Evaluation of the PARIS Project

1
Evaluation of the PARIS Project

• Preliminary Findings

2
Background

• The PARIS Evaluation has three main objectives
• Assess the effectiveness of state implementation
  efforts to determine what factors facilitate or
  hinder the implementation of PARIS.
• Examine how states have implemented PARIS, and
  how data from the various PARIS matches are used
  to support program integrity efforts.
• Identify, where possible, consistent methods by
  which costs and benefits of the PARIS system can
  be calculated.

3
Methodology

• Study used a sample of 14 participating PARIS
  states and two non-participating governmental
  entities (one county and one state).
• Sample based on diversity of approach, use of
  data, and length of time participating in PARIS.
• Interviews were conducted with PARIS coordinator,
  technical staff, follow-up staff, financial
  managers, program (Medicaid, TANF, Food Stamp)
  staff, and others as needed.

4
Focus on the Uses of PARIS Data

• Closing cases reported as active when a
  participant has moved from one state to another,
  but has not reported the move (interstate match).
  
• Examination of potential dual participation for
  the purpose of recovering overpayments
  (interstate match).
• Verifying income from the VA and other Federal
  sources (Veterans and Federal matches).
• Coordination of benefits between Medicaid and
  other Federal insurance coverage (Veterans and
  Federal matches).

5
Limitations of this Presentation

• Discussion must be more general in nature, as
  final analysis and data quality control
  activities must still be completed.
• Report needs clearance from ACF.
• States will be identified in final report when
  there is a need to demonstrate or present a best
  practice.
• No individual respondents will be identified
  related to findings.

6
Key Organizational Findings

• Almost all states required approval of high-level
  state officials to participate in PARIS, and most
  reported that the approval process was not
  difficult.
• PARIS participation was sold to state officials
  primarily based on the benefits of the interstate
  match.
• For the most part, PARIS was viewed as an
  additional tool for the detection of improper
  payments, and not a program or project unto
  itself.
• States varied in where responsibility for PARIS
  was placed within the organization.

7
Advantages of Organizational Location

• PARIS was generally located in one of three
  areas technical services, audits and
  investigations, or general program
  administration. Factors contributing to
  decisions included
• Ability to mobilize resources.
• Contribution to the ongoing work of the agency.
• Easy integration.
• Motivated staff.

8
Organizational Issues

• Lack of span of control related to local
  operations.
• Limited scope of program areas may exclude other
  programs from participating (TANF and Food Stamps
  in different agency from Medicaid).
• Communication issues.

9
Findings Related to Staffing

• All states reported adequate staffing for
  creation and preparation of the PARIS file for
  submission.
• Most states reported inadequate staff for
  follow-up activities.
• Few states reported adequate staff resources to
  reconcile PARIS match follow-up on an ongoing
  basis.

10
Findings Related to Submission of PARIS Data

• No state reported any problems with the
  submission of PARIS data.
• Decisions related to frequency of PARIS match
  submission often related to ability to conduct
  follow-up.
• Technical requirements for submission of PARIS
  data were neither burdensome nor difficult.

11
Findings Related to Use of Data

• Interstate match is used by all.
• Veterans match is used by some, but mostly for
  verification of income. Few reported using it
  for coordination of medical benefits.
• Federal match is the least used and is viewed as
  the most complicated and difficult to work with.
• Filters are an effective tool to reduce burden.

12
Findings Related to Follow-up Interstate

• States mostly used this to close cases from
  unreported moves.
• Some states have developed effective fraud
  control systems.
• Lack of staff time to conduct necessary research.
• PARIS data often not tracked separately from
  other alerts, therefore do not know
  effectiveness of PARIS matches compared to other
  information sources.
• Communication with other states.
• Lack of strong recovery systems.

13
Findings Related to Follow-up Veterans

• Only two states have developed effective
  coordination of benefits systems, but other
  states are interested.
• Some states have developed effective tools for
  sorting Veterans data for use by local offices.
• Some states report difficulty understanding the
  data in the Veterans file, and therefore do not
  use it.
• Some data not used because of Food Stamp
  simplified reporting.

