This is the accessible text file for GAO report number GAO-13-229 
entitled 'Medicaid: Enhancements Needed for Improper Payments 
Reporting and Related Corrective Action Monitoring' which was released 
on May 1, 2013. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

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

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

March 2013: 

Medicaid: 

Enhancements Needed for Improper Payments Reporting and Related 
Corrective Action Monitoring: 

GAO-13-229: 

GAO Highlights: 

Highlights of GAO-13-229, a report to congressional requesters. 

Why GAO Did This Study: 

Medicaid has the second-highest estimated improper payments of any 
federal program that reported such data for fiscal year 2011. Also, 
the Congress has raised questions about reporting and corrective 
actions related to the Medicaid program’s improper payments. The 
objectives of this report were to determine the extent to which (1) 
CMS’s methodology for estimating Medicaid improper payments follows 
OMB guidance and produces reasonable national and state-level 
estimates and (2) corrective action plans have been developed to 
reduce Medicaid payment error rates and whether these plans address 
the types of payment errors identified. To address these objectives, 
GAO analyzed CMS’s policies and procedures against federal guidance 
and standards for estimating improper payments and developing related 
corrective actions to address errors. GAO also reviewed the results of 
all state-level reviews and conducted site visits at selected states 
that either received relatively large amounts of Medicaid payments or 
had varying rates of estimated improper payments, including states 
with possible best practices. GAO also met with cognizant CMS 
officials and contractors. 

What GAO Found: 

The Centers for Medicare & Medicaid Services’ (CMS) methodology for 
estimating a national improper payment rate for the Medicaid program 
is statistically sound. However, CMS’s procedures did not provide for 
updating state data used in its methodology to recognize significant 
corrections or adjustments after the cutoff date. The Office of 
Management and Budget (OMB) requires that federal agencies establish a 
statistically valid methodology for estimating the annual amount of 
improper payments in programs and activities susceptible to 
significant improper payments. CMS developed the Payment Error Rate 
Measurement (PERM) program in order to comply with improper payment 
estimation and reporting requirements for the Medicaid program. Under 
the PERM methodology, CMS places states in one of three cycles, and 
each year one of the cycles reports new state-level data based on the 
previous year’s samples. CMS then calculates the national Medicaid 
program improper payment estimate using these new data for one-third 
of the states and older data for the other two-thirds of the states. 
CMS’s estimated national improper payment error rate for fiscal year 
2011 for the Medicaid program was 8.1 percent, or $21.9 billion. 
However, CMS’s procedures did not provide for considering revisions to 
state-level Medicaid program error rates used in the CMS methodology 
for calculating its national Medicaid program error rate. Because 
corrections to the 2 years of older data after the cutoff date are not 
officially recognized by CMS, the entire 3-year cycle could be 
affected. OMB has identified as a best practice that agencies should 
establish a policy for handling unscheduled corrections to data. Until 
CMS establishes procedures for considering changes to initially 
reported state-level error rates that would be significant to the 
national error rate, CMS is impaired in its ability to ensure that its 
reported estimate of the extent of national Medicaid improper payments 
is reliable. 

CMS and state agencies developed corrective action plans (CAP) related 
to identified PERM payment errors. However, GAO identified the 
following areas where improvements were needed in CMS’s written 
guidance to states on CAPs to ensure efficient and effective actions 
to reduce improper payments. 

* CMS’s PERM Manual did not clearly identify the circumstances under 
which states should consider, and if cost effective include, 
nonpayment errors (such as certain coding errors that could have but 
did not result in a payment error) and minimal dollar errors in their 
CAPs. 

* The PERM Manual and the associated website did not provide complete 
and consistent information on the required elements to include in a 
state CAP. 

* CMS guidance did not clearly delineate CMS officials’ roles and 
responsibilities for conducting oversight of (1) state CAP submissions 
to ensure that they contained all of the required elements and 
adequately addressed errors identified in the PERM reviews and (2) 
states’ progress in implementing CAP corrective actions. 
Although the nonpayment errors identified in PERM reviews did not 
result in improper payments, the underlying issues may result in 
improper payments in future years if not addressed. Also, complete 
information in state CAPs is necessary for CMS to analyze the progress 
and effectiveness of the CAPs. Further, clear accountability for 
continuous monitoring helps ensure that actions are taken to 
effectively reduce Medicaid improper payments. 

What GAO Recommends: 

GAO is making four recommendations to help improve CMS’s reporting of 
estimated Medicaid improper payments and its related corrective action 
process. The Department of Health and Human Services concurred with 
GAO’s recommendations and cited a number of actions under way and 
planned. 

View [hyperlink, http://www.gao.gov/products/GAO-13-229]. For more 
information, contact Beryl H. Davis at (202) 512-2623 or 
davisbh@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Estimate of National Medicaid Improper Payments Did Not Consider 
Revised Individual State Error Rates: 

Deficiencies in CMS's Guidance and Monitoring Impair Efficient and 
Effective Medicaid Program Improper Payments Reductions: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Objectives, Scope, and Methodology: 

Appendix II: Fee-for-Service, Managed Care, and Eligibility Components 
of the Medicaid Error Rate Measurement: 

Appendix III: State Error Rates for the Department of Health and Human 
Services' Fiscal Year 2011 Reporting of Medicaid Improper Payments: 

Appendix IV: Medicaid Outlays and Estimated Improper Payment Error 
Rates Reported by the Department of Health and Human Services: 

Appendix V: Additional Details for Medicaid Error Rates: 

Appendix VI: Types of Errors Identified in Payment Error Rate 
Measurement Reviews: 

Appendix VII: Comments from the Department of Health and Human 
Services: 

Appendix VIII: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: CMS's Corrective Action Strategies to Reduce PERM-Identified 
Improper Payments: 

Table 2: State Error Rates Used to Determine HHS's Fiscal Year 2011 
Reporting of National Medicaid Improper Payments: 

Table 3: Medicaid Outlays and Estimated Improper Payment Error Rates 
Reported in HHS's AFRs: 

Table 4: Margins of Error for Error Rate Data Presented in Figure 2: 

Table 5: Summary of Errors, Fiscal Years 2008 to 2010 PERM Measurement 
Cycles: 

Figures: 

Figure 1: PERM Process for Estimating and Reducing Medicaid Improper 
Payments for Fiscal Year 2011 Reporting: 

Figure 2: Medicaid Error Rate Calculation for Fiscal Year 2011 
Reporting: 

Figure 3: Types of Errors Identified in the Fiscal Years 2008 to 2010 
PERM Measurement Cycles Based on Number of Errors: 

Figure 4: CMS's Regulatory Requirements for States to Develop CAPs to 
Reduce Improper Payments: 

Abbreviations: 

AFR: agency financial report: 

CAP: corrective action plan: 

CHIPRA: Children's Health Insurance Program Reauthorization Act of 
2009: 

CMS: Centers for Medicare & Medicaid Services: 

DERM: Division of Error Rate Measurement: 

FFS: fee-for-service: 

HHS: Department of Health and Human Services: 

IPERA: Improper Payments Elimination and Recovery Act of 2010: 

IPIA: Improper Payments Information Act of 2002: 

MIG: Medicaid Integrity Group: 

OMB: Office of Management and Budget: 

PERM: Payment Error Rate Measurement: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

March 29, 2013: 

The Honorable Thomas R. Carper: 
Chairman: 
The Honorable Tom Coburn, M.D.
Ranking Member: 
Committee on Homeland Security and Governmental Affairs: 
United States Senate: 

The Honorable Claire McCaskill: 
Chairman: 
Subcommittee on Financial and Contracting Oversight: 
Committee on Homeland Security and Governmental Affairs: 
United States Senate: 

Medicaid has the second-highest estimated improper payments of any 
federal program that reported such data for fiscal year 2011.[Footnote 
1] In its fiscal year 2011 agency financial report (AFR), the 
Department of Health and Human Services (HHS) reported estimated 
improper payments for Medicaid of about $21.9 billion, based on an 
error rate of 8.1 percent,[Footnote 2] which contributed to the total 
governmentwide improper payment estimate of over $115 billion for that 
year. 

The size and diversity of the Medicaid program make it particularly 
vulnerable to improper payments--including payments made for 
treatments or services that were not covered by program rules, that 
were not medically necessary, or that were billed for but never 
provided. Medicaid, the joint federal-state health care financing 
program for certain low-income individuals, is one of the largest 
social programs in federal and state budgets, providing care to about 
70 million individuals at a cost of $436 billion in fiscal year 2011. 
[Footnote 3] The Centers for Medicare & Medicaid Services (CMS), a 
federal agency within HHS, is responsible for overseeing the program 
at the federal level, while the states administer their respective 
programs' day-to-day operations. 

We designated Medicaid as a high-risk program in 2003 in part because 
of concerns about the adequacy of fiscal oversight, which is necessary 
to prevent inappropriate program spending.[Footnote 4] Medicaid 
remains at high risk because of concerns about the adequacy of fiscal 
oversight of this large, diverse, and growing program. We have 
specifically identified improper payments to Medicaid providers 
serving program beneficiaries as an area of concern. Given the 
hundreds of billions of dollars disbursed annually for the Medicaid 
program, improper payments to providers that submit inappropriate 
claims can result in substantial financial losses to states and the 
federal government. 

As required under the Improper Payments Information Act of 2002 
(IPIA),[Footnote 5] as amended by the Improper Payments Elimination 
and Recovery Act of 2010 (IPERA),[Footnote 6] HHS has identified its 
programs that are susceptible to significant improper payments, 
including Medicaid. IPIA, as amended, requires agencies to obtain a 
statistically valid estimate, or an estimate that is otherwise 
appropriate using a methodology approved by the Office of Management 
and Budget (OMB), of the annual amount of improper payments in such 
programs. OMB has issued guidance for agencies to use in implementing 
IPIA and IPERA.[Footnote 7] 

In light of the magnitude of the Medicaid program and its related 
improper payments, you raised questions about CMS's improper payment 
estimation methodology and related corrective action plans. The 
objectives of our review were to determine the extent to which (1) 
CMS's methodology for estimating Medicaid improper payments follows 
OMB guidance and produces reasonable national and state-level 
estimates and (2) corrective action plans have been developed to 
reduce Medicaid payment error rates and whether these plans addressed 
the types of errors identified. 

To address these objectives, we reviewed applicable improper payment 
legislation and related OMB guidance, internal control standards, and 
financial reporting standards. We also reviewed CMS regulations on 
Payment Error Rate Measurement (PERM), CMS's internal written 
guidance, and results from state PERM reviews used for fiscal year 
2011 reporting. In addition, we reviewed improper payment information 
reported in HHS's fiscal year 2011 AFR. 

To further address the first objective, we compared CMS's sampling and 
statistical methods used to estimate the fiscal year 2011 Medicaid 
payment error rate with related OMB guidance. We focused on the 
Medicaid payment error rate reported for fiscal year 2011. As part of 
this assessment, we conducted interviews with CMS officials and its 
contractors to clarify our understanding of both the sampling and 
estimation methodologies. We also reviewed CMS's program manuals for 
both the payment error and eligibility payment error components of 
PERM, as well as professional statistical literature, to assess the 
statistical validity of CMS's methodology. In addition, we reviewed 
state-level payment error rates to determine whether the sample sizes 
assigned to states were in accordance with OMB statistical guidance. 
We also used the results of these reviews and analyses to identify and 
assess the reasons for any weaknesses in the estimation methodology, 
and their potential effects on identifying and reporting Medicaid 
improper payment estimates for fiscal year 2011 and going forward. In 
addition to reviewing the statistical methodology, we obtained actual 
payment error data for selected states and independently calculated 
the payment error rates. The scope of our review did not include an 
assessment of individual states' processes or payment systems. We 
assessed the reliability of the claims and error rate data by gaining 
an understanding of the processes the contractors or states use to 
perform reviews, including any use of data sharing to determine 
eligibility, and their quality controls. We determined that the data 
were sufficiently reliable for our purposes. 

To further address the second objective, we reviewed agency policies 
and procedures related to the development of PERM corrective action 
plans (CAP) and CAPs for all 50 states and the District of Columbia 
used to address the root causes of improper payments identified from 
the PERM reviews, and conducted interviews with officials from CMS. We 
also reviewed CMS's error rate reduction plans and initiatives to 
reduce Medicaid improper payments. In addition, we assessed CMS's 
process for monitoring state corrective actions and its methodology 
for measuring the effectiveness of corrective actions to reduce 
improper payments against Standards for Internal Control in the 
Federal Government related to monitoring.[Footnote 8] The scope of our 
review did not include an assessment of individual states' 
implementation of their CAPs. 

To support both objectives, we conducted site visits at seven state 
Medicaid offices (California, Florida, Illinois, Michigan, 
Pennsylvania, South Carolina, and Texas). During these site visits, we 
interviewed state personnel involved in the PERM process to gain an 
understanding of how these states compiled the universes of claims and 
beneficiaries that are sampled for the PERM reviews, how these states 
conducted eligibility reviews, and how these states developed 
corrective action plans and worked with CMS on corrective actions. We 
selected these states based on criteria such as states' federal share 
of Medicaid payments and errors identified in PERM reviews. The seven 
states we visited collectively claimed about 37 percent of the total 
federal share of Medicaid payments made in fiscal year 2010.[Footnote 
9] We also selected these states to achieve variation in the error 
rates found during PERM reviews included in the fiscal year 2011 
reporting of the Medicaid improper payment estimate. One state had the 
highest error rate for eligibility reviews as well as the highest 
combined error rate. This selection also allowed us to focus on 
certain states with noted vulnerabilities in program integrity 
efforts, as well as states with possible best practices. Although it 
does not allow us to generalize findings to all states and thus the 
program as a whole, we believe these state visits, combined with our 
analysis of CAPs for all states, enable us to determine if states' 
corrective actions are addressing the types of improper payment errors 
that have been identified. 

We conducted this performance audit from February 2012 to March 2013 
in accordance with generally accepted government auditing standards. 
Those standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe 
that the evidence obtained provides a reasonable basis for our 
findings and conclusions based on our audit objectives. See appendix I 
for additional details on our scope and methodology. 

Background: 

CMS's Process for Estimating Medicaid Improper Payments under PERM: 

The purpose of the CMS PERM program is to produce a national-level 
improper payment error rate for Medicaid.[Footnote 10] CMS developed 
PERM in order to comply with the requirements of IPIA, which was 
amended by IPERA.[Footnote 11] PERM uses a 17-state, 3-year rotation 
for measuring Medicaid improper payments. Medicaid improper payments 
are estimated on a federal fiscal year basis through the PERM process. 
The estimate measures three component error rates: (1) fee-for-service 
(FFS), (2) managed care, and (3) eligibility. FFS is a traditional 
method of paying for medical services under which providers are paid 
for each service rendered. Each selected FFS claim is subjected to a 
data processing review. The majority of FFS claims also undergo a 
medical review. Managed care is a system where the state contracts 
with health plans to deliver health services through a specified 
network of doctors and hospitals. Managed care claims are subject only 
to a data processing review. Eligibility refers to meeting the state's 
categorical and financial criteria for receipt of benefits under the 
Medicaid program. States perform their own eligibility reviews 
according to state and federal eligibility criteria. See appendix II 
for additional details on these three components. CMS uses its PERM 
Manual to provide detailed guidance for implementing CMS regulations 
on PERM. PERM regulations set forth the methodology for states to 
estimate Medicaid improper payments and outline the requirements for 
state CAPs.[Footnote 12] Figure 1 shows the PERM process for 
estimating and reducing Medicaid improper payments. 

Figure 1: PERM Process for Estimating and Reducing Medicaid Improper 
Payments for Fiscal Year 2011 Reporting: 

[Refer to PDF for image: process illustration] 

Three groups of states: 
Cycle 1: 17 states; 
Cycle 2: 17 states; 
Cycle 3: 17 states. 
        
Fifty-one states[A]are divided into three groups of 17 states. Each 
group is referred to as a cycle. Measurement data are captured once 
for each cycle during a 3-year period. As a result, CMS's Medicaid 
improper payment reporting for each year is based on new data for 17 
states and data from previous years for the other 34 states[B]. 

Three-year PERM process: 

Fiscal year: 2008; 
Year 1: Cycle 3: 17 states; 
Year 2: Cycle 2: 17 states; 
Year 3: Cycle 1: 17 states. 

Fiscal year: 2009; 
Year 1: Cycle 1: 17 states; 
Year 2: Cycle 3: 17 states; 
Year 3: Cycle 2: 17 states. 

Fiscal year: 2010; 
Year 1: Cycle 2: 17 states; 
Year 2: Cycle 1: 17 states; 
Year 3: Cycle 3: 17 states. 

The PERM process for each group of states lasts approximately 3 years. 
When a group has completed the process, it will begin the process 
again. 

PERM process details[C]: 

Year 1: 
* Develop universe: State submits claims data to CMS; 
* Begin selecting samples: States begin selecting eligibility samples. 
CMS contractors begin selecting samples for FFS and managed care 
reviews; 
* Begin reviews: States begin eligibility reviews. CMS contractors 
begin FFS and managed care reviews. 

