Supporting Statement (A-B)

SUPPORTING STATEMENT

REQUEST FOR CLEARANCE:

MEDICARE CURRENT BENEFICIARY SURVEY

ROUNDS 48 THROUGH 56

Table of Contents

A. Background 2

B. Justification 3

B..1 Need and Administrative Basis 3

B..2 Information Users 10

B..3 Improved Information Technology 14

B..4 Duplication of Similar Information 15

B..5 Small Businesses 18

B..6 Cancellation or Less Frequent Collection 18

B..7 Special Circumstances 19

B..8 Federal Register Notice / Outside Consultation 19

B..9 Payments / Gifts to Respondents 20

B..10 Confidentiality 20

B..11 Sensitive Questions 21

B..12 Burden Estimate (Total Hours & Wages) 21

B..13 Capital Costs 22

B..14 Cost to the Federal Government 22

B..15 Program Changes 23

B..16 Publication and Tabulation Dates 23

B..17 Expiration Date 25

B..18 Certification Statement 25

A. Background

This is a request to extend the existing Medicare Current Beneficiary Survey (MCBS) clearance for an additional three years. This request for continuation proposes no changes in the core instrumentation. Clearances for the MCBS (OMB No. 0938-0568) were granted in October 1990 for the pilot, August 1991 for Round 1, January 1992 for Rounds 2-6, September 1993 for Rounds 7-10, November 1994 for Rounds 10-19, December 1997 for Rounds 20-28, and April 2001 for Rounds 29-37, March 2004 for Rounds 38-47. The present clearance request encompasses all aspects of the MCBS: the community baseline and core questionnaires, the facility screener, baseline and core questionnaires, and supplementary sections such as the income and assets supplement and the access to care supplement (i.e., usual sources of care, health status and functioning, satisfaction with care and the demographic questions).

The MCBS is an ongoing, multi‑purpose survey used by the Centers for Medicare and Medicaid Services (CMS), other operating divisions of the Department of Health and Human Services (such as the Assistant Secretary for Planning and Evaluation, the National Center for Health Statistics, and the Agency for Healthcare Research and Quality), and others concerned with Medicare policy, such as Congressional Budget Office, Congressional Research Service, and the Medicare Payment Advisory Commission. The core of the MCBS is a series of interviews with a stratified random sample of the Medicare population, including aged and disabled enrollees, residing in the community or in institutions. Questions are asked about enrollees' patterns of health care use, charges, insurance coverage, and payments over time. Respondents are asked about their sources of health care coverage and payment, their demographic characteristics, their health and work history, and their family living circumstances. In addition to collecting information through the core questionnaire, the MCBS collects information on special topics through supplements. For example, questions are asked about enrollees' income and assets, access to health care, health and functional status and satisfaction with care.

Special supplements also focus on emerging trends in health care. Each of these has been the subject of an OMB clearance action. Two such supplemental clearances are the Qualified Medicare Beneficiaries (granted December 1992) and the Managed Care (granted July 1996). The Facility CAPI Conversion (granted February 1997) computerized the interview process. The benefits of this conversion are thoroughly discussed in Section B..3.

The MCBS respondents are interviewed three times per year using four-month recall periods. Round 1 for the survey, which is administered to new survey sample members, collects baseline information for the analysis of access to care and establishes a recall boundary for the next interview (the interview reference period is since the date of the previous interview). Subsequent rounds of the survey, including our proposed extension (i.e., Rounds 48-56) collect cost and utilization information (core) and administer selected supplements on income and assets and access to care. Supplemental samples of newly eligible enrollees are added in the fall round each year. (A more detailed discussion of the structure of the MCBS is contained in Section C of this Statement.)

B. Justification

B..1 Need and Administrative Basis

B..1.a Legislative and Administrative basis

The legislative authority for the MCBS is located in Section 723 of the Medicare Prescription Drug, Improvement and Modernization Act (MMA) of 2003. The following is an excerpt from Section 723 (Chronically Ill Medicare Beneficiary Research, Data, Demonstration Strategy):

(a) Development of Plan.—Not later than 6 months after the date of the enactment of this Act, the Secretary shall develop a plan to improve quality of care and reduce the cost of care for chronically ill medicare beneficiaries.

(b) Plan Requirements.—The plan will utilize existing data and identify data gaps, develop research initiatives, and propose intervention demonstration programs to provide better health care for chronically ill medicare beneficiaries. The plan shall—

(1) integrate existing data sets including, the Medicare Current Beneficiary Survey (MCBS), Minimum Data Set (MDS), Outcome and Assessment Information Set (OASIS), data from Quality Improvement Organizations (QIO), and claims data;

The administrative authority for the MCBS is described in the Statement of Organization, Functions, and Delegations of Authority for the Centers for Medicare and Medicaid Services. Part of CMS, the Office of Research, Development, and Information (ORDI) is responsible for the following functions, identified in the Federal Register (Thursday, December 29, 2005, pages 77161 and 77162):

1. Provides analytic support and information to the Administrator and the Executive Council needed to establish Agency goals and directions.

2. Performs environmental scanning, identifying, evaluating, and reporting emerging trends in health care delivery and financing and their interactions with Agency programs.

3. Designs and conducts research and evaluations of health care programs, studying their impacts on beneficiaries, providers, plans, States and other partners and customers, designing and assessing potential improvements, and developing new measurement tools.

B..1.b Program and policy issues faced by CMS

The MCBS is in a unique position to monitor effects of the Medicare program on its beneficiaries and provide the basic information needed to estimate the benefits and costs of program changes and expansions. The MCBS is used as a program management tool to assess legislative proposals (e.g., proposed expansion of home health care). Once decisions on health care reforms are made, the MCBS is in position to monitor their effects upon the Medicare population (e.g., enacted prescription drug benefit).

