Those rushing to meet the 2013 date will now have time to perform a more in-depth review of their systems to identify and improve areas of weakness in their original transition plans. Even more important, these additional 12 months will allow time for testing in a pilot environment that would not have been possible under the original implementation date. Payers and providers will now have the opportunity to run test batches of claims using the ICD-10 code sets and identify and address potential glitches in their systems prior to implementation. Successful testing will greatly reduce the potential interruptions to a provider’s revenue stream and the costs associated with re-processing incorrectly coded claims—scenarios frequently associated with the original ICD-10 implementation schedule. Viewed from this perspective, CMS’s decision to delay implementation of ICD-10 may be seen as a gift for all concerned—not just the 25 percent of providers who said they would not be ready by October 1, 2013.

What changes are you making, if any, to make the most of the extended deadline?

The current health care proposals before Congress would make Medicaid available to all adults with incomes at or below 133 percent of the poverty line. That would increase Medicaid eligibility to an additional 16 million people, bringing the total eligibility to 62 million—almost 21 percent of the entire U.S. population, according to the Kaiser Commission on Medicaid.

With the federal government expected to pay 100 percent of the costs for the first two years, the increased burden on state resources will take its toll on state Medicaid systems, both financially and administratively.

Most states are already facing serious budget shortfalls. Because Medicaid accounts for a significant portion of most state budgets, states will likely reduce their Medicaid expenditures. Any decrease in state Medicaid expenditures, coupled with increased Medicaid enrollment, translates to reduced total Medicaid payments to providers and payers—not a healthy formula.

Seems like now is the perfect time to start leveraging program integrity efforts to offset the additional costs by becoming proactive at identifying fraud and stopping it before it has the chance to drain you of already limited funds. Moving away from the pay and chase paradigm was once almost unheard of, but with the technologies available today fraud prevention can be looked upon as a cost containment function.

Does anyone have any thoughts on how organizations can leverage fraud fighting efforts to subsidize the additional costs they are incurring in other areas?

If 3 to 10 percent of health care spending is lost to fraud, that means up to $270 billion is lost annually—of which only a small fraction is recovered. The vast difference between what is recovered (the“tip of iceberg”) and what remains undetected and uncollected (what lies “below the waterline”) is due to the inability of current claims and fraud systems to look beyond their own data.

The health care community needs to bring visibility to these hidden, “below the waterline” schemes and relationships. Imagine the power of having access to 50 terabytes of public data, most of which is not in your system, which allows you to leverage four billion derived links between people, businesses, assets and providers.

Using high-value fraud detection systems—Relationship Analytics, Predictive Modeling, Identity Management and other advanced solutions—health plans can now uncover connections and identities and start recovering the losses that “lie beneath the waterline.” Just like the iceberg, these are more than likely much greater than what is currently understood.

Analytics can now take a real bite out of crime, so let’s. I’ll talk more in upcoming blog posts on the value of this data. But I’m curious to know if any of my colleagues out there are leveraging public records and analytics together and, if you are, what kind of results has it brought to your organization?

According to an April 2000 article published in JAMA, 31 percent of providers surveyed admitted they sometimes exaggerate a patient’s symptoms, report symptoms the patient did not have, or change the diagnosis so the insurance company pays the claim. This attitude along with the mandated transition from ICD-9 to ICD 10 makes for a perfect storm, diagnostic codes will increase from 14,000 to 69,000 and procedure codes will increase from 3,800 to 72,000. The more complex the coding system, the greater the opportunity for fraud & abuse. The AMA believes that ICD-10 offers no direct benefit to patient’s care, but rather creates a burden for the provider with the cost, time and resources required to implement ICD-10.