Year 2: 
* Continue and finalize reviews: CMS and states continue selecting 
samples and performing reviews, and then finalize results; 
* State error rates determined: CMS uses finalized reviews to 
determine improper payment error rates for individual states; 
* National error rate calculated: After dollar-weighting to adjust for 
differences in the size of programs, CMS combines individual state 
improper payment error rates to calculate a national improper payment 
error rate. 

Year 3: 
* Reporting: CMS shares results with states in a summary report; CMS 
reports the national improper payment error rate to HHS for us in its 
AFR; 
* Corrective action plan: Each state submits and implements a 
corrective action plan to address the underlying causes for the errors 
found during the PERM review; Each state also submits an evaluation of 
the previous cycle's corrective actions. 

Source: GAO analysis of CMS's PERM process for estimating and reducing 
Medicaid improper payments as outlined in the PERM manual. 

[A] The "51 states" refers to the 50 states and the District of 
Columbia. 

[B] For example, fiscal year 2011 reporting is based on the 
measurement cycles for fiscal years 2008 to 2010, while fiscal year 
2010 reporting is based on the measurement cycles for fiscal years 
2007 to 2009. 

[C] Based on our analysis of CMS's PERM process, it is possible that 
there may be circumstances under which certain activities from years 1 
and 2 might be delayed until years 2 and 3, respectively. 

[End of figure] 

Through its use of federal contractors, CMS measures the FFS and 
managed care components while states perform the eligibility component 
measurement.[Footnote 13] CMS contracts with two vendors--a 
statistical contractor and a review contractor--to conduct the FFS and 
managed care review components of PERM and calculate error rates. The 
statistical contractor is responsible for (1) collecting and sampling 
claims and payment data for review, including performing procedures to 
ensure that the universe is accurate and complete; (2) reviewing state 
eligibility sampling plans; and (3) calculating state and national 
error rates. The review contractor is responsible for conducting data 
processing and medical reviews after the statistical contractor 
selects the samples of claims. Beginning with the fiscal year 2011 
measurement cycle, state-specific sample sizes are calculated based on 
the prior measurement cycle's component-level error rates and 
precision. 

All payment error rate calculations for the Medicaid program (the FFS 
component, managed care component, eligibility component, and overall 
Medicaid error rate) are based on the ratio of estimated dollars of 
improper payments to the estimated dollars of total payments. The 
overall Medicaid error rate represents the combination of FFS, managed 
care, and eligibility error rates. Individual state error rate 
components and state overall Medicaid error rates are combined to 
calculate the national component error rates and national overall 
Medicaid error rate. PERM accounts for the overlap between claims and 
eligibility reviews by calculating a small correction factor to ensure 
that Medicaid eligibility errors do not get "double counted" if the 
sampled item was also tested in either the FFS or managed care 
components. National component error rates and the national overall 
Medicaid program error rate are weighted by state size in terms of 
outlays, so that a state with a $10 billion Medicaid program "counts" 
10 times more toward the national rate than a state with a $1 billion 
Medicaid program. 

For fiscal year 2011 reporting--the reporting period covered by our 
audit--CMS reported an estimated national Medicaid improper payment 
error rate of 8.1 percent or $21.9 billion ($21,448 million in 
overpayments and $453 million in underpayments).[Footnote 14] The 
weighted national component error rates are as follows: for Medicaid 
FFS, 2.7 percent; for Medicaid managed care, 0.3 percent; and for 
Medicaid eligibility, 6.1 percent.[Footnote 15] See appendix III for 
the state and national error rates for HHS's fiscal year 2011 
reporting of Medicaid improper payments. See appendix IV for the 
national Medicaid outlays and the estimated improper payment error 
rate reported in HHS's AFRs for fiscal years 2007 to 2011.[Footnote 16] 

On February 4, 2009, the Children's Health Insurance Program 
Reauthorization Act of 2009 (CHIPRA) was enacted.[Footnote 17] As 
required under Section 601 of CHIPRA, HHS published a final rule on 
August 11, 2010, effective September 30, 2010, which requires that 
PERM eligibility reviews be consistent with the state's eligibility 
verification policy rather than reviewing eligibility against a 
single, federal methodology, which was done in the past. After 
publication of the final rule, states were allowed to review cases 
under the new methodology. Figure 2 shows the roll up of the error 
rate reported for fiscal year 2011. 

Figure 2: Medicaid Error Rate Calculation for Fiscal Year 2011 
Reporting: 

[Refer to PDF for image: illustration] 

Fiscal year 2008 cycle[A], 8.7% error rate[B]: 

FFS: Reviewed by federal contractors; 
Data processing review; plus: 
Medical review = 2.6%. 

Managed care: Reviewed by federal contractors; 
Data processing review = 0.1%. 

Eligibility: Reviewed by states; 
Payment error rate = 6.7%. 

Fiscal year 2009 cycle, 9.0% error rate: 

FFS: Reviewed by federal contractors; 
Data processing review; plus: 
Medical review = 1.9%. 

Managed care: Reviewed by federal contractors; 
Data processing review = 0.1%. 

Eligibility: Reviewed by states; 
Payment error rate = 7.6%. 

Fiscal year 2010 cycle, 6.7% error rate: 

FFS: Reviewed by federal contractors; 
Data processing review; plus: 
Medical review = 3.6%. 

Managed care: Reviewed by federal contractors; 
Data processing review = 0.5%. 

Eligibility[C]: Reviewed by states; 
Payment error rate = 4.0%. 

Fiscal year 2011 reporting, 8.1% 3-year national error rate: 

FFS: 
2008: 2.6%; 
2009: 1.9%; 
2010: 3.6%; 
Weighted national average: 2.7%[D]. 

Managed care: 
2008: 0.1%; 
2009: 0.1%; 
2010: 0.5%; 
Weighted national average: 0.3%[D]. 

Eligibility: 
2008: 6.7%; 
2009: 7.6%; 
2010: 4.0%; 
Weighted national average: 6.1%[D]. 

Source: GAO analysis of CMS's PERM methodology and support for fiscal 
year 2011 error rate reporting. 

Note: All of the numbers shown in this figure are estimated from PERM 
sample data. Please see table 4 in appendix V for the associated 
margins of error at the 90 percent confidence levels. 

[A] Each measurement cycle includes 17 states. 

[B] The combined rate is a weighted average of FFS and managed care, 
with the addition of eligibility. A small correction factor ensures 
that Medicaid eligibility errors do not get "double counted" if the 
sampled item was also tested in either the FFS or managed care 
components. 

[C] Effective September 30, 2010, CMS required eligibility reviews to 
be consistent with the state's eligibility verification policy rather 
than reviewing eligibility against a single, federal methodology, 
which was done in the past. 

[D] The weighted national average is weighted by expenditures and 
therefore may not average across for 2008, 2009, and 2010. 

[End of figure] 

PERM Corrective Action Process: 

IPIA, as amended, requires the heads of federal agencies to report on 
the actions the agency is taking to reduce improper payments, 
including a description of the causes of improper payments identified, 
actions planned or taken to correct those causes, and the planned or 
actual completion date of the actions taken to address those causes. 
This law also requires heads of federal agencies to report on a 
description of the steps the agency has taken to ensure that agency 
managers, programs, and, where appropriate, states and localities are 
held accountable through annual appraisal criteria for (1) meeting 
applicable improper payment reduction targets and (2) establishing and 
maintaining sufficient internal controls, including an appropriate 
control environment that effectively prevents improper payments from 
being made and promptly detects and recovers improper payments that 
are made. According to OMB's implementing guidance for IPERA, agencies 
should utilize the results of their statistical sampling measurements 
to identify the root causes of improper payments and implement 
corrective actions to prevent and reduce improper payments associated 
with these root causes.[Footnote 18] Agencies should continuously use 
their improper payment measurement results to identify new and 
innovative corrective actions to prevent and reduce improper payments. 
Agencies should also annually review their existing corrective actions 
to determine if any existing action can be intensified or expanded, 
resulting in a high-impact, high return on investment in terms of 
reduced or prevented improper payments. 

While CMS has responsibility for interpreting and implementing the 
federal Medicaid statute and ensuring that federal funds are 
appropriately spent--including estimating improper payments--the 
program is administered at the state level with significant state 
financing. Consequently, CMS relies primarily on states to develop and 
implement CAPs to address reported PERM errors. Following each 
measurement cycle, the states included in the measurement are required 
to complete and submit a CAP based on the errors found during the PERM 
process. In addition to guidance in the PERM Manual, CMS provides 
guidance to states on the CAP process upon releasing the PERM error 
rates and throughout CAP development. 

Estimate of National Medicaid Improper Payments Did Not Consider 
Revised Individual State Error Rates: 

CMS's PERM methodology for reporting a national Medicaid program 
improper payment estimate is statistically sound and meets OMB 
requirements. However, the process for accumulating the data used in 
deriving the reported national estimate does not consider the extent 
of any significant changes in state-level improper payment data that 
occurred after the initial year-end cutoff for state reporting. The 
impact of any such significant changes in states' PERM reviews that 
were not concluded by the annual measurement cycle cutoff dates could 
significantly affect the calculation of the rolling 3-year average 
national Medicaid error rate reported each year. 

Design of CMS's PERM Methodology Meets Established OMB Requirements: 

The design of CMS's PERM methodology meets OMB requirements. CMS has 
documented the steps it took to design the sample and the steps taken 
to construct the sampling frame for the FFS, managed care, and 
eligibility review samples in its PERM Manual.[Footnote 19] The 
documentation also includes CMS's process for ensuring that each 
sampling frame was accurate, timely, and complete. For error rate 
measurement for the FFS and managed care components, as outlined in 
the PERM Manual, CMS uses a stratified random sample selected 
quarterly within each state to provide cases for the data processing 
and medical review testing. For the eligibility component, as outlined 
in CMS's PERM Manual, states use a simple random sample of eligible 
cases and negative cases, which are drawn each month during the 
measurement cycle.[Footnote 20] 

Absent an alternate methodology specifically approved by OMB, agencies 
must obtain a statistically valid estimate of the annual amount of 
improper payments in programs and activities for those programs that 
are identified as susceptible to significant improper payments. The 
estimates are to be based on the equivalent of a statistically random 
sample of sufficient size to yield an estimate with a 90 percent 
confidence interval of not more than plus or minus 2.5 percentage 
points around the estimate of the percentage of improper payments. CMS 
reports national Medicaid error rates at this 90 percent confidence 
interval to be consistent with OMB's requirements, but CMS's 
procedures provide that the sample size for PERM is to conform to OMB 
optional guidance for estimating payment errors--specifically, the 
PERM Manual specifies a target precision of plus or minus 3 percentage 
points at a 95 percent level of confidence within each state.[Footnote 
21] The PERM Manual provides for the sample size for each state to be 
based upon the previous payment error rate and the OMB optional 
standard for the precision and confidence level. To estimate the 
percentage of dollars paid in error, CMS's PERM Manual provides for 
using a ratio estimation methodology to produce the PERM estimate. 
This means the PERM payment error rate is a ratio of the estimated 
total dollars paid in error divided by the estimated total payments. 
The choice of ratio estimation methodology under these circumstances 
is statistically appropriate. 

The PERM Manual describes the data collection methods for the medical 
reviews, data processing reviews, and eligibility determinations. The 
PERM Manual also describes the statistical ratio estimation 
methodology to be used to produce the estimated percentage of dollars 
paid in error. CMS's PERM Manual also provides for the error rates and 
summary reports to be provided to each state participating in the 
measurement cycle. We found that CMS's PERM Manual is consistent with 
OMB statistical guidance. 

Procedures Used in National Error Rate Calculation Did Not Consider 
Updates to State-Level Rates: 

Although the CMS PERM methodology is statistically sound, CMS did not 
have procedures for considering the impact of any revisions to state-
level error rates in calculating the national error rate after the 
cutoff date for each of the 3 measurement years. Specifically, the 
individual state error rates used to calculate the national error rate 
are not updated to reflect activities occurring after the PERM cycle 
cutoff. Without a process to consider these more current data on 
states' reported improper payment error rates, the reliability of 
CMS's reported national estimate may be adversely affected. OMB has 
identified as a best practice that agencies should establish a policy 
for handling unscheduled corrections to data, such as including 
threshold criteria identifying conditions under which data will be 
corrected and redisseminated.[Footnote 22] 

According to the PERM Manual, a state may request a new error rate 
calculation from CMS after the cycle cutoff date for informational 
purposes and for determining sample sizes for the next cycle under 
certain circumstances.[Footnote 23] For example, states may request a 
recalculation when information supporting a claim as correctly paid 
was submitted to CMS after the cycle cutoff date--but CMS's review 
contractor did not have time to complete the review--or when a mistake 
made by the PERM contractor was identified. This request must be made 
within 60 business days of the posting date of the state's program 
error rate on the CMS review contractor's website. In such instances, 
CMS will issue a revised rate to the state. However, each state's 
official error rate--used in the calculation of the national Medicaid 
error rate--will not change as a result of this recalculation. 
According to CMS, official error rates will be calculated based on 
information received by the cycle cutoff date. While CMS aims for a 
cycle cutoff date of July 15--4 months prior to the reporting date--
the CMS cycle manager may extend the cycle cutoff date depending on 
the progress of the PERM reviews.[Footnote 24] CMS officials 
acknowledged that historically CMS has had to postpone the cycle 
cutoff to allow the process to be as complete as possible while still 
permitting CMS to report an improper payment rate timely in HHS's AFR. 
However, after the cutoff date, CMS's PERM Manual does not allow for 
any revisions to be factored into a state's official error rate. 

In reviewing the results of state PERM reviews, we identified some 
instances where CMS issued revised state Medicaid error rates. For 
example, CMS issued a revised rate to one state for its eligibility 
reviews for the fiscal year 2008 measurement cycle because in January 
2010, two months after error rate reporting, CMS and the state 
discovered that the amount of dollars in error was reported 
incorrectly by the state. This revised overall state error rate 
estimate decreased from 20.8 percent to 7.8 percent.[Footnote 25] In 
another example for the same fiscal year 2008 measurement cycle, in 
December 2009, 1 month after error rate reporting, CMS issued a post-
cutoff date revised rate to a state for its FFS reviews because CMS 
received additional documentation from providers after the cycle 
cutoff date for official error rate calculations. This revised overall 
state error rate estimate decreased from 6.4 percent to 5.9 percent. 
[Footnote 26] These revised percentages were not included in the 
official error rates used to calculate the national estimate of 
Medicaid improper payments. While these were both smaller states and 
the actual impact on the national error rate would be minimal, CMS's 
PERM Manual does not provide for CMS to consider the impact and it is 
possible that these types of changes would have had an impact on the 
national error rate reported in the subsequent 2 years if the changes 
were significant and were for states with larger levels of outlays. 

Because the national error rate is based on 3 years of data and 
corrections to the 2 years of older data after the cutoff date are not 
officially recognized by CMS, the entire 3-year cycle could be 
affected. As a result, the reported estimate of Medicaid improper 
payments may be adversely affected if needed corrections are 
significant. This potentially affects CMS's ability to accurately 
report on the extent of improper payments, evaluate program 
performance, and utilize its own resources, as well as state 
resources, effectively to identify and reduce improper payments. 

Deficiencies in CMS's Guidance and Monitoring Impair Efficient and 
Effective Medicaid Program Improper Payments Reductions: 

CMS and state agencies developed CAPs that were generally responsive 
to identified payment errors. However, CMS's PERM Manual does not 
provide for addressing all nonpayment errors either by identifying 
specific corrective actions or by analyzing these errors to determine 
whether actions, if cost effective, are needed. Also, CMS's PERM 
Manual does not identify conditions under which corrective action for 
an error should not be undertaken because the cost of state corrective 
actions would outweigh the benefit. In addition, not all required 
elements of the CAPs are being completed by all states and CMS's 
written guidance on these required elements is not clear or 
consistent. Further, CMS's internal guidance on monitoring state CAPs 
is not sufficient to help ensure that states' CAPs contain all of the 
required elements and that states prevent and reduce improper payments 
going forward. 

Corrective Action Plans Developed and Responsive to Identified PERM 
Payment Errors: 

States are responsible for developing, executing, and evaluating CAPs 
to address specific errors identified during the PERM reviews, and CMS 
has reported on other initiatives to supplement state corrective 
actions and help reduce errors. We found that state CAPs were 
generally responsive to the types of payment errors identified in the 
PERM reviews. 

Through PERM, CMS identifies and classifies types of errors and shares 
this information with each state. States are then to analyze and 
determine the root causes for their specific improper payments. 
According to CMS, in addition to the PERM Manual, it provides guidance 
to state contacts on the CAP process upon providing the PERM error 
rates and throughout the CAP development. 