In 2006 CMS Strategic Action Plan (Attachment I), which serves as a blueprint to help reach the CMS Vision: to achieve a transformed and modernized health care system. This plan spells out the CMS role in achieving HHS priorities in health information technology; Medicare prescription drug coverage; Medicaid modernization; transforming the New Orleans Health System; emergency response and Commissioned Corps renewal; prevention; pandemic preparedness; Science-driven opportunity for Management of personal health through Affordable, Reliable, and Targeted health care; and health transparency. Of the five key objectives articulated by CMS, the MCBS is instrumental in measuring and monitoring both, High-Value Health Care and Confident, Informed Consumers. Specifically, the MCBS is identified to assist policy makers and researchers in monitoring and evaluating the Medicare program and produce statistics and linked data files. This function is generalized, but not limited to the following tasks:

1. responses to MCBS questions on access to care are used to measure our beneficiaries’ ability to get the health care services in a timely manner in both fee‑for‑service and managed care settings;

2. responses to questions on satisfaction from the MCBS are used to measure the degree to which a beneficiary’s perception of care received meets or exceeds his or her expectation for care, in both fee‑for‑service and managed care settings;

3. responses to questions on utilization from the MCBS and the associated Medicare claims are used to track the percent of Medicare beneficiaries who receive preventative health services to include but not be limited to:

- an annual vaccination for influenza and a lifetime vaccination for pneumococcal;

- a screening or diagnostic mammogram within a 2‑year period; and

- diabetic eye exam for beneficiaries diagnosed with diabetes.

4. responses to questions on beneficiary needs measure the effectiveness of CMS’ dissemination of Medicare information to it’s beneficiary population; and

5. responses to questions on beneficiary knowledge measures the effectiveness of CMS’ initiative to increase beneficiary understanding of basic features of the Medicare program.

In addition to the performance measures included in CMS’ Strategic Action Plan, the Agency must be prepared to address proposed legislative changes to both the Medicare and Medicaid programs and to respond to changes in beneficiary and provider behavior. To do this, CMS needs to maintain comprehensive and timely information on the use and cost of health care for the Medicare population, coupled with beneficiary characteristics such as health status, physical functioning, age, race, sex, education, income, patterns of health insurance coverage, type of physician or specialist care, usual source or setting in which care is provided, and so on. Such information enables the Agency to administer the program more effectively, to analyze the impact of policy initiatives, to respond to legislative proposals and to project and assess the impact of program initiatives on expenditures and on Medicare enrollees.

B..1.c Shortcomings of existing data bases to address these issues

Other than the MCBS, there is no current and continuous source of data; which meets the information needs of both the Department and CMS policy makers and program actuaries. This is not an indictment of other existing surveys; each has a purpose of its own and has adapted itself to its purpose. Rather it is an argument for a Medicare specific data collection effort. Broadly speaking, other existing databases exhibit one (or more) of six deficiencies, which are discussed below (comparisons of the MCBS to specific data bases are made in a later section):

(i) Absence of financial information. To serve CMS information needs a database must include detailed financial information. Financial data are needed to analyze program costs as well as monetary burden on the beneficiaries (e.g., out of pocket expenditures). Databases such as the National Health Interview Survey measure only the use of health care and not the cost. The Medicare Health Outcomes Survey measures mental and physical health outcomes for the Medicare Advantage populations.

(ii) Lack of focus on Medicare population. This can be in terms of the population covered or the services covered. Most existing surveys exclude the institutional population. Yet that population is of most interest to Medicare actuaries to cost out possible program expansions. (The Medical Expenditure Panel Survey [MEPS] does include the institutional population, but does not primarily focus on the elderly and the disabled covered by Medicare.)

In addition, few surveys include a sizable number of Medicare respondents, making it difficult – if not impossible – to draw inferences about the experience of Medicare sub-populations of program-specific interest. Many national surveys combine the elderly into a single “over 65” group due to their limited sample of this cohort. This allows only limited analysis of the heterogeneous nature of the aging population. No other surveys provide information on the under – 65 Medicare disabled or on Medicare / Medicaid dual eligibles.

Another difference between MCBS and other surveys is our comprehensive measurement of health care expenditure. For example, the Consumer Expenditure Survey asks only about the out-of-pocket costs of households. As a result, little can be said about total health care spending, especially for older Americans. CMS has a system of Medicare administrative data, based on the claims payment files. These data are complete as far as they go, but administrative files do not contain information on sources of payment other than Medicare, nor on access to health care, nor use of and expenditure for services not covered by Medicare (e.g., long-term care, dental and vision care). The MCBS creates a complete account of coverage, use, and expenditures for the Medicare population.

During the 1980s, the absence of a continuous source of person‑level data constrained the ability of CMS’ Office of the Actuary to produce age‑specific estimates of national health expenditures requested by Congress. From 1967 through 1977, the Social Security Administration (predecessor to CMS in responsibility for the Medicare program) conducted a monthly survey of SMI enrollees. Since the end of that survey, the Department's other health expenditures surveys (the National Medical Care Expenditure Survey [NMCES], the National Medical Care Use and Expenditure Survey [NMCUES], NMES, and MEPS) were administered at broad intervals. As a result, CMS has had to rely on several different sources of information to estimate health expenditures by age. This lack of a consistent database, coupled with payment reforms between episodic surveys, creates fragmentary and indirect estimates (Waldo et al., 1989). The infrequency of health surveys also leaves gaps in the data collection process that impair policy-makers' ability to track and predict the impact of systemic changes in the U.S. health financing and delivery system.

(iii) Lack of timely information. Before the development of the MCBS, the Department’s health surveys often failed to produce timely information, producing data as much as five years after interviews were conducted. Successive surveys were conducted at intervals of as much as ten years. In the current context of a rapidly changing delivery system and an ever more rapidly developing Medicare program, this is not sufficient. We have seen improvements in current survey timeliness as a result of the example set by MCBS.

(iv) Lack of continuity. Recent methodological emphasis has been on approaches to develop coordinated analysis of surveys with heterogeneous samples and instrumentation (Corder and Manton, 1991). No survey (with the exception of the MCBS) is specifically targeted to collect health care data for the elderly using a continuous panel or series of continuous panel surveys. Although several surveys have over-sampled small population groups such as the elderly, this is costly. Over sampling by itself does not address the special problems in measuring health and functional status in the elderly. This requires a longitudinal panel design, a targeted sample and specially designed instruments to address the measurement requirements for analyses of the elderly and the oldest old populations (Corder and Manton, 1991).

Significant changes in the financing of Medicare are taking place on an ongoing basis. The National Research Council indicated that the lack of longitudinal data on the use of and expenditures for medical care would be a major impediment to understanding the impact of changes in the financing of medical care for individuals and their families. This is particularly true for increases in use of services as a person ages and as the risk increases of chronic diseases that require both acute and long-term care (Densen, 1991). CMS believes that the best way to monitor the ongoing effects of Medicare on the Medicare population is through the use of a longitudinal survey such as the MCBS.

