Testimony to House Energy and Commerce Committee …

Testimony to House Energy and Commerce Committee
              Subcommittee on Health

      "State Fiscal Relief: Protecting Health Coverage in an
                      Economic Downturn"

                       Chairman Frank Pallone (D-NJ)
                     Ranking Member Nathan Deal (R-GA)

                              By James R. Frogue
                             State Project Director
                        Center for Health Transformation

                               Tuesday, July 22, 2008



Chairman Pallone, Ranking Member Deal and Members of the Committee, thank you for

the opportunity to testify today. My oral and written remarks reflect solely my own views

and not necessarily those of the Center for Health Transformation, its staff or members.



This Committee is considering legislation that would send an additional $15 billion to the

states for Medicaid costs they have incurred. There is one simple action this Committee

could lead that would be low cost and go a very long way toward improving the care

received by 50 million people on Medicaid while eliminating much of the waste, fraud

and abuse that is largely responsible for states having chronic financial trouble with

Medicaid in the first place.
Legislation should be put forward by this Committee that would require states to post

their Medicaid patient encounter data on the Internet for all to see. Specifically, this is the

set of claims that Medicaid providers send to the state for reimbursement for treatment of

patients.



This is administratively simple, cheap, and would have a profoundly positive impact on

the quality of care delivered via Medicaid. In addition, it would dramatically increase

accountability for how Medicaid dollars are spent thereby decreasing the likelihood that

state leaders would return to seek still more money from Congress.



How many dollars the federal government sends to each state annually is a known

number. Each state's FMAP is a known number. Therefore some very simple arithmetic

gives policymakers and the taxpaying public the target figure for the sum total of

Medicaid claims, plus a reasonable amount for administrative overhead.



Of course it must be stated clearly and emphatically up front that this data should only be

made public in a patient de-identified way. Patient privacy is sacred. Fortunately there are

multiple safeguards that can and must be put in place to ensure that individual patient

names, or information that would identify an individual, are not revealed to unauthorized

persons or entities. Use of the right algorithms to scramble patient identities is routinely

successful in similar studies of large employer groups and other public programs like

Medicare.
States already collect Medicaid patient encounter data so uploading it to the Internet

would require minimal cost and effort. This incredibly rich data set would then be open to

policymakers, academics, clinicians and the widest possible range of people with

expertise in medicine, pricing practices, technology, accounting, fraud detection and a

vast array of other disciplines relevant to improving and modernizing this important

program. Call it, "Open Source Medicaid."



The data would lay bare to all whether or not Medicaid beneficiaries are getting

appropriate medical care. Among the many thousands of statistics revealed by patient

encounter data, for example, is what percentage of women over 50 are getting annual

mammograms. The figure should be 100 percent. In one state, the data revealed that only

17 percent of women on Medicaid in this age group were getting annual mammograms.

That exceedingly low figure, heretofore unknown to the public at large, means that these

women are at severe risk of undetected breast cancer. It also means that the overall cost

to taxpayers is likely to be much higher down the road because relatively low-cost

screenings today could eliminate the need for much higher-cost interventions in the

future.



The same state's claims data showed 4,000 people who had gotten six or more Oxycontin

prescriptions. Less than half of children received well child check ups. It even showed

one beneficiary who had visited the emergency room 405 times in a three year span. It

also appeared that the state was overpaying for the very expensive drug therapy this

individual was receiving, probably to the tune of hundreds of thousands of dollars.
Obviously this person was suffering unnecessarily by getting uncoordinated, haphazard

care, while costing the state millions of dollars unnecessarily.



In another claims review of a different state, a hospital was found billing Medicaid for

pneumonia treatments at a rate of 80 percent bacterial and 20 percent viral. In nature,

pneumonia tends to be 80 percent viral and 20 percent bacterial. So this study revealed

that either there was a highly unusual and worrisome outbreak of bacterial pneumonia or

there was fraud. In either situation, it is important for policymakers and the general public

to know immediately. It turned out that Medicaid reimbursed treatment for bacterial

pneumonia at a much higher rate in this state and this hospital had been engaged in fraud.



Claims data shows outliers, trends, adherence to evidence-based medicine, best practices,

disease patterns and outbreaks, and pricing, among many other key points. It is absolutely

theoretically impossible for any one state's Medicaid administration to do a better job

maximizing the value of this information than would the collective wisdom of everyone

else who may view it. Hence the need to put this information in the public domain to

leverage the potential of mass collaboration, a concept known as "wikinomics."



