Supporting Statement

SUPPORTING STATEMENT FOR PAPERWORK REDUCTION ACT SUBMISSION

0720-0031

1. JUSTIFICATION

1. Circumstances: Need and Use

This submission modifies the request and justification for the survey approved under OMB control number 0720-0031. The reason for this modification is because of congressional requirements in the National Defense Authorization Act (NDAA) for Fiscal Year (FY) 2008, Section 711 expanding and extending the previous survey methodology and sample requirements. Section 723 of the FY04 NDAA required the Department of Defense to survey civilian physicians. Section 711 of the FY06 NDAA required additional questions. An extract of Section 711 of the FY08 NDAA is provided, reflecting the continuation and extension of survey requirements. This submission requests approval of an expanded survey questionnaire (and supporting telephone script), with no expected increase in the overall approved burden hours because we will use the same approved sample size of 40,000 and devote a portion of it to non-physician health care providers and the few additional questions will not materially effect the survey estimate of 5 minutes.

In January 2008, the FY08 NDAA extended and expanded the scope of the survey requirement to :

a) continue provider surveys until 2011;

b) expand the target population to include mental health providers;

c) include surveys of beneficiaries in addition to surveys of health care and mental health care providers; and

d) modify the geographic stratification to ensure geographic areas where the TRICARE Prime benefit option is offered, as well as where it is not. TRICARE Prime Service Areas (PSAs) are those geographic areas where the TRICARE managed care support contractors (MCSCs) offer the TRICARE Prime benefit through established networks of providers. TRICARE Prime is available at military treatment facilities (MTFs), in areas around most MTFs, (called “MTF PSAs”) in a number of areas where an MTF was eliminated in the Base Realignment and Closure process (“BRAC PSAs”), and in some other areas where the MCSCs proposed in their contract bids to offer the benefit (“non-catchment PSAs”).

Also, the new legislation requires the Department to establish benchmarks for primary and specialty care providers (including mental health providers) to determine adequacy of providers available to TRICARE-eligible beneficiaries. The Government Accountability Office (GAO) continues to be required for review of the processes, procedures and analyses used by DoD to determine the adequacy of the number of health care and mental health care providers available to TRICARE-eligible beneficiaries.

Surveys of civilian healthcare providers (M.D.s and D.O.s) were conducted in TRICARE market areas to determine how many healthcare providers were accepting new patients under TRICARE Standard in each market area. This modification request seeks to expand the questionnaire, target population, and geographic areas starting in FY08 and extending through FY 2011, without changing the overall annual estimated burden hours.

2. Purpose and Users of the Information

The revised survey will gather data on providers (physicians and mental health providers) to asses the extent to which they are aware of the overall TRICARE program, accept new TRICARE Standard patients specifically, and the extent to which these physicians accept Medicare patients. The information gathered through this project will be used to generate reports to address the legislative requirements specified in Section 711 of the FY08 NDAA. Information resulting from the collection efforts of this project will assist DoD in developing policies and initiatives to improve TRICARE beneficiary access to civilian providers. The results of the previous survey efforts have been briefed to, or provided in written communication to Health Affairs, TMA and senior DoD personnel, TRICARE Regional Office Directors and their staff, members of Congress, selected state leaders (such as governors to solicit support) and selected medical societies, staff members of the Government Accountability Office, TRICARE Beneficiary Groups, at MHS Conferences and have been referred to in public media such as the Military Officers Association of America. The level of detail provided has never been at the individual provider level, but instead reflected the percentage of respondents aware of TRICARE, accepting new TRICARE standard patients, or aggregate numbers and percentages by medical specialties (e.g. Surgeons), or categories of comments (e.g. “reimbursement-related issues).