(v) Inflexibility. A primary advantage of the MCBS is its ability to respond to urgent policy issues by modifying its sample, its questionnaire, or both. Existing surveys have studied the elderly and the disabled either by over sampling the elderly (MEPS) or through supplements to existing surveys. These approaches do not facilitate the development of special supplements to assess anticipated policy changes, garner information on opinions that may impact responsiveness to the rapidly evolving variety of health care delivery systems, or assess special areas within the Medicare population, e.g., utilization of over the counter medication.

(vi) Lack of coordination. Previous surveys, for the most part, have been self-contained, unlinked to other sources of information except for such aggregate data bases such as the Area Resources File. This prevents the survey data from being augmented and validated by comparison with concurrent sources such as administrative data.

Lack of coordination across surveys also occurs in terms of definitions and survey approaches. A comparison of terms, methods, and survey content across five national surveys displayed differences in the definition, use, and content of terms, e.g., functional status and disability (Corder and Manton, 1991). This difference coupled with the complexity of sample designs and instrumentation across national surveys, creates difficulty in coordination analysis of survey data.

B..1.d Advantages of MCBS for program/policy analysis

MCBS is the appropriate response to the need for current information on the Medicare population for the following reasons:

(i) Financial Information. The central purpose of the MCBS questionnaire is to determine utilization of and charges for care, insurance coverage, and sources of payment for the sampled population. These can then be related to beneficiary characteristics – such as age, sex, marital status, income, health status, and satisfaction – to form a comprehensive picture of Medicare beneficiaries’ health care experience.

(ii) Targeted Population. The MCBS is a sample of Medicare enrollees. The sample design is based on age strata with an over-sample of the oldest old (85+) and the disabled (under 65 years of age). Over-sampling the oldest old implies over-sampling (de facto) the nursing home population. These segments of the Medicare population are of program and policy interest because of their high rates of use of medical resources.

(iii) Timely Availability of Data. The goal of the MCBS data dissemination plan is to provide the data in a rapid and timely manner. Annual files in the Access to Care series are released roughly one year following data collection (September—December). Annual files in the Cost and Use series are released roughly two years from the close of the collection year. The additional year delay for the Cost and Uses series is a result of the necessity to conduct up to two additional interviews collecting the prior year’s utilization statements. This goal has been a driving force beginning with the release of the 1991 Access to Care and the 1992 Cost and Use files (first in each series) and continuing un-interrupted to the most recent release. Below is a listing of the last three years of MCBS user files and the next three years of proposed user files:

2003 Access to Care October 2004 2006 Access to Care * October 2007

2002 Cost and Use December 2004 2005 Cost and Use * December 2007

2004 Access to Care October 2005 2007 Access to Care * October 2008

2003 Cost and Use December 2005 2006 Cost and Use * December 2008

2005 Access to Care October 2006 2008 Access to Care * October 2009

2004 Cost and Use December 2006 2007 Cost and Use * December 2009

* estimated date of completion

As a consequence of this rapid availability of data, both Med PAC and CMS have incorporated information from the MCBS into their annual reports to congress concerning the impact of physician payment reform upon Medicare enrollees.

(iv) Continuous data collection. MCBS provides information on Medicare enrollees, including the disabled and the elderly (including people residing in the traditional and non-traditional institutionalized setting). The survey continues to collect information on the entire continuum of medical services, i.e., prescribed medications to inpatient hospitalizations, as well as insurance coverage and sources of payment.

The core components (health insurance and service-specific cost and use) for the household and nursing home questionnaires provides consistent and comparable information throughout the project life. Core data collection is continuous. This provides the opportunity for continuous monitoring and assessment of the effects of policy changes phased in over several years; an example is physician and beneficiary reaction to physician payment reform.

The individual questions and scales (e.g., functional ability) in the core components have remained consistent throughout the survey. Consistency between rounds and individual questions will be maintained in Rounds 48-56. We anticipate minimal changes to the core questionnaire. Supplements will be used to augment the core components in the community and facility instruments.

The core community component also contains a summary module. The summary collects information on charges and payments that were not available during the previous interview(s) and provides respondents the opportunity to correct or modify information that was previously collected. We will continue to use the charge payment summary in Rounds 48-56.

(v) Flexibility. Utilization of supplements to augment the core components has helped the MCBS to be responsive to policy changes. For example, in conjunction with the former Office of Research and Demonstrations (ORD) (now part of the Office of Research, Development, and Information, ORDI), an HMO supplement was administered starting with Round 16 and continued each year during the September to December round. Due to the initially small universe, from 1996 to 1999 the sample was augmented to enhance CMS’ ability to address issues in its risk-contracting program. Information collected from this supplement is helping to shape and understand the evolving Medicare delivery system, through:

• an assessment of satisfaction with services for beneficiaries enrolled in managed care with chronic conditions and functional limitations, and a comparison to similar beneficiaries in Fee-For-Service.

• an examination of beneficiary satisfaction in specific geographic areas to screen for possible problems specific to those areas.

• an assessment that measures the extent to which biased selection continues to exist in Medicare’s managed care program.

(vi) Integration of data. The MCBS provides the opportunity to integrate primary and secondary data sources. MCBS information on cost and use is linked with CMS administrative files and the Medicare and private insurance statements. Thus, this survey has inherent internal validity checks that include on-site comparisons between the respondent‑reported data and available statements, calendar entries and documentation such as checks and bills. A second validity check is created by the merger of program data with respondent‑reported data, resulting in a data base that integrates survey and administrative data sources (Eppig and Chulis, 1997). In addition, it is possible that the MCBS could be linked with other existing data sources, e.g., the Area Resource File, and that sample persons retired from the survey could be tracked through Medicare claims files and the National Death Index.

This integrated capability is available for the facility component of the survey as well as the household component. The facility component relies on medical records rather than statements for the reporting of use. These records, coupled with the cost, billing and reimbursement information in administrative records provide validity cross checks for the data collected by interviewing facility staff.

B..2 Information Users

MCBS data users can assess the impact of major policy innovations and health care reform on Medicare beneficiaries (pre and post implementation). They can monitor delivery of services, sources of payment for Medicare covered and non-covered services, beneficiary cost sharing and financial protection, and satisfaction with and the access to health care services. Analysts can relate the dynamics of future aging patterns to age-specific rates of use of health care services. MCBS data also allow accurate measurement of total health care expenditures for enrollees and changes in private health insurance benefits (including long-term care insurance). The capacity to follow individuals into and out of nursing homes and hospitals provides analysts the opportunity to estimate the total cost of episodes of illness and level and type of system interventions including home health care.

