oversight

Medicaid: Data Sets Provide Inconsistent Picture of Expenditures

Published by the Government Accountability Office on 2012-10-29.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                             United States Government Accountability Office

GAO                          Report to the Ranking Member,
                             Committee on Finance, U.S. Senate



October 2012
                             MEDICAID

                             Data Sets Provide
                             Inconsistent Picture
                             of Expenditures




To access this report
electronically, scan this
QR Code.
Don't have a QR code
reader? Several are
available for free online.




GAO-13-47
                                             October 2012

                                             MEDICAID
                                             Data Sets Provide Inconsistent Picture of
                                             Expenditures
Highlights of GAO-13-47, a report to the
Ranking Member, Committee on Finance,
U.S. Senate




Why GAO Did This Study                       What GAO Found
CMS, within the Department of Health         Medicaid expenditures in the Medicaid Statistical Information System (MSIS) were
and Human Services, and state                generally less than CMS-64 amounts. National expenditures in MSIS were 86, 87,
Medicaid agencies jointly administer         and 88 percent of the amounts in CMS-64 in fiscal years 2007 through 2009,
the multibillion-dollar Medicaid             respectively. In fiscal year 2009, MSIS expenditures for states ranged from 59 to
program, which finances health care          119 percent of CMS-64. Specifically, 40 states reported lower expenditures in
for certain low-income individuals.          MSIS than CMS-64; 5 states and the District of Columbia reported higher
Medicaid is on GAO’s high-risk list          expenditures; and 5 states reported similar levels of expenditures.
because of vulnerabilities to waste,
fraud, abuse, and mismanagement.             Total Medicaid Expenditures in MSIS and CMS-64, Fiscal Years 2007-2009
CMS has two data sets that report
state Medicaid expenditures. The                                           MSIS              CMS-64 Difference between                           MSIS as a
                                              Fiscal             expenditures          expenditures     MSIS and CMS-64                      percentage of
MSIS data set is designed to report           year         (dollars in billions) (dollars in billions) (dollars in billions)           CMS-64 expenditures
individual beneficiary claims data. The
                                              2007                          $273.9                      $320.1                $46.1                   86%
CMS-64 data set aggregates states’
                                              2008                           294.2                        338.6                 44.4                    87
expenditures, which are used to
reimburse the states for their Medicaid       2009                           323.1                        366.5                 43.4                    88
expenditures. However, neither data          Source: GAO analysis of Centers for Medicare & Medicaid Services’ data.

set provides a complete picture of           Note: MSIS and CMS-64 expenditures were rounded.
Medicaid expenditures.
                                             GAO was able to quantify some, but not all, of the identified differences in
GAO was asked to compare MSIS and            expenditures between MSIS and the CMS-64.
CMS-64 data. This report (1) examines
the extent to which MSIS and CMS-64          •     GAO adjusted MSIS for expenditures that were not attributed to individual
expenditure data differ and (2) where              beneficiaries—such as prescription drug rebates. These adjustments
possible, quantifies the identified                increased MSIS to 92, 93, and 94 percent of the amounts in CMS-64 in fiscal
differences between the two data sets.             years 2007 through 2009, respectively.
GAO reviewed documents, compared
Medicaid expenditure data, and               •     GAO could not account for the remaining differences in part because of
interviewed CMS and state officials.               inconsistencies in the Centers for Medicare & Medicaid (CMS) guidance
GAO used fiscal years 2007 through                 between the two data sets. For example, CMS officials explained that
2009 data—the most-recent and most-                expenditures for inpatient services as reported by a state in MSIS and as
complete data available.                           reported in CMS-64 are not necessarily for the same services.
                                             GAO also found that states do not submit timely MSIS information. CMS requires
                                             states to submit MSIS data within 45 days and CMS-64 data within 30 days of the
                                             end of the quarter. However, states’ reporting of MSIS data can be up to
                                             3 years late, whereas CMS-64 data are consistently reported on time. Also, MSIS
                                             expenditure data are considered less reliable when compared with CMS-64.
                                             GAO has reported that CMS will need more reliable data for assessing
                                             expenditures and measuring performance in the Medicaid program. MSIS and
                                             CMS-64 have the potential to offer a robust view of the Medicaid program,
                                             enhancing CMS oversight of aggregate spending trends, per beneficiary spending
                                             growth, and cross-state comparisons, all of which could be useful in improving the
                                             financial integrity of this high-risk program. However, delays in reporting MSIS
                                             data and inconsistencies between the two data sets limit their usefulness as
                                             oversight tools. CMS has recently completed a pilot study aimed in part at
View GAO-13-47. For more information,        improving the timeliness and consistency of both systems data.
contact Carolyn L. Yocom at (202) 512-7114
or yocomc@gao.gov.                           HHS provided technical comments on a draft of this report, which were
                                             incorporated as appropriate.
                                                                                                            United States Government Accountability Office
Contents


Letter                                                                                         1
                       Background                                                              5
                       MSIS Medicaid Expenditures Amounts Are Generally Less than
                         CMS-64 Expenditure Amounts                                            8
                       Some Factors Could Be Quantified and Accounted for
                         Approximately Half of the Expenditure Difference between the
                         Data Sets                                                           13
                       Concluding Observations                                               27
                       Agency Comments                                                       28

Appendix I             Scope and Methodology                                                 30



Appendix II            Total Baseline Medicaid Expenditures in MSIS and CMS-64,
                       by State, Fiscal Year 2009                                            41



Appendix III           Total Adjusted Medicaid Expenditures in MSIS and CMS-64,
                       by State, Fiscal Year 2009                                            43



Appendix IV            Adjusted MSIS Expenditures as a Percentage of CMS-64,
                       by State, Fiscal Year 2009                                            45



Appendix V             Adjusted MSIS Expenditures as a Percentage of CMS-64,
                       by State and Expenditure Category, Fiscal Year 2009                   46



Appendix VI            GAO Contact and Staff Acknowledgments                                 49



Related GAO Products                                                                         50




                       Page i                                             GAO-13-47 Medicaid Data
Tables
          Table 1: Total Medicaid Expenditures in the Medicaid Statistical
                   Information System (MSIS) and CMS-64, Fiscal Years 2007-
                   2009                                                             9
          Table 2: Total Medicaid Expenditures Adjusted in the Medicaid
                   Statistical Information System (MSIS) and CMS-64, Fiscal
                   Years 2007–2009                                                14
          Table 3: Medicaid Statistical Information System (MSIS) and CMS-
                   64 Service Types, Fiscal Years 2007–2009                       34
          Table 4: Combined Expenditure Categories and Corresponding
                   MSIS and CMS-64 Service Types                                  36


Figures
          Figure 1: Medicaid Statistical Information System (MSIS) Medicaid
                   Expenditures as a Percentage of CMS-64 Expenditures, by
                   State, Fiscal Year 2009                                        10
          Figure 2: MSIS Medicaid Expenditures as a Percentage of CMS-64
                   Expenditures, by Expenditure Category, Fiscal Years
                   2007–2009                                                      12
          Figure 3: MSIS Medicaid Expenditures as a Percentage of CMS-64
                   Expenditures, by Expenditure Category, Fiscal Year 2009        13
          Figure 4: Comparison of MSIS Baseline to Adjusted Expenditures,
                   by Expenditure Category, as a Percentage of CMS-64,
                   Fiscal Year 2009                                               17
          Figure 5: Comparison of MSIS Baseline to MSIS Adjusted
                   Expenditures, as a Percentage of CMS-64, Fiscal Year
                   2009                                                           18
          Figure 6: Comparison of MSIS Baseline to Adjusted Medicaid
                   Expenditures, as a Percentage of CMS-64, by State, Fiscal
                   Year 2009                                                      19
          Figure 7: Adjusted Medicaid Expenditures in MSIS as a Percentage
                   of CMS-64, by Expenditure Category, Fiscal Years 2007–
                   2009                                                           22




          Page ii                                              GAO-13-47 Medicaid Data
Abbreviations

CHIP              Children’s Health Insurance Program
CMS               Centers for Medicare & Medicaid Services
DSH               Disproportionate Share Hospital
FPL               federal poverty level
HCERA             Health Care and Education Reconciliation Act of
                    2010
HHS               Department of Health and Human Services
LTSS              long-term support services
M-CHIP            Medicaid Expansion Children’s Health Insurance
                    Program
MSIS              Medicaid Statistical Information System
PPACA             Patient Protection and Affordable Care Act
S-CHIP            Separate Children’s Health Insurance Program



This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.




Page iii                                                          GAO-13-47 Medicaid Data
United States Government Accountability Office
Washington, DC 20548




                                   October 29, 2012

                                   The Honorable Orrin Hatch
                                   Ranking Member
                                   Committee on Finance
                                   United States Senate

                                   Dear Senator Hatch:

                                   The Centers for Medicare & Medicaid Services (CMS), within the
                                   Department of Health and Human Services (HHS), and state Medicaid
                                   agencies jointly administer the multibillion-dollar Medicaid program, which
                                   finances health care for approximately 67 million low-income individuals
                                   who meet specific eligibility criteria. In recent years, the Medicaid program
                                   has undergone steady growth and is expected to continue to expand in
                                   light of the Patient Protection and Affordable Care Act of 2010 (PPACA). 1
                                   PPACA expands eligibility for Medicaid to nonelderly individuals whose
                                   income does not exceed 133 percent of the federal poverty level (FPL), 2
                                   or $30,700 for a family of four in 2012. 3 However, because states may
                                   choose not to participate in the PPACA expansion, it is unclear how much
                                   the Medicaid program will grow. 4




                                   1
                                    Pub. L. No. 111-148, 124 Stat. 119, as amended by the Health Care and Education
                                   Reconciliation Act of 2010 (HCERA), Pub. L. No. 111-152, 124 Stat. 1029. For purposes
                                   of this report, references to PPACA include the amendments made by HCERA.
                                   2
                                    PPACA § 2001(a)(1), 124 Stat. 271.
                                   3
                                    We use FPL to refer to federal poverty guidelines issued by HHS each year in the
                                   Federal Register. These guidelines provide income thresholds that vary by family size and
                                   for certain states, and which are updated using the Consumer Price Index.
                                   4
                                    Under the Medicaid program, failure by a state to comply with federal requirements may
                                   result in a termination of federal Medicaid matching funds. However, the U.S. Supreme
                                   Court has ruled that states that choose not to expand Medicaid eligibility to these newly
                                   eligible individuals under PPACA will forgoe only the matching funds associated with such
                                   expanded coverage. See National Federation of Independent Business, et al., vs.
                                   Sebelius, Sec. of Health and Human Services, et al., 567 U.S. __, 2012 WL 2427810
                                   (U.S. June 28, 2012). Prior to the Supreme Court decision, the CMS Actuary estimated
                                   the PPACA Medicaid expansion would enroll an additional 15 million individuals in 2014.
                                   However, in July 2012 the Congressional Budget Office estimated that an additional
                                   7 million individuals would enroll in Medicaid and the Children’s Health Insurance Program
                                   (CHIP) in 2014.




                                   Page 1                                                           GAO-13-47 Medicaid Data
Medicaid is on our high-risk list in part because of concerns about the
adequacy of fiscal oversight, which is necessary to prevent inappropriate
spending. 5 We have previously reported that Medicaid has among the
highest estimated improper payments of any federal program reporting
such data. Consequently, we have reported that, particularly as PPACA is
implemented, CMS will need new tools and resources, including more
reliable data for assessing expenditures and measuring performance. 6

CMS has two data sets that report state Medicaid expenditures, but the
data sets have different purposes and limitations. Consequently, neither
data set provides a complete picture of Medicaid expenditures.

