CMS 10184.Supporting Statement Part A(7-6-10)

Supporting Statement:

Eligibility Error Rate Measurement in Medicaid and the Children’s Health Insurance Program

1. Background

The Improper Payments Information Act (IPIA) of 2002 requires CMS to produce national error rates for Medicaid and the Children’s Health Insurance Program (CHIP). To comply with the IPIA, CMS will use a national contracting strategy to produce error rates for Medicaid and CHIP fee-for-service and managed care improper payments. The federal contractor will review States on a rotational basis so that each State will be measured for improper payments, in each program, once and only once every three years.

Subsequent to the first publication, we determined that we will measure Medicaid and CHIP in the same State. Therefore, States will measure Medicaid and CHIP eligibility in the same year measured for fee-for-service and managed care. We believe this approach will advantage States through economies of scale (e.g. administrative ease and shared staffing for both programs reviews). We also determined that interim case completion timeframes and reporting are critical to the integrity of the reviews and to keep the reviews on schedule to produce a timely error rate. Lastly, the sample sizes were increased slightly in order to produce an equal sample size per strata each month.

As outlined in the October 5, 2005 interim final rule, CMS convened an eligibility workgroup comprised of the Department of Health and Human Services, the Office of Management and Budget (OMB) and representatives from two States. The Office of Inspector General (OIG) participated in an advisory capacity. The workgroup was charged to make recommendations for measuring Medicaid and CHIP improper payments based on eligibility errors within the confines of current statute, with minimal impact on States’ resources and considering public comments on the August 27, 2004 proposed rule and the October 5, 2005 interim final rule. Based on the eligibility workgroup’s recommendations and public comments, we developed an eligibility review methodology that we expect will provide consistency in the reviews of active (i.e., beneficiaries receiving Medicaid or CHIP) and negative cases (i.e. beneficiaries whose benefits were denied or terminated) as well as achieve the confidence and precision requirements at the national level required by the IPIA.

In response to the public comments from the October 5, 2005 IFC, we published a second interim final rule in the August 28, 2006 federal register, which reiterated our national contracting strategy to estimate improper payments in both Medicaid and CHIP fee-for-service (FFS) and managed care, and set forth and invited further comments on State requirements for estimating improper payments due to errors in Medicaid and CHIP eligibility determinations. We also announced that a State’s Medicaid and CHIP programs would be reviewed in the same year.

In the August 31, 2007 Federal Register (72 FR 50490), we published a final rule for the PERM program, which implements the IPIA requirements. The August 31, 2007 final rule responded to the public comments on the August 28, 2006 interim final rule and finalized State requirements for submitting claims to the Federal contactors that conduct FFS and managed care reviews. The final rule also finalized State requirements for conducting eligibility reviews and estimating payment error rates due to errors in eligibility determinations.

We indicated in the proposed rule and the interim final rule that States would be expected to take some part in the eligibility reviews. We determined that States shall:

• Review eligibility in the same year the States are selected for Medicaid or CHIP fee-for-service and managed care reviews;

• Submit a sampling plan;

• Select monthly samples;

• Submit monthly sample lists of those cases randomly selected for review;

• Conduct the eligibility reviews;

• Report summary and detailed findings to CMS; and

• Provide analysis of the findings and proposed actions in a corrective action plan.

The States selected for review will submit an initial eligibility sampling plan to CMS for approval 60 days prior to the fiscal year being reviewed. The sampling plan should be developed to produce an error rate that meets a 95 percent confidence interval (using the mid-point of the confidence level) with +/- three percent precision. Once the sampling plan is approved, it will serve as the basic plan and the State will only resubmit the sampling plan if it makes major changes in future years. States will not need to resubmit the plan for approval of minor changes, for example, to react to fluctuations in the universe.

These States also will submit monthly sample selection lists to CMS. States will select monthly samples and conduct the reviews using a CMS standardized review methodology. The federal contractor will calculate State and national eligibility error rates for Medicaid and CHIP based on the States’ error findings.