3. Information Collection Techniques

A multi-mode data collection method will be used through a mailed survey with internet option and a telephone follow-up survey. Because this new guiding legislation extends the survey requirement to non-physician offices which might be small practices, and without facsimile options, the data collection will also offer an internet survey option using assigned passwords and identification numbers from the survey notification letter (not previously provided). The Office of the Assistant Secretary of Defense (Health Affairs), TRICARE Management Activity (TMA) will continue to contract with a private-sector vendor to conduct the mail and telephone follow-up survey. An initial mailed survey will be sent to members of the target population within specified geographic areas, with a follow-up mail survey sent within a defined period after the first. The initial and follow-up survey includes a cover letter signed by a senior TMA director requesting the recipient’s participation and requesting a response by return mail or facsimile, as well as providing a toll-free number to call with any questions and a web address to take the survey via the internet. A scripted telephone follow-up of those from whom a survey has not been returned will be accomplished as well. The telephone survey will use a standardized Computer Assisted Telephone Interview (CATI) protocol.

Mailed surveys will be sent to the provider’s stated work address only, and not the residence, to the extent the work address is different from the home address and can be discerned. Telephone follow-up will be to the work address as well, and, similarly, to the extent the work telephone is different from the home address and can be discerned. These surveys are designed to be answered by the billing manager or person responsible for the provider’s billing practices, to minimize the burden on the provider’s practice, and to obtain data the billing expert may be most knowledgeable about. If a recipient receives multiple surveys for multiple providers in the same office or practice group, the recipient is asked to complete a separate mail survey or answer to a separate scripted telephone survey for each provider. The revised survey will ask if the recipient is other than the provider.

TMA does not anticipate deviating from the multi-mode survey methodology, approved by OMB May 16, 2005, and employed in the FY05, FY06 and FY07 survey efforts, other than possibly extending an additional internet option to the recipients. TMA’s six-site telephone mode pilot effort in FY04 yielded data from 25.9% of the target population, and the dual-mode second pilot effort of fourteen sites in FY04 resulted in a 49.5 % response rate, while the FY05, FY 2006, and FY 2007 surveys each resulted in an overall adjusted response rate of about 50 percent, each year, using the American Association for Public Opinion Research (AAPOR) methodology (AAPOR, Table 3). At this point, we do not anticipate the overall response rate to be any different, but wish to keep an internet option available.

4. Avoiding Duplication

There is no duplication of the data collection effort. No other DoD survey addresses these issues. This effort replaces/continues the previously approved survey, and, as such, will not duplicate its efforts.

5. Impact on Small Business

This collection of information may involve small business or other small entities, as we understand many non-physician practices are family run (e.g. social works, counselors, and psychologists), however, the Congressional legislation clearly requires surveys of non-physician, mental health providers.

6. Consequences to Federal program if collection is not conducted or conducted less frequently than requested

The proposed methodology responds directly to the requirements levied by the annual survey as directed by Congress. It is no more frequent than the minimum directed, and while the requirements have been extended and expanded, seeks to avoid any increase to the burden previously approved in support of earlier legislation. The survey methodology is designed to sample providers from a mixture of at least 20 TRICARE Prime Service Areas (PSA) each year and 20 geographic areas that do not offer the TRICARE Prime benefit., As in the previous three-year study geographic areas to be surveyed will be both purposively and randomly selected, with geographic areas defined by Hospital Service Areas (HSAs). TRICARE PSAs are geographic areas encompassing complete zip codes in which TRICARE has contracted with selected health care providers and institutions, but which do not necessarily mimic any other standard geographic area such as a county, HSA, SMA, voting block, etc. In addition, the design permits us to calculate regional and national estimates of access measures. The intent of this methodology is to avoid surveying all physicians in all market areas by capitalizing on randomized selection of location and provider where possible, and surveying those selected providers only once.

7. Special Circumstances

There are no special circumstances anticipated in the collection of these data where respondents would be required to submit more than the annual frequency, to prepare a written response other than in response to the actual survey question.

