneficiaries' 8-10 ratings of their health plan in the surveyed HSAs (about 63 percent) compared to those in the areas outside of the surveyed HSAs (about 66 percent). However, for the purposes of our analyses, we determined that although these were statistical differences, they were not substantial differences. [32] On the scale of 0 to 10, 0 is the worst possible and 10 is the best possible. [33] We divided the rating scale into two categories on the basis of the ratings scale used by TMA to analyze the satisfaction measures for TRICARE beneficiaries (0 to 7 and 8 to 10), where 0 is considered the worst possible and 10 is the best possible. The CAHPS commercial survey asks beneficiaries about their experiences over the last 12 months, whereas the Medicare and Medicaid surveys ask about the beneficiaries' experiences over the last 6 months. [34] We found similar results in our analysis of the first year of TMA's 2008-2011 survey data and 2008 CAHPS data for Medicare fee-for- service and commercially insured beneficiaries. Specifically, in March 2010, we reported that, although there were no statistically significant differences in the estimated ratings for nonenrolled TRICARE beneficiaries and other beneficiary types, the estimated ratings for nonenrolled beneficiaries in surveyed areas (using categories of 0-6 and 7-10) were slightly lower than estimated ratings of Medicare fee-for-service beneficiaries across three of the satisfaction measures--primary care provider, specialty care provider, and health plan. See [hyperlink, http://www.gao.gov/products/GAO-10-402]. [35] The margins of error for civilian providers' awareness of TRICARE and acceptance of new TRICARE patients are both within plus or minus 1 percentage point at the 95 percent confidence level. [36] 2008 HSC Health Tracking Physician Survey, Center for Studying Health System Change. The survey results were based on a 2008 national survey of 4,720 physicians. The margins of error for physicians' acceptance of new commercially insured beneficiaries, new Medicare beneficiaries (fee-for-service and managed care beneficiaries), and new Medicaid beneficiaries are all plus or minus 1 percentage point at the 95 percent confidence level. The differences in estimates between civilian providers' acceptance of new TRICARE patients and providers' acceptance of new commercially insured, Medicare, and Medicaid beneficiaries are significant at the 95 percent confidence level. [37] TMA's 2005-2007 civilian physician survey was sent to physicians only and did not include nonphysician mental health providers. Therefore, when comparing to TMA's 2005-2007 civilian physician survey, we show the results of TMA's 2008-2011 civilian provider survey for civilian physicians only, which consist of civilian primary care and specialty care physicians, including psychiatrists. [38] In accordance with the NDAA 2008, TMA identified benchmarks for analyzing the results of the beneficiary and provider surveys. To benchmark its provider survey, TMA compared the results of its 2008- 2011 surveys with the results of its 2005, 2006, and 2007 physician surveys. A TMA official noted that TMA was unaware of any external benchmarks that would be applicable to its surveys of providers. [39] The margins of error for civilian physicians' acceptance of any new TRICARE patients from the 2008-2011 surveys and the 2005-2007 surveys are both within plus or minus 1 percentage point at the 95 percent confidence level. [40] The margins of error for civilian mental health care, primary care, and specialty care providers' acceptance of new TRICARE patients were each within plus or minus 1 percentage point at the 95 percent confidence level. For acceptance of new TRICARE beneficiaries, the differences in estimates between provider types are significant at the 95 percent confidence level. [41] The margins of error for civilian providers' awareness of TRICARE in PSAs and non-PSAs are both within plus or minus 1 percentage point at the 95 percent confidence level. The differences in estimates are significant at the 95 percent confidence level. [42] The margins of error for civilian providers' acceptance of new TRICARE patients in PSAs and non-PSAs are both within plus or minus 1 percentage point at the 95 percent confidence level. The differences in estimates are significant at the 95 percent confidence level. [43] Each HSA is part of a PSA or non-PSA (depending on the location), and because HSAs are not mutually exclusive of the PSAs or non-PSAs, we did not compare the results from civilian providers in HSAs to civilian providers in PSAs or non-PSAs. Instead, we compared the results for the civilian providers in the surveyed HSAs to those civilian providers in the areas outside of HSAs. [44] We used the individual area's