The Children’s Health Insurance Program Reauthorization Act (CHIPRA) was enacted February 4, 2009. Sections 203 and 601 of the CHIPRA relate to the PERM program.

Section 203 of the CHIPRA establishes an error rate measurement with respect to the enrollment of children under the Express Lane Eligibility option. The law directs States not to include children enrolled using the Express Lane Eligibility option in data or samples used for purposes of complying with the MEQC and PERM requirements.

Section 601 of the CHIPRA, among other things, requires a new final rule and aims to harmonize the PERM and MEQC programs and provides States with the option to apply PERM data resulting from its eligibility reviews for meeting MEQC requirements and vice versa, with certain conditions. Our proposed rule would also codify and revise several procedural aspects of the process for estimating improper payments in Medicaid and the Children’s Health Insurance Program (CHIP).

Although the current information collection is approved for the provisions of the new PERM final rule, there is a PERM cycle of States that are still subject to the current final rule in which States are responsible for calculating their eligibility error rates and submitting them to CMS. We are adding the burden back into our estimate for this cycle of States.

2. Justification

1. Need and Legal Basis

The collection of information is necessary for CMS to produce national error rates for Medicaid and CHIP as required by Public Law 107-300, the IPIA of 2002.

The collection of information is also necessary to implement provisions from the Children’s Health Insurance Program Reauthorization Act of 2009 (CHIPRA) (Pub. L. 111-3) with regard to the Medicaid Eligibility Quality Control (MEQC) and Payment Error Rate Measurement (PERM) programs.

2. Information Users

The information collected from the States selected for review will be used by CMS to ensure States use a statistically sound sampling methodology, to ensure the States complete reviews on all cases sampled, and will be used by the federal contractor to calculate State and national Medicaid and CHIP eligibility error rates.

3. Use of Information Technology

This information collection involves the use of electronic submission of information to the extent that States have the technological capability. CMS will not require States to provide information electronically if they do not have secure systems in place to do so. The summary findings report form will require a signature and CMS will accept electronic signatures if available. The percentage of information expected not to be received electronically is less than one percent.

4. Duplication of Efforts

This information collection does not duplicate any other effort and the information cannot be obtained from any other source for CHIP.

To mitigate any duplication of effort for those States performing “traditional” Medicaid Eligibility Quality Control (MEQC), reviews and to reduce cost and burden for all States conducting pilots under the MEQC, at State option and upon CMS approval, the MEQC traditional reviews can be considered as meeting the Payment Error Rate Measurement (PERM) eligibility requirements for Medicaid and Title XXI Medicaid expansion if the MEQC reviews meet the PERM sampling confidence and precision requirements, and the review methods are in accordance with Section 431.812 and Part 7 of the State Medicaid Manual (SMM) at Chapter 3. The CHIPRA also allows States in their PERM year to apply PERM eligibility data to meet the annual MEQC requirement. The CHIPRA also requires CMS to review PERM and MEQC requirements and coordinate both sets of requirements in an effort to reduce redundancies. Based on feedback received prior to the August 2007 final rule, States in their PERM year can elect to use their PERM negative case reviews to meet their MEQC negative case action review requirement.

CMS has worked to make the active case review requirements less stringent than required under the MEQC program and the August 27, 2004 proposed rule by minimizing the verification requirements, allowing for certain case exclusions from the universe and providing that the States can cite cases where eligibility cannot be determined as “undetermined”.

5. Small Businesses

The collection of information does not impact small businesses or other small entities.

6. Less Frequent Collection

Failure to acquire this information will prevent CMS from effectively collecting State-specific eligibility payment error data on which to base State and national eligibility error rates for Medicaid and CHIP. Consequently, CMS will not be able to produce these error rates.

7. Special Circumstances

CMS does not anticipate that States would be required to submit information more often than monthly in the year the States are reviewed (once every three years per program). States will provide a sampling plan in the beginning of the year of selection, monthly selection lists at the beginning of each month, findings on the cases reviewed and a corrective action plan.