Medicare claims data has been given to select researchers and institutions for decades and

has yielded extremely valuable information about best practices while raising some red

flags about facilities that have much higher costs without corresponding better health

outcomes. The Dartmouth Health Atlas is just one good example. There are many others.

If Medicare claims data were available to the general public, anyone could study it and
the result would be exponentially more solutions for more effective and more efficient

care.



The idea to request that states release their Medicaid patient encounter data is consistent

with the transparency movement that is sweeping through government. Members of

Congress are familiar with the required transparency for campaign donations from the

Federal Elections Commission and your staff is certainly familiar with their salaries being

posted on Legistorm, as two examples.



The most conservative and most liberal United States Senators, Tom Coburn and Barack

Obama respectively, successfully pushed through the Coburn-Obama Transparency Act

in 2006 which requires the Office of Management and Budget to have a single web portal

where citizens can get information on the recipients of all federal funds including all

grants and contracts. This was an important first step. Future versions could have ever-

more granularity that would allow for real time tracking of dollars. Taxpayers have the

right to know how their money is being spent.



Medicaid also has a serious problem with fraud, waste and abuse. It is actually difficult to

know exactly the scope of the problem because data is so scarce, but examples and

vignettes we do get indicate very troublesome levels of misuse and inefficiency. The

people hurt the most by this are poor Americans who see their access to health care

services restricted or eliminated, providers who must deliver care at average
reimbursement rates that are well below even those in Medicare, and taxpayers who must

foot the excessive bill.



The Government Accountability Office has documented questionable Medicaid financing

schemes by states going well back into the 1980s. Please see the attached chart of

selected studies at the end of this testimony. The most recent report in May of 2008

requested by Senator Charles Grassley was entitled, "Medicaid: CMS Needs More

Information on The Billions of Dollars Spent on Supplemental Payments." That title

alone is cause for serious concern. Of cause for greater concern is that this fits a decades-

long pattern. There is far too little sunlight on how states spend Medicaid dollars, over

half of which are from the federal government. States posting their encounter data online

would be a major step toward rooting out intentional or unintentional misuse of money

meant to finance health care for poor Americans.



The New York Times ran a series of articles in July, 2005 that uncovered breathtaking

amounts of fraud and abuse in New York State's Medicaid program, which is the nation's

largest both in per capita and overall spending. Consider:



        James Mehmet the former inspector general estimated that up to 40 percent of all

        Medicaid claims are questionable.

        Michael Zegarelli, another former top official said the system, "almost begs

        people to steal."

        One Buffalo school official sent 4,434 kids to speech therapy in a single day.
       A single doctor in one year prescribed $11.5 million dollars of a drug intended for

       AIDS patients that was likely diverted to bodybuilders.

       One Brooklyn dentist billed for 991 claims in one day in 2003 and over $5 million

       that same year for services that were never performed (for contrast, there is not a

       McDonald's franchise anywhere on the planet that sold 991 phantom

       cheeseburgers or a Federal Express delivery truck that invented 991 packages)

       Of 400 million Medicaid claims paid in 2004, state investigators uncovered only

       37 cases of suspected fraud.



The horrific levels of fraud suggested by this New York Times series was confirmed by an

outside study of New York's Medicaid claims that was completed in 2006 and delivered

to a handful of officials in New York's health department in Albany. It found that a full

one-quarter of New York's Medicaid program cannot be explained. One-quarter of the

$44 billion spent on New York's Medicaid program in 2005 was $11 billion.



The Congressionally-created Medicaid Commission had its first meeting one week after

this New York Times series ran. One of the Commission's principle objectives was to

find $10 billion in scorable federal Medicaid savings over five years. They were literally

handed the answer to their 18 month quest by the New York Times on day one ­ that all

$10 billion could have been found in New York state fraud alone in a mere two years

(considering New York's 50 percent federal match rate). Instead the Commission

recommended a series of cuts that would have mostly impacted honest providers and

reduced access to care for Medicaid beneficiaries.
A model for what would happen if states posted their Medicaid claims is the Goldcorp

Challenge. In March of 2000 the CEO of a Canadian mining company named Rob

McEwan was frustrated by his geologists' inability to strike gold. He had recently

attended a conference and learned about Linus Torvalds who founded Linux, the open-

source software. Inspired, Mr. McEwan placed all of his geological data on-line and

announced a global contest with $500,000 in prize money. His in-house geologists were

appalled.