•   The Medicaid Statistical Information System (MSIS) was established
    as a national eligibility and claims data set, and can provide CMS a
    summary of expenditures linked to specific beneficiaries on the basis
    of their medical claims for care. 7 CMS reviews these data for
    reliability, and uses these data for policy analysis, program utilization,
    and forecasting expenditures. However, these data exclude other
    aspects of the Medicaid program that are not tied to specific
    beneficiaries. For example, the MSIS data set does not contain
    supplemental payments to providers that are separate from standard
    Medicaid payments for services.

•   The CMS-64 data set aggregates states’ expenditures, which are
    used to reimburse the states for their federal shares of Medicaid
    expenditures. 8 CMS reviews these submissions, and the data are the


5
 Our list identifies areas that are at high risk because of their greater vulnerabilities to
waste, fraud, abuse, and mismanagement or major challenges associated with their
economy, efficiency, or effectiveness.
6
 See GAO, High-Risk Series: An Update, GAO-11-278 (Washington, D.C.: February
2011).
7
 States are required to have in operation a mechanized claims-processing and
information-retrieval system based on certain federal requirements. See 42 U.S.C.
§ 1396b(r). For all claims filed on or after January 1, 1999, states have been required to
electronically transmit claims data, including detailed individual enrollee encounter data in
a format consistent with MSIS. See 42 U.S.C. § 1396b(r)(1)(F).
8
  For purposes of this report, we refer to form CMS-64 as “CMS-64.” The form CMS-64 is
titled the Quarterly Medicaid Statement of Expenditures for the Medical Assistance
Program. The information is stored in a data set called the Medicaid Budget and
Expenditure System. States are required to submit aggregate total quarterly Medicaid
expenditures on the form CMS-64 no later than 30 days after the end of each quarter.
42 C.F.R. § 430.30 (2011).




Page 2                                                                GAO-13-47 Medicaid Data
    most-reliable accounting of total Medicaid expenditures. However,
    these data exclude beneficiary-specific data and thus have limited use
    in examining program spending. For example, CMS cannot use
    CMS-64 data to conduct a beneficiary-level analysis to identify
    program spending abuses in particular service areas, such as
    prescription drugs. 9

Total Medicaid expenditure data, nationally and by state, often differ
widely between MSIS and CMS-64, even after accounting for differences
between the two in the purpose of the database and the type of
information provided. Members of Congress, industry experts, and
researchers have noted that these two sets of expenditure data cannot be
easily reconciled, thus limiting CMS oversight. For example, differences in
expenditure data between the two data sets limit CMS’s ability to conduct
cross-state comparisons of Medicaid spending. Such comparisons could
be used to analyze Medicaid spending patterns by eligibility group and
other enrollee characteristics that could be useful in improving the
financial integrity of this high-risk program. To better understand the
strengths and limitations of federal data on Medicaid expenditures, you
requested that we study the similarities and differences between MSIS
and CMS-64. In this report, we (1) examine the extent to which MSIS and
CMS-64 data on Medicaid expenditures differ nationally, by state, and by
expenditure category for fiscal years 2007 through 2009; and (2) where
possible, quantify identified differences between these two data sets.

To determine the extent to which data on Medicaid expenditures differ,
we examined CMS data on total Medicaid expenditures, as reported by
states in MSIS and CMS-64, nationally, by state, and by expenditure
category for fiscal years 2007 through 2009. 10 We used these fiscal years
because they were the most-recent and most-complete data available at
the time of our analysis. We examined total MSIS and CMS-64
expenditures, nationally and by state, by comparing the expenditures
reported by states in MSIS as a percentage of those reported in CMS-64.


9
 See GAO, Foster Children: HHS Guidance Could Help States Improve Oversight of
Psychotropic Prescriptions, GAO-12-201 (Washington, D.C.: Dec. 14, 2011).
10
  In this report, we use the term “state” to refer to the 50 states and the District of
Columbia. We do not include Medicaid administrative expenditures, state collections, or
expenditures from Puerto Rico or the U.S. territories of American Samoa, Guam, the
Northern Mariana Islands, or the U.S. Virgin Islands (which have Medicaid programs)
because they were not reported in MSIS during fiscal years 2007 through 2009.




Page 3                                                           GAO-13-47 Medicaid Data
We provide results for fiscal years 2007 through 2009, and provide
additional detail for the most recent year available, fiscal year 2009. We
also combined the individual service types in MSIS and CMS-64 into six
combined expenditure categories. 11 This was necessary because, in
many instances, there is not a one-to-one match of service types in MSIS
and CMS-64. As a result, we were able to report on expenditures by
combined expenditure categories, and thereby better identify the factors
that account for the differences between MSIS and CMS-64. We
compared expenditures by combined expenditure categories nationally
and by state.

To examine and, where possible, quantify the identified differences
between MSIS and CMS-64 data, we identified and analyzed differences
in the types of expenditures included in each data set. Where possible,
we adjusted the MSIS expenditures on the basis of the differences we
identified, adding expenditures reported in CMS-64 that were not reported
in MSIS to obtain total adjusted expenditures for MSIS. We then took the
total adjusted expenditures nationally, and by state, and calculated the
expenditures reported in MSIS as a percentage of those reported in
CMS-64. We also compared adjusted expenditures by combining
expenditure categories, nationally and by state. We identified additional
factors that accounted for differences between the two data sets, but
could not be quantified.

For both objectives, we reviewed relevant guidance and documentation,
including CMS forms and data dictionaries, and also interviewed CMS
officials and other experts familiar with Medicaid expenditure data. We
reviewed the data for reasonableness and consistency, including
screening for missing data, outliers, and obvious errors. We also
interviewed CMS officials about steps they take to ensure data reliability.
We determined that these data were sufficiently reliable for our purposes.
We also interviewed a judgmental sample of state Medicaid agencies to
help illustrate specific factors that account for the differences in reported
Medicaid expenditures. We chose the judgmental sample on the basis of


11
  While we are matching services between these two data sets, some of the expenditures
we describe here are not directly tied to a Medicaid service in MSIS, such as Medicaid
payments for Medicare premiums. Other examples of individual service types are inpatient
hospital, nursing facility, transportation, and rural health clinic screenings. Our six
combined expenditure categories are (1) hospital, (2) acute and long-term support
services—noninstitutional, (3) drugs, (4) managed care and Medicaid premium
assistance, (5) long-term support services—institutional, and (6) Medicare.




Page 4                                                         GAO-13-47 Medicaid Data
             a variety of criteria using the most-recent and most-complete data
             available, including the size of their total adjusted Medicaid expenditures,
             adjusted MSIS expenditures as a percentage of CMS-64 expenditures,
             and geographic variation. A detailed discussion of our scope and
             methodology is presented in appendix I.

             We conducted this performance audit from March 2012 through
             September 2012 in accordance with generally accepted government
             auditing standards. Those standards require that we plan and perform the
             audit to obtain sufficient, appropriate evidence to provide a reasonable
             basis for our findings and conclusions based on our audit objectives. We
             believe that the evidence obtained provides a reasonable basis for our
             findings and conclusions based on our audit objectives.


             Medicaid is a health care program jointly funded by the federal
Background   government and states to provide care for certain low-income individuals.
             The federal government oversees states’ Medicaid programs, and
             typically pays from 50 to 83 percent of each state’s allowable Medicaid
             costs. 12 Medicaid enrollees are entitled to receive a range of medical
             services, including hospital care, physician services, laboratory and other
             diagnostic tests, prescription drugs, dental care, and long-term care
             services. In addition, Medicaid provides assistance to low-income elderly
             individuals who are also eligible for Medicare, called “dual eligibles.” 13
             This assistance can include covering Medicare premiums and cost
             sharing.




             12
               In general, state Medicaid spending is matched by the federal government at a rate that
             is based, in part, on each state’s per capita income according to a formula established by
             law. The federal share of Medicaid expenditures, known as the federal medical assistance
             program, typically ranges from 50 to 83 percent. The federal government pays a larger
             portion of Medicaid expenditures in states with low per-capita income relative to the
             national average, and a smaller portion for states with higher per-capita incomes. In fiscal
             year 2012, the largest federal portion was 74.18 percent.
             13
               Medicare is the federal health insurance program that covers seniors aged 65 and older,
             and some disabled persons.




             Page 5                                                            GAO-13-47 Medicaid Data
Medicaid Expenditure   MSIS is a national Medicaid eligibility and claims data set, and is the
Data                   federal source of Medicaid expenditure data that can be linked to a
                       specific enrollee. State Medicaid agencies are required to provide CMS,
                       through MSIS, quarterly electronic files approximately 45 days after a
                       quarter has ended. These files contain: (1) persons covered by Medicaid,
                       known as “eligible files”; and (2) adjudicated claims, known as the “paid
                       claims file,” for medical services reimbursed by the Medicaid program.
                       Each state’s eligible file contains one record for each person covered by
                       Medicaid for at least 1 day during the reporting quarter. Individual eligible
                       files consist of demographic and monthly enrollment data. Paid claims
                       files contain information on medical service-related claims and capitation
                       payments. 14

                       MSIS data include enrollees’ eligibility status for Medicaid and the
                       Children’s Health Insurance Program (CHIP), types of services received
                       by enrollees, and expenditure data. MSIS data are used for policy
                       analysis, program utilization, and forecasting expenditures. However,
                       MSIS data are not used to determine the federal share of Medicaid
                       expenditures, and are not used by the states to manage the daily
                       operations of their Medicaid programs.

                       The CMS-64 data set contains program-benefit costs and administrative
                       expenses that are not linked to individual enrollees. State Medicaid
                       agencies submit this information 30 days after a quarter has ended by
                       means of the Quarterly Medicaid Statement of Expenditures for the
                       Medical Assistance Program—also known as the form CMS-64—within
                       the Medicaid Budget and Expenditure System. CMS-64 data are reported
                       at a state aggregate level, such as a state’s total expenditures for such
                       categories as inpatient hospital services and prescription drugs.
                       Therefore, unlike MSIS, these data do not include individual expenditure
                       data on the state’s enrollees or the services they received under
                       Medicaid. Also unlike MSIS, CMS-64 contains expenditures that are not
                       linked to specific enrollees, such as supplemental payments including
                       Disproportionate Share Hospital (DSH) payments. 15 CMS-64 data are the


                       14
                         Capitation payments are specified amounts of money paid to a health plan or doctor.
                       These payments are used to cover the cost of a health plan member’s health care
                       services for a certain length of time.
                       15
                         DSH payments are required to be made to hospitals that serve a disproportionate share
                       of low-income and Medicaid patients to help offset hospitals’ uncompensated costs for
                       serving these individuals. See 42 U.S.C. §§ 1396a(a)(13)(A)(iv) and 1396r-4.




                       Page 6                                                          GAO-13-47 Medicaid Data
                        most-reliable and most-comprehensive information on Medicaid
                        spending. Agency officials review expenditures submitted through
                        CMS-64, and use the data to compute the federal financial participation
                        for each state’s Medicaid program costs.


Medicaid: A High-Risk   Our reports have demonstrated the need for CMS to improve its oversight
Program                 of this growing, complex program. In particular, federal internal-control
                        standards, as documented in GAO’s Standards for Internal Control in the
                        Federal Government, state that program managers need both operational
                        and financial data to determine whether they are meeting their goals for
                        accountability and efficient use of resources in order to make operating
                        decisions, monitor performance, and allocate resources. Pertinent
                        information should be identified, captured, and distributed in a form and
                        time frame that permits people to perform their duties efficiently. 16 These
                        reports have also identified shortcomings with the MSIS and CMS-64
                        data sets, particularly in two areas: Medicaid program integrity and
                        supplemental payments. 17

                        We recently issued a report that found that the majority of CMS’s
                        Medicaid Integrity Group audits through February 2012 relied on MSIS
                        data, although CMS officials told us that they recently reduced their
                        reliance on MSIS data for these audits. 18 We noted that MSIS is an
                        extract of states’ claims data and is missing key elements, such as
                        provider names, that are necessary for auditing. Furthermore, we found
                        that the median amount of the potential overpayment identified in MSIS
                        was relatively small compared with other types of audits that used state-
                        based data sets. We recommended, and HHS partially concurred, that
                        the CMS Administrator ensure that the program-integrity group’s update
                        of its comprehensive plan provide key details about the audit program,




                        16
                         GAO, Internal Control: Standards for Internal Control in the Federal Government,
                        GAO/AIMD-00-21.3.1 (Washington, D.C.: November 1999).
                        17
                          In addition to studies regarding Medicaid program integrity and supplemental payments,
                        as discussed below, our high-risk report also indicated that Medicaid demonstration
                        projects were an area of concern. See GAO-11-278.
                        18
                          See GAO, National Medicaid Audit Programs: CMS Should Improve Reporting and
                        Focus on Audit Collaboration with States, GAO-12-627 (Washington, D.C.: June 14,
                        2012).