Analysts are able to assess the aggregate cost of short stays and long stays in nursing homes, and of combined hospital/nursing home stays. In addition, the MCBS data can be used to forecast the need and settings for selected chronic disease care and the total need for long-term care services by examining trends in the use of health care services, linked with morbidity, disability and mortality data. (A detailed analysis plan is included as Attachment II. This plan discusses the analytic potential of the data and presents studies and data uses that exceed the budgetary and staffing limitations of the MCBS.)

B..2.a Categories of users

The comprehensive nature of the MCBS makes its data appealing to a broad spectrum of users.

(i) CMS actuaries. Within CMS’ Office of the Actuary (OACT), MCBS results are used to estimate the cost or savings of proposed changes to the Medicare program. The results also aid in the creation of various models of consumer behavior to simulate the effects of changes to Medicare on spending, access and costs to beneficiaries.

The MCBS provides an ongoing source for the production of age‑specific estimates of national health expenditures. It facilitates accurate estimates of trends in health care use, such as patterns of acute care preceding institutionalization and use pre‑ and post‑hospitalization. It enhances CMS’ ability to simulate the impact of systemic changes on the Medicare enrollees (e.g., health care system reform, or the impact of proposed Medicaid reforms on the elderly population). In addition, the MCBS information is a primary data source for calculating demographic cost factors used in computing the Adjusted Average Per Capita Cost (AAPCC) HMO Payment Rates, maintained by OACT. This actuarial model is used to project health care expenditures under a variety of scenarios. Finally, MCBS data will be helpful in estimating aggregate national cost and use of long-term care, an area in which data currently are very sketchy.

(ii) Other CMS components. Survey results have been and will continue to be used by the CMS’ Office of Research, Development, and Information. ORDI is using the information collected to examine access to health care, including the preparation of a mandated Report to Congress. The survey detects any changes that may occur with the implementation of the Medicare Modernization Act and monitors any changes over time to access, availability and cost of medical care for Medicare enrollees. In addition to monitoring access to health care, ORDI uses both the household and the facility components of MCBS to assess the use of influenza and pneumonia vaccinations, to compare the use and cost of prescribed medication in both outpatient community and facility settings and to determine the impact of the Qualified Medicare Beneficiary program. ORDI uses the managed care supplement to the MCBS to compare fee-for-service with alternative delivery systems and assess beneficiary satisfaction and access to alternative delivery systems. Also, MCBS provides information on health promotion and preventive techniques from the core instrument. Additional information on these areas can be obtained by the development of supplements.

Analysis of the facility component also allows CMS to examine expenditures that are covered by Medicaid, the shifts between private pay and Medicaid, and the cost implications for both Medicare and Medicaid in the areas of spending down assets and spousal impoverishment.

Use of the MCBS provides the opportunity to quantify and establish baseline measures for several of the objectives in the CMS action plan, specifically, to involve the beneficiaries in defining their health care needs by aggregating and using data for continuous policy and process improvement, and to assess outreach and information to the beneficiaries on programs, services and health care choices.

(iii) Other government and quasi‑government components. The MCBS is a major source of information for the Medicare Payment Advisory Commission (Med PAC) (formerly the Physician Payment Review Commission) to monitor access to health care and beneficiary financial protection. CMS staff have worked closely with Med PAC staff in collecting and processing the MCBS, and the commission stated in its 1996 Annual Report to Congress that ...

“Beneficiary perceptions of access to care are addressed by the Medicare Current Beneficiary Survey. The CBS is a longitudinal survey providing information on specific aspects of beneficiary access to care, such as reasons for difficulty in finding a physician, whether care was delayed for financial reasons, availability of a usual source of care, and satisfaction with care. This important data source provides useful contextual information that complements analyses of utilization patterns.”(Chapter 12, p. 211)

The commission stated in its 1996 Monitoring Access of Medicare Beneficiaries that steps have been taken by CMS ...

“toward supplementing the MCBS questions and sample to reflect Medicare managed-care expansion.” (p. 52)

The MCBS data have been routinely included in Med PAC’s Annual Reports and data collected by the MCBS have been used both in descriptive statistics and simulation.

The MCBS is also a major source of information for the Congressional Research Service (CRS) to monitor access to health care. CMS staff has worked closely with CRS staff in collecting and processing the MCBS data to publish a CRS article on the role of supplemental health insurance (O’Sullivan, Lee, and Yang, 1996) and using the cost and use data to develop a model of spending for Medicare covered and non-covered services.

Several other agencies that have developed a partnership in using MCBS data. These include the Congressional Budget Office, Public Health Service, Medicare Payment Advisory Commission, National Institute of Mental Health, National Institute on Aging, and Advisory Council on Social Security.

(iv) Research community and policy analysts. From Rounds 1 through 43 (fall 2005), MCBS data are currently available on CD Rom and mainframe tapes for use by other health care researchers. The MCBS is the primary source of basic information for research and policy analysis of health care financing and delivery issues for the Medicare population. The survey produces data sets suitable for both longitudinal and cross‑sectional analysis.

One of the most important policy issues we face is the increasing number of aged and disabled people in long-term care, including home health care. The MCBS provides the capability to analyze detailed characteristics of transitions between institutional and community status. The survey collects information on the type of facilities used, private insurance coverage including long-term care insurance, and activities of daily living. The longitudinal nature of the survey provides the opportunity to relate changes in use with changes in coverage and health status, and to observe processes that occur over time, such as tracing the elderly through the medical care system, effects of age and service substitution issues by setting.

B..2.b Consequences of failure to collect data for Rounds 48-56

Failure to embark on Rounds 48-56 would prohibit the collection of the access supplement in Rounds 49, 52, and 55. This would severely restrict the ability of the Medicare Payment Advisory Commission to continue to consistently assess the impact of reimbursement modifications on the beneficiary population. Also, failure to complete Round 48 data consolidation would negatively impact the longitudinal analysis of various competing delivery systems (started with the Round 16 managed care supplement) and the beneficiaries’ perception of the care they received. Failure to continue the collection of this data will eliminate the continuation of statistical analysis and publication work e.g. Health and Health Care of the Medicare Population: Data from the MCBS and The Characteristics and Perceptions of the Medicare Population: Data from the MCBS.