As reported by CMS, and shown in figure 3, overall, the majority of 
the errors reported in fiscal year 2011 (about 54 percent) for the 
Medicaid program--based on the fiscal years 2008 to 2010 measurement 
cycles--were a result of cases reviewed for eligibility, where 
recipients were either not eligible (25.3 percent) or where their 
eligibility status could not be determined (28.2 percent). The most 
common causes of cases in error for the FFS medical review was 
insufficient documentation (9.2 percent) or no documentation (4.3 
percent). Our analysis of error types is shown in appendix VI. 

Figure 3: Types of Errors Identified in the Fiscal Years 2008 to 2010 
PERM Measurement Cycles Based on Number of Errors: 

[Refer to PDF for image: pie-chart] 

* Undetermined (eligibility review), or no documentation or 
insufficient documentation (FFS medical review): 
Undetermined (eligibility review): 28.2%; 
No documentation (FFS medical review): 4.3%; 
Insufficient documentation (FFS medical review): 9.2%; 
Total: 41.7%; 
* Not eligible (eligibility review): 25.3%; 
* Others[A] (remaining 27 error types): 32.9%. 

Source: GAO analysis of CMS error data. 

Note: Values do not total 100 percent because of rounding. 

[A] The other 27 error types included instances where the incorrect 
number of units was billed, the state overpaid an amount of expenses, 
an individual was eligible for the Medicaid program but not a 
particular service, and payment did not correspond with the applicable 
pricing schedule. 

[End of figure] 

As shown in figure 3, almost 42 percent of reported PERM review errors 
resulted from documentation deficiencies, including either a lack of 
or insufficient documentation, or because a definitive review decision 
could not be made because of a lack of or insufficient documentation 
(undetermined). As these are common types of errors, CMS has reported 
on certain corrective actions that states have developed to address 
them. Specifically: 

* No documentation and insufficient documentation. In about 14 percent 
of all PERM errors, reviewers identified errors because either the 
provider did not respond to the request for records within the 
required time frame (no documentation--4.3 percent) or there was not 
enough documentation to support the service (insufficient 
documentation--9.2 percent).[Footnote 27] According to CMS, because 
much of the error rate in the past was due to missing or insufficient 
documentation, the majority of states focused on provider education 
and communication methods to improve the providers' responsiveness and 
timeliness. 

* Undetermined. In about 28 percent of all PERM errors over the 3-year 
period, reviewers were unable to determine whether or not a 
beneficiary was eligible for Medicaid because the case record lacked 
or contained insufficient documentation.[Footnote 28] The PERM Manual 
outlines the due diligence a state must take before citing the case as 
"undetermined." According to CMS, specific corrective action 
strategies implemented by the states to reduce these types of 
eligibility errors have included leveraging technology and available 
databases to obtain eligibility verification information without 
client contact;[Footnote 29] providing additional caseworker training, 
particularly in areas determined by the PERM review to be error prone; 
and providing additional eligibility policy resources through a 
consolidated manual and web-based training. 

In addition to the state-specific CAPs that are developed in response 
to the PERM findings, CMS has reported on other initiatives to lower 
error rates in HHS's fiscal year 2011 AFR. For example, to help 
address the insufficient documentation errors found in medical 
reviews, CMS reported that it increased its efforts to reach out to 
providers and to obtain medical records to help resolve this problem. 
CMS also reported that it gives states more information on the 
potential impact of these documentation errors and more time for the 
states to work with providers to resolve them. Table 1 outlines CMS's 
reported overall strategies to reduce improper payments and strategies 
targeted at specific PERM error types. 

Table 1: CMS's Corrective Action Strategies to Reduce PERM-Identified 
Improper Payments: 

Corrective action focus: Facilitate communication; 
Corrective action strategy: 
* Best practice calls. CMS conducts national best practice conference 
calls to facilitate idea sharing and lessons learned among the states 
in order to decrease improper payments; 
* Provider open forum calls. CMS reported that it has sponsored a 
series of provider open forum calls for all states at the beginning of 
the PERM review cycle; 
* PERM website. CMS reported that it has enhanced the PERM website 
with up-to-date information and has included a separate web page for 
providers and an e-mail account for providers to communicate directly 
with CMS; 
* Post-CAP visits or webinars. CMS conducts post-CAP on-site visits or 
webinars with the states. The information covered during each meeting 
includes a recap of the previous PERM cycle, the disclosure of 
improper payment trends, the strategies for success in the upcoming 
PERM cycle, a review of previous CAPs submitted, a discussion of 
upcoming PERM initiatives, an overview of the various CMS work groups, 
and a summary of applicable Office of Inspector General audits. 

Corrective action focus: Training and technical assistance; 
Corrective action strategy: 
* Medicaid Integrity Institute. CMS's Medicaid Integrity Institute is 
the first national Medicaid integrity training program and offers 
state officials training and opportunities to develop relationships 
with program integrity staff from other states; 
* Systems work group. CMS formed a state systems workgroup to address 
individual state system problems that may cause payment errors. The 
work group includes representatives from HHS and state staff; 
* Provider education webinars. CMS sponsors a series of PERM provider 
education webinars to educate providers on the PERM program and 
documentation submission requirements. 

Corrective action focus: Improvements to PERM methodology; 
Corrective action strategy: 
* PERM+. CMS developed PERM+, a new method for states to submit claims 
data for the PERM review. According to CMS, PERM+ makes claims data 
submission easier for states and condenses the PERM audit timeline; 
* Aggregate payments. CMS developed an aggregate payment methodology 
that, if appropriate, allows aggregate payments to be submitted and 
sampled for PERM, rather than at the beneficiary level.[A]; 
* Eligibility instructions. CMS reported that it is working with 
states to identify areas for clarification in the eligibility 
instructions to ensure an accurate measurement across the states. 

Source: GAO analysis of CMS's corrective action strategies. 

[A] While most Medicaid payments are made at the beneficiary level, 
states also calculate and pay for some services on behalf of a group 
of beneficiaries. This is known as an aggregate payment. 

[End of table] 

Although all states developed CAPs that were generally responsive to 
the payment errors identified through PERM reviews, we were unable to 
assess the CAPs' impact on the improper payment error rate because of 
limited comparative data between PERM measurement cycle years. 

Nonpayment Errors Not Addressed in All State CAPs: 

State CAPs did not always address errors identified during PERM 
reviews that did not have a payment error amount associated with them. 
Specifically, we identified three types of these nonpayment errors 
through our analysis of the PERM process that are not consistently 
addressed in all state CAPs--negative case errors, deficiencies, and 
technical errors. 

* A negative case error occurs when a state incorrectly denies an 
application or terminates eligibility. 

* A deficiency is generally defined as an action or inaction on the 
part of the state or the provider that could have resulted in a dollar 
error but did not. 

* A technical error is an error where the eligibility caseworker did 
not act in accordance with state or federal policy, but this did not 
result in an erroneous eligibility determination or result in a 
difference between the amount that was paid and the amount that should 
have been paid. 

CMS's PERM Manual requires that states test negative cases as part of 
their eligibility reviews. However, it does not clearly require that 
states address negative case errors in their CAPs. While a payment 
error rate is not calculated because there are no payments associated 
with negative cases, a negative case error rate is calculated to 
estimate the percentage of the decisions in which eligibility was 
incorrectly denied or terminated. Our analysis showed that for fiscal 
year 2011 reporting, approximately 40 percent of the states where 
negative case errors were identified did not address negative case 
errors in their CAPs. According to CMS officials, these negative 
errors should be included in state CAPs. 

While deficiencies do not result in a dollar amount in error and 
therefore had no impact on the payment error rate for fiscal year 
2011, they may represent issues that need to be addressed to prevent 
future payment errors. Although not considered payment errors, some 
deficiencies were noted during PERM data processing and medical 
reviews.[Footnote 30] Examples of deficiencies identified in FFS and 
managed care reviews include the following: 

* A data processing deficiency in which a male was coded as a female 
in the system but because the service provided could have been 
appropriate for either sex, it did not result in a dollar difference. 

* A medical deficiency wherein although a provider billed for the 
wrong procedure code, the correct procedure code would have paid the 
same rate per unit. Therefore, it did not result in a dollar 
difference but could have under other circumstances. 

Our analysis showed that deficiencies identified in PERM reviews 
represented approximately 8 percent of the total FFS and managed care 
errors identified for the fiscal year 2011 reporting, and that 
approximately 67 percent of these deficiencies were not included or 
analyzed in state CAPs. In addition, only 10 of the 43 states with 
deficiencies addressed these deficiencies in their CAPs.[Footnote 31] 
While the PERM Manual does not clearly state that CAPs are to address 
deficiencies, CMS officials told us that states should address 
deficiencies in their CAPs. 

During eligibility reviews, states may identify technical errors. An 
example of a technical error is a failure to follow state 
administrative procedures that do not affect eligibility if acceptable 
documentation is otherwise obtained that supports beneficiary 
eligibility. According to the PERM Manual, states are not currently 
required to report these technical errors to CMS and may document 
technical errors as appropriate during the PERM reviews.[Footnote 32] 
Furthermore, the PERM Manual suggests but does not require that states 
include an analysis of technical errors and related corrective actions 
in their CAPs. 

Although these nonpayment errors did not result in improper payment 
amounts, they represent internal control deficiencies that could have 
prevented eligible beneficiaries from receiving Medicaid benefits or 
may result in improper payments in future years if not addressed. Not 
clearly requiring states to address nonpayment errors, or to document 
that sufficient analysis was performed to determine if corrective 
actions, if cost effective, are needed, may reduce the effectiveness 
of CAPs for addressing the underlying causes of improper payments. 
Further, this may inhibit ongoing efforts to prevent and reduce 
improper payments and to ensure that Medicaid is provided to all 
eligible beneficiaries. 

OMB's implementing guidance for IPERA requires agencies to implement 
corrective actions to prevent and reduce improper payments.[Footnote 
33] In addition, CMS's PERM regulations and its PERM Manual require 
each state to complete and submit a CAP based on errors found during 
the PERM process. However, while specifically allowing states to 
exclude eligibility technical errors, the PERM Manual does not clearly 
identify whether the states should consider or include deficiencies or 
negative case errors in their CAPs. While the PERM Manual does not 
clearly state that CAPs are to address both deficiencies and negative 
case errors, CMS officials told us that states should address both of 
these in their CAPs. 

Requirements for State CAPs Do Not Provide for Considering Costs 
versus Benefits of Corrective Actions: 

Although CMS's PERM Manual requires each state to complete and submit 
a CAP based on the errors found during the PERM process, this guidance 
makes no exception for small errors--sometimes caused by rounding--
which may result in states incurring costs to implement corrective 
actions that exceed the benefits of those actions. In its PERM Manual, 
CMS encourages states to use the most cost-effective corrective 
actions that can be implemented to best correct and address the root 
causes of the errors; however, it does not acknowledge that states can 
address errors by documenting situations where they determined that 
the costs of implementing the corrective action exceed the benefits. 

Officials at one state we visited told us that the cost of 
implementing a system to correct some of its errors that were less 
than a dollar would outweigh the benefits of this action. A PERM 
review in this state identified 11 pricing errors resulting from 
incorrect rounding that netted to $0.53.[Footnote 34] State officials 
informed us that they were aware of this rounding issue, as it had 
been identified in the previous PERM cycle and CMS also identified and 
reported this type of error for the fiscal year 2011 measurement 
cycle. According to this state, the original estimate for a system 
solution to correct these rounding errors was $575,000 to $1,150,000. 
State officials told us they did not believe that the cost to address 
this issue was justified as the return on investment for the system 
solution to correct the condition might never be realized. According 
to CMS, in e-mail communication with this state, it told state 
officials that if the state determines that the cost of implementing a 
corrective action outweighs the benefits then the final decision of 
implementing the corrective action is the state's decision. The state 
continued to pursue corrective actions and was ultimately able to 
obtain a revised estimate of $115,000 for changes to the system, based 
on further detailed analysis of the necessary solution. The state now 
plans to redesign its system in order to avoid these types of PERM 
errors going forward. 

According to Standards for Internal Control in the Federal Government, 
management should design and implement internal controls--in this 
case, controls to prevent and reduce improper payments--based on the 
related costs and benefits. Further, PERM regulations require states 
to evaluate their corrective action plans by assessing, among other 
things, the efficiencies that they create. However, the lack of clear 
written guidance for states on how to address situations where the 
cost of corrective actions identified by states may outweigh the 
benefits because of the low dollar amounts associated with these types 
of errors may result in an unnecessary burden on state resources. 

State CAPs Do Not Contain All Required Elements: 

Although we found that states have generally been engaged in the PERM 
CAP process and developed CAPs to address improper payment errors, not 
all required elements of the CAPs are being completed by all states. 
When developing CAPs, CMS's PERM regulations require states to perform 
five key steps to reduce improper payment errors identified through 
the PERM reviews. For CAPs subsequent to the initial measurement year, 
CMS's PERM regulations also require an update on the previous CAP. 
These requirements are summarized in figure 4. 

Figure 4: CMS's Regulatory Requirements for States to Develop CAPs to 
Reduce Improper Payments: 

[Refer to PDF for image: table] 

State CAPs - required steps: 
        
1. Conduct data analysis: 
States must conduct data analysis, such as reviewing clusters of 
errors, general error causes, characteristics, and frequency of errors 
that are associated with improper payments. 

2. Identify programmatic and root causes of errors: 
States must review the findings of the data analysis to determine the 
specific programmatic causes to which errors are attributed (for 
example, provider lack of understanding of the requirement to provide 
documentation) and to identify root error causes. 

3. Corrective action planning: 
States must determine the corrective actions to be implemented that 
address the root error causes. 

4. Evaluate plan: 
States must evaluate the effectiveness of the corrective action by 
assessing improvements in operations, efficiencies, number of errors, 
and improper payments. 
        
Subsequent state CAPs: 

1. States must submit to CMS a new CAP for each subsequent error rate 
measurement that contains an update on the status of a previous CAP. 
The update should address the effectiveness of implemented corrective 
actions, findings on short-term corrective actions, and the status of 
the long-term corrective actions. 

2. States should also address discontinued or ineffective actions, 
actions not implemented, and those actions, if any, that were 
substituted for such discontinued, ineffective, or abandoned actions. 

Source: 42 C.F.R. § 431.992. 

[End of figure] 

Not all required elements of the CAPs--such as the evaluation step or 
the update on the previous CAP--were consistently reported on by all 
states. 

* For example, for fiscal year 2011 reporting, 8 of the 51 states did 
not submit the required evaluation element of the CAP. An additional 9 
states submitted the evaluation element for some, but not all PERM 
components. 

* Furthermore, for fiscal year 2011 reporting, only 24 of the 34 
states required to submit an update of the previous CAP complied with 
this requirement. Another 5 states submitted updates for some, but not 
all, of the PERM components, and of the 29 states that submitted 
complete or partial updates of their previous CAPs, only 19 submitted 
them by the due date required by CMS. The other 10 were submitted 
after CMS followed up with the states. 

CMS officials acknowledged that some state CAPs are missing certain 
elements, and they are in the process of finalizing specific 
procedures to outline CMS's role in reviewing state CAPs and following 
up with states to obtain any missing elements, as discussed later in 
this report. 

CMS's PERM Manual, updated in September 2011, provides guidance for 
state CAP development, but it does not include specific instructions 
for completing the evaluation element or on how to report the update 
on the previous CAP. Furthermore, the CAP template included in the 
PERM Manual does not include these two required elements. However, on 
its PERM website, CMS has provided a separate example of a CAP for the 
states to utilize that includes examples of the evaluation element and 
a separate report for the update on the previous CAP. Inconsistencies 
between the PERM Manual--which includes a CAP template--and the 
example CAP on the PERM website may cause confusion regarding what 
states are to include in their CAPs. As of August 2012, CMS had 
updated its PERM Manual and the CAP template to include instructions 
and a template for reporting on the update of the previous CAP. 
However, the updated template still did not include the evaluation 
element, and the separate example of a CAP on the PERM website was not 
updated to be consistent with the updated PERM Manual guidance and 
template. Clear, consistent written guidance and instructions on all 
required elements for CAPs would assist the states in submitting 
complete CAPs, and increase the likelihood that CMS has the 
information necessary for analyzing the progress and effectiveness of 
state CAPs. The lack of clear, consistent guidance in the PERM Manual 
and the related template on the PERM website on how to develop key 
elements of the state CAP may have contributed to the missing elements 
we describe in this report. 

CMS's Monitoring of State CAPs Is Limited: 

CMS lacked a formal policy describing its role in monitoring state 
CAPs to ensure that (1) the CAPs contained all of the required 
elements and completely addressed errors identified in the PERM 
reviews and (2) states were making progress on implementing corrective 
actions. 

In our high-risk series update, we reported that CMS needs to ensure 
that states develop appropriate corrective action processes to address 
vulnerabilities to improper Medicaid payments.[Footnote 35] Our 
analysis of state CAPs continues to identify issues regarding CMS's 
coordination with states in developing and implementing their CAPs. 
Specifically, during our review and analysis of state CAPs for the 
fiscal years 2008 to 2010 PERM measurement cycles, we found that CMS 
had not conducted sufficient oversight to ensure that states submitted 
complete CAPs, took the five required steps in developing CAPs, and 
updated the status of previous CAPs. 