                        Page 7                                                          GAO-13-47 Medicaid Data
                      including its expenditures and audit outcomes, program improvements,
                      and plans for effectively monitoring the program.

                      We also recently reported that the accountability and transparency of
                      supplemental payments have been lacking using CMS-64. 19 Specifically,
                      our work found that states reported $32 billion in DSH and non-DSH
                      Medicaid supplemental payments during fiscal year 2010. However, the
                      exact amount of supplemental payments is unknown using CMS-64
                      because not all states reported their non-DSH supplemental payments
                      separately from their regular payments. In an earlier report, we also found
                      that information on non-DSH supplemental payments was incomplete
                      because states did not provide full information to CMS regarding these
                      payments. We noted that until reliable and complete information on
                      states’ supplemental payments is available, federal officials overseeing
                      the program and others will lack the information they need to review
                      payments and ensure that they are appropriately spent for Medicaid
                      purposes.


                      MSIS Medicaid expenditure amounts nationwide were generally less than
MSIS Medicaid         CMS-64 amounts. For fiscal years 2007, 2008, and 2009, total
Expenditures          expenditures based on MSIS data for the nation were 86, 87, and
                      88 percent, respectively, of the amounts shown in CMS-64. In fiscal years
Amounts Are           2007 through 2009, the difference in Medicaid expenditures between the
Generally Less than   two data sets decreased from about $46 billion in fiscal year 2007 to
CMS-64 Expenditure    $43 billion in fiscal year 2009. For fiscal year 2009, the most-recent and
                      most-complete data available, MSIS showed $323 billion in total
Amounts               expenditures compared with the $366 billion in CMS-64, a difference of
                      $43 billion. (See table 1.)




                      19
                        Supplemental payments are payments separate from and in addition to those made at a
                      state’s standard Medicaid payment rate and include DSH payments. DSH payments are
                      made to hospitals that treat large numbers of Medicaid and uninsured individuals. Other
                      supplemental payments, or non-DSH payments, are made to providers above the
                      standard Medicaid payment rates but within the Upper Payment Limit, which is the
                      estimated amount that Medicare pays for comparable services. States have made non-
                      DSH supplemental payments to hospitals, nursing homes, physician groups, and other
                      Medicaid providers. See GAO, Medicaid: States Reported Billions More in Supplemental
                      Payments in Recent Years, GAO-12-694 (Washington, D.C.: July 20, 2012) and Medicaid:
                      CMS Needs More Information on the Billions of Dollars Spent on Supplemental Payments,
                      GAO-08-614 (Washington, D.C.: May 30, 2008).




                      Page 8                                                         GAO-13-47 Medicaid Data
Table 1: Total Medicaid Expenditures in the Medicaid Statistical Information System (MSIS) and CMS-64, Fiscal Years 2007-
2009

                                                                                               Difference between              MSIS baseline as a
                                                                                                MSIS baseline and                  percentage of
Fiscal year   MSIS baseline (dollars)    CMS-64 baseline (dollars)                        CMS-64 baseline (dollars)             CMS-64 (percent)
2007                $273,925,484,790                  $320,052,492,375                                   $46,127,007,585                      86%
2008                 294,167,348,289                    338,552,036,761                                    44,384,688,472                       87
2009                 323,120,029,363                    366,486,147,093                                    43,366,117,730                       88
                                         Source: GAO analysis of Centers for Medicare & Medicaid Services’ data.



                                         MSIS Medicaid expenditures for individual states were generally less than
                                         CMS-64 amounts. In fiscal years 2007 through 2009, states’ MSIS
                                         Medicaid expenditures ranged from 59 percent to 120 percent compared
                                         with CMS-64. In fiscal year 2009 alone, states’ MSIS Medicaid
                                         expenditures ranged from 59 to 119 percent of those in CMS-64. (See
                                         fig. 1.) Specifically, MSIS Medicaid expenditures were less than CMS-64
                                         amounts in 40 states. These expenditures were greater than CMS-64
                                         expenditures in 6 states, and were similar to CMS-64 expenditures in
                                         5 states. 20 See appendix II for a table of total baseline state expenditures
                                         reported in MSIS and CMS-64 by dollar amount in fiscal year 2009.




                                         20
                                           For purposes of this report, we defined “similar” to include MSIS expenditures within 98
                                         to 102 percent of CMS-64 expenditures.




                                         Page 9                                                                             GAO-13-47 Medicaid Data
Figure 1: Medicaid Statistical Information System (MSIS) Medicaid Expenditures as a Percentage of CMS-64 Expenditures, by
State, Fiscal Year 2009




                                        Page 10                                                    GAO-13-47 Medicaid Data
We found differences in Medicaid expenditures between the two data sets
when we combined MSIS and CMS-64 services types into our six
combined expenditure categories: (1) hospital, (2) acute or long-term
support services (LTSS)-noninstitutional, (3) prescription drugs,
(4) managed care and Medicaid premium assistance, (5) LTSS-
institutional, and (6) Medicare. 21 For example, MSIS Medicaid
expenditure data in the hospital expenditure category were less than the
amounts shown in CMS-64, at 61, 62, and 60 percent, respectively, for
fiscal years 2007, 2008, and 2009. Conversely, MSIS Medicaid
expenditures related to the prescription-drug category were greater than
the amounts shown in CMS-64 data, at 147, 154, and 161 percent for the
same period. During this time, states did not report expenditures for
Medicare premiums in MSIS. However, expenditures associated with
Medicare coinsurance and deductibles were reported under individual
services throughout MSIS and therefore not identified as Medicare
expenditures. As a result, MSIS expenditures as a percent of CMS-64 in
the Medicare category are reported as 0 percent. (See fig. 2.)




21
  Medicaid provides assistance to low-income elderly individuals who are also eligible for
Medicare, called “dual eligibles.” This assistance can include covering Medicare premiums
and cost sharing.




Page 11                                                          GAO-13-47 Medicaid Data
Figure 2: MSIS Medicaid Expenditures as a Percentage of CMS-64 Expenditures, by Expenditure Category, Fiscal Years 2007–
2009




                                        Note: LTSS stands for long-term support services.
                                        a
                                         The Medicare expenditure category is reported at $0 in the Medicaid Statistical Information System
                                        (MSIS). During this period, states did not report expenditures for Medicare premiums in MSIS.
                                        However, expenditures associated with Medicare coinsurance and deductibles were reported under
                                        individual services throughout MSIS and therefore not identified as a Medicare expenditure.


                                        For fiscal year 2009, Medicaid expenditures in MSIS were lower than
                                        CMS-64 for four of the six expenditure categories: hospital, acute and
                                        LTSS-noninstitutional, LTSS-institutional, and Medicare. The hospital,
                                        prescription drugs, and Medicare categories showed the largest
                                        difference between what was reported in MSIS compared with CMS-64,
                                        for fiscal year 2009. In particular, Medicaid hospital expenditures in MSIS
                                        were 60 percent of those reported in CMS-64, in fiscal year 2009.
                                        Medicare expenditures were reported as $0 in MSIS, compared with
                                        $12 billion in CMS-64. For the remaining two categories, prescription drug
                                        expenditures in MSIS were 61 percent larger in MSIS than in CMS-64,
                                        while managed care and Medicaid premium assistance MSIS
                                        expenditures were similar to those in CMS-64. (See fig. 3.)




                                        Page 12                                                                  GAO-13-47 Medicaid Data
Figure 3: MSIS Medicaid Expenditures as a Percentage of CMS-64 Expenditures, by Expenditure Category, Fiscal Year 2009




                                        Note: LTSS stands for long-term support services.
                                        a
                                         The Medicare expenditure category is reported at $0 in the Medicaid Statistical Information System
                                        (MSIS). During this period, states did not report expenditures for Medicare premiums in MSIS.
                                        However, expenditures associated with Medicare coinsurance and deductibles were reported under
                                        individual services throughout MSIS and therefore not identified as a Medicare expenditure.



                                        Some—but not all—factors could be quantified to narrow the difference
Some Factors Could                      between MSIS and CMS-64 expenditures. In particular, we adjusted for
Be Quantified and                       expenditures that could not be attributed to individual beneficiaries—one
                                        of the key differences in the design of the data sets. However, we could
Accounted for                           not quantify the effect of other factors, such as inconsistent CMS
Approximately Half of                   guidance across the two data sets.
the Expenditure
Difference between
the Data Sets
Differences Attributable to             MSIS is designed to report claims data, and CMS-64 is designed to
the Design of the Data Sets             reimburse states for their federal share of Medicaid expenditures. As we
Could Be Quantified and                 have noted, some expenditures that are required to be reported in
                                        CMS-64 do not appear in MSIS, such as when the expenditure is not tied
Adjusted                                to an individual enrollee’s claim. After adjusting the MSIS data to include
                                        expenditures for factors not related to individual enrollees’ claims,
                                        Medicaid expenditures for the nation based on MSIS data were 92, 93,
                                        and 94 percent of amounts shown in CMS-64 data, respectively, for fiscal
                                        years 2007, 2008, and 2009. (See table 2.)




                                        Page 13                                                                  GAO-13-47 Medicaid Data
Table 2: Total Medicaid Expenditures Adjusted in the Medicaid Statistical Information System (MSIS) and CMS-64, Fiscal
Years 2007–2009

                                                                                                    Difference between                   MSIS baseline
                                                                                                    MSIS adjusted and                as a percentage of
Fiscal year     MSIS adjusted (dollars)         CMS-64 baseline (dollars)                      CMS-64 baseline (dollars)              CMS-64 (percent)
2007                  $294,595,771,198                       $320,052,492,375                                  $25,456,721,177                     92%
2008                   315,779,358,947                         338,552,036,761                                  22,772,677,814                       93
2009                   345,172,184,969                         366,486,147,093                                  21,313,962,124                       94
                                          Source: GAO analysis of Centers for Medicare & Medicaid Services’ data.



                                          For fiscal year 2009, we were able to adjust MSIS data for four factors:
                                          DSH payments, Medicare premiums, national and state rebates for
                                          prescription drugs, and Medicaid health insurance payments. None of
                                          these factors were reported in MSIS because CMS officials indicated they
                                          were not attributed to an individual enrollee.

                                          •     DSH payments were included under the hospital expenditure category
                                                in CMS-64, but not in MSIS. 22 We adjusted MSIS expenditure data to
                                                account for DSH payments to hospitals. For example, in fiscal year
                                                2009, states reported approximately $88 billion in total hospital
                                                expenditures on CMS-64, which included approximately $18 billion for
                                                DSH payments. Total hospital expenditures in MSIS were at about
                                                $53 billion. Adding the $18 billion in DSH payments to total hospital
                                                expenditures in MSIS increased the percentage of MSIS expenditures
                                                from 60 percent of CMS-64 expenditures to 81 percent. Even after
                                                this adjustment, MSIS hospital expenditures are $17 billion lower than
                                                those on CMS-64, indicating that there are additional factors that
                                                account for the difference in hospital-related expenditures between
                                                the two data sets. 23




                                          22
                                            In fiscal years 2007 through 2009, the CMS-64 included expenditures for DSH payments
                                          on line 1B for Inpatient Hospitals and 2B for Mental Health Facilities on the CMS-64, but
                                          these were not reported in MSIS.
                                          23
                                              Totals do not add up due to rounding.