8. Federal Register/Outside Consultation

Due to the emergency nature of the program announce OMB has waived the Federal Register notice requirements for this collection.

The MEQC system is discussed at regional and national meetings of the National Association for Program Information and Performance Measurement and other related groups.

9. Payments/Gifts to Respondents

There is no provision for any payment or gift to respondents associated with this reporting requirement.

10. Confidentiality

Confidentiality has been assured in accordance with section 1902(a) (7) of the Social Security Act.

11. Sensitive Questions

No questions of a sensitive nature are asked.

12. Burden Estimate (Total Hours & Wages)

The number of respondents is estimated to be 34 States (17 Medicaid and 17 CHIP States). The annualized number of hours estimated that may be required to respond to requests for information equals 15,755 hours (per State, per program).

CMS recognizes there are other costs associated with this measurement, other than labor. These include overhead costs such as supplies to complete reviews (e.g. mailing cases and verification requests, travel for possible interviews), training and manual development. These costs will vary from State to State depending on many variables including the type of program integrity practices in place, salaries and pricing. CMS includes the current fringe and overhead costs that is provided to the PERM medical review contractor (who conducts reviews on medical records) which is 25.27 percent and 26.55 percent respectively. We believe these are reasonable costs per State in the hourly burden estimates for a total computable eligibility review cost per State, per program. The fully loaded rate totals $43.19 per hour. 15,755 hours x $43.19 per hour = $680,458.45 per State, per program.

Each year, 17 States will participate in both the Medicaid error rate measurement and the CHIP error rate measurement. Therefore, estimates were calculated for 34 responses to each request for information.

It is estimated that each State will spend up to 15,755 hours of time annually (when selected), per program, to support this collection of information. The State will provide the following information, per program.

1. A sampling plan, for CMS approval, based upon the universes of beneficiaries in the program and persons whose benefits were denied or terminated. States would only resubmit the sampling plan when changes are made (responding once per year @ 1,000 hours per program);

2. Monthly sample lists detailing the active and negative cases selected for review that month (responding 12 times per year @ 100 hours in each response or 1,200 hours per program);

3. Review findings on each case following the eligibility and payment reviews (responding to each of the approximately 708 sampled cases, i.e. approximately 504 active cases and 204 negative cases for a total estimated 10,055 hours per program) in order to prepare findings, including an error rate, reviews must be completed and the burden here is inclusive of all of the associated review activities (more detail below).

4. Report individual review findings within 150 days of the end of the sample month (responding 12 times per year @ 100 hours in each response or 1,200 hours per program);

5. Report individual payment findings 60 days after initiating the claims collection (responding 12 times per year @ 100 hours in each response or 1,200 hours per program);

6. Summary of eligibility and payment review findings (100 hours); and

7. A corrective action report for purposes of reducing the payment error rate in eligibility (responding once at up to 1,000 hours per State).

It should be noted that the current regulation still requires States to calculate the eligibility error rates and submit them to CMS. Therefore we are reinstating the burden for the current cycle of States that are still administering PERM under the current regulation. However, the statistical contractor has developed a calculator for States to use for those that do not have the statistical expertise to calculate the error rates. We expect all 17 States to choose this option but States may incur additional burden if attempting to calculate the eligibility error rates using the formulas provided in our guidance and opt not to use the calculator.

Based on the new legislation (CHIPRA) we are also providing burden estimates for States’ option to elect to use MEQC data to meet the PERM eligibility requirement in their PERM year. Currently, this option is limited to 19 States (12 Medicaid and 7 CHIP), as these States are conducting a “traditional” MEQC review in accordance with section 431.812. The annualized number of hours estimated we believe is similar to the PERM eligibility requirements but we will add to the burden estimate for the MEQC review procedures to account for its more stringent requirements. The number of hours that may be required to respond to requests for information equals 21,426 hours (per State, per program).

