R & R Analytics

U.S. Health Insurers Analytic Solutions

Experience: R & R Analytics has over eight years of experience creating end-to-end analytics solutions (data mining, outlier detection, and predictive modeling) for healthcare payers. This experience includes data from Medicare (New York, New Jersey, and Florida), Blue Cross and Blue Shield plans (Michigan, Florida, Ohio, Texas, and New York), and commercial plans (WellPoint, Aetna, and CIGNA). Its solutions have recovered over $600 million and increased profitability up to 20% in HCC Reconciliations. Our solutions are compatible with SAS Enterprise Data Miner, Microsoft SQL 2005 and 2008, Oracle Analytics (OBIEE), Hyperion, Cognos, SPSS Clementine, or Business Objects.

Business Issues

1. Detection of outliers in the data

1. Medical Audit/Fraud and Abuse - 5% to 10% of the $170 trillion in yearly healthcare expenditures is determined to be fraud and abuse. This represents a total of $85 billion to $170 billion yearly that is wrongly paid in the US. On the federal level, the United States Government Accountability Office (GAO) estimates that $1 of every $10 spent on Medicare is lost to fraud and abuse, and that in 2006 alone, Medicare lost nearly $40 billion to fraudulent or unnecessary claims.¹ On the private payer side, industry experts estimate that less than 1% of reimbursement lost to fraud and abuse is ever recovered. Yet a survey conducted by the Health Insurance Association of America (HIAA) in 1999 found that health insurers save $11 for every $1 they spend fighting fraud, an average of $5.5 million per company in 1998. Meanwhile, a 2003 HIAA survey found that 2.2% of health care claims that had a claims-pending status were “pended” because of potential fraud.² In other words, when it comes to fraud and abuse, payers are just “scratching the surface” and missing a potential gold mine in savings.

   1. Medicare Part D - There were over $27 billion in payments to private prescription drug plan sponsors and between $1.35 billion to $2.7 billion in potential fraud. Medicare reports that almost 100% of all fraud and abuse compliance is reactive (responding to complaints), and that there are no proactive programs. Federal law requires that prescription drug sponsors have comprehensive fraud and abuse programs.

2. Claims Processing - The IDC estimates that there are approximately five billion claims processed every year in the U.S., at a cost (excluding payments) of $6 billion.³ The processing of healthcare claims is done by payers through a series of edits into the individual payer system that are burdensome and complicated. The reason for this complexity is that there are multiple providers and networks that require the payer to meet different contractual obligations. Contractual simplicity is something that has not been achieved in the healthcare space.

Inaccuracy in claims processing has created an environment of lack of trust between providers and payers. The consequence of this environment has resulted in class-action lawsuits by providers and state regulators against payers. The settlements of these lawsuits cost payers hundreds of millions of dollars to implement claims processing systems that reduce the error rate in claims processing. The issue, then, is what steps can payers reasonably and efficiently take to insure the accurate payment of claims.

As technology and science have expanded over the last ten years, they have given us the ability to expedite the processing of a large amount of claims and to discover trends in the data that increase the return of investment (ROI) by reducing the time that it takes to identify and correct errors during the claims process cycle. A metric commonly used by payers to determine the efficiency of their claims processing system is the average number of claims reviewed by an analyst (or representative) per day. The purpose of the R&R Claims Processing Analytics Solution is to increase the number of claims reviewed by an analyst per day by grouping potentially incorrectly paid claims by similarities. In other words, R&R Analytics increases the number of perspectives by which a claim can be analyzed through the claims process.

2. HCC Reconciliation

1. Medicare Advantage Program - Medicare Advantage programs expenditures were $75 billion in 2007. Chronic disease expenditures accounts for $56 billion a year in the privately-sponsored Medicare Advantage programs. Approximately $2.5 billion is estimated to be inaccurately paid to payers by the federal government. These underpayments are negatively impacting the revenue stream of private payers that have Medicare Advantage programs. The reason for those underpayments and missing revenue is due to providers supplying wrong, inaccurate, or missing diagnoses (WIM) to payers. The reimbursement method used for Medicare Advantage depends on accurate and complete diagnoses reported by providers; hence, the need for payers to find wrong, inaccurate, or missing diagnoses that can affect their revenue stream.

Solutions

1. Outliers Detection Analytics – Data Mining for Gold - The commercial lesson of mining for precious metals and stones is that the return of investment (ROI) is far greater doing underground mining than surface mining. The same concept applies in detecting outliers in the data to determine in which areas/providers to conduct medical audits, or to open investigations for potential fraud and abuse.

