Monday - 9/10/12 - 3:30
Vesta Kimble - Maryland - Deputy Executive Director
Description: Pre-review can prevent improper payments of SNAP and TANF and improper certification for Medicaid/CHIP. Maryland has created an easy-to-use, 5 minute pre-review system that has quadrupled the number of case reviewed vs. supervisory case reviews because it has a limited set of questions that trained peer pre-reviewers and supervisors answer after the eligibilty has been determined but before the benefits are issued.
Maryland's SNAP rates 7.68% in 10 to 3.37% in 12. Presently in sanction status for FY11. Ranked #2 in most improved for FY12. The SNAP caseload has doubled in last six years due to active outreach. 306,002 participants in '07 to 702,604 participants in '12
Meanwhile, staffing levels reduced by 12% in last 6 years. 1,841 FTE in '07 to 1,626 FTE in '12
Through QC reviewes, it was discovered that 2/3 of payment errors are related to staffing rushing through cases. FY11 Breakdown: 37% worker, 6% due to data entry, 7% wages/SSA not checked, 11% case not updated timely, 17% policy errors and 22% were customer errors.
Developed multiple strategies for reducing errors in FFY11 and FY12. 1) Mandatory on-line training and refresher quiz, 2) gold fork competitive awards for good performance with managers in attendance and 3) 12/11 Rushmore Consulting for QC made recommendations which were adapted. Trianing modules are visually pleasing and provide instantious results to the trainee. Called "Get your SNAP on"
Comparison to old process "Pay & Chase"
Who does the reviews - Pay & Chases - only supervisrs/PRE reviews - peers, clerical leads, teams
Volume - Pay & Chase - tied to quota per staff / PRE reviews -expectation linked to volume of applications approved and RDs/changes processed
Method - Pay &Chase - assumes reviewer knows their stuff / PRE reviews - asks question which require an answer based upon policy
Useful - Pay &Chase - personnel evaluations / PRE reviews - drill down to error details, can be used not only for performance evaluations but also for program performance/corrective action
Flexability - Pay &Chase - tied to legacy system which requires IS intervention for adjustments/updates / PRE - programmed in Excel - changes can be made overnight by non-IS staff
Mechanism for cleansing errors hiding in caseload.
Recruited PRE review workgroup to design comprised of field, program and QC. Work group began meeting in 2/11 and designed review questions between March and July 2011. 194 specific questions were developed by the group for SNAP, TANF, Medicaid. The questions are color coded by program on excel spreadsheet and goes through all elements of eligibility.
This was trimmed down to 19 mandatory questions for all SNAP reviews with others optional based on local office desires/clientele, etc. If some offices didn't expand the questions but trending patterns indicated they should, they were politely invited to expand.
The process was reviewed by management in 7/11 and was up by 10/11. It has an excel/visual basic front-end with SQL back end. It has a devoted SQL server.
Active Directory is on shared drives with controlled access rights.
Each office has individual folder - Generator v 2007 runs reports and selects samples of active, closures, etc. optional for offices to use.
Starts with blank review form. The review is posted with a specific title which contains details which identify case, results (labeled "efinal" if an error) Once a disposition is determined, the case "hops" automtically into folders such as "completed clean", "pending correction", "corrections updated", etc. The review doesn't count in overall results until the correction are completed on the case.
Data elements automatically load into SQL server from which reports are drawn.
Training occured in local offices in 9/11. A total of 326 reviewers were trained and PIRAMID liaisons were developed. The training lasted for 1/2 day.
(PowerPoint of training module displayed)
Main points of displayed trining module: 1) reviews are conducted on cases prior to posting cases for benefits, 2) demographics auto fills from eligibility system to begin the PIRAMID review, 3) reports of final reviews list error cases first, 4) automatically e-mail errors to offices for correction, 5)auto saves each step in process and 6) produces a top 10 list of errors identified.
The initial goal was relayed to complete as many reviews as possible before an application is finalized or immediately after a RD or interim change is processed. However, it was assigned that a minimum would be 10% of applications and 15% of RDs with efficiency measured by QC reviews and ME results, OIG audits and program appeal requests.
The results of these reviews are also used for personnel evaluations as well as policy updates and the development of training for the on-line modules.
They are starting to see shift in ME adverse findings. And as performance improves, changes can be made to the questions on the Excel spreadsheet for the specific program under review.
To date, a total of 121,919 cases have been reviewed in FFY12.
The lapsed time for reviews is under 10 minutes (11,327 in 2/12) with an average time per review of 5 minutes.
Improvement is also noted in reduced QC errors. 3,237 in FY11 to 2,078 in FY12 (but not all federal errors - some are informational processing errors such as an unsigned application)
The application allows for the tracking of the number of errors by specific questions posed on the spreadsheet and offices can pull up specifics in real time data.
Future enhancements include: 1) real time review log, 2) hyperlinks to program policy, 3) CARES 2.0 (eligibility) system incorporate the pre-review functionality into the business process thereby eliminating the PIRAMID application.
The spreadsheet contains a column with specific instructions on how to fix the error. This is what the eligibility worker follows in making case corrections.
Eligibility staff voiced concerns about QC seeing the PIRAMID spreadsheet and case narrations. The result was that they were trained on proper documentation/case narration. (detailing what was fixed on a case not what the identified problems were)
The spreadsheet filters each review template to match program and certification cycle under review.
No comments:
Post a Comment