14
Findings Related to Follow-up Federal

• Few states reported using the federal file.
• Most states found the file complicated and hard
  to understand.

15
General Findings

• States would like more technical assistance to
  streamline follow-up, make reports more user
  friendly, and help improve efficiency of data
  use.
• All states think the interstate match is a
  valuable tool for reducing improper payments.
• Communication issues between states should be
  addressed and improved.
• More states would like to use the Veterans and
  Federal matches for coordination of benefits in
  the Medicaid program, but see this as a difficult
  system to implement.

16
Costs and Benefits Standardized Cost Model

• The study will recommend a standardized cost
  model that includes three key components
• Start-up costs For new PARIS states or those
  implementing PARIS within the last two years, the
  study recommends documentation of start-up costs
  to provide potential new states with information
  that may sell joining the PARIS family.
• Ongoing annual operational costs to prepare,
  submit, filter, and distribute the PARIS data.
• Cost of conducting follow-up activities.

17
Issues Related to Follow-up Costs

• Most states do not track follow-up costs to
  PARIS, thus a special study would be necessary to
  account for costs.
• Cost of follow-up would need to be tracked
  differently for the various uses of PARIS data.
• It is likely that time studies would be needed to
  create average cost per-case data. Many states
  already do these.
• Sample data could also be used to populate local
  follow-up cost models.
• Most states indicated that, given time and
  resources, data were available to populate cost
  models.

18
Categories for Inclusion as Follow-up Cost

• Cost of staff to conduct follow-up on interstate
  matches, including time spent working with other
  states, contacting clients, preparing and sending
  notifications, and closing cases.
• Cost of making adjustments to client case files
  and taking actions based upon unreported income
  identified through the Veterans and Federal
  files.
• Cost of developing and managing a coordination of
  benefits effort.
• Cost of processing recoveries and prosecuting
  fraud.

19
Benefits

• Benefit models should be structured around the
  four major uses of PARIS data
• Closing cases based upon unreported moves.
• Perusing fraud and overpayments.
• Savings created through Medicaid coordination of
  benefits.
• Reducing benefits as a result of improper income
  reporting.

20
Issues Related to Calculation of Benefits

• Food Stamp and TANF benefits are more easily
  calculated than Medicaid.
• Medicaid benefits that are capitation savings are
  straight forward and easy to calculate.
• Medicaid savings in fee-for-service are more
  complicated, and would be based on utilization
  rates and average cost per visit.
• Recoveries must balance obligations as compared
  to actual recoveries.
• Any population of benefit models will require
  special studies.

21
Issue Related to Time Calculation

• The issue of how long a time period over which
  savings should be calculated is complex and
  divisive.
• Medicaid savings from closing cases associated
  with unreported moves, or through coordination of
  benefits are the most straight forward.
• Savings from TANF and Food Stamps are more
  complicated using end of certification period
  vs. using a standard time block.
• The more complex the methods for populating the
  models, the less likely states will be willing to
  do so.

22
Issues Related to Benefit Calculation

• Study will recommend models to populate and
  methodologies for collecting data.
• Some flexibility will be needed with regard to
  methodologies e.g. sampling and averaging vs.
  actual.
• All methods will require states to develop a link
  between the PARIS match and the follow-up action.
• Standardized methodology will benefit federal
  officials in documenting PARIS success
• Strong cost-benefit data can help to support
  state efforts to obtain resources.

23
Final Thoughts

• PARIS is well thought of by states. They have a
  desire to use PARIS data in an effective and
  efficient manner.
• Most states face challenges in expanding the
  scope of PARIS within their state to make the
  most use of the data.
• Additional help would promote expanded scope.
• States are doing interesting things and can learn
  from one another to improve the use of PARIS data
• Addressing interstate communication challenges
  will be important for future success.