In addition, failure to continue the survey would be costly and would necessitate the drawing and fielding of a completely new sample, this would be extremely costly in terms of money, loss of continuity in data collection and contract resources (e.g., field operations and data management and administration). This would prevent the continuation of the collection of longitudinal data on the sample (i.e., beginning in 1991), thus limiting the collection of patterns of use and expenditures of health care over time. It would also significantly limit analysis of transitional processes, such as Medicaid "spend down" and institutionalization. Researchers on aging have stressed the importance of collecting information on this heterogeneous population by selected characteristics including age, income, and health. The MCBS represents the only study that addresses the issues of age and impact of disability on the provisions of health care services.

In addition to interrupting the flow of the survey and nullifying the use of currently collected data for additional longitudinal analysis, failure to continue this survey for Rounds 48 through 56 of the MCBS would preclude the opportunity to understand the impact of temporal ordering, limit evaluation, and prohibit the monitoring of change with precision. Limiting the opportunity to collect Rounds 48-56 and subsequent MCBS rounds would prevent the creation of baseline information to assess impact of continued Medicare reform (pre and post implementation), on the Medicare beneficiaries, and ultimately the impact of changes to Medicare on the overall health care system.

Failure to continue to collect these data would virtually preclude the development of a consistent and continuous database of total personal health care expenditures for those over 65. This in turn will hamper CMS actuaries' ability to produce the timely and reliable estimates of health expenditures by the elderly population that are demanded annually by the Congress. It would make very difficult the task of identifying current and future trends in source of payment for health care. It would retard our understanding of transitional phases in living arrangements and their interaction with the Medicare program.

Failure to collect these data would mean that there would be no reliable, continuous and comprehensive source of information on use, cost, and expenditure for health care for Medicare enrollees that can be used to make credible estimates of the effects of program changes. Failure to continue the MCBS would prohibit the implementation and monitoring of CMS’ standards under MMA. The MCBS can provide baseline information on health status, access to health care, preventive activities to provide both baseline and short and long-term monitoring of changes for the CMS measurements standards. There would be no reliable and current source of linked use, cost, and demographic information needed for models of the impact of the Medicare program on the health sector and the total economy if the MCBS is not continued. In addition, failure to collect the MCBS would mean that other agencies will need to collect data independently, a process that would almost certainly be more expensive and generally less useful (because of the small sample size for the elderly and disabled in other national surveys).

B..3 Improved Information Technology

The MCBS takes full advantage of advances in survey techniques. Respondents living in the community are administered a personal, face to face interview using computer-assisted personal interviewing (CAPI). Though respondents residing in long-term care facilities are not administered a personal, face-to-face interview, the CAPI instrument is used in interviewing the designated facility staff proxy. CAPI, which functions with programmed edit checks, reduces respondent burden by minimizing the potential for double reporting and inconsistent responses. CAPI enables the interviewer to move through complex skip patterns quickly, which reduces respondent burden by shortening the interview and eliminating the need for call backs to correct errors. CAPI greatly increases the efficiency of the questionnaire during the interview in the following ways:

• CAPI tailors the sequence of questions to the responses of the interviewee, resulting in few – if any – interviewer skip errors. The natural flow of the interview is maintained even when the pattern of questions is complex.

• CAPI automatically provides “fills”, or word choices within questions. One keystroke can insert “you”, “he”, or “she”, as appropriate, for the duration of the interview. The sample person’s name, date of the last interview, and other items can also be filled as needed.

• CAPI maintains rosters or lists created during the interview, such as household members, health insurance plans, medical conditions, providers, visit dates, prescription drugs, and people who help with daily activities. These rosters can be used to structure questions, e.g., cycling through a series of doctor visits and checking for missing information. Interviewers can select items from a roster, add items, or correct them. Rosters are carried over from one interview to the next.

• CAPI edits entries for range and consistency. The interviewer can make corrections immediately. Information missing from a previous round can be inserted in the questionnaire.

• CAPI allows instantaneous calculations to be made, such as the amount remaining to be paid on a medical bill after totaling several payments.

• Interviewers use the computer to electronically transmit completed cases to the central office over the telephone.

CAPI has strategic implications for survey design and planning, some of which are not easily predicted. CAPI strongly influences the shape of the questionnaire itself: It can be more complex, more tailored to the characteristics of respondents, with more cross-references and rosters, and have more error checks.

B..4 Duplication of Similar Information

During the development and initial administration of the MCBS, a number of people inside and outside the Federal government were consulted. This consultation included issues of design, content, and statistical methodology and analysis. None of the people contacted were aware of duplicative information, nor were they aware of any other survey that duplicates the efforts of MCBS. In addition, both CMS and its data collection contractor have undertaken exhaustive reviews of the literature and other data sources. In no instance have we identified another source of data, which would be an effective substitute for the MCBS. Continuing interagency collaboration insures against the likelihood of duplicative data collection processes now and in the future.

B..4.a Absence of similar information

Health surveys, especially expenditure and use surveys, integrate or build upon other health and medically related surveys. MCBS is an amalgam of data collection approaches that proved effective in other health and health related surveys. By design, parts of the MCBS replicate and draw from previous surveys on health care and use for the U.S. noninstitutional civilian population, as well as those of the institutional civilian population. However, no single survey has included all of the design features that are the MCBS. No combination of surveys, or survey components will address the purposes identified in section B..2 nor in the analytical plan contained in Attachment II. With the exception of the Current Medicare Survey, the forerunner of the MCBS, the studies summarized below lacked the sample and cell size to reliably estimate use, cost, and expenditures for the Medicare population.

The household component of the MCBS builds upon surveys that have collected health services use and expenditure information. The MCBS reflects the cumulative experiences garnered in:

• The 1977 National Medical Care Expenditure Survey (NMCES), sponsored by the National Center for Health Services Research (NCHSR); designed and conducted by both NCHS and NCHSR. NMCES was a panel study consisting of six household interviews including provider and third-party record checks.

• The 1980 National Medical Care Use and Expenditure Survey (NMCUES), co-sponsored by the National Center for Health Statistics and the Health Care Financing Administration in 1980. NMCUES was a panel study consisting of five interviews, plus supplemental surveys of Medicaid enrollees in four States, coupled with Medicare and Medicaid administrative records.

