Child Care Bureau: CCDF: Measuring Improper Payments in the Child Care Program

Second Error Rate Pilot Report

Download this document in Word (1,076 KB) or PDF (1.28 MB)

IV. FINDINGS AND STATES' RESPONSES

This chapter reviews findings from the record review process and error rate analysis for the five pilot States—Florida, Kansas, New Jersey, Oregon, and West Virginia. It also includes actions of States, or steps they plan to take, in response to the causes of errors. The review of findings for each State includes a brief description of the record review process followed by a presentation of the error rate findings. The findings include four measures computed for each State as follows:

Percentage of cases with an error—This percentage is based on the number of sampled cases with an error, regardless of whether it resulted in an improperly authorized payment or not, compared to the total number of cases in the sample. This percentage is determined by dividing the number of sampled cases with an error by the total number of cases reviewed in the sample and then multiplying by 100.
Percentage of cases that have an improperly authorized payment¹⁶ This percentage is based on the number of sampled cases with errors that have an improperly authorized payment, compared to the total number of cases in the sample. The percentage is determined by dividing the number of cases with an error that resulted in an improperly authorized payment by the total number of cases in the sample and then multiplying by 100.
Payment Error Rate (Percentage of authorized payments in error for the review period)—The payment error rate is the percentage of the gross amount of improperly authorized payments (overpayments plus underpayments) for the review period compared to the total amount of authorized payments in the sample. This rate is determined by dividing the gross amount of improperly authorized payments by the total dollar amount of authorized payments in the sample cases and then multiplying by 100.
Average amount of improperly authorized payment—The average amount of improperly authorized payments is the average amount of money the State authorized for payment improperly on a per child basis during the designated review month. This rate is determined by dividing the gross amount (overpayments plus underpayments) of improperly authorized payments in the sample by the number of cases that had an improperly authorized payment.

Following the discussion of the error rate findings, the chapter outlines State explanations of the error causes and summarizes the corrective actions to be taken as a result of the error rate findings. The source of this information is State responses to the State Response Form included as Appendix I and discussed within the Next Steps section of each State’s findings.

A. Florida

In Florida, the Review Team consisted of two Managers, two Performance Analysts, and three Program Analysts. The Project Team made substantial modifications to the Record Review Worksheet template. The review process occurred centrally, with all records sent to Tallahassee. A second member of the team re-reviewed all records for inter-rater reliability to ensure that findings and coding in Columns 3 and 4 of the Record Review Worksheet would be the same regardless of reviewer.

Following completion of the review, Florida learned from a finding in a IV-E Federal audit that only the State’s auditor general or Department of Children and Family’s (DCF’s) Family Safety Unit could access case information related to any protective services placement. The review had found six cases in error due to missing protective services documentation. Florida submitted amended review sheets on these six cases, changing the finding from ineligible to correct.

1. Results

As shown in Exhibit 3, the data indicated that 53 records (35% of the sample) had one or more administrative errors and all 53 cases had administrative errors that resulted in an improperly authorized payment. The total of authorized payments to the 150 sampled cases for the review month was $37,571 of which $6,908 was the result of error. This error amount represents 18 percent of all authorized payments for the sample.

Exhibit 1. Florida Findings (N=150)


Percentage of cases with errors	35%
Percentage of cases that have an improperly authorized payment	35%
Percentage of improperly authorized payments	18%
Average amount of improperly authorized payment	$130.35

Missing documentation as an error cause accounted for the largest percentage (36%) of the total dollars in error in the Florida sample. All of the errors attributed to missing documentation resulted in ineligibility. Missing documentation occurred in a wide range of eligibility elements, including Application Forms (Element 100), Income (earned and unearned), work activity, TANF, transitional TANF, and age.