As discussed previously, not all required elements of the CAPs--such 
as the evaluation step or the update on the previous CAP--were being 
completed by all states. Once the CAPs are submitted, officials in the 
seven states we visited noted that there was minimal monitoring of 
implementation by CMS. For example, officials in one state told us 
that CMS did not follow-up with the state on the implementation of the 
corrective actions until the state submitted the CAP related to its 
next error rate measurement 3 years later. According to CMS officials, 
they do not track the progress of the states' implementation of CAPs 
and are not required to do so.[Footnote 36] However, CMS officials 
told us that they review the implementation information that the 
states provide in their CAPs, specifically in the update of their 
previous CAPs, and hope to see a reduction in error rates as the CAPs 
are implemented. Additionally, based on our analysis of state CAPs for 
fiscal year 2011 reporting, we also noted that approximately 5 percent 
of all payment errors identified during the PERM reviews were not 
fully addressed by all states in their CAPs. Improved monitoring by 
CMS would help ensure that state CAPs contain all of the required 
elements and are addressing all types of errors identified through the 
PERM process, and that the actions identified are appropriate to 
reduce those types of errors going forward. 

The responsibility for oversight of the states' development, 
implementation, and evaluation of their CAPs rests with the Division 
of Error Rate Measurement (DERM) within CMS's Office of Financial 
Management. These efforts include coordinating the CAP process with 
the states and other agency offices. The Medicaid Integrity Group 
(MIG)[Footnote 37] within CMS's Center for Program Integrity is 
responsible for reviewing the state CAPs, with assistance from the 
agency's regional offices. According to CMS, MIG reviews the state 
CAPs to (1) ensure the plans address the errors identified during the 
PERM reviews, (2) provide feedback to the states for improvements, and 
(3) review the implementation status of the state's previous CAP. 

Oversight through continuous monitoring helps ensure that actions are 
taken to effectively work toward reducing improper payments. According 
to OMB's implementing guidance, agencies must ensure that their 
managers and accountable officers, program and program officials, and 
where applicable states and local partners are held accountable for 
reducing improper payments.[Footnote 38] Therefore, although the 
states are responsible for developing, implementing, and monitoring 
their CAPs, CMS should be responsible for monitoring states' 
compliance with CMS's regulations related to the PERM process. 

We also found that the roles and responsibilities of DERM and MIG are 
not formally outlined in policies and procedures for the PERM review 
and corrective action process. CMS officials told us that they are in 
the process of developing protocols to address the CAP review process. 
Specifically, CMS officials told us that they have developed a draft 
policy describing each party's role in the different stages of the 
PERM CAP process as well as a review guide to outline CMS's procedures 
for coordinating reviews of state CAPs. CMS plans to review state CAPs 
submitted in February 2013 using this new collaborative process for 
the first time for the states that are part of the fiscal year 2011 
measurement cycle and were reported on in HHS's fiscal year 2012 AFR. 
According to CMS officials, they plan to review the CAPs to ensure 
that all of the attributes outlined in the PERM regulations are 
addressed and, as needed, notify the states of any missing elements. 
After reviewing the fiscal year 2011 cycle CAPs, CMS officials told us 
that they plan to further refine the standard operating procedures and 
CAP review guide before the documents are finalized. CMS's draft 
policy and review guide were not finalized before the completion of 
our fieldwork, and we did not examine any interim drafts. Thus, we are 
unable to determine whether the planned revisions to existing 
procedures will fully address the deficiencies we identified 
concerning CMS's monitoring of state CAPs. Monitoring is CMS's 
opportunity to ensure that states are appropriately implementing the 
corrective actions that they have identified to help reduce improper 
payments. If states are not addressing all applicable issues or are 
not effectively implementing the actions outlined in their CAPs, 
future reductions in the Medicaid error rate may be limited. 
Additional monitoring by CMS would help hold the states accountable 
for developing, implementing, and evaluating corrective action 
strategies in support of CMS's efforts to prevent and reduce Medicaid 
improper payments. 

Conclusions: 

The design of CMS's PERM methodology is statistically sound. However, 
refining the required PERM process for estimating and reporting 
national Medicaid improper payments so that the impact of corrections 
to the data after the cutoff date is considered would help ensure that 
the reported estimates are reasonably accurate and complete. As CMS 
reports its estimated Medicaid improper payments based on a rolling 3-
year estimate, adjustments made to any of these 3 years can affect 
yearly reporting and potentially affect the accuracy of the reported 
national estimate. Given the importance of providing HHS management, 
OMB, and the Congress with accurate information on the extent of 
improper payments in federal programs, it is imperative that CMS 
ensure that its reported estimates of Medicaid improper payments are 
reliable. 

Corrective actions are critical for preventing and reducing improper 
payments. While states have developed corrective action plans to 
address payment errors identified in PERM reviews, not all nonpayment 
errors were addressed in these plans, which could hinder the 
prevention of future improper payments. Also, while states are 
currently required to address all errors, clear written guidance that 
permits states to document why an action is not being implemented 
would help ensure the most efficient and effective use of state 
resources for errors that do not pose a risk of significantly 
affecting future improper payments. Further, ensuring that states have 
clear written guidance for developing corrective action plans is key 
to CMS's ability to oversee states' corrective action processes. 
Strengthening CMS's required procedures for monitoring the state-level 
corrective actions is critical to help ensure that states make 
progress in preventing and reducing improper payments. 

Recommendations for Executive Action: 

In order to ensure the accuracy of reported improper payment estimates 
for the Medicaid program, we recommend that the Secretary of HHS 
direct the CMS Administrator to take the following action: 

To help ensure that corrective action strategies effectively address 
identified types of improper payments and reduce Medicaid improper 
payments in a cost-effective manner, we recommend that the Secretary 
of HHS direct the CMS Administrator to take the following three 
actions: 

* Update PERM Medicaid improper payment reporting procedures to 
provide for considering any corrections to state-level improper 
payment error data subsequent to the cutoff date that would have a 
significant impact on any of the 3 years used to develop the rolling 
average for the reported national Medicaid improper payment estimate. 

* Revise the PERM Manual to provide that states (1) analyze all 
deficiencies, negative case errors, technical errors, and minimal 
dollar errors identified in PERM reviews to determine if any 
corrective actions, if cost effective, are needed to prevent such 
errors in the future and (2) document the results of their analysis. 

* Clarify guidance in the PERM Manual, and on the PERM website, on the 
required elements to be included in a CAP and the specific actions 
states are to take each measurement cycle to (1) effectively prepare 
and evaluate their current cycle's CAPs and (2) provide updates to 
their previous cycle's CAPs. 

* Finalize draft policies and procedures to clarify specific CMS 
officials' roles and responsibilities for monitoring states' 
corrective actions to ensure, at a minimum, that (1) the CAPs contain 
all of the required elements and completely address errors identified 
in the PERM reviews and (2) states are making progress on implementing 
corrective actions. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to the Secretary of HHS for 
comment. In its written comments, reprinted in appendix VII, HHS 
concurred with the four recommendations in our report. HHS cited a 
number of actions already taken and other initiatives planned or under 
way related to our recommendations. For example, with respect to our 
three recommendations to help ensure that corrective action strategies 
effectively address identified types of improper payments and reduce 
Medicaid improper payments in a cost-effective manner, HHS cited CMS's 
plans to update its PERM Manual and other relevant documents 
consistent with our recommended actions to clarify and standardize 
guidance. HHS also cited action under way to finalize policies and 
procedures related to monitoring states' corrective actions. 

HHS also concurred with our recommendation to update procedures for 
considering the impact of any corrections to state-level improper 
payment errors on reported national error rates. HHS stated that it 
will consider revising its procedures in this area. HHS also expressed 
concern that the draft suggests that past reported national Medicaid 
error rates were unreliable. We acknowledged in our draft report that 
the prior year post-cutoff date error rate revisions we reviewed were 
not sufficient to have had an impact on the national error rate for 
fiscal year 2011 reporting. Rather, our recommendation is focused on 
augmenting procedures to help ensure the reliability of future 
national error rate reporting. HHS also expressed concern about our 
suggestion that OMB's Standards and Guidelines for Statistical Surveys 
should be used to determine how to handle PERM-related data 
corrections. In our draft report, we characterized this as a best 
practice. HHS noted, and we agree, that OMB did not include guidance 
for handling unscheduled corrections to data in its implementing 
guidance for IPERA. However, taking action, as we recommended, to 
establish procedures to consider the extent to which any corrections 
to state-level improper payment data subsequent to the cutoff date 
would affect the reported national Medicaid improper payment estimate 
would best ensure the reliability of reported national error rates 
going forward. HHS also expressed concern about our suggestion that 
states may request a recalculation of the state-level error rate when 
records for a medical claim were received prior to the cycle cutoff 
date but CMS's review contractor did not have time to complete the 
review. HHS cited that CMS's review contractors will complete all 
reviews for claims where the documentation was received prior to the 
cycle cutoff date and that states may request a recalculation when 
information supporting a claim as correctly paid was submitted to CMS 
after the cycle cutoff date. We agreed with HHS's point and modified 
the report accordingly. 

HHS also expressed concern about including the state error rates 
identified in appendix III of the draft. HHS commented that readers 
may use the rates to make state-to-state comparisons that are 
inappropriate because of variations in states' sizes and programs and 
in states' implementation and administration of their programs. We 
acknowledged HHS's concerns in our draft report by including language 
in appendix III to caution readers about using these state-level rates 
to make state-to-state comparisons. However, it is important to 
present these state-level error rates for transparency regarding the 
results of state PERM reviews. 

In addition, HHS provided technical comments that we incorporated as 
appropriate and discussed in our additional evaluation in appendix VII. 

As agreed with your offices, 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 to the 
appropriate congressional committees, the Secretary of Health and 
Human Services, and other interested parties. In addition, the report 
will be available at no charge on the GAO website at [hyperlink, 
http://www.gao.gov]. 

If you or your staffs have any questions about this report, please 
contact me at (202) 512-2623 or davisbh@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. GAO staff members who made key 
contributions to this report are listed in appendix VIII. 

Signed by: 

Beryl H. Davis: 
Director: 
Financial Management and Assurance: 

[End of section] 

Appendix I: Objectives, Scope, and Methodology: 

The objectives of this report were to determine the extent to which 
(1) the Centers for Medicare & Medicaid Services' (CMS) methodology 
for estimating Medicaid improper payments follows Office of Management 
and Budget (OMB) guidance and produces reasonable national and state-
level estimates and (2) corrective action plans (CAP) have been 
developed to reduce Medicaid payment error rates and whether these 
plans addressed the types of payment errors identified. 

To address these objectives, we reviewed the Improper Payments 
Information Act of 2002 (IPIA),[Footnote 39] the Improper Payments 
Elimination and Recovery Act of 2010 (IPERA),[Footnote 40] and related 
OMB guidance effective for fiscal year 2011.[Footnote 41] We also 
reviewed CMS regulations on Payment Error Rate Measurement (PERM) 
[Footnote 42] and CMS's internal written guidance on PERM. In 
addition, we reviewed results from state PERM reviews for fiscal years 
2006 through 2011, prior GAO and Department of Health and Human 
Services (HHS) Office of Inspector General reports, and internal 
control standards. Further, we reviewed improper payment information 
reported in the Improper Payments Section of HHS's fiscal year 2011 
agency financial report (AFR). We reviewed these documents to 
understand CMS's efforts to address IPIA and IPERA requirements and to 
identify previously reported issues with CMS's improper payment 
reporting. 

To further determine the extent to which CMS's methodology for 
estimating Medicaid improper payments follows OMB guidance and 
produces reasonable national and state-level estimates, we compared 
the following components of CMS's methodology for estimating the 
fiscal year 2011 payment error rate with related OMB guidance: (1) 
sampling methods, including the sample size, sample selection, sample 
representation, and precision of the estimates, and (2) statistical 
methods used to estimate the error rates and precision. As part of 
this assessment, we did the following: 

* Conducted interviews with CMS officials and its contractors to 
clarify our understanding of both the sampling and estimation 
methodologies. 

* Reviewed the program manuals for both the payment error and 
eligibility payment error components of PERM to assess the statistical 
validity of CMS's methodology. 

* Reviewed professional statistical literature to validate the 
suitability of stratified random sampling and ratio estimation to 
address the particular characteristics of the payment and eligibility 
data in the state-administered Medicaid program. 

* Reviewed state-level payment error rates from the most recent year 
available to determine whether the sample sizes assigned to states met 
the precision level for payment error sampling in OMB statistical 
guidance. 

We also used the results of these reviews and analyses to identify and 
assess the reasons for any weaknesses in the estimation methodology 
and their potential effects on identifying and reporting Medicaid 
improper payment estimates for fiscal year 2011 and going forward. 

In addition to reviewing the statistical methodology, we obtained 
actual payment error data from CMS for the seven states selected for 
our site visits and independently calculated the payment error rates 
to confirm the calculations done by CMS using the statistical 
methodology specified in the program manuals. The basis for our site 
visit selection is discussed later in this appendix. 

The scope of our review did not include an assessment of individual 
states' processes or payment systems. We assessed the reliability of 
the claims and error rate data by gaining an understanding of the 
processes the contractors or states use to perform their reviews, 
including any use of data sharing to determine eligibility, and their 
quality controls. We determined that the data were sufficiently 
reliable for our purposes. 

To further determine the extent to which CAPs have been developed to 
reduce Medicaid payment error rates and whether these plans addressed 
the types of errors identified, we did the following: 

* Reviewed agency policies and procedures related to the development 
of PERM CAPs and CAPs for all 50 states and the District of Columbia, 
which are used to address the root causes of improper payments 
identified from the PERM reviews. 

* Conducted interviews with officials from CMS related to its 
oversight role and its own initiatives for reducing Medicaid improper 
payments. 

* Reviewed CMS's error rate reduction plans and initiatives to reduce 
Medicaid improper payments. 

* Reviewed the reported causes of improper payments as outlined in 
HHS's fiscal year 2011 AFR. 

* Assessed CMS's process for monitoring state corrective actions and 
its methodology for measuring the effectiveness of corrective actions 
to reduce improper payments. 

As part of our review of states' CAPs, we assessed whether they: 

* addressed issues identified in fee-for-service, managed care, and 
eligibility reviews; 

* included the required elements as outlined by CMS; and: 

* evaluated the effectiveness of implemented corrective actions. 

The scope of our review did not include an assessment of individual 
states' implementation of their CAPs. 

In addition, we conducted site visits at seven state Medicaid offices 
(California, Florida, Illinois, Michigan, Pennsylvania, South 
Carolina, and Texas). During these site visits, we interviewed state 
personnel involved in the PERM process to gain an understanding of how 
states compile the universes of claims and beneficiaries that are 
sampled for the PERM reviews, how eligibility reviews are conducted, 
and how the states develop corrective action plans and work with CMS 
on corrective actions. We selected these states based on criteria such 
as the states' federal share of Medicaid payments and errors 
identified in PERM reviews. The seven states we visited collectively 
claimed about 37 percent of the total federal share of Medicaid 
payments made in fiscal year 2010, the most recent data available at 
the time of our review for site visit selection. We also selected 
these states to achieve variation in the error rates found during PERM 
reviews included in the fiscal year 2011 reporting of the Medicaid 
improper payment estimate.[Footnote 43] One state had the highest 
error rate for eligibility reviews as well as the highest combined 
error rate. This selection also allowed us to focus on certain states 
with noted vulnerabilities in program integrity efforts, as well as 
states with possible best practices. Although it does not allow us to 
generalize findings to all states and thus the program as a whole, we 
believe these state visits, combined with our analysis of CAPs for all 
states, enable us to determine if states' corrective actions are 
addressing the types of improper payment errors that have been 
identified. 

We conducted this performance audit from February 2012 to March 2013 
in accordance with generally accepted government auditing standards. 
Those standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe 
that the evidence obtained provides a reasonable basis for our 
findings and conclusions based on our audit objectives. 

[End of section] 

Appendix II: Fee-for-Service, Managed Care, and Eligibility Components 
of the Medicaid Error Rate Measurement: 

The Payment Error Rate Measurement (PERM) program uses a 17-state, 3-
year rotation for measuring Medicaid improper payments. Medicaid 
improper payments are estimated on a federal fiscal year basis through 
the PERM process. The estimate measures three component error rates: 
(1) fee-for-service (FFS), (2) managed care, and (3) eligibility. 

FFS and Managed Care Components: 

FFS is a traditional method of paying for medical services under which 
providers are paid for each service rendered. Managed care is a system 
where the state contracts with health plans to deliver health services 
through a specified network of doctors and hospitals. The health plan 
is then responsible for reimbursing providers for specific services 
delivered. 

