Dr. Vishal James Makker could be a medical miracle worker, saving Medicare patients from perpetual back pain, or he could be a fraudster, bent on bilking Medicare of hundreds of thousands of dollars.
The Portland, Ore., spinal surgeon’s higher-than-normal rate of spinal fusions jumped off the page when The Wall Street Journal conducted a survey of a Medicare database. Analysis of the data revealed he performed more spinal fusions on Medicare patients than any other surgeon in the country performing the same surgery on 20 or more Medicare patients in 2008 and 2009.
Makker said his higher rates were because he took difficult cases other doctors declined due to the patient’s Medicare status. The Oregon Medical Board disagreed, and in 2006 levied a complaint and notice of proposed disciplinary action, claiming many of the spinal fusions weren’t medically necessary.
Miracle worker or fraudster, Makker exemplifies the many issues the Centers for Medicare and Medicaid Services confront in its efforts to stem improper payments and wasteful spending. He’s also a perfect example of how predictive analytic tools are helping CMS stop the flow of dollars lost to waste, fraud and abuse.
Politicians Take Aim at CMS Spending
As the debate over the 2012 budget heats up, CMS finds itself in the crosshairs. Budget proposals across party lines suggest major changes to the system are
necessary to control waste, fraud and abuse.
Rep. Paul Ryan’s (R-Wisc.) “Path to Prosperity” proposal would effectively privatize Medicare and allow states to tailor Medicaid block grants to fit individual state needs. The Obama administration’s proposal builds on the healthcare reform law, requiring state Medicaid agencies to track spending and analyze data for patterns of waste, fraud and abuse. The White House proposal also sets aside $581 million in discretionary funding for program integrity to reduce payment errors and better equip CMS to pursue criminal and civil charges when fraud is discovered.
Often, the source of CMS’ monetary bleeding is found in the agency itself. Its size alone makes it a target for fraud, and waste is almost a foregone conclusion.
The Government Accountability Office gave testimony in March 2011 before the House Subcommittee on Oversight and Investigations, Committee on Energy and Commerce on Medicare’s high-risk status. Confirming the internal bleeding of money, the GAO study found Medicare to be a high risk for improper payments because of the agency’s size and complexity.
The report estimated Medicare paid out a stunning $48 billion in improper payments in 2010 alone. With figures like that, it’s no surprise political parties, grassroots advocacy groups, industry experts, medical professionals and CMS itself can agree the system is unsustainable and change is necessary even if they can’t agree on how and what to fix.
CMS Overwhelmed?
The general consensus it that CMS has to stop the waste, fraud and abuse, but each element offers unique challenges. Sorting out which to tackle first is a challenge in itself — in the fight against fraud, the number of scams is overwhelming “The primary challenge in preventing and detecting healthcare fraud is sheer volume,” said Kristine Martin Anderson, senior vice president at Booz Allen Hamilton.
“There are so many schemes operating at a relatively low dollar value that it can exhaust all of CMS’ resources to detect and prevent these types of fraud, and in many instances the resulting criminal cases fall below minimum federal dollar thresholds to prosecute.”
While tackling fraud is sexy and politically expedient, waste and abuse are the real killers, experts say. Dr. Andy Friede, chief medical officer for SRA, likened CMS’ challenges to an iceberg.
“The part you’re seeing and the part that gets the most press is fraud,” he said. “Abuse is sort of right under the water, but waste is the biggie. That’s where the real action is, and that’s the biggest challenge that CMS faces.”
CMS traditionally follows a pay-and-chase model. Claims are submitted and providers are paid before any records are audited. Only then are discrepancies found, and the chase to recoup misspent dollars begins. The model is ineffective and expensive.
Data Holds Keys to Winning Battle
CMS is turning to health IT contractors for a better model. And as it turns out, the devil’s in the details. But for CMS, awash in data, that’s proving to be the agency’s best weapon.
“CMS needs to move to the point where it’s examining claims for fraudulent behavior before it pays them,” said Kerry Weems, senior vice president and general manager of health solutions for Vangent. He compared healthcare fraud with credit card fraud.
When credit card companies see purchase behavior that deviates from the norm, they cut off the card.
“The same kind of thing happens in the healthcare fraud context,” Weems said. “It can be seen, and it can be stopped before making the payment. It’s just a matter of being able to understand the behavior and predict it.”
Enter data analytics, a combination of modeling tools and data designed to predict patterns before fraud emerges.
“Predictive analytics is something that’s really become much more prevalent for a couple of reasons,” said Tom Romeo, senior executive of federal healthcare for Accenture.