The R&R Outlier Detection Analytics is a provider centric comprehensive solution that allows organizations to safely, economically, and with as little waste as possible identify patterns of outliers in the data. It uses R&R knowledge and resources in infrastructure, business intelligence, and data mining technologies to allow organizations to “dig deep” for gold. The analysis of outliers is used in the detection of healthcare insurance fraud.⁴ An outlier is an observation that lies outside the overall pattern of a distribution in the data. Usually, the presence of an outlier indicates some sort of problem.⁵ Nevertheless, an outlier, in and of itself, is not an indication of potential abuse. For the outlier to be considered representative of abuse activity, the variable must take into consideration the purpose and means used by those committing the abuse.

A significant outlier variable analyzes the relationship between the purpose and means used in a potentially fraudulent scheme to create meaningful variables. The supposition of this formula is to create variables that maximize the differences in independent variables, while simultaneously minimizing the similarities in the data to detect outliers that are indicators of potential fraud and abuse. This effect could be described as data mining for the purpose of “squeezing and pulling out” the potential outliers from the data set.⁶

The R&R Outlier Detection Analytics solution uses state-of-the-art statistical methodologies such as regression, classification, correlation, and cluster analyses to create a powerful engine to identify potential outlier claims and abuse patterns. Our solution is vendor-neutral and allows organizations to apply the technology that best fits the job.

 

2. HCC Reconciliation Analytics – This predictive modeling solution is a patient centric forecasting data mining model that allows payers to make decisions through the evidence-based medicine approach, or comparative effectiveness. It produces a probability score for each Medicare beneficiary to determine whether the HCC for a patient is wrong, incomplete, or missing by predicting diabetes, cardiovascular, or mental health diagnoses.

The HCC reconciliation analytics model acts as a magnetic field by bringing together the similarities in the claims data, measuring the strength of those similarities, and separating the differences. The magnetic field represents the unique variables created by our analytics data mining process that simultaneously bring the similarities together and separates the differences in the data.

The R&R predictive modeling solution can also be utilized in comparative effectiveness studies, as well as in reconciliation of wrong, incomplete, or missing diagnosis in Medicare Advantage programs. Comparative effectiveness refers to a “rigorous evaluation of the impact of different options that are available for treating a given medical condition for a particular set of patients. Such a study may compare similar treatments, such as competing drugs, or it may analyze very different approaches, such as surgery and drug therapy. The analysis may focus on the relative medical benefits and risks of each option only, or it may weigh both the costs and the benefits of those options.”⁷

Return of Investment (ROI)

R&R outlier detection methodology has recovered over $600 million for payers in the areas of: infusion drugs, sleep studies, cardiovascular procedures, blood agents, gastrointestinal procedures, physical therapy, orthopedic procedures, durable medical equipment, and radiology.

Our HCC reconciliation methodology has increased the revenue stream of payers up to 20% in Medicare Advantage programs in the areas of diabetes, cardiovascular disease, and mental health.

Our Payer Analytics Model includes web-based portals for payer staff, members, providers, and employers. It also includes over 300 reports using basic, complex, and advanced analytics.

R&R business model reduces costs of analytics implementation by payers because we have accelerators and by having the flexibility of using payers’ IT resources. Also we can use offshore resources that do up to 70% of the development using HIPAA compliance data that has been blinded to comply with the Standards for Privacy of Individually Identifiable Health Information (Privacy Rule) and do not contain any protected health information.⁸

We can provide our analytic solutions at a fixed price and have actionable deliverables within 2-3 months.

Healthcare Payer Analytics

High-Level Solution Architecture

R & R – Rates

Geographical Area	Rate*	Expenses*
Continental U.S. (includes Alaska, Hawaii, Puerto Rico, and the US Virgin Islands)	$139 per hour	Plus actual expenses
Continental U.S. all inclusive (includes Alaska, Hawaii, Puerto Rico, and the US Virgin Islands)	$169 per hour	Included in hourly rate
International	$250 per hour	Plus actual expenses
International all inclusive	$350 per hour	Included in hourly rate
Dashboard or Scorecard Project Fixed Cost – up to 3 Months9	$150,000 to $250,000

* Payable seven (7) days upon receipt of invoice. The client agrees to a simple interest rate of 10.5% each month on any outstanding balance at the end of the month.