                                          Page 14                                                                                GAO-13-47 Medicaid Data
•    Medicare premiums were included in CMS-64, but CMS did not
     require states to report them in MSIS. 24 In addition, Medicaid
     payments for enrollees with Medicare coinsurance and deductibles
     are included in MSIS, but within the various service types, and cannot
     be distinguished from other expenditures. 25 In fiscal year 2009, total
     expenditures for Medicare as reported on CMS-64 were
     approximately $11 billion, whereas MSIS expenditures for the
     Medicare category were $0. Adjusting for the approximately
     $11 billion reported in fiscal year 2009 for Medicare premiums in
     CMS-64 increased the percentage of total Medicare expenditures in
     MSIS from 0 percent of CMS-64 expenditures to 92 percent, or a
     difference of $908 million. Thus, the primary factor that accounted for
     the difference in Medicare expenditures between the two data sets
     can be attributed to the absence of Medicare premium expenditures in
     MSIS.

•    Prescription drug rebates were included in CMS-64, but CMS did not
     require states to report them in MSIS. 26 Prescription drug rebates are
     made by drug manufacturers to states in a lump sum payment for
     Medicaid enrollees who use specific drugs, and therefore are not
     connected to individual claims. In fiscal year 2009, total expenditures
     for the prescription drugs category, as reported on CMS-64, were
     initially about $25 billion. However, there was a reduction to $16 billion
     when $10 billion dollars in national and state prescription drug rebates



24
   In fiscal years 2007 through 2009, the CMS-64 included expenditures for Medicare on
lines 17A–D of the CMS-64, including rows for Medicare premiums, but the expenditures
were not reported in MSIS. See line 17A for Medicare Health Insurance Payments Part A
premiums, line 17B for Medicare Health Insurance Payments Part B premiums, line 17C1
for 120–134 percent of the poverty level, line 17C2 for 135–175 percent of the poverty
level.
25
  In fiscal years 2007 through 2009, while MSIS did not include Medicare premiums, both
CMS-64 and MSIS did include Medicaid expenditures for Medicare coinsurance and
deductibles. However, MSIS includes these expenditures within the various expenditure
types. Therefore, these expenditures cannot be tracked as a Medicare coinsurance or
deductible using the MSIS annual person summary file. As a result, since expenditures
from Medicare coinsurance and deductibles exist in MSIS, but are not captured in the
Medicare category, we did not adjust for this, as doing so would double count these
expenditures. In fiscal year 2009, the amount for Medicare coinsurance and deductibles in
CMS-64 was approximately $908 million.
26
  In fiscal years 2007 through 2009, the CMS-64 included expenditures for prescription
drug rebate offsets on rows 7A1 for national agreements and 7A2 for state side bar
agreements of the CMS-64, but these were not reported in MSIS.




Page 15                                                         GAO-13-47 Medicaid Data
      were included. 27 Total prescription drug expenditures in MSIS were
      approximately $25 billion. Adjusting for the $10 billion reported for
      prescription drug rebates in CMS-64 decreased the percentage of
      MSIS expenditures from 161 percent of CMS-64 expenditures to
      99 percent of CMS-64 expenditures. Thus, the difference in reported
      prescription drug expenditures can be almost entirely attributed to the
      rebates. 28

•     Some Medicaid health insurance payments were included in CMS-64
      that CMS did not require states to report in MSIS. 29 In fiscal year
      2009, total expenditures for the managed care and Medicaid premium
      assistance expenditure category, as reported on CMS-64, were
      approximately $82 billion, of which $3 billion were for Medicaid health
      insurance payments not reported in MSIS. Total managed care and
      Medicaid premium assistance expenditures, as reported in MSIS,
      were approximately $85 billion. Adjusting for the roughly $3 billion for
      Medicaid health insurance payments increased the percentage of
      MSIS expenditures from 100 percent of CMS-64 expenditures to more
      than 103 percent. Therefore, total managed care and Medicaid
      premium assistance expenditures in MSIS were greater than those
      reported in CMS-64. (Fig. 4 compares the MSIS baseline to the MSIS
      adjusted expenditures as a percentage of CMS-64, by expenditure
      category, for fiscal year 2009.)




27
    Totals do not add up due to rounding.
28
  CMS officials told us that some states did include rebates in their MSIS totals. We were
unable to adjust for these since MSIS is not set up to separate rebates from other reported
data.
29
  In fiscal years 2007 through 2009, the CMS-64 included expenditures for Medicaid
health insurance payments on rows 18C (Medicaid health insurance payments–Group
Health Plan Payments), 18D (Medicaid health insurance payments–Coinsurance and
Deductibles) and 18E (Medicaid health insurance payments–Other), but these were not
reported in MSIS.




Page 16                                                          GAO-13-47 Medicaid Data
Figure 4: Comparison of MSIS Baseline to Adjusted Expenditures, by Expenditure Category, as a Percentage of CMS-64,
Fiscal Year 2009




                                        a
                                         Adjustments were made by including reported CMS-64 dollars in the Medicaid Statistical Information
                                        System (MSIS).


                                        Overall, for fiscal year 2009, the difference in Medicaid expenditures, after
                                        adjustments, decreased from $43 billion to $21 billion nationally. As a
                                        result, MSIS expenditures as a percentage of CMS-64 for the nation
                                        increased 6 percentage points from 88 percent to 94 percent—thereby
                                        reducing the difference in reported expenditures by half. (See fig. 5.)




                                        Page 17                                                                 GAO-13-47 Medicaid Data
Figure 5: Comparison of MSIS Baseline to MSIS Adjusted Expenditures, as a Percentage of CMS-64, Fiscal Year 2009




                                        a
                                         Adjustments were made by including reported CMS-64 dollars in the Medicaid Statistical Information
                                        Systems (MSIS).

                                        After adjustments, MSIS Medicaid expenditures for individual states
                                        increased slightly, but generally remained less than CMS-64 amounts.
                                        For individual states, Medicaid expenditures in MSIS ranged from
                                        55 percent to 131 percent of CMS-64 in fiscal years 2007 through 2009
                                        after adjustments were made. For fiscal year 2009, 29 states reported
                                        lower expenditures in MSIS than in CMS-64, 9 states reported higher
                                        expenditures in MSIS than in CMS-64, and 13 states reported similar
                                        Medicaid expenditures in MSIS when compared with CMS-64. (See
                                        fig. 6.) See appendix III for a table of adjusted state expenditures reported
                                        in MSIS and CMS-64, by dollar amount, for fiscal year 2009. See
                                        appendix IV for a map showing the percentages of adjusted Medicaid
                                        expenditures in MSIS compared with CMS-64, by state, for fiscal year
                                        2009.




                                        Page 18                                                                 GAO-13-47 Medicaid Data
Figure 6: Comparison of MSIS Baseline to Adjusted Medicaid Expenditures, as a Percentage of CMS-64, by State, Fiscal Year
2009




                                        Page 19                                                    GAO-13-47 Medicaid Data
Note: Adjustments were made by including reported CMS-64 dollars in the Medicaid Statistical
Information System (MSIS).




Page 20                                                                GAO-13-47 Medicaid Data
Differences between MSIS and CMS-64 Medicaid expenditures in six
expenditure categories were also reduced after we adjusted the data in
fiscal years 2007 through 2009. Two of the six expenditure categories—
hospital and Medicare—continued to show the largest percentage
difference in Medicaid expenditures between MSIS and CMS-64 data.
After adjustments were made, MSIS Medicaid expenditures in the
hospital category were 80, 83, and 81 percent of the amounts shown in
CMS-64, respectively for fiscal years 2007, 2008, and 2009, compared
with 61, 62, and 60 percent before the adjustments were made.
Expenditures for Medicare increased from 0 percent to more than
90 percent for all 3 fiscal years when adjustments were made for
Medicare premiums. (See fig. 7.)




Page 21                                             GAO-13-47 Medicaid Data
Figure 7: Adjusted Medicaid Expenditures in MSIS as a Percentage of CMS-64, by Expenditure Category, Fiscal Years 2007–
2009




                                        Note: Adjustments were made by including reported CMS-64 dollars in the Medicaid Statistical
                                        Information System (MSIS). No adjustments were made for acute and long-term support services
                                        (LTSS)-noninstitutional and LTSS-institutional.



Inconsistent Guidance and               Inconsistent CMS guidance and states’ practices resulted in differences
States’ Practices Result in             between MSIS and CMS-64 data. However, we could not adjust for these
Differences That Could                  factors because we could not quantify the extent to which these factors
                                        resulted in differences.
Not Be Quantified
                                        Three of the factors that account for the differences in expenditures result
                                        from inconsistent CMS guidance regarding expenditure definitions,
                                        reporting dates, and reporting supplemental payments during fiscal years
                                        2007 through 2009.




                                        Page 22                                                              GAO-13-47 Medicaid Data
•    Inconsistent MSIS and CMS-64 Definitions: CMS guidance to states
     for reporting expenditures in MSIS and CMS-64 contained
     inconsistent definitions of services, making it impossible to do a one-
     for-one match of similar expenditures. 30 CMS officials indicated that
     the definitions used in guidance to states on completing the CMS-64
     are based on definitions included in federal regulations, 31 while MSIS
     definitions are based on separate CMS guidance that was not derived
     from the regulations. These definitions are sometimes inconsistent.
     CMS officials explained that, for example, expenditures for inpatient
     services, as reported by a state in CMS-64, cannot be assumed to be
     the same services reported by the state in MSIS, despite the service
     having the same name in both data sets. Therefore, expenditures can
     be measured and reported by a state inconsistently across the data
     sets, making comparisons problematic. Specifically, even after
     adjustments were made, 13 states reported hospital-related
     expenditures in MSIS that were 75 percent or less of what they
     reported in their CMS-64 data.

     Additionally, inconsistent definitions across the two data sets make it
     difficult to examine Medicaid spending. For example, states may vary
     in their interpretation of the definitions in MSIS and CMS-64 guidance,
     and this inconsistency creates a challenge for examining total
     expenditures by service nationally and across states. (See app. V for
     reported expenditures by service, for all 50 states and the District of
     Columbia in fiscal year 2009.)

•    Reporting Dates Differed: CMS officials recognized that CMS
     guidance to states for reporting the date of expenditures and adjusting
     payments made in prior years is different between MSIS and CMS-64.
     MSIS expenditures were based on each claim’s date of adjudication—
     the date the payment was approved—while CMS-64 expenditures
     were reported as of the date of payment. The two data sets also differ
     in how they adjust payments made in prior years, according to CMS
     officials. Adjustments in MSIS allow the state to show changes to net



30
  MSIS Guidance: The CMS MSIS File Specifications and Data Dictionary provides
information, including definitions, for MSIS reporting. CMS-64 Guidance: The Category
of Service Line Definitions for the 64.9 provides information, including definitions, for
CMS-64 reporting of net expenditures.
31
  In general, service type definitions used for reporting expenditures on the CMS-64 are
derived from federal regulation. See 42 C.F.R. part 440. (2011).




Page 23                                                            GAO-13-47 Medicaid Data
      totals for the year, no matter when the adjustment was reported into
      the data. In contrast, adjustments to CMS-64 were entered for the
      fiscal year in which they were identified and the original year was not
      modified. MSIS and CMS-64 expenditures, therefore, may vary by
      quarter and by fiscal year if the dates associated with an expenditure
      differ. Since MSIS and CMS-64 report expenditures using different
      dates, we cannot adjust for this difference in the data. CMS officials
      indicated their concerns with this inconsistency and noted that it
      makes comparing expenditure data between the two data sets
      difficult.

