This document discusses using predictive analytics to detect, predict, and change behavior related to fraud, waste and abuse (F/W/A) in healthcare. It defines F/W/A and provides statistics on the large financial impact of F/W/A. A case study shows that a predictive modeling approach identified over $8 million in savings opportunities from a large Medicare program. The document recommends that healthcare organizations re-examine their current F/W/A practices, question if they are sufficiently reducing costs, and work proactively using predictive analytics to prevent losses from F/W/A.

1. Fraud, Waste & Abuse:
Capture low hanging fruit and more…
Using Analytics to Detect, Predict and Change Behavior
ACAP Webinar
May 19, 2016

2. Leon Edelsack
President
TPG Data Services
Craig Stern, RPh, PharmD, MBA
Senior Consultant
ProData Analytics
Marjorie Zimmerman MS, BS Pharm
Senior Consultant
The Pharmacy Group
.
2

3. Introduction to
TPG Data Services (TPG-DS) has the knowledge, expertise and resources
to provide your organization with the tools necessary to efficiently
manage your prescription drug program and potential FWA issues.
Our team helps clients:
• Manage the financial performance of healthcare transactions to help
payors
• Continually monitor the marketplace, arming us with up-to-date
business intelligence to deliver unbiased information tailored to
individual client needs
• Mitigate F/W/A via an innovative and multidisciplinary approach
necessary to effectively combat losses and change behavior
3

4. Webinar Overview
• Defining F/W/A
• F/W/A Statistics
• Market Challenges
• Case Study
• Predictive Analytics
• Solutions
• Key Takeaways
4

5. Defining F/W/A
Fraud: “… an intentional deception or misrepresentation made by a person
with the knowledge that the deception could result in some unauthorized
benefit to himself or some other person. It includes any act that constitutes
fraud under applicable Federal or State law.”
Waste: Not explicitly defined by rules but “generally understood to
encompass over- utilization, underutilization or misuse of resources, and
typically is not a criminal or intentional act.”
Abuse: “... provider practices that are inconsistent with sound fiscal,
business, or medical practices, and result in an unnecessary cost to the
Medicaid program, or in reimbursement for services that are not medically
necessary or that fail to meet professionally recognized standards for health
care. It also includes beneficiary practices that result in unnecessary cost to
the Medicaid program.”
Definitions, 42 C.F.R. § 455.2 5

6. Types of Fraud, Waste and Abuse
• Overutilization
• Billing for unnecessary services or items
• Billing for services or items not rendered
• Upcoding
• Unbundling
• Billing for non-covered services or items
• Beneficiary fraud
6

7. F/W/A By the Numbers
• Every hour, the United States loses $28.5
million to healthcare F/W/A.
• That is over ten times the amount an average
American will earn in their lifetime.
• Totaling as much as $234 billion annually,
F/W/A is becoming a growing problem which
is too large to overlook.
7

8. Market Challenges
• Growth in F/W/A driving up healthcare costs
and increasing risk for plans and members
• Absence of real-time approach to identify
fraud before it happens
• Fraud management processes lack
sophistication and preventative measures
• Outdated technology systems
• Lack of technical knowledge at Plans to know
how to assess the vast amounts of data
8

9. F/W/A Players and Flow of Information
• By combining point of sale, patient and prescriber information, you gain a
total situational view
• Resulting in efficient identification and timely resolution of problems
9

10. Multidisciplinary Approach
• Assessment of the nexus of activity between
the member/patient, the prescriber and the
pharmacy
• Evaluation of transactions and activity among
those three entities
• Map the relationship between these different
constituents
10

11. F/W/A Program Elements
Capture low-hanging fruit and more..
• Protect
• Predict
• Monitor
• Change Behavior
Impacting the cycle of behavior is critical to the
success of the program…
11

12. Predictive Analytics
• Predictive analytics can identify patterns of
behavior that indicate areas of potential F/W/A
• Technology by itself will not change behavior
• By proactively developing and implementing
targeted communications programs to reach out
and suggest corrective action we have seen
results
12

13. Preventing Payments
• Ways we work to identify a F/W/A profile as
soon as the transaction happens
• Leveraging predictive tools to prevent
payment
• Aligning with the current direction of CMS –
to prevent wrongful payments before they
happen
13

14. Integrating Analytics &
Behavior Modification
• Specific analytics
• Technical & Clinical support
• Results help change the behavior of the
doctor or a pharmacy, and potentially even
members
14

15. Case Study – Industry Example
• Large Medicare program
• Concerns over PBM detection and internal
Strategic Investigation Unit (SIU) focus
• Focus on independent pharmacies
• Budget year impact net savings >$8M
15

16. Predictive Model Considerations
16
• Analytical flags cover financial, utilization,
geographic, severity, day/time, risk flags
• Input based on financial threshold/provider
• Analyze prescribers, pharmacies, patients – roll up
results to one or more, contact points
• Output based on:
• Severity, geography, provider type
• Flag findings weighted by frequency and probability of
risk
• Contact providers selected by threshold of risk

17. Predictive Modeling Output
Downtown Drugs has been identified in Fraud-Waste-Abuse audits as an outlier with more than
75% above multiple FWA metrics. Downtown Drugs was compared to other Independent
pharmacies with similar severity populations so that results indicate significant above average
variance.
State and Federal agencies have increased their vigilance on Healthcare Fraud, Waste, and Abuse
(FWA). This has prompted implementation of FWA Alarm Trigger Programs. The prescription(s)
below have initiated a FWA Excess Alarm Trigger for excess quantity (Qty), an excess daily dose
(OD), an excess dollar amount (Amt), excess day supply (OS), etc. This Alarm notice serves to
identify patients and medications that trigger an FWA Audit.
Recommendation: In addition to high risk from multiple edits, the claims below indicate paid
amounts that vary significantly over comparable claims filled by network pharmacies. These issues
require reconciliation and corrective action plans.
Note to Physician Providers: Please review the patient's chart and/ or medical condition and
consider the necessity for the trigger entity. In many cases, doses, quantities, and/or medication
can be adjusted.
This report relies on the validity of
the Medical and Pharmacy claims
submitted by Clients and PBMs. 17

18. Predictive Modeling Output
18
Name: DOWNTOWN DRUGS Type: INDEPENDENT Severity: Mid Severity Pharmacy ID: 2318663

19. Summary Potential Financial Impact(s)
• F/W/A metrics impact 14% to 25% of total paid
• F/W/A financial impact ranges from $1.5M - $6M
for pharmacy only
19

20. Segmented Potential Financial Impact
20

21. Lessons Learned
• Capture the low hanging fruit and more…
• Validated and surpassed visibility of SIU and PBM
• Fraud is legal issue and resource intensive
• Indicated greater than 4X ROI
• Demonstrated Waste & Abuse as significant as
Fraud
• Changing behavior takes time
21

22. Our Recommendations & Solutions
• Considerations include:
• Re-examine current internal practices for F/W/A
• Question whether you are sufficiently bending
the cost curve
• Work proactively, not reactively to mitigate
losses...
22

23. Key Takeaways
• Focus on prevention (CMS emphasis)
• Identifying and managing Waste & Abuse
should be high priority
• Bending the cost curve takes more than
technology
• Achieving high rate of return will take time
23

24. Questions?
24

25. THANK YOU
For additional information, please contact:
Leon Edelsack
President, TPG Data Services
leon.edelsack@tpg-ds.com
(c) 412-720-8955
www.tpg-ds.com
25