Goldcorp's data was downloaded 1,400 times in the next several weeks. It became clear

that those people who eventually sent in their contest entries spent combined time and

resources that were orders of magnitude beyond the $500,000 purse. The winners were

from a small Australian company, none of whom had ever even been to Canada.

Goldcorp ended up finding an astounding eight million ounces of gold and the company

quickly catapulted from a $100 million sleeper into a $9 billion juggernaut.



Medicaid's chronic financial problems are well known and guaranteed to continue

unabated absent real change. If Congress chooses to bailout states again as it did five

years ago then at the very least it should require states to prove that they are using

taxpayer dollars optimally. The best, easiest and cheapest way to do this is to require

states to post their Medicaid patient encounter data on the Internet for all to see. Congress

should require the same for SCHIP. State officials and providers with nothing to hide

should have no objection.
Again, thank you Chairman Pallone and Ranking Member Deal for the invitation to be

here today. I look forward to your questions.
                     Medicaid Financing Schemes Used to Inappropriately Generate
                   Federal Payments and Federal Actions to Address Them (partial list)
                                            Source: GAO
Financing Arrangement Description                          Action Taken
Excessive payments to state    States made excessive Medicaid                  In 1987, the Health Care Financing Administration
health facilities              payments to state-owned health facilities,      (HCFA) issued regulations that established payment
                               which subsequently returned these funds         limits specifically for inpatient and institutional facilities
                               to the state treasuries.                        operated by the state.
Provided taxes and donations   Revenues from provider-specific taxes           The Medicaid Voluntary Contribution and Provider-
                               on hospitals and other providers, and           Specific Tax Amendments of 1991 essentially barred
                               from provider `donations,' were matched         certain provider donations, placed a series of restrictions
                               with federal funds and paid to the              on provider taxes, and set other restrictions for state
                               providers. These providers could then           contributions.
                               return most of the federal payment to the
                               states.
Excessive disproportionate     DHS payments are meant to compensate            The Omnibus Budget Reconciliation Act of 1993 placed
share hospital (DSH)           those hospitals that care for a                 limits on which hospitals could receive DSH payments
payments                       disproportionate number of low-income           and capped both the amount of DSH payments states
                               patients. Unusually large DSH payments          could make and the amount individual hospitals could
                               were made to certain hospitals, which           receive.
                               then returned the bulk of the payment to
                               the state.
Excessive DSH payments to      A large share of DSH payments were              The Balanced Budget Act of 1997 limited the proportion
state mental hospitals         paid to state-operated psychiatric              of a state' DSH payments that can be paid to state
                               hospitals, where they were used to pay          psychiatric hospitals.
                               for services not covered by Medicaid or
                               were returned to state treasuries.
Upper payment limit (UPL)      In an effort to ensure that Medicaid            The Medicare, Medicaid, and SCHIP Benefits
for local government health    payments are reasonable, federal                Improvement and Protection Act of 2000 required
facilities                     regulations prohibit Medicaid from              HCFA to issue a final regulation that established a
                               paying more than a reasonable estimate          separate payment limit for each of several classes of
                               of the amount that would be paid under          local government health facilities. In 2002, CMS issued
                               Medicare payment principles for                 a regulation that further lowered the payment limit for
                               comparable services. This UPL applies           local public hospitals.
                               to payments aggregated across a class of
                               facilities and not for individual facilities.
                               As a result of the aggregate upper limit,
                               states were able to make large
                               supplemental payments to a few local
                               public health facilities, such as hospitals
                               and nursing homes. The local
                               government health facilities then
                               returned the bulk of the state and federal
                               payments to the states.
Federal financial              FFP is the funding mechanism used to            In May 2007, CMS enacted a rule that would place a
participation (FFP) rates      reimburse agencies with federal funds for       ceiling on payments to governmental healthcare
                               certain Medicaid activities. States would       providers, not to exceed the costs of services provided,
                               overpay governmental healthcare                 serving as an additional check for UPLs. The rule would
                               providers, above and beyond the costs of        also prohibit states from requiring non-governmental
                               services provided to Medicaid                   providers (e.g., non-profit hospitals) to return part of
                               beneficiaries.                                  their Medicaid payments to the State.