8. Federal Register Notice/Consultations

The Federal Register Notice for this collection of information was published November 29, 2007 (72 FR 67595). No comments have been received. The survey questions have been presented to MHS beneficiary representatives and to the government representative of the American Medical Association.

9. Payment/Gift to Respondents: None.

10. Confidentiality

Civilian providers selected for the sample receive complete protection, except as required by law. Respondent names will not be released when data and analytic results are reported. This will also be conveyed in the telephone script and the internet-based instrument. The relevant statutory authority for protection of identifiable data within DoD and the Federal Government is the Privacy Act of 1974. This survey does not request any Protected Health Information (as defined by HIPAA), and the data being collected similarly are not patient or beneficiary-specific. Further, the sample frame source is from publicly available provider data sources.

11. Sensitive Questions

There are no questions of a sensitive nature on the survey instrument.

12. Burden Estimated (hours)

This modification holds the burden at the current rate. 40,000 providers (including mental health) x 100% response rate (worse case basis for calculating burden) x 5 minutes = 3,333 hours.

The estimated average cost of this burden to providers for billing managers to answer the survey is $ 1.67. This was based on an estimate of 5 minutes per completed survey (.0833 of an hour) and a conservative estimate of the average billing manager wage of $20 per hour (OPM GS 7, Step 5, middle of the row is $17.67 Base, and $19.99 with locality adjustment (for “rest of the United States” other than cities cited). A Google search also reflects the median hurly rate by years experience for billing mangers in the United States ranges from $16.84/hour for those with 10-19 years experience to $17.52/hour for those with 20 years or more (http://www.payscale.com/research/US/Job=Billing_Manager/Hourly_Rate, updated 4/17/2008; web site accessed 4/18/2008).

13. Cost to Respondents

There are no start-up, O&M, or capital costs to respondents.

14. Cost to Federal Government

The total estimated cost to the Department of Defense for the effort is $555,000 for contract support to be awarded for the purchase of the physician and mental health provider databases, development of sample frame and provider locater data, including address updates and corrections, fielding the mail and telephone survey, coding the internet survey option, and subsequent collection and analysis of the respondent data and weighting of results including adjustments for non-response.

15. Change in Burden

There is no change in burden as a result of the new language/modification.

16. Publication/Tabulation

Information gathered through this project will be aggregated and analyzed for the purpose of addressing the questions identified in previous legislation (Section 723 of the FY04 NDAA, and Section 711 of FY06 NDAA) and current guiding legislation (FY08 NDAA Section 711). The FY08 survey effort began with public notification in the Federal Register and data gathering will formally commence only with OMB approval. Survey mail fielding is expected to begin within days of OMB approval, with a desired completion of the first year’s survey fielding prior to September 30, 2008, and annually by September 30 of each year thereafter through 2011. After respondent data is collated and quality controlled, it will be analyzed during the remainder of the calendar year, with results presented from the end of the calendar year through spring of the following (i.e. Dec 31 2008 through March 30, 2009). The annual schedule, to include survey fielding, data consolidation and de-duplication, analysis and reporting will replicate the approximate dates above: survey fielding between Jun-Sep; data cleansing and analysis between Sep-Nov; reporting and presentation from Dec-Mar of the following year.

17. Expiration Date

No exception is sought for displaying the expiration date of the data collected through this project.

18. Exceptions

There are no exceptions to the certification statement in Item 19 of OMB Form 83-I.

2. COLLECTION OF INFORMATION EMPLOYING STATISTICAL METHODS

1. Background: Congress has mandated that the Secretary of Defense conduct a survey of civilian physician and mental health providers in at least 20 areas where the Prime option is offered, and 20 different areas where it is not to determine the extent of access problems created by reluctance of civilian providers to see TRICARE Standard and Extra patients. The Congress left to DoD to identify the Prime and non-prime areas. The TRICARE Management Activity has identified the locations where health care support contractors have formally developed networks of providers contracted to provide the Prime Benefit, and permitted to extend their practice for the Extra benefit, called Prime Service Areas (PSAs).