•     Supplemental Payments: For CMS-64 data, CMS officials required
      states to report supplemental payments within the corresponding
      expenditure categories, because states were not required to report
      these expenditures separately. 32 However, CMS officials indicated
      states reported supplemental payments in MSIS as a lump sum, and
      therefore they were excluded from the MSIS annual person summary
      file used in our analysis. 33 Therefore, supplemental payments,
      excluding DSH payments, were reported inconsistently between the
      two data sets during fiscal years 2007 through 2009. 34 As such, MSIS
      expenditure data are likely to show a lower amount than CMS-64 for
      several Medicaid expenditure categories. GAO and others have
      reported concerns about supplemental payments over the last
      decade, including the use of supplemental payment arrangements to
      increase federal funding without a commensurate increase in state
      funding. 35 Absent improved reporting, CMS cannot adequately
      oversee states’ use of supplemental payments.

Three additional factors that accounted for the differences in expenditures
result from inconsistent practices among states. Specifically, states are
inconsistent in MSIS compared with CMS-64 regarding the timeliness of



32
  CMS added several new lines to the form CMS-64 in fiscal year 2010, including lines for
states to report supplemental payments within each category of care for several
expenditure types.
33
  CMS officials indicated that, on a rare occasion, states may report supplemental
payments at the individual claims level. Overall, states report few supplemental payments
in MSIS.
34
    See GAO-12-694.
35
    See GAO-11-278.




Page 24                                                          GAO-13-47 Medicaid Data
MSIS data, the quality of MSIS data reported, and payments for local
government providers during fiscal years 2007 through 2009.

•    Timeliness of MSIS data: States often delay reporting MSIS data, but
     report CMS-64 expenditure data on time. CMS guidance requires
     states to report data in CMS-64 within 30 days of the end of the
     quarter and within 45 days for MSIS. However, states have often been
     late reporting MSIS data, with some states delaying reporting of MSIS
     data for as long as 3 years. For example, as of July 2012, 37 states
     had submitted their fiscal year 2010 MSIS data, even though CMS
     requirements would indicate that MSIS data for 2010, 2011, and the
     first two quarters of 2012 should have already been provided.
     Alternatively, states submit their expenditures by means of CMS-64
     on time and have a strong incentive to do so promptly because CMS
     uses this information to reimburse states for the federal share of
     Medicaid expenses. In contrast, states have less incentive to submit
     MSIS data promptly because the data are not tied to their federal
     reimbursement, and MSIS data are not used in the daily operations of
     their Medicaid programs. CMS officials told us that states that delay
     reporting can have issues with both the timeliness and quality of their
     submissions. If states submit poor quality data, CMS may reject the
     submission, resulting in further delays. 36

•    MSIS Data Quality: MSIS expenditure data are considered less
     reliable when compared with CMS-64. CMS-64 is reviewed regularly
     by CMS officials and considered the most-reliable and most-
     comprehensive data on Medicaid expenditures. Alternatively, while
     CMS conducts routine quality checks on MSIS data, problems still
     remain. CMS publishes a report of anomalous MSIS claims—those
     that are not consistent with what is expected to be reported by the
     state. 37 On the basis of our interviews with state Medicaid officials, the
     quality of MSIS data, as indicated by an anomalous claim, may be an
     additional factor that accounts for the difference between MSIS and
     CMS-64 reported expenditures. For example, Medicaid officials in one
     state we interviewed told us that the state incorrectly reported Home &



36
  CMS officials indicated that some states have also not reported their data in prior years.
Additionally, the MSIS delay did not affect our ability to analyze the data because we used
the most-recent and most-complete data available, fiscal years 2007 through 2009.
37
  The MSIS State Data Characteristics/Anomalies Report identifies MSIS data that are not
reconciled after quality checks are completed between CMS and the state.




Page 25                                                           GAO-13-47 Medicaid Data
     Community-Based 1915(c) waivers payments in MSIS in fiscal year
     2009, and therefore CMS noted this in the anomalies report. 38 As a
     result of this error, the state underreported its MSIS expenditures for
     Home & Community-Based services compared with those reported on
     CMS-64. Having more timely data could be a mechanism for helping
     identify anomalies between the two systems and correcting
     discrepancies earlier.

•    Local Government Provider Payments: CMS officials indicated that
     states may report in MSIS the amount paid to the local government for
     the medical service reported on the claim. However, the local
     government provider may also receive payment from the local
     government through prior alternative arrangements, and thus would
     not link these expenditures to a specific service provided or even
     report them at all in MSIS. These expenditures arise when states and
     localities finance Medicaid services through certified public
     expenditures. 39 In contrast, these additional expenditures would be
     captured in the total expenditures reported by states in CMS-64.
     Consequently, states may report a smaller amount in MSIS
     associated with the claim than reported by the state in total Medicaid
     expenditures on CMS-64. CMS officials indicated that neither CMS-64
     nor MSIS provides a link to expenditures from government providers;
     as a result, we were unable to adjust for this difference.

CMS has indicated plans to improve MSIS through a pilot program called
Transformed-MSIS. There are a number of goals for this pilot, including
plans to establish MSIS as a robust, flexible, and repeatable data-
collection process, to collect valid, timely information for decision making,
and to integrate data between management information systems. These
goals include a link between states’ reporting of MSIS with reporting of
CMS-64 data. However, CMS officials have indicated that there is no



38
  The 1915(c) waivers are one of many options available to states to allow the provision of
long-term care services in home and community-based settings under the Medicaid
program. States’ programs can offer a combination of standard medical services, and
nonmedical services, including homemaker, home health aide, personal care, adult day
health services, habilitation, and respite care. See Social Security Act §1915(c), codified
at 42 U.S.C. § 1396n(c).
39
  Under a Certified Public Expenditure arrangement, a government provider, such as a
county hospital, certifies to a state the amount of expenditures for a Medicaid-covered
service provided to a Medicaid beneficiary. The state obtains federal Medicaid matching
funds based on the amount of the expenditure.




Page 26                                                          GAO-13-47 Medicaid Data
               timeline for implementing this goal, whereas other aspects of the initiative
               are expected to be implemented in all states by fiscal year 2014.
               Moreover, such integration will only be of value if improvements are made
               in the timeliness and quality of MSIS data. One state in the pilot indicated
               that data-quality issues remain prevalent. Additionally, this state noted
               that the size and complexity of MSIS creates challenges with
               submissions’ timeliness and availability. As of June 2012, CMS officials
               indicated that the initial results for the pilot program are being considered
               by management, yet until these plans have been fully implemented it is
               unclear what the outcome of these efforts will be.


               In fiscal year 2009, the difference between MSIS and CMS-64 was
Concluding     $43 billion. Much of the difference was primarily the result of the different
Observations   designs of each data set. CMS uses MSIS data for beneficiary-specific
               expenditures, while CMS-64 data are used to compute the federal
               financial participation for each state’s Medicaid program costs. However,
               even after adjusting for DSH payments, Medicare premiums, prescription
               drug rebates, and Medicaid health insurance payments, differences
               remain. In fiscal year 2009, total MSIS expenditure data, after
               adjustments, showed MSIS at 94 percent of CMS-64 expenditures, which
               left billions of dollars unexplained.

               The remaining differences between the two data sets are potentially
               explained by inconsistencies in CMS guidance and states’ reporting
               practices, neither of which can be quantified. In fiscal years 2007 through
               2009, CMS provided states with inconsistent MSIS and CMS-64 guidance
               regarding expenditure definitions, reporting dates, and reporting of
               supplemental payments. Additionally, when compared with CMS-64, state
               MSIS data were often delayed beyond the time frames established by
               CMS, inconsistent in reporting payments for local government providers,
               and were of poor quality.

               Taken together, these two data sets have the potential to offer a robust
               view of the Medicaid program, enhancing CMS oversight of aggregate
               spending trends, per beneficiary spending growth, and cross-state
               comparisons, all of which could be useful in improving the financial
               integrity of this high-risk program. This is critical given that Medicaid, a
               program that GAO identified on our high-risk list, has among the highest
               estimated improper payments of any federal program reporting such data.
               However, the usefulness of these data sets as oversight tools is limited
               because of delays in reporting and unnecessary inconsistencies between
               the two data sets, both of which are inconsistent with federal internal


               Page 27                                                 GAO-13-47 Medicaid Data
                  control standards. The 3-year lag in states’ reporting of MSIS data
                  prevents its use for timely oversight of beneficiary-related utilization and
                  other spending trends. For example, identifying a difference in hospital
                  expenditures between MSIS and CMS-64 is of limited use when detected
                  3 years later. If states were meeting the current requirement of providing
                  MSIS data 45 days after each quarter, then such comparisons could
                  provide more useful and timely information.

                  CMS has recently completed a pilot study aimed in part on improving
                  MSIS data. CMS has indicated that it will begin implementing aspects of
                  this initiative in all states by fiscal year 2014. One goal of this initiative is
                  to integrate state reporting of MSIS with the reporting of CMS-64 data.
                  However, CMS officials have indicated they have yet to determine a
                  timeline for this goal. While the initial results of this pilot have not been
                  finalized, improving the timeliness and consistency of MSIS and CMS-64
                  data through this effort could aid CMS’s understanding and oversight of
                  this high-risk program.


                  HHS reviewed a draft of this report and provided technical comments,
Agency Comments   which we incorporated as appropriate.


                  As agreed with your office, unless you publicly announce the contents of
                  this report earlier, we plan no further distribution until 30 days from the
                  report date. At that time, we will send copies of this report to the
                  Secretary of Health and Human Services, the Administrator of CMS, and
                  other interested parties. In addition, the report will be available at no
                  charge on the GAO website at http://www.gao.gov.




                  Page 28                                                    GAO-13-47 Medicaid Data
If you or your staff have any questions about this report, please contact
me at (202) 512-7114 or yocomc@gao.gov. Contact points for our Offices
of Congressional Relations and Public Affairs may be found on the last
page of this report. GAO staff who made key contributions to this report
are listed in appendix VI.

Sincerely yours,




Carolyn L. Yocom
Director, Health Care




Page 29                                              GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology
             Appendix I: Scope and Methodology




             We conducted two types of analysis of Medicaid expenditures for this
             report to compare the Medicaid Statistical Information System (MSIS) and
             CMS-64 data. First, we conducted a baseline analysis to compare total
             expenditures reported to MSIS and CMS-64 nationally and by state. We
             also compared reported expenditures by expenditure category. Secondly,
             we conducted an adjustment analysis. Specifically, we attempted to
             reconcile the expenditures by making adjustments to MSIS on the basis
             of the differences we identified in the MSIS and CMS-64 data sets. To the
             extent possible, we added expenditures reported in CMS-64, but not
             reported in MSIS, to total expenditures in MSIS.

             I. Baseline Analysis

             To determine the extent to which MSIS and CMS-64 data sets on
             Medicaid expenditures differ, nationally and by state, 1 for fiscal years
             2007 through 2009, we conducted a multistep analysis. 2 With total
             expenditures from both data sets, we calculated the expenditures
             reported in MSIS as a percentage of those reported in CMS-64. We also
             determined the percentage difference in total expenditures between the
             two data sets, by expenditure category.

             Step 1: Obtain MSIS and CMS-64 Data

             In order to identify the baseline of total Medicaid expenditures for fiscal
             years 2007 through 2009, we obtained from the Center for Medicare &
             Medicaid Services (CMS) MSIS and CMS-64 data for each of the 50
             states and the District of Columbia. 3




             1
              In this report, we use the term “state” to refer to the 50 states and the District of
             Columbia. We do not include Puerto Rico or the U.S. territories of American Samoa,
             Guam, the Northern Mariana Islands, or the U.S. Virgin Islands (which have Medicaid
             programs), Medicaid administrative expenses, or state collections in the CMS-64 because
             they were not reported in MSIS during fiscal years 2007 through 2009.
             2
              We used these fiscal years because they were the most-recent and most-complete data
             available at the time of our analysis.
             3
              We define the baseline to be our initial calculations of total Medicaid expenditures as
             reported by CMS-64 and MSIS for all 50 states and the District of Columbia. Therefore,
             the baseline does not include any adjustments to the data. There is a baseline
             expenditure amount for CMS-64 and another for MSIS. Additionally, U.S. territories were
             excluded from our analysis because they do not report any Medicaid claims in MSIS.