estimate and margin of error at the 95 percent confidence level to determine whether it was above or below the national estimates. Specifically, for nonenrolled beneficiary problems, we used the lower confidence limit of the estimate: If the individual area's lower confidence limit was equal to or greater than the national estimate, then we included it as an area. Additionally, for civilian providers' acceptance of TRICARE, we used the upper confidence limit of the estimate: If the upper limit of the estimate was equal to or less than the national estimate, then we included it as an area. [45] A particular geographic area's exclusion from the lists of problem areas below does not necessarily indicate that nonenrolled beneficiaries were not experiencing access problems in that area. Because we took a conservative methodological approach and used the margins of error at the 95 percent confidence limit to determine whether a geographic area met our criteria of a problem area, there may be other areas where nonenrolled beneficiaries are experiencing access problems. [46] For the 2008-2011 beneficiary survey, 80 PSAs, 80 non-PSAs, and 55 HSAs were surveyed. Because the beneficiary survey did not include the 16 HSAs selected to be surveyed in 2011, they are not included in this analysis. However, the 2011 civilian provider survey did include these 16 HSAs. See appendix V to see a list of these 16 HSAs and civilian providers' acceptance of any new TRICARE patients in these areas. [47] Austin, Texas, HSA is part of the Eastern-Central Texas PSA, and the Dallas/Ft. Worth, Texas, HSA is part of the Northeastern Texas PSA. [48] We do not present the estimates for the percentage of civilian primary care providers in the two HSAs that were accepting any new TRICARE patients, if they were accepting new Medicare patients, because the number of responses was below 50. [49] One of the nine areas, the Alaska non-PSA, had less than 50 civilian specialty care provider respondents to the question that asked about acceptance of any new TRICARE patients. Therefore, its estimate is not included in this range. [50] In order for nonenrolled beneficiaries to respond to the question that asked "in the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan?," they needed to have answered "yes" to the question that asked "in the last 12 months, did you need any treatment or counseling for a personal or family problem?" Additionally, nonenrolled beneficiaries had to have responded that their mental health care provider was a civilian provider. [51] TRICARE Prime is an option that includes the use of civilian provider networks and requires enrollment. TRICARE beneficiaries who do not enroll in this option may obtain care from non-network providers through TRICARE Standard, or from network providers through TRICARE Extra. We included TRS beneficiaries in our definition of nonenrolled beneficiaries because, although they must enroll in the plan, they can receive care from non-network or network providers similar to TRICARE Standard and Extra beneficiaries. We did not include TRICARE Young Adult-Standard Option beneficiaries in our analysis because this plan did not become available until May 2011. [52] PSAs are geographic areas determined by the Assistant Secretary of Defense (Health Affairs) that are defined by a set of five-digit zip codes, usually within an approximate 40 mile radius of a military treatment facility. The managed care support contracts require the contractor to develop civilian provider networks at all Base Realignment and Closure (BRAC) sites, which are military installations that have been closed or realigned as a result of decisions made by the Commission on Base Realignment and Closure. [53] The Hospital Referral Region designation is derived from a Dartmouth College study that groups HSAs into distinct sets by documenting where patients were referred for major cardiovascular surgical procedures and for neurosurgery. Each HSA was examined to determine where most of its residents went for these services. The result was the aggregation of the more than 3,000 HSAs into 306 Hospital Referral Regions. A TMA official noted that TMA endorsed the Hospital Referral Region methodology in part because it is based on the medical observations of all Medicare beneficiaries, and TRICARE reimbursement rates are based on Medicare reimbursement rates. In addition, TMA used this methodology in its survey of civilian providers during fiscal years 2005 through 2007. In 2006, we reviewed the methodology TMA used for the 2005 civilian provider survey. GAO, Defense Health Care: Access to Care for Beneficiaries Who Have Not Enrolled in TRICARE's Managed Care Option, [hyperlink, http://www.gao.gov/products/GAO-07-48] (Washington, D.C.: Dec. 22, 2006). [54] HSAs