• The 1987 National Medical Expenditure Survey (NMES), conducted by Westat for what the National Center for Health Services Research. NMES was a five round study, which collected data for a one-year reference period.

• The Current Medicare Survey of Medicare enrollees sponsored by the Social Security Administration until 1977. Conducted by the Census Bureau, the Current Medicare Survey was a monthly interview of SMI enrollees.

• The 1970, 1974, 1975 and 1977 Family Medical Expense Supplement included in the National Health Interview Survey, a retrospective self-administered instrument that collected categories of out-of-pocket expenses.

With the exception of the Current Medicare Survey, the number of Medicare recipients in each of these surveys was too small to produce reliable estimates on the subgroups of current interest to CMS. As explained earlier in this Statement, these surveys, even taken in combination, fail to meet CMS’ needs. These surveys are repeated too infrequently to support CMS’ need to produce estimates of health use and cost or to assess the impact of actual and proposed legislative changes on the Medicare beneficiary population. In addition, while there was some consistency between the questionnaires used, the scope and emphasis of these surveys varied over time and the Medicare population was not a priority focus in any of these surveys.

Though not a survey, CMS maintains administrative records on the Medicare program. However, these claims data do not provide detailed demographic or health status information, information on the cost or use of services not covered by Medicare, or sources and amounts of payment for services other than Medicare.

The nursing home component of the MCBS also builds upon the experience gained through other surveys. These surveys include the National Long-term Care Surveys, the 1982 and 1984 Supplement on Aging (NHIS), and its 1986 and 1988 follow-ups. In addition, the nursing home component builds on the experiences of the National Nursing Home Survey and the NMES Institutional Population Component (IPC). Since 1997, the facility component of the MCBS has adopted a CAPI questionnaire derived from AHRQ’s (formerly known as AHCPR) National Nursing Home Survey.

The 1982, 1984, 1989 and 1990 DHHS National Long‑term Care Surveys provides a database on a sample of the Medicare population at selected points in time. These surveys do not include continuous information on service use. In addition, the universe consisted of elderly (age 65 and over) people with functional limitations. The surveys allow only cross‑sectional analysis of a subgroup of the elderly, i.e., those with functional limitations. Therefore, use of this data is limited by its capacity to generalize to the Medicare population, to predict or monitor the impact of legislative changes, and assess cohort or historical effects.

The 1992 National Home and Hospice Care Survey (NCHS), examines selected characteristics, including health and functioning status of persons receiving home health care and hospice care, along with various measures of utilization.

The 1984 Supplement on the Aging (SOA) collected information on the elderly, as well as those individuals who would become elderly (ages 55-65). Extensive information was collected on medical conditions, impairments, and visits to nursing homes.

The 1986 and 1988 Longitudinal Study on Aging (LSOA), conducted by the NCHS from a sample identified in the 1984 NHIS Supplement on Aging, is another important database. However, the LSOA does not collect detailed information on service use and source of payment, nor does its sample include institutionalized people except those sampled people who lived in households in 1984 and subsequently became institutionalized. In addition, the 1986 survey was a re-interview survey, generated from the 1984 NHIS Supplement on Aging. The population consisted of those individuals who were aged 70 or older, restricting the generalizability of this data to the Medicare population. Neither the SOA or the LSOAs can provide full-year use and expenditure information, provide up-to-date information nor information about the transitions between care received in institutions and while living at home, e.g., nursing home and community settings.

The NMES IPC examined service use, sources of payment and charges in a longitudinal sample of residents and admissions to inpatient facilities. It was designed to be generalizable to the overall institutional population in nursing and personal care homes, facilities for the mentally retarded and psychiatric hospitals. MCBS is more restrictive in the sample frame and findings will be generalizable to the Medicare population in nursing homes and board and care and personal care homes.

The Medical Expenditure Panel Survey (MEPS), which follows in the sequence of NMEES, NMCUES, and NMES, provides information on the health and long--term care experience of a cross--section of the national population. The MEPS complements the MCBS by providing comparable data on a general sample, but it will not duplicate the functions of the MCBS for many of the same reasons that NMES did not: only MCBS provides the focus on the Medicare population that is needed to address a broad range of Medicare program issues.

The focus of the Medicare Health Outcomes Survey (HOS) is to provide information on the health status of the Medicare managed care population for use in quality improvement activities, plan accountability, public reporting, and improving health. As noted above, only the MCBS provides the focus on the Medicare population that is needed to address a broad range of Medicare program issues.

Other panel studies, such as the Consumer Expenditures Survey and the Survey of Income and Program Participation, do not collect sufficiently detailed information on health care use and expenditures. In addition, these surveys do not collect data from both the institutional and the noninstitutional population.

We found that all other government (i.e., existing and planned) and private sources of data for research and policy analysis either include an insufficient number of elderly respondents or do not collect information on the cost, use, expenditures, sources of payment, reimbursement patterns and socio-demographic characteristics of the elderly on a continuous basis.

B..5 Small Businesses

Most of the data collected for the MCBS will be from individuals in households. However, in any given round about 900 sample persons will reside in government‑sponsored, non‑profit, and for‑profit institutions such as nursing and personal care homes. Some of these institutions qualify as small businesses. All data collected on sample persons in these institutions will be collected from small businesses, their employees serve as proxies for each sample person in their care. Interviewers in long-term settings make every effort to determine, for each type of question, which staff members are most able to answer them. The data collection procedures are designed to minimize the burden on all respondents, with skip instructions so that respondents will be asked only questions that apply to them. Facility respondents are not asked to answer questions of which they are not likely to have ready knowledge, such as attitudes and perceptions. Furthermore, the MCBS questionnaire mirrors the Minimum Data Set questions where applicable, to further reduce respondent burden.

All enrollees will be selected from the Medicare Enrollment Database and sample persons residing in nursing homes will be randomly distributed. Therefore, the burden for the collection of this information will be proportional to the size of nursing homes (on average), thereby reducing the burden on small nursing homes.

