Effective Screening with Provider Risk Scores
The National Health Care Anti-Fraud Association (NHCAA) estimates that financial losses due to healthcare fraud are in the tens of billions of dollars each year. A small minority of dishonest health care providers commits majority of the fraud. Health care payers have stepped up the due diligence upfront while enrolling providers in their network.
Need for provider risk score
There is no common industry definition of what makes a provider potentially fraudulent or one with high risk. The Centers for Medicare & Medicaid Services (CMS) advocates “provider type” as the basis of assigning risk category. Classifying risk based on provider characteristics like provider type may not really indicate the “real risk” associated with the provider. In short, lack of what factors contribute to provider risk and a common scale of ascertaining the same makes provider risk determination a very subjective exercise today.
Health care organizations can take cues from the financial industry on how the lenders successfully used the concept of “credit score” to determine the likelihood of a debtor repaying the dues, based on the latter’s past borrowing behavior. Health care payers need to implement a similar concept of “provider risk score” which can indicate the likelihood of provider committing fraud, based primarily on past and current exceptions found against the provider. Exceptions can be of varying types and significance that include the exclusion lists such as OIG LEIE, SAM, etc. and criminal records such as felonies, sanctions, infractions and misdemeanors.
What makes a risk score effective
While the concept may sound interesting, it is extremely important that the risk score includes all the necessary elements to make it a true indicator of provider risk.
- Coverage of all the necessary provider exceptions and a few key provider characteristics which qualify as risk factors
- Quantification of score against various risk factors, appropriate aggregation and normalization on a common scale
- Generation of scores at underlying party level – direct and related, with relevant contribution to the overall score
- Flexibility to adjust the system computed risk score to reflect the outcome of manual investigation and impact of exceptions
How does the risk score help
Provider screening involves verification of a wide range of data attributes across multiple parties associated with the provider. Providing risk score carries several benefits:
- Captures the big picture and drives further verification of the provider. For example a low risk score would generally imply favorable screening outcome and in turn a faster processing time
- Provides a standard basis for interpreting provider risk across all types of Providers. Lack of composite score makes risk determination subjective, whereas risk score on a standard scale (say 0 to 1000) makes it easier to classify providers as high, medium and low risk.
- Ensures effective research and investigation by analysts, esp. when the analyst decision is contrary to the risk score based suggestion.
- Good indicator of changing risk profile of a provider. Provider risk score is not a one time measurement. With subsequent applications that a provider submits after initial enrollment, its risk profile can change and an increasing risk trend may trigger further investigation.
- Effective metric for reporting. Risk scores of provider applications can be classified by different data attributes to provide meaningful insights such as – which provider types or counties have higher risk scores, correlation between risk score and the screening decision, etc.
An accurate provider risk score can be a very simple yet powerful capability that can guide analysts in doing the right level of investigation and making an informed decision about the provider. At Digital Harbor, we have developed a comprehensive and accurate model for provider risk scoring. Risk scores generated by the system have increasingly become a standard for our customers to ascertain provider risk and proceed with resulting investigation.