are collections of zip codes organized into over 3,000 geographic regions in which Medicare beneficiaries seek the majority of their care from one hospital or a collection of hospitals. HSAs have nonoverlapping borders and contain all U.S. zip codes without gaps in coverage. [55] Because of timing issues, the 21 HSAs were not identified in time to be included with the 2008 fielding of the nonenrolled beneficiary survey. Therefore, TMA surveyed these 21 HSAs in the 2009 fielding of the nonenrolled beneficiary survey, along with the 9 HSAs scheduled to be surveyed during the 2009 fielding. Although the 21 HSAs were not actually surveyed during the 2008 fielding, TMA included them when it presented the results of the 2008 nonenrolled beneficiary survey. The civilian provider survey was not affected by these issues. [56] Of the 71 HSAs, all were included for the civilian provider survey, but only 55 HSAs were included for the beneficiary survey. According to TMA officials, the 16 HSAs that were included in the 2011 civilian provider survey were not included in the 2011 beneficiary survey because of funding issues. [57] See GAO, Defense Health Care: 2008 Access to Care Surveys Indicate Some Problems, but Beneficiary Satisfaction Is Similar to Other Health Plans, [hyperlink, http://www.gao.gov/products/GAO-10-402] (Washington, D.C.: Mar. 31, 2010). [58] DEERS is a database that contains the service-related and demographic data that are used to determine eligibility for military benefits, including health care, for all active duty servicemembers, military retirees, and the dependents and survivors of active duty servicemembers and military retirees. As individuals join the military, the various agencies enter information about them into DEERS and update this information as an individual's status changes. The individual servicemember is responsible for providing information to DEERS on dependents, and for reporting changes concerning dependents. [59] TMA's sample included retirees not enrolled in Medicare, dependents of active duty personnel, and beneficiaries enrolled in TRS in fiscal year 2008. [60] TMA desired a sample error of plus or minus 5 percent at the 95 percent confidence level. [61] This reduction was achieved by eliminating the HSAs from the 2011 nonenrolled beneficiary survey area selection. [62] According to a TMA official, using TRICARE claims data would help to increase the proportion of TRICARE users to those that used other health insurance. [63] TMA did not include all physician specialist types, such as epidemiologists and pathologists, in its survey. [64] TMA desired a sample error of plus or minus 5 percent at a 95 percent confidence level. [65] The Paperwork Reduction Act requires that all federal agency activities that involve collecting information from the public involving 10 or more people be approved by OMB to ensure that collection of this information will have a minimum burden on the public. See 44 U.S.C. §§ 3507 and 3508. [66] OMB's guidance suggests that if response rates are below 80 percent, agencies should conduct a nonresponse analysis. Such an analysis is used to verify that nonrespondents to the survey would not answer differently from those who did respond and that the respondents are representative of the target population, thus ensuring that the results can be generalized to the population from which the sample was chosen. [67] We include TRS beneficiaries in our definition of nonenrolled beneficiaries because, although they must enroll in the plan, they can receive care from network or non-network providers similarly to TRICARE Standard and Extra beneficiaries. We did not include TRICARE Young Adult-Standard Option beneficiaries in our analysis because this plan did not become available until May 2011. [68] We include TRS beneficiaries in our definition of nonenrolled beneficiaries because, although they must enroll in the plan, they can receive care from network or non-network providers similarly to TRICARE Standard and Extra beneficiaries. We did not include TRICARE Young Adult-Standard Option beneficiaries in our analysis because this plan did not become available until May 2011. [69] Nonphysician mental health providers include: (1) certified marriage and family therapists, (2) mental health counselors, (3) pastoral counselors, (4) certified psychiatric nurse specialists, (5) clinical psychologists, and (6) certified clinical social workers. [70] "Any civilian provider" means the nonenrolled beneficiary had problems finding a civilian primary, specialty, or mental health care provider who would accept TRICARE patients. [71] For the beneficiary survey, 80 Prime Service Areas (PSA), 80 non- Prime Service Areas (non-PSA), and 55 Hospital Service Areas (HSA) were surveyed. Because the