             Page 30                                                          GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology




•   MSIS data: We obtained the MSIS Annual Person Summary File data,
    which CMS officials confirmed were appropriate for our analysis. The
    summary file includes monthly enrollment data, but only includes
    annual expenditure data by type of service. 4 It is most comparable to
    CMS-64 because both data sets provide total expenditures for the
    fiscal year. The summary file also includes some enrollment and
    expenditure information on the Children’s Health Insurance Program
    (CHIP). 5

•   CMS-64 data: We used the CMS-64 net expenditures financial
    management report within the Medicaid Budget and Expenditure
    System. The financial management report is an annual account of
    states’ program and administrative Medicaid expenditures, including
    federal and state expenditures by expenditure category. The financial
    management report also contains information solely related to
    Medicaid.

Step 2: Determine Medicaid Expenditures in MSIS for the Fiscal Year on
the Basis of the Number of Medicaid Enrollees

Because MSIS includes both Medicaid and CHIP expenditures, we
separated these expenditures to the extent possible. To do this, we first
determined whether a beneficiary was eligible for Medicaid or CHIP by
using the monthly eligibility code in MSIS. 6 The Annual Person Summary
File ties an entire month’s expenditures to an individual on the basis of his


4
 MSIS includes some lump sum expenditures that are not captured using the annual
person summary file.
5
 MSIS data are available in several file formats, including (1) full files with detailed claims
and enrollment data by individual, such as encounters, name of provider, and type of care
as reported directly by the states; (2) annual person summary files of expenditure data,
which summarize all the data for a person for the fiscal year, but exclude some encounter
details included in the full file (for example, the summary file may not include details
regarding the care encounter, such as individual cost per encounter; however, it does
include monthly enrollment data); (3) Medicaid Analytic eXtract (MAX) files, which are
considered to have the highest-quality data, but also have the greatest delay in
availability—up to 3 years; and (4) the State Summary DataMart, which is a public online
data system that provides state aggregated Medicaid claims, but not by the individual
person.
6
 The monthly eligibility categories include both Medicaid and CHIP. The eligibility
categories and their MSIS coding are: 0=ineligible, 1=Medicaid, 2=Medicaid Expansion
CHIP (M-CHIP), 3=CHIP non-Medicaid (also known as S-CHIP), and 4=CHIP unknown
(also interpreted to mean that the eligibility overall is unknown).




Page 31                                                              GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology




or her monthly eligibility code. The next step was to remove, to the extent
possible, the CHIP expenditures in order to match MSIS expenditures
with CMS-64. (CMS-64 does not include CHIP expenditures.)

CMS officials indicated that states were able to report CHIP expenditures
within MSIS if they had a Medicaid expansion-CHIP program (i.e., a CHIP
program that operates as part of the Medicaid program), also known as
M-CHIP. 7 Consequently, we were able to distinguish between Medicaid
and M-CHIP spending. 8 We thus removed from all MSIS totals any
M-CHIP expenditures to the extent possible. We determined the number
of months in the fiscal year a beneficiary received benefits under
Medicaid or M-CHIP. On the basis of this count, we prorated the
enrollees’ total expenses for the fiscal year on the basis of the proportion
of the year a person was enrolled in either M-CHIP or Medicaid. If a
person was enrolled in the M-CHIP program for part of the year, then a
portion of the annual spending was apportioned toward M-CHIP. For
example, if a person was enrolled in M-CHIP for 3 months, then a quarter
of the expenditures would be considered M-CHIP, and the remaining
three-quarters of expenditures would be considered Medicaid.
Consequently, the Medicaid expenditures include prorated amounts for
Medicaid and M-CHIP enrollees.

In contrast, CMS officials told us that until fiscal year 2010, states with a
stand-alone CHIP program (i.e., one that is operated separately from a
Medicaid program by the state; also known as S-CHIP) were not
supposed to report these expenditures into MSIS despite the presence of
the S-CHIP eligibility code. CMS officials indicated that any expenditures
associated with S-CHIP should be assumed to be Medicaid expenditures
despite the CHIP eligibility code. Therefore, we included any expenditures
associated with the S-CHIP code as Medicaid expenditures. 9




7
 Instead of a M-CHIP program, a state may elect to have a stand-alone CHIP program
that is operated separately from the Medicaid program. Stand-alone programs are referred
to as S-CHIP.
8
 Although M-CHIP programs operate as expansions of Medicaid programs, expenditures
are considered CHIP expenditures rather than Medicaid expenditures.
9
 Expenditures in MSIS for enrollees whose eligibility was listed as S-CHIP totaled
$481 million in fiscal year 2009, or less than 1 percent of total Medicaid expenditures
reported in MSIS.




Page 32                                                            GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology




Lastly, we included as Medicaid expenditures any other expenditures
associated with all other eligibility categories. These other eligibility
variables included “unknown” and “ineligible,” as well as Medicaid. 10
Consequently, for states with standalone S-CHIP programs and those
that did not separate out M-CHIP expenditures, the total Medicaid
expenditures may be inflated in MSIS. Similarly, any expenditures
reported under the unknown and ineligible categories that were not
Medicaid expenditures may also inflate Medicaid expenditures.

Step 3: Matching Inconsistent Definitions of Services between MSIS and
CMS-64

MSIS and CMS-64 data consist of expenditures broken down by service
types. 11 To compare MSIS and CMS-64 expenditures, we reviewed the
definitions of Medicaid service types used in each data set. This is
necessary because, in many instances, a one-to-one match of service
types in MSIS to those in CMS-64 is not possible. For example, in fiscal
years 2007 through 2009, the MSIS Annual Person Summary File had 29
service types, whereas CMS-64 had 43. (See table 3 for a list of MSIS
and CMS-64 service types used in fiscal years 2007 through 2009.)




10
  Expenditures in MSIS for enrollees whose eligibility was listed as unknown or ineligible
totaled approximately $28 billion and $6 billion, respectively, in fiscal year 2009; or
approximately 9 percent and 2 percent of total Medicaid expenditures reported in MSIS.
11
  While we are matching services between these two data sets, some of the expenditures
we describe here are not directly tied to a Medicaid service in MSIS, such as Medicaid
payments for Medicare premiums.




Page 33                                                           GAO-13-47 Medicaid Data
                                          Appendix I: Scope and Methodology




Table 3: Medicaid Statistical Information System (MSIS) and CMS-64 Service Types, Fiscal Years 2007–2009

MSIS service types                                    CMS-64 line numbers and service types
Inpatient hospital                                    1A.    Inpatient hospital
Inpatient mental health–aged                          1B.    Inpatient hospital–DSH
Inpatient mental health–under 21
Intermediate care facility–mentally retarded          2A.    Mental health facility
Nursing facility                                      2B.    Mental health facility–DSH
Physician
Dental                                                3.     Nursing facility services
Other practitioner                                    4A.    Intermediate care facility–mentally retarded (public providers)
Outpatient hospital
Clinic                                                4B.    Intermediate care facility–mentally retarded (private providers)
Home health                                           5.     Physicians services
Lab and x-ray
Drugs                                                 6.     Outpatient hospital services
Other services                                        7.     Prescribed drugs (gross spending)
HMO-capitation
                                                      7A1.   Drug rebates offset (national agreement)
Prepaid Health Plan (PHP)–capitation
Primary Care Case Management (PCCM)–capitation        7A2.   Drug rebates offset (state sidebar agreement)
Sterilization                                         8.     Dental
Abortion
Transportation                                        9.     Other practitioner services
Personal care                                         10.    Clinic services
Targeted case management
Rehabilitative services                               11.    Laboratory and radiological services
Physical, occupational, speech, and hearing therapy   12.    Home health services
Hospice
Nurse Midwife                                         13.    Sterilizations
Nurse Practitioner                                    14.    Abortions
Private duty nurse
                                                      15.    Early Periodic Screening, Diagnosis, and Treatment (EPSDT) services
Unknown
                                                      16.    Rural health clinic screenings
                                                      17A.   Medicare health insurance payments–Part A premiums
                                                      17B.   Medicare health insurance payments–Part B premiums
                                                                                                           a
                                                      17C1. 120 percent–134 percent of the poverty level
                                                                                                           b
                                                      17C2. 135 percent–175 percent of the poverty level
                                                      17D. Medicare–coinsurance and deductibles
                                                      18A. Medicaid health insurance payments–managed care organizations
                                                      18B1. Medicaid health insurance payments–prepaid ambulatory health plan
                                                      18B2. Medicaid health insurance payments–prepaid inpatient health plan
                                                      18C. Medicaid health insurance payments–group health plan payments
                                                      18D. Medicaid health insurance payments–coinsurance and deductibles
                                                      18E. Medicaid health insurance payments–other
                                                      19.    Home and community based services
                                                      20.    Home and community based care for functionally disabled elderly
                                                      21.    Community supported living services




                                          Page 34                                                              GAO-13-47 Medicaid Data
                     Appendix I: Scope and Methodology




MSIS service types                    CMS-64 line numbers and service types
                                      22.      Programs of All-Inclusive Care for the Elderly (PACE)
                                      23.      Personal care services
                                      24.      Targeted case management services
                                      25.      Primary Care Case Management (PCCM) Services
                                      26.      Hospice benefits
                                      27.      Emergency services for undocumented aliens
                                      28.      Federally-qualified health center
                                      29.      Other care services
                     Source: GAO analysis of Centers for Medicare & Medicaid Services data.
                     a
                      Medicaid expenditures reported on line 17C1 include premiums paid for Medicare Part B for
                     individuals whose income is 120-134 percent of the poverty level.
                     b
                      Medicaid expenditures reported on line 17C2 include the percentage of the Medicare Part B
                     premium attributable to the Home Health Benefit transferred from Part A to Part B for individuals
                     whose income is 135-175 percent of the poverty level.


                     Because of the lack of a one-to-one match between MSIS and CMS-64
                     service types, we combined MSIS and CMS-64 service types into six
                     combined expenditure categories. (See table 4 for the list of the combined
                     expenditure categories, and MSIS and CMS-64 service types included in
                     each.) As a result, we were able to report on total expenditures by
                     combined expenditure categories.