2. Overall Design Strategy: The overall strategy for FY08 continues the effort of

the last three years, FY 2005- FY 2007, and is unchanged from that articulated in the submission leading to the May 16 2005 OMB approval. The overall strategy is to randomly sample 40,000 civilian providers (compared to 40,000 physicians previously, so fewer physicians will be sampled to extend the sample frame to non-physicians). If we obtain the desired response rate of 60 percent, this should yield approximately 24,000 provider responses each year, for a four-year total yield of approximately 106,000 responding providers by the end of the final survey year in 2011. Survey at least 20 PSA and 20 non-PSA health care markets each year; so that we:

(1) Ensure all beneficiary-identified PSAs and non-PSAs are given high priority within available resource constraints and the four-year survey period.

(2) Randomly sample MDs/Dos, and five types of non-physician mental health providers in each geographic area (PSAs and non-PSAs): psychologists, mental health counselors, pastoral counselors, social workers, and marriage or family counselors.

(3) Allow generalization of results to provide state-wide estimates, regional and national estimates, in addition to selected market level estimates.

(b) Survey sub-market areas to provide precise estimates at the local level of health care our beneficiaries most identify with.

3. Sample frame:

A. The physician population universe will be all licensed office-based civilian Medical Doctors (M.D.’s) or licensed civilian Doctors of Osteopathy (D.O.’s) within specified locations who are engaged in more than twenty hours of patient care per week. Physicians will be identified using the American Medical Association’s (AMA) physician master file. We have relied on the AMA Masterfile because it is widely recognized as one of the best commercially available lists of physicians in the U.S. containing over 600,000 active physicians along with locating information (addresses and telephone numbers), as well as key practice and physician characteristics useful for determining both eligibility for the study as well as for analytic purposes, including examination of non-response bias. The AMA states its Physician Masterfile includes current and historical data on all physicians, including AMA members and nonmembers, and graduates of foreign medical schools who reside in the United States and who have met the educational and credentialing requirements necessary for recognition as physicians (http://www.ama-assn.org/ama/pub/category/2673.html, as of Dec 12, 2004). The AMA updates physician addresses monthly and other elements through a rotating census methodology involving approximately one-third of the physician population each year.

1. However, although the Masterfile is considered to have a high level of coverage, there are still deficiencies in coverage and inaccuracies in detail. This may be due, in part, to the rotating database update methodology the AMA uses and may be more pronounced for physicians who have recently relocated. As such, we require the vendor to supplement as the AMA Masterfile with information from other sources such as state licensing data bases, local commercial lists, professional society and association lists. The vendor will also compare the AMA Masterfile with the National Change of Address (NCOA) and equivalent address databases to verify and update business addresses. We also require the vendor to establish a toll-free telephone number for respondents’ use, as well as provide a toll-free fax number for receiving written address corrections, status updates, re-mailing requests, or completed surveys during the survey fielding period.

2. Consistent with prior year analyses and AMA coding, for analysis, physicians will be identified by one of two practice types: Primary Care Physicians (PCPs) or Specialists (SPECs). PCPs consist of those physicians whose primary area of practice is Internal Medicine, Pediatrics, Obstetrics and Gynecology, or General Family Practice. Specialists (SPECs), will consist of all other areas of practice. This classification of PCP and SPEC will be used to classify the population and sample of physicians.

We will use the physician’s office practice zip codes taken from the AMA master file to identify the location of their practices. The zip codes have been classified into HSAs and hospital referral regions (HRRs) as specified by the Dartmouth Center for the Evaluative Clinical Sciences. HSAs are collections of zip codes in which resident Medicare beneficiaries seek the majority of their care from the same hospital or collection of hospitals. The HSA geographic units are considered because they were developed independent of TRICARE concerns, have been reported extensively in the scientific literature, were constructed via a scientific methodology, and were based on a population whose reimbursement rates provide the foundation for the TRICARE reimbursement rates (i.e. the TRICARE maximum allowable charges (TMAC).