States submit quarterly adjudicated claims data from which a randomly 
selected sample of FFS and managed care claims are drawn each quarter. 
Each selected FFS claim is subjected to a data processing review. The 
majority of FFS claims also undergo a medical review.[Footnote 44] 
Managed care claims are subject only to a data processing review. 
[Footnote 45] 

* A data processing error is a payment error that can be determined 
from the information available from the claim or from other 
information available in the state Medicaid system, other related 
systems, as well as outside sources of provider verification (except 
medical reviews and eligibility reviews). Data processing errors 
include, but are not limited to, the following: payment for duplicate 
items, payment for noncovered services, payment for FFS claims for 
managed care services, payment for services that should have been paid 
by a third party but were inappropriately paid by Medicaid, pricing 
errors, logic edit errors, data entry errors, and managed care payment 
errors. 

* A medical review error is an error that is determined from a review 
of the medical documentation in conjunction with state and federal 
medical policies and information presented on the claim. Medical 
review errors include, but are not limited to, the following: lack of 
documentation, insufficient documentation, procedure coding errors, 
diagnosis coding errors, number of unit errors, medically unnecessary 
services, policy violations, and administrative errors. 

Eligibility Component: 

Eligibility refers to meeting the state's categorical and financial 
criteria for receipt of benefits under the Medicaid program. States 
perform their own eligibility reviews according to state and federal 
eligibility criteria. An eligibility error occurs when a person is not 
eligible for the program or for a specific service and a payment for 
the service or a capitation payment covering the date of service has 
been made.[Footnote 46] An eligibility error can also occur when a 
beneficiary has paid the incorrect amount toward an assigned liability 
amount or cost of institutional care. The results from the eligibility 
reviews will include eligibility errors based on erroneous decisions 
as well as payment errors. The Centers for Medicare & Medicaid 
Services (CMS) combines the state-reported eligibility component 
payment error rates to develop a national eligibility error rate for 
Medicaid. This rate is calculated from the active case payment review 
findings.[Footnote 47] For fiscal year 2011 reporting, CMS estimated 
that the active case error rate was 8.2 percent while the weighted 
eligibility component error rate was 6.1 percent. 

Eligibility reviews are also performed on a sample of negative cases. 
[Footnote 48] Negative cases contain information on a beneficiary who 
applied for benefits and was denied or whose program benefits were 
terminated based on the state agency's eligibility determination in 
the month that eligibility is reviewed. CMS calculates only a case 
error rate for negative cases, because no payments were made. The 
negative case error rate estimates the percentage of the decisions in 
which eligibility was incorrectly denied or terminated. For fiscal 
year 2011 reporting, CMS estimated that the negative case error rate 
was 4.9 percent. The results of all PERM reviews, including the 
negative case reviews, are used to determine future sample sizes. 

[End of section] 

Appendix III: State Error Rates for the Department of Health and Human 
Services' Fiscal Year 2011 Reporting of Medicaid Improper Payments: 

According to the Centers for Medicare & Medicaid Services (CMS), 
states' Medicaid improper payment error rates identified through the 
Payment Error Rate Measurement (PERM) program may vary because of 
multiple factors related to differences in how states implement and 
administer their programs and should be considered in the context of 
these differences and operational realities. CMS provides each state 
its specific error rate and data analysis reports to use to develop 
corrective actions designed to reduce major error causes and to 
identify trends in errors or other factors for purposes of reducing 
improper payments. Also, according to CMS, because of the variation of 
states' sizes, overall program variations, and different ways that 
each state's rate affects the national rate, CMS does not encourage 
comparisons based solely on error rates. PERM is designed to produce 
precise error rates at the national level. Therefore, according to 
CMS, sample sizes per state are relatively small and the precision of 
state-specific error rates varies significantly. 

In addition, during the fiscal years 2008 and 2009 measurement cycles, 
CMS noted instances where some states' policies differed from CMS's 
policies for determining PERM errors. For example, according to CMS, 
in the review of some eligibility cases, policy and operational 
differences among states may have affected the degree to which states 
and providers could obtain documentation to validate payments and 
eligibility decisions for PERM purposes. According to CMS, states that 
have simplified eligibility documentation rules through use of self-
declaration and administrative renewal often found it harder to obtain 
necessary documentation for PERM reviews, which were treated as errors 
for PERM. In its fiscal year 2011 agency financial report (AFR), the 
Department of Health and Human Services (HHS) reported that as 
required under Section 601 of the Children's Health Insurance Program 
Reauthorization Act of 2009,[Footnote 49] it published a final rule on 
August 11, 2010, effective September 30, 2010, which required the 
eligibility reviews to be consistent with the state's eligibility 
verification policy rather than reviewing eligibility against a 
single, federal methodology, which was done in the past. After 
publication of the final rule, states were allowed to review cases 
under the new methodology. HHS also reported that based on current 
regulations, certain cases from the fiscal years 2008 and 2009 
measurement cycles, included in the error rates below, would no longer 
be considered as errors. 

Table 2 provides a list of state error rates used to determine HHS's 
fiscal year 2011 reporting of national Medicaid improper payments. 

Table 2: State Error Rates Used to Determine HHS's Fiscal Year 2011 
Reporting of National Medicaid Improper Payments: 

Measurement cycle[A]Fiscal year 2008: [Empty]. 

Measurement cycle[A]: Fiscal year 2008; 

State: Alaska; 
Combined[B]: Error rate: 0.6%; 
Combined[B]: Margin of error: 0.5%; 
Fee-for-service (FFS): Error rate: 0.5%; 
Fee-for-service (FFS): Margin of error: 0.5%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 0.1%; 
Eligibility: Margin of error: 0.2%. 

State: Arizona; 
Combined[B]: Error rate: 2.6%; 
Combined[B]: Margin of error: 1.9%; 
Fee-for-service (FFS): Error rate: 2.7%; 
Fee-for-service (FFS): Margin of error: 1.2%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 2.2%; 
Eligibility: Margin of error: 1.9%. 

State: District of Columbia; 
Combined[B]: Error rate: 20.1%; 
Combined[B]: Margin of error: 16.0%; 
Fee-for-service (FFS): Error rate: 6.2%; 
Fee-for-service (FFS): Margin of error: 2.1%; 
Managed care: Error rate: 0.2%; 
Managed care: Margin of error: 0.4%; 
Eligibility: Error rate: 16.0%; 
Eligibility: Margin of error: 16.8%. 

State: Florida; 
Combined[B]: Error rate: 14.6%; 
Combined[B]: Margin of error: 13.0%; 
Fee-for-service (FFS): Error rate: 7.5%; 
Fee-for-service (FFS): Margin of error: 2.1%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 9.2%; 
Eligibility: Margin of error: 13.7%. 

State: Hawaii; 
Combined[B]: Error rate: 16.8%; 
Combined[B]: Margin of error: 5.8%; 
Fee-for-service (FFS): Error rate: 5.7%; 
Fee-for-service (FFS): Margin of error: 2.2%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 13.4%; 
Eligibility: Margin of error: 5.9%. 

State: Indiana; 
Combined[B]: Error rate: 17.2%; 
Combined[B]: Margin of error: 10.5%; 
Fee-for-service (FFS): Error rate: 4.4%; 
Fee-for-service (FFS): Margin of error: 2.0%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 14.3%; 
Eligibility: Margin of error: 10.8%. 

State: Iowa; 
Combined[B]: Error rate: 4.9%; 
Combined[B]: Margin of error: 4.6%; 
Fee-for-service (FFS): Error rate: 1.7%; 
Fee-for-service (FFS): Margin of error: 1.7%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 3.3%; 
Eligibility: Margin of error: 4.4%. 

State: Louisiana; 
Combined[B]: Error rate: 4.0%; 
Combined[B]: Margin of error: 3.1%; 
Fee-for-service (FFS): Error rate: 2.5%; 
Fee-for-service (FFS): Margin of error: 2.9%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 1.5%; 
Eligibility: Margin of error: 1.1%. 

State: Maine; 
Combined[B]: Error rate: 5.7%; 
Combined[B]: Margin of error: 2.4%; 
Fee-for-service (FFS): Error rate: 3.8%; 
Fee-for-service (FFS): Margin of error: 1.7%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 2.0%; 
Eligibility: Margin of error: 1.8%. 

State: Mississippi; 
Combined[B]: Error rate: 3.5%; 
Combined[B]: Margin of error: 3.2%; 
Fee-for-service (FFS): Error rate: 3.4%; 
Fee-for-service (FFS): Margin of error: 3.1%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 0.1%; 
Eligibility: Margin of error: 0.3%. 

State: Montana; 
Combined[B]: Error rate: 4.4%; 
Combined[B]: Margin of error: 6.5%; 
Fee-for-service (FFS): Error rate: 0.9%; 
Fee-for-service (FFS): Margin of error: 0.7%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 3.6%; 
Eligibility: Margin of error: 6.5%. 

State: Nevada; 
Combined[B]: Error rate: 7.3%; 
Combined[B]: Margin of error: 2.6%; 
Fee-for-service (FFS): Error rate: 4.9%; 
Fee-for-service (FFS): Margin of error: 1.8%; 
Managed care: Error rate: 0.1%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 3.3%; 
Eligibility: Margin of error: 2.2%. 

State: New York; 
Combined[B]: Error rate: 7.8%; 
Combined[B]: Margin of error: 4.4%; 
Fee-for-service (FFS): Error rate: 1.5%; 
Fee-for-service (FFS): Margin of error: 1.0%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 6.7%; 
Eligibility: Margin of error: 4.4%. 

State: Oregon; 
Combined[B]: Error rate: 20.8%; 
Combined[B]: Margin of error: 12.0%; 
Fee-for-service (FFS): Error rate: 1.7%; 
Fee-for-service (FFS): Margin of error: 1.7%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 20.0%; 
Eligibility: Margin of error: 12.1%. 

State: South Dakota; 
Combined[B]: Error rate: 0.9%; 
Combined[B]: Margin of error: 0.7%; 
Fee-for-service (FFS): Error rate: 0.8%; 
Fee-for-service (FFS): Margin of error: 0.7%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 0.0%; 
Eligibility: Margin of error: 0.1%. 

State: Texas; 
Combined[B]: Error rate: 5.1%; 
Combined[B]: Margin of error: 3.4%; 
Fee-for-service (FFS): Error rate: 0.4%; 
Fee-for-service (FFS): Margin of error: 0.5%; 
Managed care: Error rate: 0.4%; 
Managed care: Margin of error: 0.7%; 
Eligibility: Error rate: 4.7%; 
Eligibility: Margin of error: 3.4%. 

State: Washington; 
Combined[B]: Error rate: 6.4%; 
Combined[B]: Margin of error: 4.8%; 
Fee-for-service (FFS): Error rate: 3.5%; 
Fee-for-service (FFS): Margin of error: 1.8%; 
Managed care: Error rate: 0.9%; 
Managed care: Margin of error: 1.2%; 
Eligibility: Error rate: 3.6%; 
Eligibility: Margin of error: 4.7%. 

Measurement cycle[A]: Fiscal year 2009; 

State: Arkansas; 
Combined[B]: Error rate: 4.2%; 
Combined[B]: Margin of error: 2.0%; 
Fee-for-service (FFS): Error rate: 3.0%; 
Fee-for-service (FFS): Margin of error: 1.3%; 
Managed care: Error rate: N/A; 
Managed care: Margin of error: N/A; 
Eligibility: Error rate: 1.2%; 
Eligibility: Margin of error: 1.6%. 

State: Connecticut; 
Combined[B]: Error rate: 3.3%; 
Combined[B]: Margin of error: 1.3%; 
Fee-for-service (FFS): Error rate: 1.5%; 
Fee-for-service (FFS): Margin of error: 1.0%; 
Managed care: Error rate: 0.1%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 2.0%; 
Eligibility: Margin of error: 1.0%. 

State: Delaware; 
Combined[B]: Error rate: 5.0%; 
Combined[B]: Margin of error: 2.1%; 
Fee-for-service (FFS): Error rate: 3.5%; 
Fee-for-service (FFS): Margin of error: 1.5%; 
Managed care: Error rate: 0.4%; 
Managed care: Margin of error: 0.8%; 
Eligibility: Error rate: 2.8%; 
Eligibility: Margin of error: 1.9%. 

State: Idaho; 
Combined[B]: Error rate: 1.6%; 
Combined[B]: Margin of error: 1.1%; 
Fee-for-service (FFS): Error rate: 1.2%; 
Fee-for-service (FFS): Margin of error: 1.0%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 0.4%; 
Eligibility: Margin of error: 0.5%. 

State: Illinois; 
Combined[B]: Error rate: 3.8%; 
Combined[B]: Margin of error: 2.1%; 
Fee-for-service (FFS): Error rate: 1.5%; 
Fee-for-service (FFS): Margin of error: 1.1%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 2.3%; 
Eligibility: Margin of error: 1.8%. 

State: Kansas; 
Combined[B]: Error rate: 10.4%; 
Combined[B]: Margin of error: 8.0%; 
Fee-for-service (FFS): Error rate: 1.1%; 
Fee-for-service (FFS): Margin of error: 0.9%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 9.6%; 
Eligibility: Margin of error: 8.0%. 

State: Michigan; 
Combined[B]: Error rate: 69.9%; 
Combined[B]: Margin of error: 20.1%; 
Fee-for-service (FFS): Error rate: 1.7%; 
Fee-for-service (FFS): Margin of error: 1.5%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 69.6%; 
Eligibility: Margin of error: 20.3%. 

State: Minnesota; 
Combined[B]: Error rate: 2.0%; 
Combined[B]: Margin of error: 1.5%; 
Fee-for-service (FFS): Error rate: 0.6%; 
Fee-for-service (FFS): Margin of error: 0.5%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 1.6%; 
Eligibility: Margin of error: 1.4%. 

State: Missouri; 
Combined[B]: Error rate: 2.6%; 
Combined[B]: Margin of error: 1.2%; 
Fee-for-service (FFS): Error rate: 2.0%; 
Fee-for-service (FFS): Margin of error: 1.2%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 0.9%; 
Eligibility: Margin of error: 0.8%. 

State: New Mexico; 
Combined[B]: Error rate: 1.9%; 
Combined[B]: Margin of error: 1.1%; 
Fee-for-service (FFS): Error rate: 3.8%; 
Fee-for-service (FFS): Margin of error: 2.0%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 0.4%; 
Eligibility: Margin of error: 0.7%. 

State: North Dakota; 
Combined[B]: Error rate: 3.2%; 
Combined[B]: Margin of error: 2.4%; 
Fee-for-service (FFS): Error rate: 3.1%; 
Fee-for-service (FFS): Margin of error: 2.4%; 
Managed care: Error rate: 6.7%; 
Managed care: Margin of error: 3.4%; 
Eligibility: Error rate: 0.1%; 
Eligibility: Margin of error: 0.1%. 

State: Ohio; 
Combined[B]: Error rate: 9.8%; 
Combined[B]: Margin of error: 11.5%; 
Fee-for-service (FFS): Error rate: 1.4%; 
Fee-for-service (FFS): Margin of error: 1.2%; 
Managed care: Error rate: 0.4%; 
Managed care: Margin of error: 0.7%; 
Eligibility: Error rate: 8.8%; 
Eligibility: Margin of error: 11.6%. 

State: Oklahoma; 
Combined[B]: Error rate: 1.2%; 
Combined[B]: Margin of error: 0.9%; 
Fee-for-service (FFS): Error rate: 1.2%; 
Fee-for-service (FFS): Margin of error: 0.9%; 
Managed care: Error rate: N/A; 
Managed care: Margin of error: N/A; 
Eligibility: Error rate: 0.0%; 
Eligibility: Margin of error: 0.0%. 

State: Pennsylvania; 
Combined[B]: Error rate: 4.1%; 
Combined[B]: Margin of error: 3.6%; 
Fee-for-service (FFS): Error rate: 3.8%; 
Fee-for-service (FFS): Margin of error: 4.2%; 
Managed care: Error rate: 0.2%; 
Managed care: Margin of error: 0.4%; 
Eligibility: Error rate: 2.0%; 
Eligibility: Margin of error: 2.8%. 

State: Virginia; 
Combined[B]: Error rate: 17.4%; 
Combined[B]: Margin of error: 11.8%; 
Fee-for-service (FFS): Error rate: 0.7%; 
Fee-for-service (FFS): Margin of error: 0.6%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 16.9%; 
Eligibility: Margin of error: 11.9%. 

State: Wisconsin; 
Combined[B]: Error rate: 5.7%; 
Combined[B]: Margin of error: 7.7%; 
Fee-for-service (FFS): Error rate: 2.3%; 
Fee-for-service (FFS): Margin of error: 1.6%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 4.4%; 
Eligibility: Margin of error: 7.8%. 

State: Wyoming; 
Combined[B]: Error rate: 8.3%; 
Combined[B]: Margin of error: 5.4%; 
Fee-for-service (FFS): Error rate: 2.8%; 
Fee-for-service (FFS): Margin of error: 1.5%; 
Managed care: Error rate: N/A; 
Managed care: Margin of error: N/A; 
Eligibility: Error rate: 5.7%; 
Eligibility: Margin of error: 5.3%. 