B..6 Cancellation or Less Frequent Collection

Only a continuous survey can fully meet CMS’ information needs for operating the Medicare program. And, address many of the limitations of existing databases for collecting information on the elderly and disabled including the oldest old portion of the population (85+). Large changes in the Medicare delivery system and program itself have been occurring with increasing frequency, e.g., The Medicare Prescription Drug, Improvement, and Modernization Act. While such extensive changes as contained in MMA are the exception rather than the rule, adjustments to the Medicare program continue to remain an ongoing part of the program administration. Only a continuous survey will produce the data that support accurate evaluation of the effects of all such changes.

By re‑interviewing the same respondents and replenishing the sample annually, the MCBS will continue to support longitudinal as well as cross‑sectional analyses. Longitudinal data provide the material for models that analyze quantitative change over time. Longitudinal data are needed for causal analysis, such as the relationship between legislative and program changes and their effects on enrollees, providers, and costs. Longitudinal data can also help clarify the relationship between changes in health status and functioning and levels of health care service use and expenditures (i.e., temporal ordering for analysis). These and other uses for longitudinal data are described more fully in the Analysis Plan at Attachment II.

Research on survey methodology shows that as the length of time between interviews grows, there is a marked decline in the quality of the data reported (Cohen and Burt, 1985; Cohen and Cohen, 1986). The literature seems to support spacing interviews for a continuous survey of medical care use up to four months apart. The Current Medicare Survey employed a one‑month recall period, the 1977 NMCES an average of three months, and the 1987 NMES an average of four months. In the first round of interviewing, the sample person in the community is provided with a calendar and asked to record all visits to health care providers, expenditures and to retain all statements including private insurance and Medicare statements/bills. Thus after the initial round, the recall period for the MCBS is since the time of the last interview. Making MCBS rounds any farther apart would significantly jeopardize the quality of the results, because of the strong potential for recall problems with the elderly. In addition, less frequent data collection may negate the opportunity to obtain additional information by using supplements; more time spent on repetition of the core items might not allow time for the inclusion of proposed supplements questions to assess legislative and program changes.

B..7 Special Circumstances

None of the special circumstances listed by OMB apply to the MCBS.

B..8 Federal Register Notice / Outside Consultation

A 60-day Federal Register Notice was published on November 24 , 2006 (Attachment III). Since this collection is not new, we have not gone out to solicit outside consultation; however, during the 60-day Federal Register Notice the public is free to comment.

B..8.a General Collection Guidelines. This data collection effort is consistent with the guidelines in 5 CFR 1320.6.

B..8.b Consultation with parties outside CMS. Three types of consultations outside CMS where held during the start-up period of the survey. These included:

(i) Other Federal agencies. These contacts include the DHHS Office of the Assistant Secretary for Planning and Evaluation, the Medicare Payment Advisory Commission, the Agency for Healthcare Research and Quality (formerly known as the Agency for Health Care Policy and Research), the National Center for Health Statistics, the Social Security Administration, and the National Institute on Aging.

(ii) Other policy, research, statistical and data collection consultants. This Panel, comprised of government, academic, and industry consultants, has met twice. This panel is convened to assess the progress of the survey, review data collection, and analysis plans and initiate planning of future dynamics of the MCBS.

(iii) Interested groups. Contact with provider organizations, insurers and other interested organizations, has been timed to facilitate data collection activities. Consultations with the American Association of Retired Persons and the Administration on Aging, has occurred throughout the survey.

While we have maintained direct contact with provider agencies, Social Security Administration District Offices, American Association of Retired Persons, and Agencies for Aging, both state and local agencies throughout the data collection process, this contact is in the form of outreach rather than consultation. As the MCBS approaches its annual sample notification, applicable organizations are reminded of the continued effort. The sample design and much of the survey content are now established in a Medicare program support mode. The longitudinal nature of the survey dictates a certain degree of comparability across year.

B..9 Payments / Gifts to Respondents

The MCBS does not provide payments or gifts as incentives to respond. The most important incentive we use is to persuade the respondent that his or her participation is a service to the future of Medicare. We use standard refusal conversion techniques, letters, and newsletters.

B..10 Confidentiality

This is a person-based survey. The cover letter to the respondent contains a reference to the Privacy Act of 1974, as amended. A handout sheet (provided to the household respondent at the door and the nursing home administrator and proxy respondents) contains a statement of confidentiality consistent with the Privacy Act of 1974 and the Paperwork Reduction Act of 1995. In fielding selected supplements, e.g., Income and Assets, an additional hand-out sheet is provided explaining the purpose of the supplement and restating the Privacy Act. Interviewer training stresses the importance of maintaining confidentiality. The household interviewer's manual specifically addressees this and it is part of the training for the interviewers (both household and nursing home). Procedures have been established to maintain and insure confidentiality. These include computer security procedures (password protection for each interview and interviewer), electronic mail submission and next of kin consent forms for following nursing home people (discharged residents).

In addition, the administration of the facility component is preceded by a statement that CMS will collect only the information necessary to perform the system's functions. Any data published will exclude information, which might lead to the identification of specific individuals (e.g., ID number, claim numbers, and location codes). CMS will take precautionary measures to minimize the risks of unauthorized access to the records and the potential harm to the individual privacy or other personal or property rights of the individual.

All MCBS survey staff directly involved in MCBS data collection and/or analysis activities are required to sign confidentiality agreements. Furthermore, all MCBS patient-level data are protected from public disclosure in accordance with the Privacy Act of 1974, as amended.

B..11 Sensitive Questions

None of the data to be obtained by this study are considered to be sensitive.

B..12 Burden Estimate (Total Hour and Wages)

The cover letter to the respondent and a handout presented at the beginning of the interview contain the following disclosure statement:

cording to the Paperwork Reduction Act of 1995, no persons are required to respond to a collection of information unless it displays a valid OMB control number. The valid OMB control number for this information collection is 0938‑0568 (valid until April 30, 2007). The time required to complete this information collection is estimated to average 1 hour per response, including the time to review instructions, search existing data resources, gather the data needed, and complete and review the information collection. Additionally, a randomly selected 10 percent of the respondent population (per round) are contacted for a validation interview averaging 10 minutes. If you have any comments concerning the accuracy of the time estimate(s) or suggestions for improving this form, please write to: CMS, C5-14-03, 7500 Security Boulevard, Baltimore, Maryland 21244-1850.

Attachment IV is a copy of the advance letter and MCBS brochure and introduction to the MCBS sheet for the household.

Table B..12 shows estimates of the annual respondent burden. For the national survey, the annual burden for the MCBS is based on three interviews per respondent. We currently estimate that our annual burden will be 50,325 hours for the survey.