Twenty-two of the 53 total errors occurred in Payments (Element 420); however, these errors resulted in a much smaller percentage of total dollars in error (11%). Reasons for Payments errors included the incorrect use of the fee/rate schedule or the authorization of full-time payments for part-time attendance. The ten errors in Application Forms represented 36 percent of the total dollars in error. The five errors in Household Members accounted for 19 percent of the total dollars in error. An additional 9 errors occurred in Earned Income (Element 400), which accounted for 18 percent of the total dollars in error. The remaining 7 errors were dispersed among several review elements, such as Residency, Qualifying Child and Qualifying Care.

2. Florida Responses

This section identifies actions taken by the Project Team in response to the causes and corrective actions taken based on the findings of the pilot:

Developed a desk reference tool for eligibility procedures;
Initiated focused monitoring and training on eligibility procedures on an annual basis;
Provided each Early Learning Coalition (ELC) with its individual results of the pilot review process along with recommendations for corrective action;
Created mechanisms to track the results of the eligibility monitoring and other contract monitoring audits;
Implemented an annual eligibility review process of all 31 ELCs;
Implemented a Data Quality Initiative to standardize data collection and to identify data anomalies throughout the State; and
Implemented procedures to conduct post audits on reimbursement requests.

B. Kansas

The Kansas Review Team consisted of three Quality Assurance Supervisors (Topeka, Wichita, Kansas City) and three Quality Assurance Specialists. The Review Team also conducted a re-review of all cases by having a second team member conduct the review. An error summary was sent to the related local Social Rehabilitation Services (SRS) office whenever an error was found. The local SRS office corrected the case based on the review information. The Review Team reviewed all records centrally in Topeka. The Project Team modified the Record Review Worksheet to include “Yes” or “No” questions to reduce the amount of written narration for the reviewers.

1. Results

As shown in Exhibit 4, the data show that 52 (35%) of the 150 sample records had one or more errors. All of the 52 cases with errors resulted in an improperly authorized payment. The total authorized payments to the sampled cases for the review month was $40,623, of which $4,650 was the result of error. This error amount represents 11 percent of all authorized payments for the sample.

Exhibit 2. Kansas Findings (N=150)


Percentage of cases with errors	35%
Percentage of cases that have an improperly authorized payment	35%
Percentage of improperly authorized payments	11%
Average amount of authorized improper payment	$89.42

While 92 percent of the total dollars in error were in Hours of Care (Element 340), the causes varied. The most common error causes included:

Failure to allow or the miscalculation of time for meals and travel;
Failure to reconcile the hours of care needed with the hours verified in the employment documentation or the school schedule; and
Allowing, without proper documentation, hours that exceeded need to satisfy a provider minimum.

There were several cases where the reviewer was unable to attribute an error to a specific cause because the case documentation was nonexistent.

2. Kansas Responses

This section identifies actions taken by Kansas in response to the causes and corrective actions taken based on the findings of the pilot:

Prior to the pilot, Kansas had completed child care quality control (QC) reviews for the month of October 2005, the second month of statewide payments using the new child care EBT payment system. Kansas was already aware of the error causes found in the pilot and had conducted statewide refresher training in June 2006 to address these errors. The training focused on hours of care, use of child care plan hour worksheets, computation of income, use of income worksheets to document computations, and general case documentation.
Kansas continues to complete monthly Child Care QC reviews and uses the results of these reviews to inform training needs. Kansas is updating the Child Care Personnel Trainer and Training Academy to emphasize case documentation and computation of hours of care needed and income. Kansas requires this training for all new workers and for others as needed. Supervisors now complete monthly case reviews for the Child Care Assistance Program.
Kansas hired a child care trainer in July 2006 to revise training materials and conduct several training sessions for eligibility staff. The QC unit compiles results monthly for regional administrators to keep them informed about findings and areas needing improvement. Supervisory case reviews at the regional level now include child care cases.
Kansas contracted with a firm to develop software to track all case review findings and provide aggregate review data. The web based system tracks aggregate case review data for the mandatory Food Stamp reviews and the child care reviews. The child care reviews began in July 2006. The software costs $75,000 with an 18 percent maintenance costs agreement.