B. For non-physician providers, we will rely on two sources for obtaining estimates of the availability of provider numbers for defining local geographic sample sizes, as well as for obtaining postal and telephone contact and provider name and prefix information. The first is the National Plan and Provider Enumeration System (NPPES) database maintained by CMS. The second is a list of names with contact information assembled from state licensing boards. These sources have been identified as the most comprehensive databases for these health care providers.

1. Psychologists:

2. Social Workers:

3. Mental Health Counselors:

4. Pastoral Counselors

5. Marriage and family therapists

Similar to the process for physician contact information, the vendor will also compare the selected non-physician sample files with the National Change of Address (NCOA) and equivalent address databases to verify and update business addresses, as well as offer a toll-free telephone number for respondents’ use, as well as provide a toll-free fax number for receiving written address corrections, status updates, re-mailing requests, or completed surveys during the survey fielding period.

4. Stratification and precision objectives:

1. The guiding legislation calls for two surveys: one of health care providers, and one of TRICARE beneficiaries eligible for TRICARE Standard. While this request focuses on the provider survey as members of the public, our analysis must necessarily identify the potential and relevant TRICARE beneficiary population requiring access to those providers. In addition, the physician survey must also account for the distribution of selected reserve who are eligible for TRICARE Reserve Select. As of the end of 2007, there are two million non-active duty TRICARE eligible beneficiaries under the age of 65 who are not enrolled in Prime (or other enrollment program such as the Designated Provider program), who reside in the United States and do not have an APO or FPO address. Another two million are members of the selected reserve or their eligible family members. These beneficiary counts, at the zip code level, and then aggregated to PSAs or Hospital Service Areas (HSAs) will identify health market areas affecting the largest numbers of TRICARE Standard or Extra beneficiaries and potential TRICARE Standard/Extra beneficiaries. In addition, health care market areas designated by beneficiary representatives will be incorporated into the study design. Sampling strata will consist of: physicians, mental health providers, PSA market areas and non-PSA market areas. In addition, mental health providers will be stratified according to whether they are Social Workers or not.

To create the PSA market areas, all PSAs will be divided into 80 sampling strata. Each stratum will contain a roughly equal number of eligible beneficiaries: non-Prime active duty family and retired, and selected reserve. Non-Prime market areas will be divided into 80 similar sampling strata. These strata are created by combining Hospital Referral Regions (HRRs). HRRs are groups of HSAs within which hospitals refer to a designated high tech hospital or group of high tech hospitals. As much as possible, these non-Prime market areas will contain equal numbers of eligible beneficiaries. The strata will be arranged randomly, and each list divided into four equal groups, that is 20 PSA strata and 20 non-Prime market areas in each group. We will survey providers in one group each year for four years. Eighty percent of the sample will be allocated to these groups: 800 providers per stratum, 400 each of physicians and mental health providers.

In addition beneficiary groups will suggest cities and towns where access should be measured. HSAs corresponding to each city and town will be identified. Based on those groups’ recommendations a list will be created, and sorted in priority order. Each HSA is allocated a sample of 800 or all of the providers in that market, whichever is less. The HSAs are compared to the group of strata for the current and previous survey years. When an HSA is contained in a market area selected in the first stage of sampling, providers that have already been sampled will be included in that HSA’s allocation. HSAs will be added to the sample in order of decreasing priority until the remaining sample is allocated. Those HSAs not sampled in the current year may be included in a future year’s list.

In each year, providers in the randomly selected markets (including the designated HSAs if they are contained in those markets, though with a lower sampling weight) contribute to national and regional estimates. HSAs from outside the markets randomly selected in the current year or previous year will be included in the higher level estimates when their markets are randomly selected, once again with a lower sampling weight. The attached contains precision predicted for local, national and regional estimates from each year.

