Random Sampling

1
Random Sampling Extrapolation

• NAMPI Conference
• Portland, Maine

2
Schedule of Extrapolated Audit Adjustment

• Paid Claim Universe 3,879
• Sample Size
  100
• Overpayments in Sample 18
• Underpayments in Sample 1
• Total Errors in Sample
  19
• Total Sample Error Dollars 1,389.48
• Avg. Error per Sample Claim 13.5948
• Extrapolated Error Amount 52,747.82

3
Schedule of Extrapolated Audit Adjustment

• ( Information not Given to Provider )
• Standard Deviation 39.701615
• Confidence Interval 7.680585
• Lower End
  22,947
• Midpoint
  52,748
• Upper End
  82,548
• Precision is 56.5 which is not great

4
Audit Process

• Random sample of 100 claims replaced using sample
  size chart
• Hospital audits use a random sample of 200 claims
• Pharmacy audits have two universes under 1,000
  per claim and over 1,000 per claim

5
Audit Process

• Consideration for stratifying an error if
  quantifiable
• Consideration for not extrapolating an error if
  the error is deemed to be non representative of
  the universe

6
Public Act 05-195

• Effective July 1, 2005, Public Act 05-195 placed
  restrictions on the use of extrapolation and
  other audit processes
• Thirty day notice
• Clerical errors shall not constitute a willful
  program violation

7
Public Act 05-195

• Extrapolations projections shall not be used
  unless
• There is a sustained or high level of payment
  error
• Documented educational intervention has failed
• Or, the value of claims in aggregate exceed
  150,000 per year

8
Public Act 05-195

• In addition to other minor requirements this Act
  provides a mechanism for providers to request a
  review of any specific item of aggrievement
  they have with the final audit report.