C. New Jersey

After receipt of technical assistance during the site visit, New Jersey's Project Team substantially customized the Record Review Worksheet template to include coding and specific boilerplate language to guide the reviewer’s written case narrative. The Project Team submitted a universe to the contractor to select a statewide sample that included only cases using a child care voucher and appearing in the Child Care Automated Resources Eligibility System (CARES). This universe did not include contracted child care slots because these cases did not meet pilot requirements for the unit of analysis of an authorized payment on a per child basis utilizing the voucher system. The Project Team trained its Review Team centrally and then reviewers conducted the reviews in their assigned regions at the local Child Care Resource and Referrals (CCR&R) agencies. The Review Team consisted of three supervisors and 13 Child Care Specialists.

1. Results

As shown in Exhibit 5, the data show that 28 (19%) of the 150 sample records had one or more errors. All of the 28 cases with errors had an improperly authorized payment. The total authorized payments for the sampled cases for the review month was $45,807, of which $6,043 was considered to be in error. This error amount represents 13 percent of all authorized payments for the sample.

Exhibit 3. New Jersey Findings (N=150)


Percentage of cases with errors	19%
Percentage of cases that have an improperly authorized payment	19%
Percentage of improperly authorized payments	13%
Average amount of improperly authorized payment	$215.82

Calculation of Household Income (Element 400) accounted for the largest percentage (56%) of the total dollars in error. Errors in Application Forms (Element 100) resulted in an additional 27 percent of the total dollars paid in error. The most common errors included:

Authorizing a payment for hours that exceeded the documented need (half-time vs. full-time and full-time for school-aged children);
Using an incorrect household size;
Allowing income without verification;
Ignoring income from a second job;
Failing to react to a reported income change;
Applying the fee schedule incorrectly; and
Paying for days of non-attendance.

2. New Jersey Responses

This section identifies actions taken by New Jersey in response to the causes and corrective actions taken based on the findings of the pilot:

Allocate time during meetings with the Child Care Resource and Referral (CCR&R) Directors meeting to provide feedback and direction on implementing corrective measures.
Institute the Error Rate Pilot monitoring tool for use of all future file reviews, in addition to increasing the number of files for review.
Provide technical assistance to CCR&R agency staff in those areas where improperly authorized payments were detected in file reviews.
Improve lines of communication between Division of Youth and Family Services (DYFS), the lead State child protective service agency and the Division of Family Development, the lead State child care agency and the CCR&R agencies at State and local levels.
Look into implementing changes in the CARES system to better detect data entry errors or create error messages when data entry errors occur.
Implement a process to conduct electronic matching of the automated child care client database with records on other wage, SSI, and child support information systems to identify and reduce the number of improperly authorized payments.
Hire quality assurance staff or allocate the FTE needed to enable the Division to conduct file reviews, using a methodology that guarantees a statistically valid sample size of randomly selected cases.
Link reduced improperly authorized payments to penalties in Child Care Resource and Referral contracts.

D. Oregon

Oregon's Project Team made very few modifications to the Record Review Worksheet template. The Review Team consisted of five Quality Control (QC) Staff Persons who have responsibility for TANF and Food Stamp case reviews. Following training on child care policy and procedures, the Review Team reviewed the cases centrally. The Project Team Leader reviewed all cases with errors a second time and discovered that the Review Team went beyond the level of a desk audit, expecting greater detail in documentation than what had been anticipated or required of the child care review process. As a result, the Review Team reviewed a subset of cases a third time and changed three review results.

1. Results

As shown in Exhibit 6, the data show that 40 (27%) of the 150 sample cases had one or more errors; however, only 16 of the cases with errors resulted in an improperly authorized payment. The total authorized payment for the sampled cases for the review month was $81,757, of which $1,925 was considered to be in error. This error amount represents 2 percent of all authorized payments for the sample.