“There’s more data available. There’s more capability to process large volumes of data and the knowledge in what to look for as far as patterns go, and how to predict potential improper payments has increased significantly.”
Romeo said the model is built around a set of “rules” against specific variables, which then reveal outliers. Running a model built to find standard payments for tonsillectomies, for example, might show 90 percent of physicians charge $2,000 for the procedure.
When a charge for $10,000 pops up, that should raise a red flag.
“At Accenture, we’ve built an analytic workbench that includes tools but allows you to build rules that would look at patterns in the data, call out the patterns
that look suspicious for review and use multiple tools,” Romeo said. “Each one of them builds their own pattern for use.”
Romeo said CMS’ approach so far has been to use multiple tools, strung together to cover as much as possible.
“Our workbench allows you to do that, to string the tools together, and also refresh and update the rules as you find new patterns,” he said.
Anderson places great stock in data analytics. She said she believes the modeling tools are a good replacement for the pay-and-chase model the agency
currently uses.
“IT solutions can support and strengthen CMS’ prospective approach to fraud, waste and abuse,” she said. “Those solutions should allow CMS and their stakeholders to make more data-driven, analytically defensible and timely decisions to help prevent fraud, waste and abuse.”
As CMS starts to dip its toe in the predictive analytic market, there’s already evidence the tools are paying off. According to the Obama administration, CMS recovered $4 billion in fraudulent payments made in 2010.
So far, the focus has been on detecting patterns in billing. CMS is also considering the use of predictive analytics to look directly at the service provider.
“When a service provider signs up, could we predict at that point maybe this will not be a good provider to have [or] potentially fraudulent?” Romeo said.
Analytics Improve Care
While he sees the value in fighting fraud, Friede wants CMS to go a step further.
From a healthcare perspective, he would like to see CMS use its dataanalytic models to combat not only upfront waste, fraud and abuse, but also on the back end with improved medical treatment.
“The real money is in understanding the data that CMS has and using the data to develop better guidelines to do more research in effective healthcare and to guide physicians and other healthcare providers in providing the best treatment at the lowest possible cost,” he said.
Friede’s suggestion is on par with the goals of the Obama administration. By emphasizing preventative care and more efficient medical care, many industry experts believe CMS would save a vast amount.
“In healthcare reform, they’re trying to move to a method that pays for outcomes, so to prevent you from being sick,” Romeo said. “To take a diabetic and stop the progression of the disease before it gets to the later stages that are more expensive to deal with.”
Analytic tools can help caregivers improve medical care, reduce costs and receive better compensation for better outcomes. Systems can notify caregivers of potential drug interactions, suggest alternative treatments and track patient history to avoid duplication of tests. They’re also credited with lowering rates of infections and re-admissions.
“The funny thing is that the best medical care compared to average or poor medical care costs half,” Friede said. “It’s not only more cost effective; it’s just
plain, old cheaper.”
CMS is acutely aware of its challenges. Hamstrung by an antiquated system and bloated bureaucracy, the agency is doing its best to wage the battle against waste, fraud and abuse. In 2009, the Department of Health and Human Services and the Justice Department created the interagency Health Care Fraud Prevention and Enforcement Action Team. HEAT teams fanned out across the country, reinforcing existing programs such as the Medicare Fraud Strike Force. The agency is also investing in new analytic models to detect and prevent fraud, waste and abuse, looking at ways to prevent them before they occur and restructuring to better position itself to be proactive rather than reactive.
“Our goal is to keep those individuals and companies that intend to defraud Medicare, Medicaid and Children’s Health Insurance Program out of these programs in the first place,” Deputy Administrator and Director of Program Integrity for CMS Peter Budetti told the Senate Finance Committee in March. “Not to pay fraudulent claims when they are submitted and to remove such individuals and companies from our programs if they do get in.”
In the case of Makker, the doctor accused of padding the bill, predictive analytics quickly zeroed in on his high rate of repeat surgeries. Medicare responded by reducing his compensation rate from the industry standard 21 percent to 11 percent, saving the agency more than $4 million over a three-year period.
“Application of predictive modeling and trend-analysis techniques support making accurate payments at the outset, a ‘getting it right the first time’ philosophy that will then permit a focused and more efficient application of resources on complex schemes,” Anderson said. “As schemes are detected and halted through these methods, increased effectiveness in spotting successive schemes is built through these upfront reviews.”
The Congressional Budget Office estimated Medicare and Medicaid will spend around $846 billion in 2011. With budget hawks circling and political winds
not blowing in its favor, CMS continues to advance its directive to provide efficient and effective healthcare to seniors and the needy. In answer to its critics and calls for reduced spending, CMS remains an opportunity for the government-contractor solutions that advance its directives.