Common practices of provider fraud include:

• Falsifying diagnosis
• Misrepresenting procedures
• Up-coding
• Waiving co-pays/deductibles and over paying
• Billing for services not performed
• False membership/identification

Most fraud today is detected by looking for non-conforming billing patterns in claims. After the transition to ICD-10, almost everything will be non-conforming; payers will need to consider the following:

• How will ICD-10 affect the way we view FW&A?
• How can we build FW&A detection into ICD-10?
• How can we use ICD-10 to assist in finding FW&A?

We’ll be at HIMSS will you? If so stop by one of our sessions on this very topic or booth #5225.  We look forward to seeing you there!

Recently, we were asked to develop a social network analytics model from a list that contained only 509 Medicaid beneficiaries and their addresses. All that we were told to start the investigation was that it was likely they had assets that made it look like they shouldn’t be on Medicaid. What our model found was well beyond what a normal beneficiary profile would look like; more importantly, it “uncovered” additional remora–seemingly unrelated but interesting and potentially predictive activity–such as extremely active mortgage turnover, which can only be discovered by leveraging data sets outside the one we were given.

This will hold true whenever supposed social network analytics is done without leveraging external data.   That approach will only “net” you a little different look at what you already know, not reveal the collusion you can’t see.  Much of the fraud, waste and abuse that plagues health care is the result of organized, collusive activity among providers and between providers and patients. Not every player will be a member or network provider and therefore it might be impossible for you to really understand what is happening with your money. In the above example, for instance, it appeared people outside of the client’s data were driving the collusion.

Tracy Kitten, managing editor of Bank Info Security, writes that 111 were arrested in a $13M ID fraud scheme in what amounted to the biggest ID theft bust in history. It would be fascinating to have a look at those 111 with good SNA analytics and see where the “remora” lead—how much more we would “net” and what “sharks” are still too deep to see.

Social network analytics can bring visibility to these hidden relationships. Imagine building a graph (more on graphs in later posts), leveraging not just your data, but 50 terabytes of public data, 250 million disambiguated and linked unique identities, one billion unique business contacts and more than four billion derived associations—that’s the framework of powerful, cost-saving social network analytics modeling.

A USA Today article discusses the fact that health care fraud prosecutions are on pace to rise 85% over last year. The article gives partial credit for this increase in health care fraud prosecutions to ramped-up enforcement efforts under the Obama administration. The article also includes statistics from the non-partisan Transactional Records Access Clearinghouse (TRAC), which show a 24% increase in prosecutions over the total for all of fiscal year 2010. There have been 903 prosecutions so far this year, compared with last year when 731 people were prosecuted for health care fraud through federal agencies across the country. TRAC has also found that prosecutions have gone up 71% from five years ago.

Another article, from Healthcare Payer News, details recent fraud charges that were filed against 91 individuals for $295M in false Medicare billing. The story references “our highly coordinated, nationwide Strike Force operations” that are working aggressively to combat Medicare fraud. The article quotes Attorney General Eric Holder as saying that, “…our anti-healthcare fraud efforts have never been more innovative, collaborative, aggressive – or effective. We will continue to work with our law enforcement partners and partners across government to fight against healthcare fraud.”

One of the tools that is enabling better identification, and later on prosecution, of health care fraud is a new technology called social network analytics. Social network analytics can help identify relationships and interactions within clusters of individuals, including:

• Patient relationships with known perpetrators of healthcare fraud;

• Links between recipients, businesses, and assets, as well as relatives and associates;
• Links between licensed and non-licensed providers; and
• Inappropriate relationships between employees, suppliers, and partners with patients and providers.

Social network analytics leverages massive data sets, powerful computing and advanced association algorithms, to reveal organized activity, even when those engaged it in it are trying to stay hidden.  The technology couples vast public records databases that go well beyond phone and address and other “credit header” type information with advanced analytics to expose connections between entities, assets, and people that might otherwise go unnoticed. At its core, social network analytics uses algorithms that aggregate linkages into high-value clusters of interest, illustrated by graph analysis. Examples of graph analyses include maps of transportation systems or disease epidemics, or of connections that the web itself makes among subject matter.