Exhibit 4. Oregon Findings (N=150)


Percentage of cases with errors	27%
Percentage of cases that have an improperly authorized payment	11%
Percentage of improperly authorized payments	2%
Average amount of authorized improper payment	$120.33

The 24 cases with errors with no improperly authorized payment involved discrepancies such as missing applications or income verification, miscalculation of hours, and failure to document need for care. Fifty-six percent of the total dollars in error occurred in the area of calculation of household Income (Element 400). A general summary of the error causes includes:

Use of unearned income that had ended;
Failure to react when the client reported a job change;
Use of outdated pay stubs;
Failure to include regular overtime pay in the income calculations; and
Improper treatment of biweekly vs. bimonthly income.

The remainder of the dollars in error occurred in Qualifying Care (Element 340). The causes of these errors were the miscalculation of hours of care and either an understatement or overstatement of the hours needed.

2. Oregon Responses

This section identifies actions taken by Oregon in response to the causes and corrective actions taken based on the findings of the pilot:

Oregon plans to modify the Food Stamp targeted review process and database that contains review information to ensure that income errors identified in any Food Stamp case are corrected in the companion ERDC case.
Each year, the Oregon Department of Human Services (DHS) conducts in all areas of the State a series of “Accuracy Summits” that focus on techniques to reduce errors in the Food Stamp program. DHS plans to include workshops to improve ERDC accuracy in the next series of summits that begins in July 2007.
DHS produces a monthly newsletter that focuses on techniques to reduce Food Stamp errors. The newsletter will now include information about other programs, including ERDC. A recent newsletter included an article about correctly determining co-pay amounts in ERDC cases. Future newsletters will contain information about ERDC payment accuracy, including an article about the results of the pilot review.
The Governor’s recommended budget for the next biennium would allow DHS to extend Targeted Review outcome measures to the Child Care programs in local offices by FY2008. This comprehensive Program Integrity infrastructure provides a method for gathering timely, local performance data and will enable the Department to develop corrective action measures.

E. West Virginia

West Virginia's Project Team made substantial modifications to the Record Review Worksheet, using check-off boxes, “Yes” and “No” fields, and coding to eliminate the need for extensive narrative recording. The Review Team consisted of two Field Consultants and a Policy Specialist. All of the reviews occurred on site at the regional CCR&R agencies. The Review Team had already read 10 cases during one of their regular CCR&R agency audits. To improve the review process, the three reviewers completed several reviews jointly and consulted by telephone frequently. Because the Review Team reviewed the cases in regions, travel time was a considerable factor with several hours of driving time from one CCR&R to another.

1. Results

As can be seen in Exhibit 5, West Virginia found 33 (22%) of the 150 sample records to have one or more errors; however, only 20 (14%) of the cases with errors resulted in an improperly authorized payment. The total authorized payment to the sample cases for the review month was $34,506, of which $2,522 was found to be in error. This error amount represents 7 percent of all authorized payments for the sample.

Exhibit 5. West Virginia Findings (N=150)


Percentage of cases with errors	22%
Percentage of cases that have an improperly authorized payment	14%
scope="row"Percentage of improperly authorized payments	7%
Average amount of improperly authorized payment	$126.14

Thirteen of the cases with errors did not result in an improperly authorized payment. These case errors involved discrepancies such as failure to enter mandatory information into the Family and Child Tracking System (FACTS), computation errors that did not result in a fee change, and missing documentation. Errors in Application Forms (Element 100) represented 35 percent of total dollars in error. Household Members (Element 320) accounted for 28 percent of the total dollars in error. Earned Income (Element 400) and Income Eligibility (Element 410) represented a combined 19 percent of the total dollars in error.

A general summary of the error causes includes:

Signatures on attendance sheets;
Missing parent agreements;
Failure to enter the work schedule into the FACTS system.

Failure to include documented earned and unearned income;
Use of an incorrect conversion factor (4 vs. 4.3, weekly vs. biweekly, etc.);
Use of net rather than gross pay; and
Failure to reconcile the hours of care allowed with the hours verified in the employment documentation or a school schedule.