States that elect to substitute MEQC data for PERM will provide the following information:

1. A current MEQC sampling plan. The burden estimate for this requirement is approved under OMB control number 0938-0146 and instrument CMS-317;

2. Two six month MEQC error rates from the previous fiscal year. The burden estimate for this requirement is approved under OMB control number 0938-0246 and instrument CMS-301;

3. Modified PERM sampling plan that incorporates MEQC sampling elements while meeting PERM precision requirements (500 hours per program);

4. Monthly MEQC sample selection lists. This requirement is approved under OMB control number 0938-0147 but a new instrument based on the CMS-10184A has been modified for the purpose of data substitution and is attached to this statement;

5. MEQC Review findings on each case following the eligibility and payment reviews (responding to each of a maximum of 1,204 sampled cases, i.e. approximately 1,000 active cases and 204 negative cases for a total estimated 17,040 hours per program) in order to prepare findings, reviews must be completed and the burden here is inclusive of all of the associated review activities (more detail below).

6. Report individual MEQC review findings, consistent with PERM reporting, within 150 days of the end of the sample month (responding 12 times per year @ 100 hours in each response or 1,200 hours per program). A new instrument based on CMS-10184B has been modified for the purpose of substituting data and is attached to this statement;

7. Report individual MEQC payment findings, consistent with PERM reporting, 60 days after initiating the claims collection (responding 12 times per year @ 100 hours in each response or 1,200 hours per program). A new instrument based on CMS-10184D has been modified for the purposes of substituting data and is attached to this statement;

8. Summary of eligibility and payment review findings consistent with PERM reporting, in which a modified version of CMS-10184E will provide necessary information and is attached to this statement (100 hours per program); and

9. A corrective action report for purposes of reducing the payment error rate in eligibility (responding once at up to 1,000 hours per program).

10. We’ve also added an additional 2 hours per required form to reformat MEQC data into the appropriate forms. We are adding an additional 98 hours for each State to reformat MEQC data into the appropriate PERM eligibility forms.

Sample Size Development

This measurement will be a case based sample with approximately 504 active cases and 204 negative cases, per program. For PERM, an active case means an individual beneficiary and for MEQC and active case means an individual beneficiary or family hat is enrolled in the Medicaid or CHIP program in the month that the case is sampled. Negative case means a beneficiary or family that has completed an application for benefits and is denied for program benefits or is terminated based on the State agency’s completed redetermination.

These 708 cases will be sampled over the period of one fiscal year. The approximately 504 active cases will be further stratified into three equal strata (estimated at 168 cases each). The Medicaid active universe consists of all active Medicaid cases funded through Title XIX for the sample month. The following cases should be excluded from the Medicaid universe:

• Cases for which the Social Security Administration, under a Section 1634 agreement with a State, determines Medicaid eligibility for Supplemental Security Income recipients;

• All foster care and adoption cases under Title IV-E of the Social Security Act;

• All State only funded cases (excluded from the PERM universe);

• Cases currently under an active beneficiary fraud investigation;

• Suspended cases or cases that have not met applicable spend down in the sample month.

• Cases enrolled in Medicaid using States’ Express Lane Eligibility option under Section 1902(e) (13) of the Social Security Act.

Although express lane cases are excluded from the universe, the number of cases excluded due to Express Lane Eligibility must be reported separately for PERM purposes.

The CHIP active universe consists of all active CHIP cases funded through title XXI for the sample month, with the exception of cases enrolled in CHIP using States’ express lane eligibility option under section 1902(e)(13) of the Social Security Act. CHIP cases will also be stratified into three strata. The negative case samples will not be stratified.

Given these parameters and that States’ sampling plans must estimate a sample size to achieve a payment error rate at +/- three percent precision and 95 percent confidence (using the mid-point of the confidence interval) for the active cases; we anticipate that sampling plans will take up to 1,000 hours per State, per program.