Measurement cycle[A]: Fiscal year 2010; 

State: Alabama; 
Combined[B]: Error rate: 2.4%; 
Combined[B]: Margin of error: 1.1%; 
Fee-for-service (FFS): Error rate: 1.5%; 
Fee-for-service (FFS): Margin of error: 1.0%; 
Managed care: Error rate: 2.1%; 
Managed care: Margin of error: 1.6%; 
Eligibility: Error rate: 0.8%; 
Eligibility: Margin of error: 0.7%. 

State: California; 
Combined[B]: Error rate: 1.6%; 
Combined[B]: Margin of error: 0.9%; 
Fee-for-service (FFS): Error rate: 1.7%; 
Fee-for-service (FFS): Margin of error: 1.0%; 
Managed care: Error rate: 0.5%; 
Managed care: Margin of error: 1.0%; 
Eligibility: Error rate: 0.2%; 
Eligibility: Margin of error: 0.3%. 

State: Colorado; 
Combined[B]: Error rate: 6.9%; 
Combined[B]: Margin of error: 2.5%; 
Fee-for-service (FFS): Error rate: 6.8%; 
Fee-for-service (FFS): Margin of error: 2.1%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 1.0%; 
Eligibility: Margin of error: 1.7%. 

State: Georgia; 
Combined[B]: Error rate: 4.7%; 
Combined[B]: Margin of error: 2.7%; 
Fee-for-service (FFS): Error rate: 4.1%; 
Fee-for-service (FFS): Margin of error: 1.9%; 
Managed care: Error rate: 1.3%; 
Managed care: Margin of error: 1.5%; 
Eligibility: Error rate: 1.5%; 
Eligibility: Margin of error: 2.4%. 

State: Kentucky; 
Combined[B]: Error rate: 2.0%; 
Combined[B]: Margin of error: 1.0%; 
Fee-for-service (FFS): Error rate: 2.3%; 
Fee-for-service (FFS): Margin of error: 1.2%; 
Managed care: Error rate: 0.4%; 
Managed care: Margin of error: 0.7%; 
Eligibility: Error rate: 0.0%; 
Eligibility: Margin of error: 0.0%. 

State: Maryland; 
Combined[B]: Error rate: 3.2%; 
Combined[B]: Margin of error: 1.9%; 
Fee-for-service (FFS): Error rate: 1.8%; 
Fee-for-service (FFS): Margin of error: 1.1%; 
Managed care: Error rate: 0.2%; 
Managed care: Margin of error: 0.3%; 
Eligibility: Error rate: 2.0%; 
Eligibility: Margin of error: 1.8%. 

State: Massachusetts; 
Combined[B]: Error rate: 13.4%; 
Combined[B]: Margin of error: 2.2%; 
Fee-for-service (FFS): Error rate: 17.7%; 
Fee-for-service (FFS): Margin of error: 3.0%; 
Managed care: Error rate: 1.0%; 
Managed care: Margin of error: 1.0%; 
Eligibility: Error rate: 0.0%; 
Eligibility: Margin of error: 0.1%. 

State: Nebraska; 
Combined[B]: Error rate: 2.1%; 
Combined[B]: Margin of error: 1.1%; 
Fee-for-service (FFS): Error rate: 2.2%; 
Fee-for-service (FFS): Margin of error: 1.1%; 
Managed care: Error rate: 0.4%; 
Managed care: Margin of error: 0.7%; 
Eligibility: Error rate: 0.0%; 
Eligibility: Margin of error: 0.0%. 

State: New Hampshire; 
Combined[B]: Error rate: 1.5%; 
Combined[B]: Margin of error: 1.1%; 
Fee-for-service (FFS): Error rate: 1.5%; 
Fee-for-service (FFS): Margin of error: 1.1%; 
Managed care: Error rate: N/A; 
Managed care: Margin of error: N/A; 
Eligibility: Error rate: 0.0%; 
Eligibility: Margin of error: 0.0%. 

State: New Jersey; 
Combined[B]: Error rate: 2.6%; 
Combined[B]: Margin of error: 1.6%; 
Fee-for-service (FFS): Error rate: 2.0%; 
Fee-for-service (FFS): Margin of error: 1.6%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 1.0%; 
Eligibility: Margin of error: 1.1%. 

State: North Carolina; 
Combined[B]: Error rate: 11.9%; 
Combined[B]: Margin of error: 15.3%; 
Fee-for-service (FFS): Error rate: 3.4%; 
Fee-for-service (FFS): Margin of error: 2.5%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 8.9%; 
Eligibility: Margin of error: 15.7%. 

State: Rhode Island; 
Combined[B]: Error rate: 15.6%; 
Combined[B]: Margin of error: 5.8%; 
Fee-for-service (FFS): Error rate: 6.1%; 
Fee-for-service (FFS): Margin of error: 2.1%; 
Managed care: Error rate: 0.4%; 
Managed care: Margin of error: 0.8%; 
Eligibility: Error rate: 11.8%; 
Eligibility: Margin of error: 5.9%. 

State: South Carolina; 
Combined[B]: Error rate: 18.8%; 
Combined[B]: Margin of error: 15.8%; 
Fee-for-service (FFS): Error rate: 2.6%; 
Fee-for-service (FFS): Margin of error: 1.3%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 17.2%; 
Eligibility: Margin of error: 16.1%. 

State: Tennessee; 
Combined[B]: Error rate: 3.6%; 
Combined[B]: Margin of error: 4.6%; 
Fee-for-service (FFS): Error rate: 1.7%; 
Fee-for-service (FFS): Margin of error: 2.9%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 2.8%; 
Eligibility: Margin of error: 4.5%. 

State: Utah; 
Combined[B]: Error rate: 8.2%; 
Combined[B]: Margin of error: 4.9%; 
Fee-for-service (FFS): Error rate: 4.5%; 
Fee-for-service (FFS): Margin of error: 1.9%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 4.5%; 
Eligibility: Margin of error: 4.8%. 

State: Vermont; 
Combined[B]: Error rate: 8.0%; 
Combined[B]: Margin of error: 2.7%; 
Fee-for-service (FFS): Error rate: 6.8%; 
Fee-for-service (FFS): Margin of error: 2.4%; 
Managed care: Error rate: 0.9%; 
Managed care: Margin of error: 1.2%; 
Eligibility: Error rate: 1.4%; 
Eligibility: Margin of error: 1.4%. 

State: West Virginia; 
Combined[B]: Error rate: 32.7%; 
Combined[B]: Margin of error: 32.2%; 
Fee-for-service (FFS): Error rate: 4.2%; 
Fee-for-service (FFS): Margin of error: 2.2%; 
Managed care: Error rate: 0.0%; 
Managed care: Margin of error: 0.0%; 
Eligibility: Error rate: 30.2%; 
Eligibility: Margin of error: 33.4%. 

Source: CMS data on state improper payment error rates for the 
Medicaid program (unaudited). 

Legend: N/A = not applicable (the state did not have a managed care 
component to measure during the related cycle). 

Note: These rates reflect the states' official error rates used to 
calculate the national error rates and do not reflect any state error 
rates that were recalculated, upon a state's request, for 
informational purposes and to determine sample sizes for the next 
measurement cycle. 

[A] HHS reported the results of the fiscal years 2008 through 2010 
measurement cycles in its fiscal year 2011 AFR. 

[B] The combined rate is a weighted average of FFS and managed care, 
with the addition of eligibility. A small correction factor ensures 
that Medicaid eligibility errors do not get "double counted" if the 
sampled item was also tested in either the FFS or managed care 
components. 

[End of table] 

[End of section] 

Appendix IV: Medicaid Outlays and Estimated Improper Payment Error 
Rates Reported by the Department of Health and Human Services: 

Table 3 provides a list of Medicaid outlays and estimated improper 
payment error rates reported in the Department of Health and Human 
Services' (HHS) agency financial reports (AFR). 

Table 3: Medicaid Outlays and Estimated Improper Payment Error Rates 
Reported in HHS's AFRs: 

Dollars in millions: 

Fiscal year: 2007[C]; 
Outlays: $139,896; 
Fee-for-service (FFS) error rate: 4.7%; 
Managed care error rate: N/A; 
Eligibility error rate: N/A; 
Combined error rate[A]: 4.7% ; 
Total estimated improper payments: $6,575; 
Estimated overpayments[B]: N/A; 
Estimated underpayments[B]: N/A. 

Fiscal year: 2008[D]; 
Outlays: $177,547; 
Fee-for-service (FFS) error rate: 8.9%; 
Managed care error rate: 3.1%; 
Eligibility error rate: 2.9%; 
Combined error rate[A]: 10.5% ; 
Total estimated improper payments: $18,642; 
Estimated overpayments[B]: N/A; 
Estimated underpayments[B]: N/A. 

Fiscal year: 2009; 
Outlays: $188,286; 
Fee-for-service (FFS) error rate: 5.7%; 
Managed care error rate: 1.5%; 
Eligibility error rate: 4.9%; 
Combined error rate[A]: 9.6%; 
Total estimated improper payments: $18,075; 
Estimated overpayments[B]: N/A; 
Estimated underpayments[B]: N/A. 

Fiscal year: 2010; 
Outlays: $239,012; 
Fee-for-service (FFS) error rate: 4.4%; 
Managed care error rate: 1.0%; 
Eligibility error rate: 5.9%; 
Combined error rate[A]: 9.4%[E]; 
Total estimated improper payments: $22,500; 
Estimated overpayments[B]: N/A; 
Estimated underpayments[B]: N/A. 

Fiscal year: 2011; 
Outlays: $269,241; 
Fee-for-service (FFS) error rate: 2.7%; 
Managed care error rate: 0.3%; 
Eligibility error rate: 6.1%; 
Combined error rate[A]: 8.1%[E]; 
Total estimated improper payments: $21,900; 
Estimated overpayments[B]: $21,448; 
Estimated underpayments[B]: $453. 

Source: HHS's AFR reporting, fiscal years 2007 to 2011, and Centers 
for Medicare & Medicaid Services' data on Payment Error Rate 
Measurement error rates (unaudited). 

Legend: N/A = not applicable. 

[A] The combined rate is a weighted average of FFS and managed care, 
with the addition of eligibility. A small correction factor ensures 
that Medicaid eligibility errors do not get "double counted" if the 
sampled item was also tested in either the FFS or managed care 
components. 

[B] The breakdown of estimated overpayments and underpayments was not 
reported for fiscal years 2007 through 2010. 

[C] This information for fiscal year 2007 reflects revised information 
reported in HHS's fiscal year 2008 AFR. Also, the combined error rate 
only represents the Medicaid FFS component as it was the only 
component measured during this cycle, and the review was limited to 17 
states for the cycle. 

[D] Fiscal year 2008 reporting was limited to 34 states. 

[E] HHS reported that as required under Section 601 of the Children's 
Health Insurance Program Reauthorization Act of 2009, it published a 
final rule on August 11, 2010, which required the eligibility reviews 
to be consistent with the state's eligibility verification policy 
rather than reviewing eligibility against a single, federal 
methodology, which was done in the past. HHS also reported that based 
on current regulations, certain cases from the measurement cycles for 
fiscal years 2008 and 2009 would no longer be considered as errors. 

[End of table] 

[End of section] 

Appendix V: Additional Details for Medicaid Error Rates: 

Table 4 provides the margins of error at the 90 percent confidence 
level for error rate data presented in figure 2. 

Table 4: Margins of Error for Error Rate Data Presented in Figure 2: 

Measurement cycle or reporting year: Fiscal year 2008 cycle; 

Component: Combined[B]; 
Error rate: 8.7%; 
Margin of error[A]: 2.5%. 

Component: Fee-for-service (FFS); 
Error rate: 2.6%; 
Margin of error[A]: 1.0%. 

Component: Managed care; 
Error rate: 0.1%; 
Margin of error[A]: 0.1%. 

Component: Eligibility; 
Error rate: 6.7%; 
Margin of error[A]: 2.4%. 

Measurement cycle or reporting year: Fiscal year 2009 cycle; 

Component: Combined[B]; 
Error rate: 9.0%; 
Margin of error[A]: 5.3%. 

Component: FFS; 
Error rate: 1.9%; 
Margin of error[A]: 0.6%. 

Component: Managed care; 
Error rate: 0.1%; 
Margin of error[A]: 0.1%. 

Component: Eligibility; 
Error rate: 7.6%; 
Margin of error[A]: 5.3%. 

Measurement cycle or reporting year: Fiscal year 2010 cycle; 

Component: Combined[B]; 
Error rate: 6.7%; 
Margin of error[A]: 3.0%. 

Component: FFS; 
Error rate: 3.6%; 
Margin of error[A]: 1.4%. 

Component: Managed care; 
Error rate: 0.5%; 
Margin of error[A]: 0.3%. 

Component: Eligibility; 
Error rate: 4.0%; 
Margin of error[A]: 2.9%. 

Measurement cycle or reporting year: Fiscal year 2011 reporting (3-
year national error rate); 

Component: Combined[B]; 
Error rate: 8.1%; 
Margin of error[A]: 2.2%. 

Component: FFS; 
Error rate: 2.7%; 
Margin of error[A]: 0.6%. 

Component: Managed care; 
Error rate: 0.3%; 
Margin of error[A]: 0.1%. 

Component: Eligibility; 
Error rate: 6.1%; 
Margin of error[A]: 2.2%. 

Source: GAO analysis of support for fiscal year 2011 error rate 
reporting. 

[A] According to the Centers for Medicare & Medicaid Services, the 
estimated error rates presented in this table have margins of error at 
the 90 percent confidence level of no more than plus or minus these 
percentage points. 

[B] The combined rate is a weighted average of FFS and managed care, 
with the addition of eligibility. A small correction factor ensures 
that Medicaid eligibility errors do not get "double counted" if the 
sampled item was also tested in either the FFS or managed care 
components. 

[End of table] 

[End of section] 

Appendix VI: Types of Errors Identified in Payment Error Rate 
Measurement Reviews: 

Table 5 provides a list of error types identified during the fiscal 
years 2008 to 2010 Payment Error Rate Measurement (PERM) measurement 
cycles. 

Table 5: Summary of Errors, Fiscal Years 2008 to 2010 PERM Measurement 
Cycles: 

Eligibility review: 

Error type description: 1. Undetermined; 
Number of errors: 910; 
Percentage of total number of errors: 28.2%; 
Dollars in error: $349,677; 
Percentage of total dollars in error: 9.3%. 

Error type description: 2. Not eligible; 
Number of errors: 814; 
Percentage of total number of errors: 25.3%; 
Dollars in error: $299,920; 
Percentage of total dollars in error: 8.0%. 

Error type description: 3. Eligible with ineligible services; 
Number of errors: 122; 
Percentage of total number of errors: 3.8%; 
Dollars in error: $75,979; 
Percentage of total dollars in error: 2.0%. 

Error type description: 4. Liability understated; 
Number of errors: 145; 
Percentage of total number of errors: 4.5%; 
Dollars in error: $30,459; 
Percentage of total dollars in error: 0.8%. 

Error type description: 5. Liability overstated; 
Number of errors: 37; 
Percentage of total number of errors: 1.1%; 
Dollars in error: $8,034; 
Percentage of total dollars in error: 0.2%. 

Error type description: 6. Managed care error - ineligible for managed 
care; 
Number of errors: 8; 
Percentage of total number of errors: 0.2%; 
Dollars in error: $1,270; 
Percentage of total dollars in error: 0.0%. 

Error type description: 7. Managed care error - eligible for managed 
care but improperly enrolled; 
Number of errors: 4; 
Percentage of total number of errors: 0.1%; 
Dollars in error: $181; 
Percentage of total dollars in error: 0.0%. 

Fee-for-service (FFS) review: Medical review: 

Error type description: 8. Insufficient documentation; 
Number of errors: 298; 
Percentage of total number of errors: 9.2%; 
Dollars in error: $940,585; 
Percentage of total dollars in error: 25.0%. 

Error type description: 9. Diagnosis coding error; 
Number of errors: 59; 
Percentage of total number of errors: 1.8%; 
Dollars in error: $454,793; 
Percentage of total dollars in error: 12.1%. 

Error type description: 10. No documentation; 
Number of errors: 140; 
Percentage of total number of errors: 4.3%; 
Dollars in error: $450,961; 
Percentage of total dollars in error: 12.0%. 

Error type description: 11. Number of units error; 
Number of errors: 161; 
Percentage of total number of errors: 5.0%; 
Dollars in error: $158,263; 
Percentage of total dollars in error: 4.2%. 

Error type description: 12. Policy violation; 
Number of errors: 55; 
Percentage of total number of errors: 1.7%; 
Dollars in error: $112,657; 
Percentage of total dollars in error: 3.0%. 

Error type description: 13. Procedure coding error; 
Number of errors: 55; 
Percentage of total number of errors: 1.7%; 
Dollars in error: $94,708; 
Percentage of total dollars in error: 2.5%. 