2. West Virginia Responses

This section includes a summary of the West Virginia response to the error causes and corrective actions taken based on the findings of the pilot:

West Virginia will conduct refresher policy training with the CCR&R directors. The directors will then provide the training to their staff. West Virginia feels that, by training the directors rather than the staff, the directors will become more involved in eligibility activities.
Technical Assistance staff will continue to perform random quarterly case audits. Prior to the pilot, the FACTS electronic record was the object of the review. An expanded review will now also include the physical case record.
CCR&R agencies must now develop and implement Quality Assurance Plans. The plans will now require inclusion of payment accuracy initiatives if a CCR&R agency has an error rate that is above the State mean.
Family child care providers must submit sign-in and sign-out sheets to verify attendance along with their billing forms. CCR&R agencies must audit billing forms and compare work and school schedules to times shown on the sheets to verify that the child care usage complies with time approved.
Child care providers who submit incorrect or improper billing forms must attend a retraining session on proper billing procedures.
Two State level child care consultants will continue to sample CCR&R agency cases to ensure compliance with appropriate policies and provide training and technical assistance on policy, procedures, and the use of FACTS. The consultants also will provide follow-up to ensure the CCR&R agency corrective action plans are completed.

F. Conclusion

This section provides a summary of the findings and conclusions based on States’ responses to the questions about causes of errors and corrective actions. The summary also includes a discussion of the similarities and differences between findings from the four States in the initial pilot and the five States in the second pilot.

This was the second pilot in an exploratory effort to develop and pilot a methodology that States could use to compute an error rate. Similar to the first pilot conducted in Arkansas, Colorado, Illinois, and Ohio, this pilot yielded extremely useful findings.

1. Summary of Results

Exhibit 6 presents a summary of the findings for the five States involved in the second pilot and for the four States involved in the first pilot. The results from the second pilot demonstrate a very similar range to the findings in the first pilot. As can be seen in the summary table below, the estimated percentage of improperly authorized payments in the five pilot States ranged from a low of 2 percent to a high of 18 percent. These results are very similar to the percentage of improperly authorized payments for the four States in the first pilot, which ranged from 4 percent to 20 percent.

Exhibit 6. Summary of Findings

Second Pilot


Measure	Florida (N=150)	Kansas (N=150)	New Jersey (N=150)	Oregon (N=150)	West VA (N=150)
Measure	Florida (N=150)	Kansas (N=150)	New Jersey (N=150)	Oregon (N=150)	West VA (N=150)
Percentage of cases with errors	35%	35%	19%	27%	22%
Percentage of cases that have an improperly authorized payment	35%	35%	19%	11%	14%
Percentage of improperly authorized payments	18%	11%	13%	2%	7%
Average amount of improperly authorized payment	$130.35	$89.42	$215.82	$120.33	$126.14

First Pilot


Measure	Arkansas (N=150)	Illinois (N=150)	Ohio (N=123)	Colorado (N=150)
Percentage of cases with errors	12%	24%	44%	35%
Percentage of cases that have an improperly authorized payment	12%	6%	32%	25%
Percentage of improperly authorized payments	14%	4%	20%	8%
Average amount of improperly authorized payment	$289.53	$150.64	$194.28	$88.99

For the estimated percentage of sampled cases determined to contain administrative errors related to eligibility, the five pilot States' percentages ranged from 19 percent to 35 percent. The four States in the first pilot had a wider range, with the percentages of cases with errors ranging from 12 percent to 44 percent.

For the percentage of sampled cases with administrative errors that resulted in an improperly authorized payment, the five pilot States percentages ranged from 11 percent to 35 percent. The four States in the first pilot had a slightly lower range with the percentage of cases with an improperly authorized payment ranging from 6 percent to 32 percent.

The average amount of an improperly authorized payment in the five pilot States ranged from a low of $89.42 to a high of $215.82. The four States in the first pilot had a slightly higher range with the average amount of an improperly authorized payment ranging from $88.99 to $289.53.