                     Page 35                                                                   GAO-13-47 Medicaid Data
                                            Appendix I: Scope and Methodology




Table 4: Combined Expenditure Categories and Corresponding MSIS and CMS-64 Service Types

Combined
expenditure
categories         MSIS service types                                  CMS-64 line numbers and service types
Hospital           Inpatient hospital                                  1A.    Inpatient hospital
                   Outpatient hospital                                 1B.    Inpatient hospital–Disproportionate Share Hospitals
                   Inpatient mental health facility for under age 21
                                                                               (DSH)
                   Mental health facility for the aged
                                                                       2A.    Mental health facility
                                                                       2B.    Mental health facility–DSH
                                                                       6.     Outpatient hospital services
                                                                       27.    Emergency services for undocumented aliens
Acute and long     Physician                                           5.     Physicians services
term support       Dental                                              8.     Dental
services (LTSS)-   Nurse midwife
noninstitutional   Nurse practitioner                                  9.     Other practitioner services
                   Other practitioner                                  10.    Clinic services
                   Home health
                   Clinic                                              11.    Laboratory and radiological services
                   Lab and x-ray                                       12.    Home health services
                   Sterilizations
                   Abortions                                           13.    Sterilizations
                   Personal care                                       14.    Abortions
                   Private duty nursing
                   Targeted case management                            15.    Early Periodic Screening, Diagnosis, and Treatment
                   Rehabilitative services                                    (EPSDT) Services
                   Hospice                                             16.    Rural health clinic screenings
                   Other services
                                                                       19.    Home and community based services
                   Physical, occupational, speech, and hearing
                    therapy                                            20.    Home and community based care for functionally
                   Transportation services                                     disabled elderly
                                                                       21.    Community supported living services
                                                                       23.    Personal care services
                                                                       24.    Targeted case management services
                                                                       26.    Hospice benefits
                                                                       28.    Federally Qualified Health Center (FQHC)
                                                                       29.    Other care services
Drugs              Drugs (gross spending)                              7.     Prescribed drugs (gross spending)
                                                                       7A1.   Drug rebates offset (national agreement)
                                                                       7A2.   Drug rebates offset (state sidebar agreement)




                                            Page 36                                                            GAO-13-47 Medicaid Data
                                           Appendix I: Scope and Methodology




Combined
expenditure
categories         MSIS service types                                            CMS-64 line numbers and service types
Managed care and HMO capitation                                                  18A.      Medicaid health insurance payments–managed care
Medicaid premium Prepaid Health Plan (PHP)                                                  organizations
assistance       Primary care case management                                    18B1. Medicaid health insurance payments–Prepaid
                                                                                        Ambulatory Health Plan
                                                                                 18B2. Medicaid health insurance payments–Prepaid Inpatient
                                                                                        Health Plan
                                                                                 18C.      Medicaid health insurance payments–Group Health
                                                                                           Plan Payments
                                                                                 18D.      Medicaid health insurance payments–Coinsurance and
                                                                                           Deductibles
                                                                                 18E.      Medicaid health insurance payments–other
                                                                                 25.       Primary Care Case Management (PCCM) Services
                                                                                 22.       Program of All-Inclusive Care for the Elderly (PACE)
Long term support Nursing facility                                               3.        Nursing facility services
services-         Intermediate care facility–mentally retarded                   4A.       Intermediate care facility-mentally retarded (public
institutional
                                                                                            providers)
                                                                                 4B.       Intermediate care facility-mentally retarded (private
                                                                                            providers)
Medicare                                                                         17A.      Medicare health insurance payments–Part A
                                                                                            premiums
                                                                                 17B.      Medicare health insurance payments–Part B
                                                                                            premiums
                                                                                 17C1. 120–134 percent of the poverty level
                                                                                 17C2. 135–175 percent of the poverty level
                                                                                 17D.      Medicare coinsurance and deductibles
                                           Source: GAO analysis of Centers for Medicare & Medicaid Services’ (CMS) data.

                                           Notes: The combined expenditure categories listed are based on analysis of the Medicaid Statistical
                                           Information System (MSIS) and CMS service types. We combined acute care and long term support
                                           services because CMS officials told us there was overlap between services in each type of care.
                                           The MSIS “unknown” service type was proportionally distributed across 5 categories, excluding
                                           Medicare.
                                           While states could report expenditures under “Religious non-medical” in CMS-64 for fiscal years 2007
                                           through 2009, no state did so. There is no corresponding category in MSIS.


                                           Step 4: Excluded Expenditures

                                           For certain CMS-64 expenditures, instead of adding them to MSIS, we
                                           excluded them from CMS-64. Specifically, expenditures from Puerto Rico
                                           or the U.S. territories from CMS-64 were excluded because these are not
                                           in MSIS data.




                                           Page 37                                                                         GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology




Additionally, we excluded all administrative expenses from CMS-64
reported amounts, because these are not reported in MSIS data. We also
excluded all state collections from CMS-64, because these are not in
MSIS. 12

Step 5: Comparison of MSIS and CMS-64 Expenditures

We compared total Medicaid expenditures in MSIS and CMS-64 for fiscal
years 2007 through 2009, nationally and for all 50 states and the District
of Columbia, to determine the extent of the difference between these two
data sets. With total expenditures from both data sets, we calculated the
expenditures in MSIS as a percentage of those reported in CMS-64. We
also compared total Medicaid expenditures by our six combined
expenditure categories, as reported by MSIS and CMS-64 for fiscal years
2007 through 2009.

II. Adjustment Analysis

To conduct our adjustment analysis, we determined the factors that
account for the difference between MSIS and CMS-64 data on Medicaid
expenditures. We identified and analyzed factors that account for the
differences in each data set in fiscal years 2007 through 2009. We
attempted to reconcile the expenditures by making adjustments to the
MSIS expenditures on the basis of the differences we identified. In
attempting to reconcile these differences, we made adjustments to the
baseline analysis described above. Specifically, we added expenditures
reported in CMS-64 that were not reported in MSIS to the MSIS amounts.
We added expenditures to MSIS rather than subtracting them from CMS-
64, because CMS-64 contains total Medicaid spending, and the additions
make MSIS a better approximation of this spending. For certain service
types, MSIS and CMS-64 reported different amounts of Medicaid
expenditures because of differences in the way they capture the
expenditure information. We identified four factors that account for the
differences that we were able to quantify. Therefore, we adjusted the
MSIS baseline amount and reduced the gap between the two data sets.
Specifically, MSIS does not include Disproportionate Share Hospital
(DSH) payments, Medicare premiums, prescription drug rebates, and



12
  State collections include offsetting collections from third-party liability, estate, and other
recoveries.




Page 38                                                               GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology




Medicaid health insurance payments. We then compared total adjusted
expenditures nationally and by state. We did this by comparing the
expenditures reported in MSIS as a percentage of those reported in
CMS-64. We also compared adjusted expenditures by combined
expenditure category, nationally and by state.

Step 1:

We added expenditures from DSH payments as reported on line 1B
(Inpatient Hospital Service–DSH Adjustment Payments) in CMS-64 to the
MSIS baseline amount within the Combined Hospital expenditure
category.

Step 2:

We added expenditures related to Medicare premiums from lines 17A
(Medicare Health Insurance Payments–Part A premiums), 17B (Medicare
Health Insurance Payments–Part B premiums), 17C1 (120–134 percent
of the poverty level), and 17C2 (135–175 percent of the poverty level)
from CMS-64 to the MSIS baseline amount within the combined Medicare
expenditure category. 13

Step 3:

We subtracted prescription drug rebate amounts reported on line 7A1
(Drug Rebate Offset–National Agreement) and 7A2 (Drug Rebate Offset–
State Sidebar Agreement) in CMS-64 within the prescription drugs
expenditure category from the MSIS baseline amount within the
Combined Prescription Drug expenditure category. 14




13
  Medicaid expenditures reported on line 17C1 include premiums paid for Medicare Part B
for individuals whose income is 120–134 percent of the poverty level. Medicaid
expenditures reported on line 17C2 include the percentage of the Medicare Part B
premium attributable to the Home Health Benefit transferred from Part A to Part B for
individuals whose income is 135–175 percent of the poverty level.
14
  Unlike the other adjustments, the adjustment for prescription drug rebates decreases the
spending reported in MSIS. Prescription drug spending is overreported in MSIS because
rebates are not included in the data set. Consequently, for this category we subtracted the
amounted reported in CMS-64 for prescription drug rebates from the amount in MSIS.




Page 39                                                          GAO-13-47 Medicaid Data
Appendix I: Scope and Methodology




Step 4:

We added expenditures related to Medicaid health insurance payments
from lines 18C (Medicaid health insurance payments–Group Health Plan
Payments), 18D (Medicaid health insurance payments–Coinsurance and
Deductibles) and 18E (Medicaid health insurance payments–Other) in
CMS 64 to the MSIS baseline amount within the combined managed care
and Medicaid premium assistance category.

Step 5:

After making the adjustments in steps 1 through 4, we compared the total
adjusted Medicaid expenditures as reported in MSIS and CMS-64 for
fiscal years 2007 through 2009, nationally and for all 50 states and the
District of Columbia. With total expenditures from both data sets, we
calculated the expenditures reported in MSIS as a percentage of those
reported in CMS-64.

We also compared total adjusted Medicaid expenditures by our six
expenditure categories, as reported by MSIS and CMS-64 for fiscal years
2007 through 2009.

After making the expenditure adjustments, we determined any remaining
expenditure differences between MSIS and CMS-64. We then identified
additional factors that help account for the difference between data sets.
However, because these were not systematic quantitative differences,
they could not be used to adjust the expenditures.

We conducted this performance audit from March 2012 through
September 2012 in accordance with generally accepted government
auditing standards. Those standards require that we plan and perform the
audit to obtain sufficient, appropriate evidence to provide a reasonable
basis for our findings and conclusions based on our audit objectives. We
believe that the evidence obtained provides a reasonable basis for our
findings and conclusions based on our audit objectives.




Page 40                                               GAO-13-47 Medicaid Data
Appendix II: Total Baseline Medicaid
                           Appendix II: Total Baseline Medicaid
                           Expenditures in MSIS and CMS-64, by State,
                           Fiscal Year 2009


Expenditures in MSIS and CMS-64, by State,
Fiscal Year 2009

                                                                                            MSIS baseline as a
State                  MSIS baseline (dollars)            CMS-64 baseline (dollars)         percent of CMS-64
Alabama                        $3,625,540,937                           $4,415,801,712                     82%
Alaska                          1,040,288,855                            1,069,624,574                      97
Arizona                         8,628,093,458                            8,665,269,493                     100
Arkansas                        3,486,489,877                            3,451,516,646                     101
California                     35,002,258,860                           41,389,559,881                      85
Colorado                        3,288,276,426                            3,555,155,172                      92
Connecticut                     5,289,134,290                            6,035,281,239                      88
Delaware                        1,263,983,534                            1,211,814,329                     104
District of Columbia            1,928,483,265                            1,626,139,477                     119
Florida                        14,051,386,231                           15,088,582,200                      93
Georgia                         7,376,411,423                            7,693,345,212                      96
Hawaii                          1,172,582,338                            1,308,211,194                      90
Idaho                           1,330,846,516                            1,276,526,247                     104
Illinois                       11,658,850,516                           13,140,383,274                      89
Indiana                         5,312,111,931                            5,906,490,283                      90
Iowa                            2,870,172,181                            2,960,114,156                      97
Kansas                          2,315,819,323                            2,443,675,864                      95
Kentucky                        4,927,054,393                            5,400,899,512                      91
Louisiana                       5,230,654,120                            6,513,211,836                      80
Maine                           1,480,509,448                            2,517,981,111                      59
Maryland                        6,124,798,441                            6,523,939,093                      94
Massachusetts                   9,928,887,946                           12,480,644,429                      80
Michigan                       10,145,744,904                           10,583,215,243                      96
Minnesota                       7,029,437,421                            7,387,421,506                      95
Mississippi                     3,199,788,959                            3,947,805,053                      81
Missouri                        5,679,022,531                            7,747,665,625                      73
Montana                           714,263,803                             875,768,845                       82
Nebraska                        1,542,139,671                            1,616,257,729                      95
Nevada                          1,196,285,725                            1,383,003,611                      86
New Hampshire                     994,261,996                            1,327,164,314                      75
New Jersey                      7,924,068,702                            9,667,209,281                      82
New Mexico                      2,633,955,686                            3,290,379,397                      80
New York                       44,882,629,782                           49,368,510,253                      91
North Carolina                  9,589,597,797                           11,506,119,180                      83
North Dakota                      581,137,250                             572,101,389                      102



                           Page 41                                                       GAO-13-47 Medicaid Data
                     Appendix II: Total Baseline Medicaid
                     Expenditures in MSIS and CMS-64, by State,
                     Fiscal Year 2009




                                                                                                         MSIS baseline as a
State            MSIS baseline (dollars)                          CMS-64 baseline (dollars)              percent of CMS-64
Ohio                      13,655,184,427                                           14,150,220,981                        97
Oklahoma                    3,441,451,894                                            3,937,604,747                       87
Oregon                      2,797,080,333                                            3,677,976,463                       76
Pennsylvania              14,206,934,568                                           17,231,560,151                        82
Rhode Island                1,492,164,387                                            1,893,290,969                       79
South Carolina              4,640,158,035                                            5,098,527,910                       91
South Dakota                  716,509,192                                               713,353,957                     100
Tennessee                   7,197,427,171                                            7,290,231,215                       99
Texas                     18,542,742,821                                           23,704,821,993                        78
Utah                        1,867,632,519                                            1,628,633,714                      115
Vermont                       969,805,347                                            1,190,678,054                       81
Virginia                    5,518,557,224                                            5,774,994,043                       96
Washington                  5,733,929,571                                            6,603,087,308                       87
West Virginia               2,588,751,602                                            2,434,058,051                      106
Wisconsin                   5,754,921,788                                            6,684,081,412                       86
Wyoming                       551,809,947                                               526,237,765                     105
Total                  $323,120,029,363                                         $366,486,147,093                        88%
                     Source: GAO analysis of Centers for Medicare & Medicaid Services’ data.