Table B..12: Estimates of the Annual Respondent Burden

Expected

Time per Number of Number of Burden

Response Interviews Completed Hours

Rounds 47-49 1.0 hour 3 16,500 49,500

Validation per 10% of population 10 minutes 3 1,650 825

---------

Total Annual 50,325

Total Estimate for Rounds 47-55 (3 years) 150,975

B..12.a Changes in burden. There are no changes to respondent burden proposed in this clearance package. This project is in CMS's FY 2007 Budget. This reflects a burden of 50,325 hours per year.

B..13 Capital Costs

There will be no cost to respondents other than their time to participate. Because of CAPI, we will be asking the respondent (household) if we can use their electricity. If they agree, we offer paying a nominal fee ($3.00) for the use of the electricity in each household interview.

The data that we are collecting includes normal cost, use, expenditure and condition-specific data. In addition, for those individuals in facilities, we are asking for information that nursing home personnel routinely maintain. In order to minimize any unanticipated costs of providing these data, we will accept estimates or approximations made by the respondent.

B..14 Cost to the Federal Government

The estimated cost to the government for collecting these data includes the Westat contract, and direct CMS expenses for labor and travel.

B..14.a Contract cost. The estimated cost for the planning, sampling, data collection and analysis for the MCBS in each year is: (The following estimated costs per Fiscal Year are approximated amounts based on the current MCBS contract and include the cost associated with conversion from a DOS based platform to a Windows based platform.)

• FY 2007: Data collection and analysis $14,000,000

• FY 2008: Data collection and analysis $14,400,000

• FY 2009: “transition” to phase 4 contract $14,800,000

These costs include all labor hours, materials and supplies, reproduction, postage, telephone charges and indirect costs.

B..14.b Direct cost. CMS personnel involved in Medicare HOS include approximately 10.5 FTEs broken out by pay grade in table B..14.b.

Table B.14.b: CMS Personnel

Grade FTE 2006 Annual Salary Cost to Government

GS12 step 5 2.0 $73,720 $147,440

GS13 step 5 6.0 $87,664 $525,984

GS14 step 5 2.5 $103,594 $258,985

$932,409

CMS staff costs will be approximately $932,409. In addition, staff travel is budgeted for $2,000. The MCBS releases its documentation on CD Rom thus eliminating its printing budget. Thus, in-house CMS cost will be $935,000.

B..15 Program Changes

There are no program changes to the survey.

B..16 Publication and Tabulation Dates

The analyses described in Attachment II are illustrative of those expected from MCBS data. Three major types of analyses are planned: descriptive, explanatory and predictive. In addition, a number of published methodological studies have been conducted (bibliography). The descriptive and explanatory analysis is similar in content and format to the Health Care Financing Review.

Special analytical reports and related publications have been issued by CMS on an ad hoc basis. These findings will also be presented at annual meetings of professional associations, at specific health policy meetings and in professional journals.

A variety of methodological issues have been explored using MCBS data. These are similar in content and format to the Methodological Series of NCHS and the Instruments and Procedures series of AHRQ. In addition, data files will continue to be prepared over the course of the survey. For a listing of data files created over the past ten year, which are currently available through CMS, refer to section B..1.d(iii) of this document.

B..16.b Schedule for information collection and dissemination

05/01/2007 Data collection begun for MCBS Round 48.

09/01/2007 Data collection begun for MCBS Round 49.

10/01/2007 Public Use File available for 2006 Access to Care series.

12/01/2007 Public Use File available for 2005 Cost and Utilization series.

01/01/2008 Data collection begun for MCBS Round 50.

05/01/2008 Data collection begun for MCBS Round 51.

09/01/2008 Data collection began for MCBS Round 52.

10/01/2008 Public Use File available for 2007 Access to Care series.

12/01/2008 Public Use File available for 2006 Cost and Utilization series.

01/01/2009 Data collection begun for MCBS Round 53.

02/29/2009 Award new MCBS 5 year contract.

05/01/2009 Data collection begun for MCBS Round 54.

09/01/2009 Data collection began for MCBS Round 55.

10/01/2009 Public Use File available for 2008 Access to Care series.

10/01/2009 Prepare clearance package for Rounds 57 – 65.

12/01/2009 Public Use File available for 2007 Cost and Utilization series.

01/01/2010 Data collection begun for MCBS Round 56.

05/01/2010 Data collection begun for MCBS Round 57.

09/01/2010 Data collection begun for MCBS Round 58.

10/01/2010 Public Use File available for 2009 Access to Care series.

12/01/2010 Public Use File available for 2008 Cost and Utilization series.

B..16.c Plans for statistical analysis and publication

Health and Health Care of the Medicare Population: Data from the Medicare Current Beneficiary Survey and Characteristics and Perceptions of the Medicare Population: Data from the Medicare Current Beneficiary Survey are two of the special reports that are being issued on an annual basis by CMS. The chartbooks contain a comprehensive set of statistical tables describing the Medicare population in terms of its demographic and socioeconomic characteristics, use of and expenditures on health care, perceived health and functional status, and access to and satisfaction with health care. The tables, which were created from Access to Care and Cost and Use files released to the public, are intended as references for persons interested in using MCBS data to analyze the health and health care of the Medicare population.

CMS has disseminated these chartbooks to federal and state governmental agencies, for profit companies, nonprofit organizations, and academic institutions. The data in these tables are being updated as new files become available, and are released to the public in the two series of chartbooks. In addition to providing statistics from the annual MCBS releases, the chartbooks can be used to follow trends and measure change in the Medicare population. Subsequent

chart books have illustrated the types of longitudinal analysis that can be performed with MCBS data.

Annual updates to the two chartbooks are available for the following years. (For years were the chartbooks are still unavailable an asterisk is used as a denotation and an estimated date of completion follows.)

Health and Health Care of the Medicare Population: Data from the MCBS

1992 – 2003, 2004* (December 2007), 2005* (December 2008), and 2006* (December 2009)

Characteristics and Perceptions of the Medicare Population: Data from the MCBS

1995 – 2004, 2005* (August 2007), 2006* (August 2008), and 2007* (August 2009)

B..17 Expiration Date

Not applicable. The collection of Medicare HOS is an ongoing endeavor. Therefore, an expiration date is not practical.

B..18 Certification Statement

There are no exceptions to the certification statement.

1