Both pilots’ findings suggest that there is little, if any correlation, between States’ percentage of cases in error, the percentage of cases in error that had an improperly authorized payment, and the percentage of improperly authorized payments. In comparison to New Jersey, Kansas had relatively high numbers of cases with errors and cases with an improperly authorized payment. Yet the Kansas percentage of improperly authorized payments was the lower of the two States. In the first pilot, Arkansas had a percentage of improperly authorized payments that exceeded both its percentage of cases with errors and percentage of cases with errors that had improperly authorized payments. State improper payment strategies need to focus on both the most common error types and those error causes which produce errors of greater amounts, such as ineligibility.

A major difference in the findings of the two pilots is the ratio of the percentage of cases with errors and the percentage of cases with errors that resulted in an improperly authorized payment. Of the four States in the first pilot, only Arkansas found all cases with errors to have an improperly authorized payment, whereas in the second pilot, three of the five States–Florida, Kansas, and New Jersey–found all cases with errors to have an improperly authorized payment. This finding should not be interpreted as an indication of the quality of the case work in a specific State, but rather as a product of the variances in the State eligibility policies. More importantly, it has implications for the methodology of conducting a consistent review process across States.

2. Most Common Errors

Exhibit 7 presents the most common errors within various eligibility elements, ordered from highest to lowest, for the five pilot States. Qualifying Care (Element 340), Income (Element 400), Payments (Element 420), Application Forms (Element 100) and Household Members (Element 320) were the five most common errors representing 91 percent of all errors. The remaining 9 percent (16) of the errors occurred in Other review elements, such as: Residency, Qualifying Child and Computations. All five pilot States experienced errors in Qualifying Care and Income. Three of the four States in the first pilot also had Income errors. Kansas had the highest frequency of errors in Qualifying Care, 43 of the 54 total errors. Florida had the highest frequency of errors in Payments, 22 of the 24 total errors. In comparing the frequency of the error elements, Kansas and Florida had the greatest frequency of Qualifying Care and Payment errors. Only Florida and New Jersey had errors in all five of the most common elements.

Exhibit 7. Frequency of Elements in Error


	Florida	Kansas	New Jersey	Oregon	West Virginia	Total
Qualifying Care (Element 340)	1	43	4	5	1	54 (32%)
Income (Element 400)	9	8	13	11	4	45 (27%)
Payments (Element 420)	22		2			24 (14%)
Application Forms (Element 100)	10		3		7	20 (12%)
Household Members (Element 320)	5		3		2	10 (6%)
Other	6	1	3		6	16 (9%)
Total	53	52	28	16	20	169

Missing documentation was the primary cause of the most common errors displayed in Exhibit 7. The States, in both pilots, differed in their coding or attribution of error types. Some States assigned all errors of missing documentation to Applications (Element 100). Others assigned the error to the element pertinent to the missing documentation, for example, assigning missing pay stubs to Income (Element 400). This is an important difference when comparing data between States.

Exhibit 8 presents the most costly errors in total dollars from highest to lowest amount. As with error frequency (Exhibit 7), Income and Qualifying Care were the two most costly errors representing 28 and 26 percent respectively of all improper authorized payments. Application Forms represent an additional 23 percent of the improper authorized payments. Household Members and Payments errors were 11 percent and 4 percent of the total respectively. The remaining 8 percent ($1,791) of the total improper payments occurred in other review elements such as Residency, Qualifying Child and Computations.

Exhibit 8. Dollar Amounts of Elements in Error


	Florida	Kansas	New Jersey	Oregon	West Virginia	Total
Income (Element 400)	$1,271	$342	$3,391	$1,082	$176	$6,262 (28%)
Qualifying Care (Element 340)	$30	$4,278	$470	$843	$63	$5,684 (26%)
Application Forms (Element 100)	$2,499		$1,638		$889	$5,026 (22%)
Household Members (Element 320)	$1,331		$434		$708	$2,473 (11%)
Payments (Element 420)	$733	0	$79		0	$812 (3%)
Other	$1,044	$30	$31		$686	$1,791 (8%)
Total	$6,908	$4,650	$6,043	$1,925	$2,522	$22,048

3. Interpretation of Missing Documentation Errors

A detailed review of the error findings in the area of missing documentation shows that States had similar findings with very dissimilar error results. Great latitude was given to the pilot States in defining errors. Some States interpreted missing or out-of-date verification as causing improperly authorized payments, while other States ruled the missing or out-of-date verification as a procedural error or no error. This inconsistency between reviewer’s interpretations also occurred within a State’s findings, despite States' efforts to achieve inter-rater reliability.