Error type description: 14. Medically unnecessary service; 
Number of errors: 13; 
Percentage of total number of errors: 0.4%; 
Dollars in error: $68,758; 
Percentage of total dollars in error: 1.8%. 

Error type description: 15. Administrative/other; 
Number of errors: 26; 
Percentage of total number of errors: 0.8%; 
Dollars in error: $14,635; 
Percentage of total dollars in error: 0.4%. 

Error type description: 16. Unbundling; 
Number of errors: 1; 
Percentage of total number of errors: 0.0%; 
Dollars in error: $92; 
Percentage of total dollars in error: 0.0%. 

Fee-for-service (FFS) review: Data processing review: 

Error type description: 17. Administrative/other; 
Number of errors: 43; 
Percentage of total number of errors: 1.3%; 
Dollars in error: $195,937; 
Percentage of total dollars in error: 5.2%. 

Error type description: 18. Noncovered service; 
Number of errors: 78; 
Percentage of total number of errors: 2.4%; 
Dollars in error: $188,242; 
Percentage of total dollars in error: 5.0%. 

Error type description: 19. Third-party liability; 
Number of errors: 12; 
Percentage of total number of errors: 0.4%; 
Dollars in error: $62,634; 
Percentage of total dollars in error: 1.7%. 

Error type description: 20. Logic edit; 
Number of errors: 9; 
Percentage of total number of errors: 0.3%; 
Dollars in error: $59,281; 
Percentage of total dollars in error: 1.6%. 

Error type description: 21. FFS claim for a managed care service; 
Number of errors: 6; 
Percentage of total number of errors: 0.2%; 
Dollars in error: $50,772; 
Percentage of total dollars in error: 1.4%. 

Error type description: 22. Pricing error; 
Number of errors: 128; 
Percentage of total number of errors: 4.0%; 
Dollars in error: $45,233; 
Percentage of total dollars in error: 1.2%. 

Error type description: 23. Duplicate item; 
Number of errors: 11; 
Percentage of total number of errors: 0.3%; 
Dollars in error: $45,186; 
Percentage of total dollars in error: 1.2%. 

Error type description: 24. Data entry error; 
Number of errors: 7; 
Percentage of total number of errors: 0.2%; 
Dollars in error: $14,778; 
Percentage of total dollars in error: 0.4%. 

Error type description: 25. Managed care payment error; 
Number of errors: 1; 
Percentage of total number of errors: 0.0%; 
Dollars in error: $0; 
Percentage of total dollars in error: 0.0%. 

Managed care review: Data processing review: 

Error type description: 26. Noncovered service; 
Number of errors: 40; 
Percentage of total number of errors: 1.2%; 
Dollars in error: $22,146; 
Percentage of total dollars in error: 0.6%. 

Error type description: 27. Logic edit; 
Number of errors: 1; 
Percentage of total number of errors: 0.0%; 
Dollars in error: $9,053; 
Percentage of total dollars in error: 0.2%. 

Error type description: 28. Duplicate item; 
Number of errors: 5; 
Percentage of total number of errors: 0.2%; 
Dollars in error: $3,493; 
Percentage of total dollars in error: 0.1%. 

Error type description: 29. Managed care payment error; 
Number of errors: 31; 
Percentage of total number of errors: 1.0%; 
Dollars in error: $533; 
Percentage of total dollars in error: 0.0%. 

Error type description: 30. Pricing error; 
Number of errors: 1; 
Percentage of total number of errors: 0.0%; 
Dollars in error: $166; 
Percentage of total dollars in error: 0.0%. 

Error type description: 31. Administrative/other; 
Number of errors: 1; 
Percentage of total number of errors: 0.0%; 
Dollars in error: $31; 
Percentage of total dollars in error: 0.0%. 

Error type description: Total; 
Number of errors: 3,222; 
Percentage of total number of errors: 100.0%; 
Dollars in error: $3,758,457; 
Percentage of total dollars in error: 100.0%. 

Source: GAO analysis of Centers for Medicare & Medicaid Services' 
error data. 

[End of table] 

[End of section] 

Appendix VII: Comments from the Department of Health and Human 
Services: 

Note: GAO comments supplementing those in the report text appear at 
the end of this appendix. 

Department of Health & Human Services: 
Office of The Secretary: 
Assistant Secretary for Legislation: 
Washington, DC 20201: 

March 12, 2013: 

Beryl H. Davis: 
Director, Financial Management and Assurance: 
U.S. Government Accountability Office:
441 G Street NW:
Washington, DC 20548: 

Dear Ms. Davis: 

Attached are comments on the U.S. Government Accountability Office's 
(GAO) report entitled, "Medicaid: Enhancements Needed for Improper 
Payments Reporting and Related Corrective Action Monitoring" (GAO-13-
229). 

The Department appreciates the opportunity to review this report prior 
to publication. 

Sincerely, 

Signed by: 

Jim R. Esquea: 
Assistant Secretary for Legislation: 

Attachment: 

General Comments Of The Department Of Health And Human Services (HHS) 
On The Government Accountability Office's (GAO) Draft Report Entitled, 
"Medicaid: Enhancements Needed For Improper Payments Reporting And 
Related Corrective Action Monitoring" (GAO-13-229): 

The Department appreciates the opportunity to review and comment on 
this draft report. 

Although HHS agrees with many of GAO's recommendations, we have 
provided extensive technical comments to GAO's draft report to clarify 
several important issues regarding CMS's management and implementation 
of the Payment Error Rate Reduction (PERM). Throughout the report, GAO 
makes statements implying that until CMS establishes procedures for 
considering changes to initially reported state-level error rates, CMS 
is impaired in its ability to ensure its reported Medicaid improper 
payment estimates are reliable. HHS believes such statements are 
misleading and suggest that past Medicaid error rates reported by CMS 
have been unreliable. The reported Medicaid improper payment estimate 
for fiscal year 2012 was 7.1 percent. This estimate included reviews 
of fiscal years 2009, 2010, and 2011 Medicaid payments. CMS 
recalculated the 2012 national Medicaid improper payment rate to 
incorporate any state-level error rate corrections from fiscal years 
2009 and 2010 and the rate remained 7.1 percent. These state-level 
recalculations did not impact the national Medicaid improper payment 
rate. [See comment 1] 

The draft report also references OMB's Standards and Guidelines for 
Statistical Surveys and suggests this guidance be the basis for 
determining how to handle PERM related data corrections. CMS points 
out that OMB guidance is not applicable to the PERM program; the PERM 
program is subject to OMB implementing guidance for Improper Payments 
Elimination and Recovery Act of 2010, Memorandum M-1 1-16. OMB did not 
include guidance for handling unscheduled corrections to data in OMB 
Memorandum M-11-16. [See comment 1] 

The draft report also references the inability of CMS's review 
contractor to review medical claims not submitted during the audit 
cycle due to time constraints. It should be noted, the review 
contractor will complete the review of all medical records received by 
the cycle cut-off prior to error rate calculation. According to the 
PERM manual, if a Review Contractor receives late documentation prior 
to the cycle cut-off date for error rate calculation and reporting 
purposes, it will review the records and, if justified, revise the 
error finding. This characterization is inaccurate and we have asked 
GAO to modify this statement in our technical comments. [See comment 1] 

HHS continues to caution GAO about including the state error rates 
identified in Appendix III, as we are concerned that readers will use 
the rates for comparison purposes when they should not. Due to the 
variation of states' sizes, overall program variations, and how states 
implement and administer their program, we do not encourage 
comparisons based solely on state rates. [See comment 1] 

CMS's implementation of PERM program complies with OMB's implementing 
guidance for the Improper Payments Elimination and Recovery Act of 
2010. CMS coordinates Medicaid improper payment measurement closely 
with the states. CMS frequently communicates with states through 
regularly scheduled cycle calls, webinars, monthly calls, emails, the 
PERM website, and ad-hoc calls as necessary. We continuously improve 
and refine the PERM process with the states as we gain experience and 
feedback from entities such as GAO. 

The GAO recommendations and HHS response to those recommendations are 
discussed below. 

Recommendation 1: 

Update PERM Medicaid improper payment reporting procedures to provide 
for considering any corrections to state-level improper payment error 
data subsequent to the cut-off date that would have significant impact 
on any of the 3 years used to develop the rolling average for the 
reported national Medicaid improper payment estimate. 

HHS Response: 

HHS concurs and will consider revising the improper payment reporting 
procedures to take into account corrections to state-level improper 
payment error data. In order to calculate and report improper payment 
rates in HHS's Agency Financial Report (AFR) each November, it is 
essential that CMS set a cut-off date so that the cycle can end prior 
to the improper payment rate reporting. 

HHS believes it is important to note that state-level error rate 
corrections that occurred after the improper payment reporting in the 
AFR did not affect the accuracy of the reported national error rate. 
The reported Medicaid improper payment estimate for fiscal year 2012 
was 7.1 percent. This estimate included reviews of fiscal years 2009, 
2010, and 2011 Medicaid payments. Based on this GAO review, we 
recalculated the 2012 national Medicaid improper payment rate to 
incorporate any state-level error rate corrections for fiscal years 
2009 and 2010 and the rate remained 7.1 percent. These state-level 
recalculations did not impact the national Medicaid improper payment 
rate. [See comment 1] 

Recommendation 2: 

Revise the PERM manual to provide that states (1) analyze all 
deficiencies, negative case errors, technical errors, and minimal 
dollar errors identified in PERM reviews to determine if any 
corrective actions, if cost-effective, are needed to prevent such 
errors in the future, and (2) document the results of their analysis. 

HHS Response: 

HHS concurs and CMS will modify the PERM manual accordingly. CMS 
relies on frequent communication with states through regularly 
scheduled cycle calls, webinars, emails, the PERM website, and ad-hoc 
calls as necessary to provide guidance relating to Corrective Action 
Plans (CAP). The PERM manual specifies that states should use the most 
cost effective corrective actions when situations arise where the 
costs of implementing a corrective action exceed the benefits. HHS 
agrees that formally documenting this guidance in writing may be 
beneficial to states and CMS will update the PERM manual accordingly. 

Recommendation 3: 

Clarify guidance in the PERM manual, and on the PERM website, on the 
required elements to be included in a CAP and the specific actions 
states are to take each measurement cycle to (1) effectively prepare 
and evaluate their current cycle's CAP, and (2) provide updates to 
their previous cycle's CAP. 

HHS Response: 

HHS concurs and will update the PERM manual and relevant documents on 
the PERM website accordingly. Supplying consistent and clear written 
guidance to states through our PERM manual and website will better 
assist states in complying with the PERM CAP requirements. 

Recommendation 4: 

Finalize draft policies and procedures to clarify specific CMS 
officials' roles and responsibilities for monitoring states' 
corrective actions to ensure, at a minimum, that (1) the CAPs contain 
all of the required elements and completely address errors identified 
in the PERM reviews, and (2) states are making progress on 
implementing corrective actions. 

HHS Response: 

HHS concurs and, as mentioned in the report, these documents are 
currently under development. The Division of Error Rate Measurement 
(DERM) within CMS is working collaboratively with CMS's Medicaid 
Integrity Group (MIG) and CMS Regional Offices to develop standard 
operating procedures outlining the roles of DERM, MIG, and the 
Regional Offices during state CAP review. CMS is also developing check 
lists and review guides that will be utilized when reviewing state 
CAPs to ensure the CAPs contain all the required elements, adequately 
address errors identified through PERM, and specify implementation 
progress. These documents are currently in draft stage and CMS is 
testing these procedures while reviewing fiscal year 2011 state CAPs. 
Based on this experience, CMS will refine the protocols as necessary 
and finalize the policies and procedures. 

1. On the highlights page and throughout the report, GAO makes 
statements implying that until CMS establishes procedures for 
considering changes to initially reported state-level error rates, CMS 
is impaired in its ability to ensure its reported Medicaid improper 
payment estimates are reliable. GAO specifically states that "Until 
CMS establishes procedures for considering changes to initially 
reported state-level error rates that would be significant to the 
national error rate, CMS is impaired in its ability to ensure its 
reported estimate on the extent of national Medicaid improper.payments 
is reliable." CMS believes these statements may mislead the reader to 
believe that past Medicaid error rates reported by CMS have been 
unreliable. The reported Medicaid improper payment estimate for fiscal 
year 2012 was 7.1 percent. This estimate included reviews of fiscal 
years 2009, 2010, and 2011 Medicaid payments. CMS recalculated the 
2012 national Medicaid improper payment rate to incorporate any state-
level error rate corrections from fiscal years 2009 and 2010 and the 
rate remained 7.1 percent. These state-level recalculations did not 
impact the national Medicaid improper payment rate. HHS suggests that 
GAO revise these statements to specify that the majority of past state-
level error rate corrections have not been significant and that 
without a process to consider this more current data on states' 
reported improper payment error rates, the reliability of CMS's future 
reported national estimates may be adversely affected. [See comment 1] 

2. While the PERM manual is not the only avenue by which CMS provides 
guidance on PERM, the PERM manual and website are the primary forms of 
guidance recognized in the report. CMS also uses frequent 
communication with the states through regularly scheduled cycle calls, 
webinars, monthly calls, emails, the PERM website, and ad-hoc calls as 
necessary. HHS suggests that GAO specify throughout the report that 
improvements are needed to CMS's written guidance to states on CAPs 
since some PERM guidance is currently only given to states through 
verbal and email communication. [See comment 2] 

3. GAO states in this report that "OMB has identified as a best 
practice that agencies should establish a policy for handling 
unscheduled corrections to data such as including threshold criteria 
identifying conditions under which data will be corrected and 
redisseminated." GAO's references to OMB's Standards and Guidelines 
for Statistical Surveys are inapplicable to improper payment rate 
measurements. Improper payment rate measurements are required to 
follow OMB implementing guidance for the Improper Payments Elimination 
and Recovery Act of 2010, OMB Memorandum M-11-16. OMB did not include 
guidance for handling unscheduled corrections to data in OMB 
Memorandum M-11-16. HHS recommends deleting these references from the 
report. [See comment 1] 

4. Figure 1 on page 6 of the report inaccurately categorizes PERM 
process details by year. The correct activities by year are as follows: 

* Year 1: Develop universe; Select samples; Begin eligibility reviews; 
* Year 2: Continue drawing samples; Begin FFS and managed care reviews; 
* Year 3: Continue and finalize reviews; State error rates determined; 
National error rate calculated; Reporting; Corrective action plans. 
[See comment 1. Page numbers in the draft report may differ from those 
in this report] 

5. Footnote 15 on page 7 states, "The weighted national component 
error rates do not total the national Medicaid improper payment error 
rate due to the correction factor." This statement is incorrect. The 
component error rates do not sum to the national Medicaid rate because 
the national Medicaid rate is a weighted average of fee-for-service 
(FFS) and managed care plus eligibility. A small correction factor is 
then subtracted and is an additional reason the three component rates 
do not add up. [See comment 2] 

6. Footnote on page 8 states, "The weighted national average for 
managed care does not average across for 2008, 2009, and 2010 due to 
rounding." However, the weighted national average for managed care 
does not average across because the rolling rates are not arithmetic 
averages. Rather, they are weighted by expenditures. By extension, the 
component rates cannot be simply averaged. For example, the FFS 
rolling rates are three years of rolling rates weighted by FFS 
expenditures. In this instance, it is simple luck that the FFS, 
eligibility, and overall rates happened to average. [See comment 2] 

7. The statement "For example, states may request a recalculation when 
medical records for a claim were received prior to the cycle cut-off 
date but CMS' review contractor did not have time to complete the 
review..." on page 11 is inaccurate. The review contractor will 
complete the review of all medical records received by the cycle cut-
off prior to error rate calculation. Per the PERM manual, "If the 
Review Contractor receives late documentation prior to the cycle cut-
off date for error rate calculation and reporting purposes (generally 
the second July 15 of a measurement cycle), it will review the records 
and, if justified, revise the error finding." HHS suggests rewording 
this sentence to say, "For example, states may request a recalculation 
when information supporting a claim as correctly paid was submitted to 
CMS after the cycle cut-off date..." [See comment 1] 

8. Page 12 of the report states, "While CMS aims for a cycle cut-off 
date of July 15 - 4 months prior to the reporting date, the CMS cycle 
manager may extend the cycle cut-off date depending on the progress of 
the PERM reviews." HHS suggests specifying that the PERM cycle has 
never been cut-off on July 15. Delays are inevitable and PERM due 
dates are not always met for various reasons (i.e. if a state is 
implementing a new Medicaid Management Information System during the 
PERM cycle they may encounter issues that lead to late submission of 
data to the PERM federal contractor). Therefore, for every cycle CMS 
has postponed the cycle cut-off to allow the process to complete as 
much as possible while still permitting CMS to report an improper 
payment rate timely in the agency final report. [See comment 2] 

9. Page 12 of the report gives two examples of CMS issuing revised 
error rates to states. HHS believes it is important for the reader to 
understand the timing of these discoveries compared to error rate 
reporting. 