                     Note: MSIS = Medicaid Statistical Information System




                     Page 42                                                                          GAO-13-47 Medicaid Data
Appendix III: Total Adjusted Medicaid
                           Appendix III: Total Adjusted Medicaid
                           Expenditures in MSIS and CMS-64, by State,
                           Fiscal Year 2009


Expenditures in MSIS and CMS-64, by State,
Fiscal Year 2009

                                                                                            MSIS adjusted as a
State                  MSIS adjusted (dollars)             CMS-64 baseline (dollars)        percent of CMS-64
Alabama                        $4,149,081,465                           $4,415,801,712                     94%
Alaska                           1,044,650,976                           1,069,624,574                       98
Arizona                          8,950,429,121                           8,665,269,493                     103
Arkansas                         3,566,963,553                           3,451,516,646                     103
California                     37,800,029,701                           41,389,559,881                       91
Colorado                         3,469,734,861                           3,555,155,172                      98
Connecticut                      5,565,391,332                           6,035,281,239                       92
Delaware                         1,243,966,462                           1,211,814,329                     103
District of Columbia             1,998,599,795                           1,626,139,477                     123
Florida                        14,754,686,816                           15,088,582,200                       98
Georgia                          7,835,521,037                           7,693,345,212                     102
Hawaii                           1,209,812,682                           1,308,211,194                       92
Idaho                            1,320,590,808                           1,276,526,247                     103
Illinois                       12,053,935,834                           13,140,383,274                       92
Indiana                          5,488,175,214                           5,906,490,283                       93
Iowa                             2,895,286,487                           2,960,114,156                       98
Kansas                           2,391,474,252                           2,443,675,864                      98
Kentucky                         5,170,171,436                           5,400,899,512                       96
Louisiana                        6,117,964,332                           6,513,211,836                       94
Maine                            1,530,105,787                           2,517,981,111                       61
Maryland                         6,308,462,604                           6,523,939,093                       97
Massachusetts                  10,223,698,086                           12,480,644,429                       82
Michigan                       10,671,661,360                           10,583,215,243                     101
Minnesota                        9,661,603,386                           7,387,421,506                     131
Mississippi                      3,458,893,939                           3,947,805,053                       88
Missouri                         6,348,322,455                           7,747,665,625                       82
Montana                              730,414,065                          875,768,845                        83
Nebraska                         1,559,900,885                           1,616,257,729                       97
Nevada                           1,305,273,428                           1,383,003,611                       94
New Hampshire                    1,210,132,665                           1,327,164,314                       91
New Jersey                       9,240,294,628                           9,667,209,281                       96
New Mexico                       2,720,776,943                           3,290,379,397                       83
New York                       47,601,163,307                           49,368,510,253                       96
North Carolina                 10,037,581,856                           11,506,119,180                       87
North Dakota                         580,335,888                          572,101,389                      101



                           Page 43                                                       GAO-13-47 Medicaid Data
                     Appendix III: Total Adjusted Medicaid
                     Expenditures in MSIS and CMS-64, by State,
                     Fiscal Year 2009




                                                                                                         MSIS adjusted as a
State            MSIS adjusted (dollars)                           CMS-64 baseline (dollars)             percent of CMS-64
Ohio                       14,470,087,279                                           14,150,220,981                      102
Oklahoma                     3,541,362,801                                            3,937,604,747                       90
Oregon                       2,950,571,584                                            3,677,976,463                      80
Pennsylvania               15,215,057,372                                           17,231,560,151                        88
Rhode Island                 1,630,860,669                                            1,893,290,969                      86
South Carolina               5,117,449,890                                            5,098,527,910                     100
South Dakota                    720,846,348                                             713,353,957                     101
Tennessee                    7,363,722,376                                            7,290,231,215                     101
Texas                      20,378,285,311                                           23,704,821,993                        86
Utah                         1,881,512,781                                            1,628,633,714                     116
Vermont                      1,000,491,043                                            1,190,678,054                      84
Virginia                     5,749,185,155                                            5,774,994,043                     100
Washington                   6,150,885,067                                            6,603,087,308                       93
West Virginia                2,582,401,847                                            2,434,058,051                     106
Wisconsin                    5,653,921,737                                            6,684,081,412                       85
Wyoming                         550,456,264                                             526,237,765                     105
Total                   $345,172,184,969                                        $366,486,147,093                        94%
                     Source: GAO analysis of Centers for Medicare & Medicaid Services’ data.

                     Note: Adjustments were made by adding expenditures reported in CMS-64 to those reported in the
                     Medicaid Statistical Information System (MSIS). Specifically, we added Disproportionate Share
                     Hospital payments, national and state rebates for prescription drugs, Medicaid health insurance
                     payments, and Medicare premiums to the expenditures reported in MSIS.




                     Page 44                                                                          GAO-13-47 Medicaid Data
Appendix IV: Adjusted MSIS Expenditures as
               Appendix IV: Adjusted MSIS Expenditures as a
               Percentage of CMS-64, by State, Fiscal Year
               2009


a Percentage of CMS-64, by State, Fiscal Year
2009




               Note: Adjustments were made by adding expenditures reported in CMS-64 to those reported in the
               Medicaid Statistical Information System (MSIS). Specifically, we added Disproportionate Share
               Hospital payments, national and state rebates for prescription drugs, Medicaid health insurance
               payments, and Medicare premiums to the expenditures reported in MSIS.




               Page 45                                                               GAO-13-47 Medicaid Data
Appendix V: Adjusted MSIS Expenditures as
              Appendix V: Adjusted MSIS Expenditures as a
              Percentage of CMS-64, by State and
              Expenditure Category, Fiscal Year 2009


a Percentage of CMS-64, by State and
Expenditure Category, Fiscal Year 2009




              Page 46                                       GAO-13-47 Medicaid Data
Appendix V: Adjusted MSIS Expenditures as a
Percentage of CMS-64, by State and
Expenditure Category, Fiscal Year 2009




Page 47                                       GAO-13-47 Medicaid Data
Appendix V: Adjusted MSIS Expenditures as a
Percentage of CMS-64, by State and
Expenditure Category, Fiscal Year 2009




Note: Four states—Maine, Mississippi, Vermont, and Wyoming—reported $0 in expenditures related
to the managed care and Medicaid premium assistance category.
Adjustments were made by adding expenditures reported in CMS-64 to those reported in the
Medicaid Statistical Information System (MSIS). Specifically, we added Disproportionate Share
Hospital payments, national and state rebates for prescription drugs, Medicaid health insurance
payments, and Medicare premiums to the expenditures reported in MSIS. No adjustments were made
for acute and long-term support services (LTSS)-noninstitutional and LTSS-institutional.




Page 48                                                              GAO-13-47 Medicaid Data
Appendix VI: GAO Contact and Staff
                  Appendix VI: GAO Contact and Staff
                  Acknowledgments



Acknowledgments

                  Carolyn L. Yocom, (202) 512-7114 or yocomc@gao.gov
GAO Contact
                  In addition to the contact named above, Robert Copeland, Assistant
Staff             Director; Muriel Brown; Shaunessye Curry; Greg Dybalski; Sandra
Acknowledgments   George; Giselle Hicks; Drew Long; Jessica Morris; and Monica Perez
                  Nelson made key contributions to this report.




                  Page 49                                            GAO-13-47 Medicaid Data
Related GAO Products
             Related GAO Products




             Medicaid: States Reported Billions More in Supplemental Payments in
             Recent Years. GAO-12-694. Washington, D.C.: July 20, 2012.

             National Medicaid Audit Programs: CMS Should Improve Reporting and
             Focus on Audit Collaboration with States. GAO-12-627. Washington,
             D.C.: June 14, 2012.

             High Risk Series: An Update. GAO-11-278. Washington, D.C.:
             February 16, 2011.

             Medicaid: CMS Needs More Information on the Billions of Dollars Spent
             on Supplemental Payments. GAO-08-614. Washington, D.C.: May 30,
             2008.

             Medicaid Demonstration Waivers: Recent HHS Approvals Continue to
             Raise Cost and Oversight Concerns. GAO-08-87. Washington, D.C.:
             January 31, 2008.

             Medicaid Financial Management: Steps Taken to Improve Federal
             Oversight but Other Actions Needed to Sustain Efforts. GAO-06-705.
             Washington, D.C.: June 22, 2006.

             Major Management Challenges and Program Risks: Department of
             Health and Human Services. GAO-03-101. Washington, D.C.:
             January 1, 2003.

             Medicaid Financial Management: Better Oversight of State Claims for
             Federal Reimbursement Needed. GAO-02-300. Washington, D.C.:
             February 28, 2002.




(291025)
             Page 50                                            GAO-13-47 Medicaid Data
GAO’s Mission         The Government Accountability Office, the audit, evaluation, and
                      investigative arm of Congress, exists to support Congress in meeting its
                      constitutional responsibilities and to help improve the performance and
                      accountability of the federal government for the American people. GAO
                      examines the use of public funds; evaluates federal programs and
                      policies; and provides analyses, recommendations, and other assistance
                      to help Congress make informed oversight, policy, and funding decisions.
                      GAO’s commitment to good government is reflected in its core values of
                      accountability, integrity, and reliability.

                      The fastest and easiest way to obtain copies of GAO documents at no
Obtaining Copies of   cost is through GAO’s website (http://www.gao.gov). Each weekday
GAO Reports and       afternoon, GAO posts on its website newly released reports, testimony,
                      and correspondence. To have GAO e-mail you a list of newly posted
Testimony             products, go to http://www.gao.gov and select “E-mail Updates.”

Order by Phone        The price of each GAO publication reflects GAO’s actual cost of
                      production and distribution and depends on the number of pages in the
                      publication and whether the publication is printed in color or black and
                      white. Pricing and ordering information is posted on GAO’s website,
                      http://www.gao.gov/ordering.htm.
                      Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
                      TDD (202) 512-2537.
                      Orders may be paid for using American Express, Discover Card,
                      MasterCard, Visa, check, or money order. Call for additional information.
                      Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Connect with GAO      Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts.
                      Visit GAO on the web at www.gao.gov.
                      Contact:
To Report Fraud,
Waste, and Abuse in   Website: http://www.gao.gov/fraudnet/fraudnet.htm
                      E-mail: fraudnet@gao.gov
Federal Programs      Automated answering system: (800) 424-5454 or (202) 512-7470

                      Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 512-
Congressional         4400, U.S. Government Accountability Office, 441 G Street NW, Room
Relations             7125, Washington, DC 20548

                      Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
Public Affairs        U.S. Government Accountability Office, 441 G Street NW, Room 7149
                      Washington, DC 20548




                        Please Print on Recycled Paper.