4. Coding of Missing Documentation Error Types

In addition to differing definitions of errors, the States in both pilots differed in their coding or attribution of error types. Some States assigned all errors of missing documentation to Applications (Element 100). Others assigned the error to the element pertinent to the missing documentation, for example, assigning missing pay stubs to Income (Element 400). This is an important difference when comparing data between States. Although the treatment of missing documentation varied between the States, missing documentation was the primary error cause and accounted for 28 percent of the total improperly authorized payments in the second pilot and 57 percent of the total improperly authorized payments in the first pilot.

5. Review Process and Missing Documentation Error

States determined whether to review records centrally or in a local office. A concern specific to centralized reviews is the shipment of case records. It is not uncommon for record material to be missed when a record is shipped, for example, failure to obtain retired material or recently received but un-filed documentation. However, the findings from the two pilots do not indicate that the decision to review the records centrally, rather than at the local office, was related to the frequency of missing documentation as a cause for errors. New Jersey reviewed records at the local CCR&R agencies and had as high a frequency of missing documentation errors as Florida, which reviewed records centrally. Ohio reviewed some records centrally and those from the larger jurisdictions, such as Cuyahoga County, on site. Cuyahoga County accounted for 30 of the 42 missing documentation errors.

As an error type, missing documentation is not affected by some corrective action strategies, such as exception reports, system edits, and data matches. Training to increase staff awareness of the problem, knowledge of policy, interviewing skills, and quality of routine case reviews are the most effective strategies States can use to prevent or reduce procedural or policy errors. The State responses to the error findings in both pilots include numerous initiatives to improve the quality of case work and the frequency of reviews.

6. Interpretation of Income Errors

Income errors were the only error type observed in all nine States. Income errors accounted for 28% of improperly authorized payments in the second pilot and was a major error prone area in all five pilot States. Similar percentages existed in the first pilot, where income accounted for 23 percent of the improperly authorized payments. Income error causes included: conversion, averaging, deductions, failure to include documented income, and inclusion of income, following job termination. The State responses to the error findings included initiatives targeting income verification and calculation policies.

7. Coding of Income Error Types

States in both pilots differed in their coding of income computation errors. Some reviewers assigned the error to Income (Element 400), while others assigned it to Payments/Computations (Element 430). Another inconsistency involved the connection between an Income error (Element 400) and the effect it has on the Rate/Fee schedule (Element 340). Some reviewers coded an income error and others assigned it to both review elements. Oregon’s responses to the error findings in the second pilot included an initiative regarding training of reviewers on error interpretation.

8. Interpretation of Hours of Care Errors

The calculation of hours of care was the third most costly error in the second pilot, accounting for 26 percent of the improper authorized payments. Unlike missing documentation and income, the significance of errors in the calculation of hours of care varied widely across the five States. West Virginia had one hour of care error that accounted for 2 percent ($63) of the improper authorized payments within the sampled cases. New Jersey also had a relatively small number of hours of care errors that accounted for only seven percent ($470) of the improper authorized payments. These findings contrast with Kansas which had 43 case errors accounting for 92 percent ($4,277) of the improper authorized payments. The hours of care error causes were most often the failure to allow travel/meals, hours that were inconsistent with work or school schedules, and the incorrect use of half time vs. full time. Based on previous child care case reviews, Kansas anticipated having cases with errors based on the fact that the sample month of October 2005 was the second month following conversion to the EBT system. Florida had the additional issue of the closure of centers due to hurricanes. As with missing documentation, the State responses to these error findings include training and case reviews.