* HHS suggests revising the sentence, "...CMS issued a revised rate to 
one state for its eligibility reviews for the fiscal year 2008 
measurement cycle because CMS and the state discovered that the amount 
of dollars in error was reported incorrectly by the state." to the 
following, "CMS issued a revised rate to one state for its eligibility
reviews for the fiscal year 2008 measurement because in January 2010, 
two months after error rate reporting, CMS and the state discovered 
the amount of dollars in error was reported incorrectly by the state." 
[See comment 2] 

* HHS recommends changing the sentence, "CMS issued a post cut-off 
date revised rate to a state for its FFS reviews because CMS received 
additional documentation from providers after the cycle cut-off date" 
to "In December 2009, one month after error rate reporting, CMS issued 
a post cut-off date revised rate to a state for its FFS reviews 
because CMS received additional documentation from providers after the 
cut-off date for official error rate calculations." [See comment 2] 

10. HHS recommends deleting the following two statements on page 12: 
"This decreased error rate would have resulted in an approximately 
$400 million decrease to the estimated improper payments for this 
state" and "This decreased error rate would have resulted in an 
approximately $31 million decrease to the estimated improper payments 
for this state," The improper payment rate percentage point change 
shows the real impact of error rate revisions and is already stated in 
the report. CMS does not report state-level estimated improper 
payments and this statistic is solely used to calculate the improper 
payment rate itself. Including these statements in the report may 
cause the reader to misinterpret the true impact of the revisions 
which GAO has already stated in the report. [See comment 2] 

11. The definition of technical error on page 17 should be revised to 
read, "A technical error is an error where the eligibility caseworker 
did not act in accordance with state or federal policy, however, it 
did not result in an erroneous eligibility determination or result in 
a difference between the amount that was paid and the amount that 
should have been paid." [See comment 2] 

12. Page 19 of the report states, "Although CMS's PERM manual requires 
states to complete and submit a CAP based on the errors found during 
the PERM process, this guidance makes no exception for small errors—
sometimes due to rounding—which may result in states incurring costs 
to implement corrective actions that exceed the dollar amount of the 
payment error." HHS believes this statement may mislead the reader for 
two reasons: 

* There are many instances when a corrective action may exceed the 
dollar amount of the payment error identified in the PERM sample. PERM 
may sample just one claim that was affected by, say, a system issue. 
However, the issue identified could affect thousands of other claims 
that are being processed by the state. Therefore, the corrective 
action may exceed the dollar amount of the PERM error, but prevent a 
large sum of improper payments. [See comment 2] 

* While the PERM guidance requires that states address all errors in 
their CAPs, it does not require states to implement particular 
corrective actions for every error. CMS does not believe requiring a 
state to address every error in their CAP would result in states 
incurring costs to implement non-effective corrective actions. [See 
comment 4] 

13. Page 19 of the report states, "While this is a sizable investment 
to address these minimal dollar errors, the state now plans to 
redesign its system in order to avoid these types of PERM errors going 
forward." HHS believes this may mislead the reader to think that the 
state continues to pursue a non-cost effective corrective action under 
the advisement of CMS. Before the fiscal year 2011 PERM cycle was 
complete, CMS stated the following through email communication with 
this state, "State's Corrective Action Plans should take into 
consideration the cost benefit analysis for any corrective action when 
developing their Corrective Action Plan... If the State determines 
that the cost of implementing a corrective action outweighs the 
benefits then the final decision of implementing the corrective action 
is the State's decision. This cost benefit analysis and final decision 
can be documented in the State's Corrective Action Plans submitted to 
CMS." [See comment 2] 

14. Page 24 of the report states, "Also, while states are currently 
required to address all errors, doing so may not be the most efficient 
and effective use of state resources for errors that do not pose a 
risk of significantly impacting future improper payments." HHS would 
like to point out that there is a difference between addressing all 
errors and implementing specific actions for each error. CMS requires 
states to address each error in their written CAP which includes 
stating what actions are being implemented or why an action is not 
being implemented for each error. CMS does not, however, require 
states to implement particular actions for each error. [See comment 4] 

GAO Comments: 

The following are GAO's comments on the Department of Health and Human 
Service's (HHS) letter dated March 13, 2013. 

1. See the "Agency Comments and Our Evaluation" section of this report. 

2. We agree with HHS's comment and modified the report as appropriate. 

3. We agree in part with HHS's comment and incorporated clarifying 
language to the figure source and Payment Error Rate Measurement 
(PERM) process details. Also, we added a figure note to acknowledge 
that certain year 1 and year 2 activities may be delayed until years 2 
and 3, respectively. 

4. We clarified the report to acknowledge that the Centers for 
Medicare & Medicaid Services' (CMS) written guidance does not indicate 
that states could address an error by stating why an action is not 
being implemented. This relates to our second recommendation, with 
which HHS concurred, that such guidance should be formally documented 
in CMS's PERM Manual. 

[End of section] 

Appendix VIII: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Beryl H. Davis, (202) 512-2623 or davisbh@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Phillip McIntyre (Assistant 
Director), Gabrielle Fagan, Kerry Porter, and Carrie Wehrly made key 
contributions to this report. Also contributing to this report were 
Carl Barden, Sharon Byrd, Francine DelVecchio, Patrick Frey, Wilfred 
Holloway, Jason Kelly, and Jason Kirwan. 

[End of section] 

Footnotes: 

[1] An improper payment is any payment that should not have been made 
or that was made in an incorrect amount (including overpayments and 
underpayments) under statutory, contractual, administrative, or other 
legally applicable requirements. This definition includes any payment 
to an ineligible recipient, any payment for an ineligible good or 
service, any duplicate payment, any payment for a good or service not 
received (except where authorized by law), and any payment that does 
not account for credit for applicable discounts. Office of Management 
and Budget guidance also instructs agencies to report as improper 
payments any payments for which insufficient or no documentation was 
found. It is important to recognize that improper payment estimates 
reported by federal agencies are not intended to be an estimate of 
fraud in federal agencies' programs and activities. 

[2] According to the Centers for Medicare & Medicaid Services (CMS), 
the estimated error rate of 8.1 percent for fiscal year 2011 has a 
margin of error at the 90 percent confidence level of no more than 
plus or minus 2.2 percentage points. The reported estimate of Medicaid 
improper payments for fiscal year 2012 is $19.2 billion based on an 
estimated error rate of 7.1 percent. According to CMS, the estimated 
error rate of 7.1 percent for fiscal year 2012 has a margin of error 
at the 90 percent confidence level of no more than plus or minus 2.0 
percentage points. HHS's fiscal year 2012 AFR was issued in November 
2012. Because of the timing of our engagement, the fiscal year 2012 
improper payment data were not included in our review. 

[3] The federal government matches states' expenditures for most 
Medicaid services using a statutory formula based on each state's per 
capita income. The federal share of Medicaid expenditures typically 
ranges from 50 to 83 percent. In fiscal year 2011, the federal share 
of Medicaid spending was $275 billion, while the state share was $161 
billion. 

[4] GAO, High-Risk Series: An Update, [hyperlink, 
http://www.gao.gov/products/GAO-13-283] (Washington, D.C.: February 
2013). 

[5] Pub. L. No. 107-300, 116 Stat. 2350 (Nov. 26, 2002), codified, as 
amended, at 31 U.S.C. § 3321 note. 

[6] Pub. L. No. 111-204, 124 Stat. 2224 (July 22, 2010), codified, as 
amended, at 31 U.S.C. § 3321 note. The changes made by this law to 
IPIA estimation and reporting were first implemented for fiscal year 
2011 reporting. IPIA and IPERA were further amended in January 2013 by 
the Improper Payments Elimination and Recovery Improvement Act, Pub. 
L. No. 112-248, 126 Stat. 2390 (Jan. 10, 2013). 

[7] OMB, Memorandum M-11-16, Issuance of Revised Parts I and II to 
Appendix C of OMB Circular A-123 (Apr. 14, 2011). For programs 
administered at the state level like Medicaid, this OMB guidance 
allows for state-level estimates to be used to generate a national 
improper payment dollar estimate and rate. However, agencies are to 
submit plans to OMB for approval to provide national-level estimates 
for state-administered programs based on a systematic selection of 
such programs each year. The justification to use this type of 
approach must include a description of the states to be selected each 
year, the methodology for generating annual national estimates, and a 
justification for using the proposed plan rather than an estimate 
based on a random statistical sample. 

[8] GAO, Standards for Internal Control in the Federal Government, 
[hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] 
(Washington, D.C.: November 1999). 

[9] The fiscal year 2010 data were the most recent data available at 
the time of our review for site visit selection. 

[10] PERM is also used to produce a national-level improper payment 
error rate for the Children's Health Insurance Program. 

[11] CMS operated pilot projects--Payment Accuracy Measurement and 
PERM--in fiscal years 2002 through 2005, published interim final rules 
in 2005 and 2006, and implemented PERM in a final rule published on 
August 31, 2007. 

[12] 42 C.F.R. part 431, subpart Q. 

[13] The Division of Error Rate Measurement (DERM) within CMS's Office 
of Financial Management is responsible for implementing PERM. DERM is 
also responsible for implementing the Comprehensive Error Rate Testing 
process to estimate improper payments in the Medicare FFS program. 

[14] According to CMS, the estimated error rate of 8.1 percent for 
fiscal year 2011 has a margin of error at the 90 percent confidence 
level of no more than plus or minus 2.2 percentage points. Also, this 
breakdown of overpayments and underpayments was not reported in fiscal 
year 2010 or previous years. 

[15] The weighted national component error rates do not total the 
national Medicaid improper payment error rate because the combined 
rate is a weighted average of FFS and managed care error rates, with 
the addition of the eligibility error rate. Also, a small correction 
factor ensures that Medicaid eligibility errors do not get "double 
counted" if the sampled item was also tested in either the FFS or 
managed care component. According to CMS, at the 90 percent confidence 
level, the estimated FFS error rate of 2.7 percent has a margin of 
error of no more than plus or minus 0.6 percentage points; the 
estimated managed care error rate of 0.3 percent has a margin of error 
of no more than plus or minus 0.1 percentage points; and the 
eligibility error rate of 6.1 percent has a margin of error of no more 
than plus or minus 2.2 percentage points. 

[16] Fiscal year 2007 was the first year that HHS reported the results 
of PERM reviews for estimating the Medicaid improper payment error 
rate. 

[17] Pub. L. No. 111-3, 123 Stat. 8 (Feb. 4, 2009). 

[18] OMB Memorandum M-11-16. 

[19] A sampling frame is a list or set of procedures for identifying 
all elements of a target population. 

[20] Negative cases are cases where eligibility was denied or 
terminated. They are discussed in more detail later in this report. 

[21] The optional OMB standard of a 3 percentage point precision at 
the 95 percent level of confidence is more rigorous in terms of the 
required sample size for the states than the regular OMB standard of a 
2.5 percentage point precision at the 90 percent level of confidence. 
An increased sample size achieves greater precision in the estimate at 
the state level. However, according to CMS, the estimated error rates 
included in fiscal year 2011 reporting at the national level have 
margins of error at the 90 percent confidence level of no more than 
plus or minus 2.5 percentage points. 

[22] OMB, Standards and Guidelines for Statistical Surveys (September 
2006). Although this guidance is not incorporated into OMB's improper 
payment estimation guidance, we cite it as a best practice because it 
describes techniques that can make government statistical sampling and 
reporting more accurate and reliable. 

[23] State-level sample sizes went into effect for fiscal year 2012 
reporting (i.e., starting with the fiscal year 2011 PERM measurement 
cycle). Beginning in the fiscal year 2011 PERM measurement cycle, 
individual state sample sizes are calculated based on that state's 
prior year's error rates for each of the three PERM components. Any 
revised error rates will be used for this purpose. 

[24] According to CMS officials, a cutoff date is needed to ensure 
that CMS has sufficient time to calculate and report the national 
Medicaid error rate. 

[25] According to CMS, this estimate has a margin of error at the 95 
percent confidence level of no more than plus or minus 4.9 percentage 
points. 

[26] According to CMS, this estimate has a margin of error at the 95 
percent confidence level of no more than plus or minus 4.7 percentage 
points. 

[27] These errors may include claims that are valid but for which 
appropriate documentation was not received as of the cutoff date, and 
therefore these are considered errors. 

[28] During eligibility reviews, these cases were coded as 
"undetermined," meaning that the case record lacks or contains 
insufficient documentation, in accordance with the state's documented 
policies and procedures, to make a definitive review decision for 
eligibility or ineligibility. "Undetermined" errors may include cases 
of beneficiaries who are in fact eligible for Medicaid but their 
eligibility was not confirmed, and therefore these are considered 
errors. 

[29] We have previously reported that when effectively implemented, 
data sharing can be particularly useful in confirming initial or 
continuing eligibility of participants in benefit programs and in 
identifying any improper payments that have already been made. GAO, 
Improper Payments: Remaining Challenges and Strategies for 
Governmentwide Reduction Efforts, [hyperlink, 
http://www.gao.gov/products/GAO-12-573T] (Washington, D.C.: Mar. 28, 
2012). In January 2013, the Improper Payments Elimination and Recovery 
Improvement Act of 2012, Pub. L. No. 112-248, 126 Stat. 2390 (Jan. 10, 
2013), was enacted, which, among other things, established a "Do Not 
Pay Initiative" to facilitate data sharing between federal agencies 
for payment-verification purposes. 

[30] Prior to the fiscal year 2010 measurement cycle, these were not 
coded as deficiencies but were classified as errors. Beginning with 
the fiscal year 2010 cycle, CMS documents the deficiencies in a 
state's final PERM cycle report for the state's consideration. 

[31] There were a total of 51 states measured for the fiscal years 
2008 to 2010 time period, and 8 states did not have any deficiencies 
identified. 

[32] According to CMS officials, beginning with the fiscal year 2013 
measurement cycle, states will be required to report technical errors 
to CMS. 

[33] OMB Memorandum M-11-16. 

[34] Pricing errors accounted for 4 percent of all of the errors 
identified across states for the fiscal years 2008 to 2010 measurement 
cycles, and about 1 percent of all dollars in error. 

[35] [hyperlink, http://www.gao.gov/products/GAO-11-278]. 

[36] While PERM regulations outline the elements that states are 
required to include in their CAPs, PERM regulations explicitly state 
that CAPs do not have to be approved by CMS. 

[37] The Deficit Reduction Act of 2005, Pub. L. No. 109-171, title VI, 
§ 6034, 120 Stat. 4, 74 (Feb. 8, 2006), codified, as amended, at 42 
U.S.C. § 1396u-6, established the Medicaid Integrity Program, which 
contracts for audit and educational services and provides long-term 
planning to combat fraud, waste, and abuse in Medicaid. To implement 
the Medicaid Integrity Program, CMS created MIG. 

[38] OMB Memorandum M-11-16. 

[39] Pub. L. No. 107-300, 116 Stat. 2350 (Nov. 26, 2002), codified, as 
amended, at 31 U.S.C. § 3321 note. 

[40] Pub. L. No. 111-204, 124 Stat. 2224 (July 22, 2010), codified, as 
amended, at 31 U.S.C. § 3321 note. Subsequent to the completion of our 
fieldwork, IPIA and IPERA were further amended in January 2013 by the 
Improper Payments Elimination and Recovery Improvement Act, Pub. L. 
No. 112-248, 126 Stat. 2390 (Jan. 10, 2013). 

[41] OMB, Memorandum M-11-16, Issuance of Revised Parts I and II to 
Appendix C of OMB Circular A-123 (Apr. 14, 2011). 

[42] 42 C.F.R. part 431, subpart Q. 

[43] The Medicaid improper payment estimate reported for fiscal year 
2011 included the results of the fiscal years 2008 to 2010 measurement 
cycles. CMS uses a 17-state 3-year rotation for measuring Medicaid 
improper payments, and therefore each state is reviewed once every 3 
years. 

[44] According to the Centers for Medicare & Medicaid Services, 
certain types of FFS claims do not go through this medical necessity 
review because of the nature of the claims. These types of claims 
include zero paid claims, fixed payments, Medicare premium payments, 
Medicare crossover claims, and denied claims. 

[45] While the number of Medicaid beneficiaries enrolled in managed 
care has grown, and is expected to continue to expand, as of fiscal 
year 2010, the federal share of managed care claims represented about 
24 percent of the total federal share of claims in dollars. 

[46] A capitation payment is a fixed payment, usually made on a 
monthly basis, for each beneficiary enrolled in a managed care plan or 
for each beneficiary eligible for a specific service or set of 
services. 

[47] Active cases are those in which an individual or family is 
enrolled in Medicaid in the month of the sample. 

[48] Negative cases are those that are denied or have a termination 
effective date in the month of the sample. 

[49] Pub. L. No. 111-3, § 601, 123 Stat. 8, 96 (Feb. 4, 2009). 

[End of section] 

GAO’s Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

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

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO’s 
website, [hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information. 

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

Congressional Relations: 

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

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

[End of document]