While health care fraud remains a serious problem, and will likely never be completely eradicated, I believe that every identification and prosecution of fraudulent activity moves us one step in the right direction. Do you believe that we are starting to turn the tide on health care fraud, waste and abuse?

Clearly current provider screening and post-payment fraud detection models don’t pass the test. Real Identity management leveraging and linking financial and derogatory data is a must. Using public data is absolutely critical, but, what is really needed is a “Pre-Pay” and Identity Management, or Enterprise Fraud Control, approach.

In another article, George Tubin is quoted as saying, “Like most employee fraud, using basic monitoring software could have identified the transaction anomalies well before the fraud reached this magnitude,” For health care providers, this can be accomplished with proper employee, member and provider screening, real-time claims review, and comprehensive identity access management.

Seven people will spend a considerable amount of time behind bars and over $30M was lost due to the “insider” fraud outlined in these two pieces. This doesn’t address the damage done to the respective company’s reputations. How would these breaches affect our sought after “trusted environment” for health care data exchange? Until we have a holistic and transparent approach to screening, claims processing and identity management, fraudsters will be able to exploit the health care system and drive cost up, while impacting the quality of care.

Also read our article about the importance of the use of technology for identity management.

The two articles:
“Prison Sentence for Insider Crimes”: http://www.bankinfosecurity.com/articles.php?art_id=4058
“Former Citi Insider Pleads Guilty”: http://www.bankinfosecurity.com/articles.php?art_id=4024

Spending on health care is continuing to increase and unfortunately so is the amount of money lost to health care fraud, waste and abuse. While the Centers for Medicare and Medicaid Services (CMS) estimates the United States spent $2.5 trillion (17.6 percent of GDP) on health care in 2009, the National Health Care Anti-fraud Association (NHCAA) conservatively estimates that 3 percent of all health care spending, or $60 billion, is lost to fraud. As part of the fraud that plagues our health care system, cases of medical identity theft are increasing as well, with more than 1.5 million people victimized by medical identity theft so far in 2011.

The one thing most industry players can agree on is that technology and advances in Health IT have the potential to make a big impact in reducing health care fraud and abuse. One of the biggest opportunities for both government and private payers is to integrate fraud risk controls at the front end of their claims payment workflow processes. Identity management, data and analytics and predictive modeling are some of the various tools that health care payers should examine.

While talking about the problem is only the first step, my hope is that with discussion comes knowledge and power. In order to jump start the conversation, I recently wrote two articles, one published in Healthcare IT News and the other in Healthcare Finance News, which examine the problem of health care fraud and abuse as well as the solution of pre-pay fraud control. I hope you will make the time to read these articles and to comment on this blog or engage with us via our Twitter handle, @LexisHealthCare. Also, check out some of the latest health care fraud and abuse stats in one of our recent blog posts.

1949 = ICD – 6 introduced in 1949
1976 = ICD – 9 introduced in 1976
10/1/2013 = ICD -10 to be implemented by 10/1/2013
14k–69k = Diagnostic codes will increase from 14,000 to 69,000
3.8k–72k = Procedure codes will increase from 3,800 to 72,000
GEMS = General equivalency mapping
$5.5 B–$13.5B = Estimated cost of ICD-10 transformation
1.6% = Only 1.6% of procedure codes have a 1:1 match from 9 to 10

So now you know some of the basic numbers and terms related to the transition of ICD-10. The real question is, what are the right questions to be asking and how do we deal with those questions? How do we leverage this vastly increased complexity to improve care and lower administrative health care costs? Next week we’ll discuss, or at least open the discussion on, how ICD-10 will impact health care fraud detection.

What do these cryptic numbers and letters mean to you? If you’re doing your homework, they mean we all have lots to do before 10/1/2013. We’ll be talking about what these numbers mean quite a bit in the coming months, and years for that matter, since they will affect every aspect of health care. Answers next week along with how they relate to health care fraud detection.