States that elect to substitute MEQC data to meet the PERM eligibility requirement must submit their MEQC sampling plan and 6 month error rates from the previous fiscal year to make a comparison between State MEQC sampling and error rate data to historical PERM sampling and error rate data. The State will work with CMS to develop a revised sampling plan that will meet PERM requirements while using sampling elements provided by the MEQC sampling plan. The maximum number of cases that will be sampled by State substituting MEQC data is 1,000 active cases. Both MEQC and PERM universe exclusions apply with the exception of State-only funded cases not subject to review under PERM and cases in which the beneficiary cannot be contacted, located or has moved out of State. The latter exceptions are cases that are considered “undetermined” for PERM purposes and must be reported. MEQC sampling plan and error rate burden is approved under OMB control numbers 0938-0146 and 0938-0246 respectively. We anticipate that developing sampling plans to allow for MEQC data substitution will take up to 500 hours per State, per program.

Case Reviews

Based on the PAM Year 2 cost and efficiency study, we estimated it took an average of 12.4 hours to complete a case review. Except for one State participant, PAM Year 2 States conducted full eligibility reviews.

In the PERM measurement, active cases are divided into three strata: stratum 1 is completed applications for the sample month, stratum 2 is completed redeterminations for the sample month, and stratum 3 is all other active cases for the sample month. We believe that strata 1, 2 and negative case reviews will take a bit less time due to the ease of reviewing a recent State action on the case and stratum 3 will take a bit more time due to varying timeframes when eligibility is reviewed, i.e. either when the last State action occurred or the sample month if the last action occurred prior to 12 months from the sample month. We estimated that 540 cases (204 negative, 168 stratum 1 active cases and 168 stratum 2 active cases) will take 10 hours to complete the eligibility review and 168 stratum 3 case reviews will take 15 hours to complete the eligibility review for a total of 7,920 hours for reviews.

We included an additional 2,135 hours to the 7,920 case review estimated hours (for a total of 10,055 hours) for supporting functions like training, supervision, quality assurance and creation of review tools, etc. Therefore, the 10,055 hours represents the burden to complete review findings to show the disposition of each case selected for review and includes all of the review supporting functions. CMS will use the detailed findings to compare to the monthly sample lists to determine that the State completed its reviews of the selected cases.

The following assumptions were used:

• The estimated number of programs needed to produce a national eligibility error rate with the confidence and precision to meet the IPIA requirements is 34 annually; 17 for Medicaid and 17 for CHIP;

• The estimated number of cases needed from each State to produce a State specific eligibility error rate with the confidence and precision needed to have a national rate meet IPIA standards is estimated to be 504 per program;

• The 504 active cases per program are going to be equally stratified on a monthly basis in three (3) strata: (1) applications approved, (2) cases where eligibility was redetermined (3) all other active cases. The 204 negative cases per program are not stratified;

• The 708 cases (total active and negative cases) will be sampled over a full fiscal year;

• Review eligibility as of the last action the State took unless, for stratum 3 cases, that action was more than 12 months from the sample month. If so, review eligibility as of the sample month;

• Attach payments for services received:

  • In the review month or the first 30 days of eligibility (according to State policy on full month or date specific eligibility coverage) for cases in strata 1 and 2, and

  • Within the sample month for cases in stratum 3;

• Review payments and verify whether the payments were made appropriately based on the eligibility review findings. The payment review may include determining if the beneficiary met his/her liability amount or cost of institutional care.

• Programs will submit State specific active case error findings and payment error findings;

• Programs will identify the number and percent of cases and payment amounts for undetermined cases (cases where eligibility could not be verified);

• Programs will submit State specific negative case error findings;

• Programs will exclude from the universe or the sample (if these cases cannot be excluded from the universe), cases under active beneficiary fraud investigation;

• Programs will conduct reviews in accordance with the State’s eligibility criteria that are in effect as of the month eligibility is being verified; and

• There is no administrative period.

Finally, CMS will provide States with the option, in those years when selected for the PERM review and subject to CMS approval, to use the eligibility review requirements in part 431, subpart P to meet the requirements for the PERM eligibility reviews. The eligibility measurement sampling process under either program must meet the PERM confidence and precision requirements.