beneficiary survey did not include the 16 HSAs selected to be surveyed in 2011, we cannot include them in this analysis. However, the 2011 civilian provider survey did include these 16 HSAs. See appendix V to see a list of these 16 HSAs and civilian providers' acceptance of any new TRICARE patients in these areas. [72] An estimated 31 percent of nonenrolled beneficiaries experienced problems finding any civilian provider nationally (i.e., a civilian primary, specialty, or mental health care provider). To determine whether an area had at least 31 percent of nonenrolled beneficiaries who experienced problems finding any type of civilian provider who would accept TRICARE, we used the margins of error at the 95 percent confidence level to determine the lower limit of the estimate. If the lower limit was 31 percent or above, then we included it as an area. [73] To determine whether an area had less than 31 percent of nonenrolled beneficiaries who experienced problems finding any type of civilian provider who would accept TRICARE, we used the margins of error at the 95 percent confidence level to determine the upper limit of the estimate. If the upper limit was below 31 percent, then we included it as an area. [74] Twelve areas (all HSAs) were not included because they had less than 30 respondents. [75] Nationwide, the estimated percentage of nonenrolled beneficiaries who experienced problems finding a civilian primary care provider was 25 percent. To determine whether an area had 25 percent or more of nonenrolled beneficiaries who experienced problems finding a provider who would accept TRICARE, we used the margins of error at the 95 percent confidence level to determine the lower limit of the estimate. If the lower limit was 25 percent or above, then we included it as an area. [76] To determine whether an area had fewer than 25 percent of nonenrolled beneficiaries who experienced problems finding a provider who would accept TRICARE, we used the margins of error at the 95 percent confidence level to determine the upper limit of the estimate. If the upper limit was below 25 percent, then we included it as an area. [77] Fifteen areas (1 PSA and 14 HSAs) were not included because they had less than 30 respondents. [78] Nationwide, the estimated percentages of nonenrolled beneficiaries who experienced problems finding a civilian specialty care provider was 25 percent. To determine whether an area had 25 percent or more of nonenrolled beneficiaries who experienced problems finding a civilian specialty care provider who would accept TRICARE, we used the margins of error at the 95 percent confidence level to determine the lower limit of the estimate. If the lower limit was 25 percent or above, then we included it as an area. [79] To determine whether an area had fewer than 25 percent of nonenrolled beneficiaries who experienced problems finding a civilian specialty care provider who would accept TRICARE, we used the margins of error at the 95 percent confidence level to determine the upper limit of the estimate. If the upper limit was below 25 percent, then we included it as an area. [80] Twenty-eight areas (2 PSAs, 2 non-PSAs, and 24 HSAs) were not included because they had less than 30 respondents. [81] In order for nonenrolled beneficiaries to respond to the question that asked "In the last 12 months, how much of a problem, if any, was it to get the treatment or counseling you needed through your health plan?," they needed to have answered "yes" to the question that asked "In the last 12 months, did you need any treatment or counseling for a personal or family problem?" Additionally, nonenrolled beneficiaries had to have responded that their mental health care provider was a civilian provider. [End of section] 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 website [hyperlink, http://www.gao.gov]. Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to [hyperlink, http://www.gao.gov] and select “E-mail Updates.” Order by Phone: The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, [hyperlink, http://www.gao.gov/ordering.htm]. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. Connect with GAO: Connect with GAO on facebook, flickr, twitter, and YouTube. Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts. Visit GAO on the web at [hyperlink, http://www.gao.gov]. To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; E-mail: fraudnet@gao.gov; Automated answering system: (800) 424-5454 or (202) 512-7470. Congressional Relations: Katherine Siggerud, Managing Director, siggerudk@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, DC 20548. Public Affairs: Chuck Young, Managing Director, youngc1@gao.gov: (202) 512-4800: U.S. Government Accountability Office: 441 G Street NW, Room 7149: Washington, DC 20548. [End of document]