Table 1 Allocation in Year 1

Stratum	Size	Design Effect*	Precision**
National MD Mental Health Provider	40,000 20,000 20,000	1.7 1.7 1.7	0.9 1.3 1.3
Regional MD Mental Health Provider	Min 10,668 Min 5,334 Min5,334	1.7 1.7 1.7	1.8 2.5 2.5
State MD Mental Health Provider	Min 1,600 Min 800 Min 800	1.3 1.3 1.3	2.9 4.0 4.0
PSA/nonPSA market MD Mental Health Provider	20,000 10,000 10,000	1.5 1.5 1.5	1.3 1.7 1.7
Market MD Mental Health Provider	800 400 400	1.0 1.0 1.0	4.9 6.9 6.9

*Estimated from 2005, 2006, and 2007 surveys of physicians.

**Precision levels, presented has a half-length confidence interval, assume a 50 percent response rate and the corresponding design effect.

Table 2 Allocation after 4 years

Stratum	Size	Design Effect*	Precision**
National MD Mental Health Provider	160,000 80,000 80,000	1.7 1.7 1.7	0.5 0.6 0.6
Regional MD Mental Health Provider	Min 42,672 Min 21,336 Min 21,336	1.7 1.7 1.7	0.9 1.2 1.2
State MD Mental Health Provider	Min 1,600 Min 800 Min 800	1.3 1.3 1.3	2.9 4.0 4.0
PSA/nonPSA market MD Mental Health Provider	80,000 40,000 40,000	1.5 1.5 1.5	0.6 0.9 0.9
Designated market MD Mental Health Provider	800 400 400	1.0 1.0 1.0	4.9 6.9 6.9

*Estimated from the combined weights of the 2005, 2006, and 2007 surveys of physicians.

**Precision levels, presented has a half-length confidence interval, assume a 50 percent response rate and the corresponding design effect.

5. Non-response analysis: To evaluate bias due to non-response, we will survey non-respondents and compare the characteristics and responses of these “non-responders” to similar characteristics of physicians and mental health providers who responded. We will specifically look at characteristics to include practice type in terms of primary care or specialty, across strata, and by TRICARE network affiliation status.

6. Procedures for Information Collection: All target population members within each market (state) and sub-market (HSA) will receive a survey with an accompanying cover letter that provides information about the survey effort, its purpose, and to encourage participation. An initial and, if necessary, one follow-up mailed survey will be sent to each target population member’s listed business address, and will be addressed care-of the Billing Manager for the target population member. Telephone surveying will begin after the conclusion of the mailed survey fielding period. Telephone contact will be attempted only to the target physician’s office; no contact will knowingly be attempted to a place of residence. Once contacted by telephone, respondents may either complete the interview or refuse participation. A maximum of ten telephone attempts will be made to contact those who have not completed a survey and who have not refused participation in the survey effort. At present, a 30-day field period is planned for the telephone survey portion of this project.

7. Methods for Maximizing Response Rate: The cover letter that accompanies each mailed survey is the primary method used to encourage participation in the survey effort. Both the cover letter and telephone script appeal to the respondent’s patriotism, and will include information about the purpose of the survey and a brief description of how the information will be used by TMA. For offices with multiple selected physicians and mental health providers, the billing manager recipient will receive separate surveys for each requested physician, and will be asked to complete one survey for each. In addition, telephone interviewers will be trained in interviewing techniques designed to minimize incidences of respondent refusals to participate in the survey. They too will ask respondents to answer separately for each physician in cases where multiple doctors are being surveyed in the same office. We do not anticipate changing the telephone surveying procedures and CATI protocols employed previously from FY 2005 to FY 2007.

8. Points of Contact (POCs)

1. TRICARE Management Activity (TMA), Health Program Analysis and Evaluation (HPA&E)

Richard R. Bannick, Ph. D.

Project Manager

(703) 681-3636 ext. 3638

OMB Submission: TRICARE Standard Survey-CP 13