States that elect to use MEQC data to meet their PERM eligibility requirement do not have to stratify the eligibility universes into the three strata as MEQC does not require stratification. We believe that the majority of sampled cases for MEQC would be classified as stratum 3 as continuing cases make up the majority of the Medicaid and CHIP universes as a whole. The MEQC review methodology is also more stringent than the PERM review methodology since in many cases, a face to face interview must be conducted. An administrative period, or the month prior to the sample month is applied to the review. As such, these cases will take a bit more time. We estimate that it will take 15 hours to complete an MEQC review. The maximum sample size for PERM or MEQC (if substituting MEQC data) is 1,000 cases, resulting in a maximum burden hours estimate for the MEQC reviews of 15,000 hours.

We also propose adding additional burden as stated above. States must report PERM and MEQC findings separately and will use an estimated 2 hours per required form to reformat PERM or MEQC data into the appropriate forms. We are adding an additional 98 hours for each State to reformat MEQC data into the appropriate PERM eligibility forms for a total of 15,098 hours.

13. Capital Cost

There are no capital costs associated with this collection of information.

14. Cost to the Federal Government

There are no additional costs.

15. Changes to Burden

Although the current information collection is approved for the provisions of the new PERM final rule, there is a PERM cycle of States that are still subject to the current final rule in which States are responsible for calculating their eligibility error rates and submitting them to CMS. Therefore we are reinstating the burden for the current cycle of States that are still administering PERM under the current regulation. However, the statistical contractor has developed a calculator for States to use for those that do not have the statistical expertise to calculate the error rates. We expect all 17 States to choose this option but States may incur additional burden if attempting to calculate the eligibility error rates using the formulas provided in our guidance and opt not to use the calculator.

16. Publication/Tabulation Dates

States selected for the FY 2010 measurement will submit the Medicaid and CHIP sampling plan prior to the fiscal year measurement. The States will sample at least 708 cases over a twelve month period beginning with October 2009. States will report sample lists on the 15^th of the month following the sample month. The detailed case review findings for 100 percent of the cases reviewed in a sample month are due 150 days from the end of the sample month. Claims collection will begin in the fifth month following the sample month and will be reported within 60 days of the first day of the month in which the claims collection process begins. The final summary report is due July 1^st after the end of the fiscal year being measured. The calculated national program error rate for both Medicaid and CHIP will be published annually in the Agency Financial Report (AFR).

17. Expiration Date

This collection does not lend itself to the displaying of an expiration date.

18. Certification Statement

There are no exceptions to the certification form

C. Collections of Information Employing Statistical Methods

1. Describe potential respondent universe.

The universe for this project is the 50 States’ and the District of Columbia’s Medicaid and CHIP programs.

The potential respondent universe is 34 unique programs (17 Medicaid and 17 CHIP). We estimate that approximately 504 active cases will be randomly selected for review by each of the 17 States in each program to achieve a State specific, program specific eligibility payment error rate. These results will be used to calculate a national eligibility component error rate. We estimate States will randomly select 204 denied and terminated cases for the negative case reviews.

The anticipated response rate is 100 percent due to the statutory requirements at section 1902(a)(6) of the Act and Section 2107(b)(1) of the Act that require States to provide information necessary for the Secretary to monitor program performance.

2. Describe procedures for collecting information.

In the first year of each State’s eligibility measurement, we determined a case sample size of 504 active and 204 negative cases (per State using an assumed error rate of 5 percent). In subsequent years, the actual sample size for each State will be estimated to achieve a 95 percent confidence level (using the mid-point of the confidence interval) within three percent precision.

In order to meet the requirements of IPIA, all selected States must participate.

3. Describe methods to maximize response rates.

We will depend on States to provide reliable data. The States are reporting findings monthly and on an annual basis for the year selected for review (once every three years).

4. Describe any tests of procedures or methods.

Not applicable.

5. Provide the names and telephone numbers of individuals consulted on the statistical aspects.

Livanta LLC, The Lewin Group, and the CMS Region 2 statistician consulted on the statistical methodology of this project and the applicable data substitution provisions.