9. Summary of State Responses

As a result of this pilot, each of the five States has planned action steps or has implemented several new systematic changes to improve monitoring and reduce improper payments. The State strategies include:

Strengthen supervision of new eligibility workers;
Clarify selected standards with eligibility workers;
Improve IT system elements to:

Prevent or decrease calculation errors,
Generate exception reports to highlight areas of potential problems or concern,
Implement automatic income calculation, and
Enhance the capability of extracting data from other data systems;

Conduct extensive technical assistance in counties to address error-prone areas;
Institute changes in the monitoring process;
Introduce statutory changes to simplify access to other State databases; and
Examine State policies to determine what changes may be necessary to provide more consistent application of policies and procedures.

Some of the State specific actions are:

West Virginia Technical Assistance staff will continue to perform random quarterly case audits. Prior to the pilot, the FACTS electronic record was the object of the review. An expanded review will now include the physical case record.
Oregon continues to conduct a series of “Accuracy Summits” in all areas of the State that focus on techniques to reduce errors in the Food Stamp program. Beginning with the next series of summits in July 2007, workshops to improve payment accuracy in child care will be included in the summits.
Oregon plans to modify the Food Stamp targeted review process and database that contains review information to ensure that income errors identified in the Food Stamp case are corrected in the companion Employment Related Day Care (ERDC) case.
Kansas continues to complete monthly Child Care QC reviews and uses the results of these reviews to inform training needs. Kansas is updating the Child Care Personnel Trainer and Training Academy to emphasize case documentation and computation of hours of care needed and income. Kansas supervisors now complete monthly case reviews for the Child Care Assistance Program.
Kansas contracted with a firm to develop software to track all case review findings and provide aggregate review data. The web based system tracks aggregate case review data for the mandatory Food Stamp reviews and the child care reviews. The child care reviews began in July 2006. The software cost was $75,000 and there is an 18 percent annual maintenance costs agreement.
Florida developed a desk reference tool for eligibility procedures, initiated focused monitoring and training on eligibility procedures on an annual basis and is providing each Early Learning Coalition (ELC) with its individual results of the pilot review process along with recommendations for corrective action.
New Jersey is implementing a process to conduct electronic matching of the automated child care client database with records on other wage, SSI, and child support information systems to identify and reduce the number of improper payments.
New Jersey plans to continue conducting child care record reviews utilizing the pilot monitoring tool, using a methodology that guarantees a statistically valid sample size of randomly selected cases, and will hire or allocate quality assurance staff needed.
Recognizing the limitations of its legacy automated system, KIDS (Key Information Delivery System), Arkansas designed and developed a new automated eligibility system called Keying in Day Care Accurately, Reliably, and Efficiently (KIDCare), to be fully operational as of July 2005. Arkansas designed KIDCare to determine eligibility based on program specific guidelines and has incorporated numerous edits to prevent inaccuracies from occurring on the front end of eligibility determination.
Illinois developed a resource guide for workers to outline acceptable forms of documentation or verification needed to determine eligibility accurately. The guide provides clarification for workers to use with clients during the eligibility process when the client cannot produce the required documentation. A worker can place a case in a pending status while awaiting necessary documentation.
As a direct result of the first pilot, Ohio began to retool its monitoring and technical assistance processes with the county agencies, implemented a quality control process for the child care program, and examined policy for possible revisions to strengthen and clarify procedures.
In response to the first pilot, Colorado planned to implement an automatic income calculation into Colorado’s automated Child Care Tracking System. State staff also conducted training and feedback with all participating counties to address deficiencies identified in the pilot. Staff members then shared the results of the error rate analysis at the State child care conference and changed statutory language to simplify access to other State databases such as those for new hires and unemployment compensation. Colorado also examined State policies to determine where changes may be necessary to provide more consistent application, particularly in the area of self-employment.

________

¹⁶ For both pilots, the payment refers to the amount authorized for payment. Return to text.

Cost Analysis >>

June, 2007