The RAC Racket

In an effort to reduce the cost of healthcare, the Tax Relief and Health Care Act of 2006 mandated the nationwide proliferation of Recovery Audit Contractor (RAC) audits by the year 2010. Since the initial rumor of the RAC concept propagated, the program has been exceedingly effective at reducing Medicare payment to providers.

By definition, RAC auditors presume that inappropriate payments have already occurred — their contracted task is to identify and recover these payments. In return for the successful recovery of funds, the RACs are paid a contingency fee ranging between 9% and 12.5%. Not surprisingly, RAC auditors are primarily focused on big-ticket items, such as indications for cardiac rhythm management devices and the appropriateness of inpatient admissions.

Virtually all physicians practice evidence-based medicine; they make decisions based on the latest clinical trials and they dedicate sizable amounts of time and money to keep themselves up to date. Unfortunately, Medicare coverage policy evolves slowly and it frequently contains stipulations not addressed in clinical trials.

For example, on March 21, 2002, findings published in the New England Journal of Medicine trumpeted the early termination of the MADIT II clinical trial as the results had already surpassed the pre-specified efficacy boundary of 27%. Moss et al indicated that “the effect of defibrillator therapy on survival was similar in subgroup analyses stratified according to age, sex, ejection fraction, New York Heart Association class, and the QRS interval.”¹

This was big news on March 21, 2002. However, CMS did not cover any of the MADIT II patients until over a year and a half later, in October 2003.

From October 2003 until January 27, 2005 (15 additional months), CMS coverage excluded MADIT II patients with Class IV heart failure and patients with a QRS shorter than 120 milliseconds (those with a prior MI and an EF of 30% or less). These temporary exclusions were in spite of data analysis illustrating no differences among these subgroups.

While the speed at which CMS policy covered the MADIT II population might be described as glacial, the manner in which random qualifiers were used to discriminate against certain MADIT II patients was arguably reckless in my opinion.

Currently, the National Coverage Determination (NCD) for defibrillator implant requires that the non-ischemic dilated cardiomyopathy segment of the SCD-HeFT population wait nine months after diagnosis before receiving an implant (Section B-7 of the NCD). Then again, the NCD establishes that we only need to wait three months for these patients if they are enrolled in a CMS-approved data registry (Section B-9). Since CMS mandates that all primary prevention patients be enrolled in the ACC National Cardiovascular Data Registry (Section B-8), the required wait period is really just three months for all of these non-ischemic dilated cardiomyopathy patients.

Medicare coverage is a riddle wrapped up in an enigma — and if you don’t decipher it the same way as your RAC auditor, you could be accused of fraud.

Medicare officials assert that RAC auditors “merely review claims for compliance with existing policy.”² Unfortunately, there is a widening gap between existing CMS policy and the standard of practice dictated by clinical trial data and specialty society recommendations. Readers should become familiar with the CMS policies most likely to trigger a negative RAC audit finding: documentation requirements for inpatient admission, justification for dual chamber, rather than single chamber, device implantation, and wait periods for primary prevention defibrillator implant following an MI or revascularization.

Inpatient Admission Criteria

To stimulate the establishment of community-based hospitals, CMS pays twice as much for many procedures performed on an inpatient rather than outpatient basis. This holds true even if the procedure is performed in the same lab and the patient sleeps overnight in the same hospital bed. Because of this, one of the first initiatives RAC auditors focused on was inappropriate inpatient admissions. The penalties and refunds triggered by the initial wave of RAC audits fueled the expansion of the program, but conversely caused many hospitals to declare bankruptcy.

Surprisingly, there is no existing policy establishing what conditions or scenarios justify inpatient admission as opposed to outpatient treatment. The 23-hour rule, InterQual criteria, and specialty society published “hospitalization criteria” are unofficial guidelines that are not endorsed nor enforced by CMS.

According to CMS officials, the decision “whether a patient should be admitted or be treated in the hospital outpatient department is a decision for the physician in the exercise of his/her best professional medical judgment in light of the patient’s condition.”² It was added, “The order for inpatient services should be signed by a physician and the medical necessity for the admission should be documented in the patient’s medical record.”²

CMS also clarified that the concerns identified by RAC auditors were not related to clinical issues but rather the lack of paperwork. “The situations in which a RAC determines that an inpatient admission should have been billed as an outpatient service do not involve RACs substituting their professional judgment for that of the admitting physician, but rather finding that the patient records either do not contain a signed order from a physician admitting the patient or documentation in the patient record of the condition requiring inpatient care.”²

RAC auditors only consider the CMS policy and the documentation submitted by providers in response to their record requests when conducting audits. If a facility were to respond to a RAC audit for a complex procedure by supplying a history and physical, the operative report, and discharge summary, it may fail a RAC audit because the submitted documentation doesn’t include the one thing the auditor is looking for: the handwritten order for inpatient admission. Fully responding to RAC record requests with all available documentation (handwritten orders, dictations, nurse instructions, etc.) will substantially reduce your risk of a negative audit finding.

Single vs. Dual Chamber Devices

The NCD for pacemaker implantation is probably the most outdated of all CMS coverage policies; it was last updated in April 2004. This NCD provides a list of indications for single and dual chamber pacemakers. To justify the additional cost of a dual chamber system, the medical record must illustrate that one of the following four conditions have been met:^3,4

1. Patients in whom single-chamber (ventricular pacing) at the time of pacemaker insertion elicits a definite drop in blood pressure, retrograde conduction, or discomfort.
2. Patients in whom the pacemaker syndrome (atrial ventricular asynchrony), with significant symptoms, has already been experienced with a pacemaker that is being replaced.
3. Patients in whom even a relatively small increase in cardiac efficiency will importantly improve the quality of life, e.g., patients with congestive heart failure despite adequate other medical measures.
4. Patients in whom the pacemaker syndrome can be anticipated, e.g., in young and active people, etc.

Fortunately, the NCD also recognizes that “expert physicians differ in their judgments about what constitutes appropriate criteria for dual-chamber pacemaker use. The judgment that such a pacemaker is warranted in the patient meeting accepted criteria must be based upon the individual needs and characteristics of that patient, weighing the magnitude and likelihood of anticipated benefits against the magnitude and likelihood of disadvantages to the patient.”⁴

The policy establishes that dual chamber pacemakers are indicated for any patient qualifying for a single chamber pacemaker if “medical necessity is sufficiently justified through adequate claims development.”⁴ Therefore, if the implanting physician documents his/her justification for inserting a dual chamber system (heart failure, chronotropic incompetence, etc.), it should be sufficient to pass a RAC audit.

The NCD for defibrillator implant contains similar guidelines. It establishes that “providers must be able to justify the medical necessity of devices other than single lead devices. This justification should be available in the patient’s medical record.”³

Since “expert physicians” differ in their judgment (according to the NCD), there must be an array of conditions that justify implanting a dual chamber system rather than a single chamber system. Since RAC auditors cannot make clinical judgment calls, they “merely review claims for compliance with existing policy” — pretty much any logical, documented justification for a dual chamber implant should be acceptable.

Primary Prevention Wait Period

The primary prevention defibrillator implant guidelines published in CMS’ NCD specify that patients must not have had a “CABG or PTCA within the past 3 months” or “an acute MI within the past 40 days.” For these patients, physicians can prescribe an automatic external defibrillator to monitor the patient and serve as a bridge through the mandated waiting period. Defibrillator implants performed while in this waiting period are not covered by existing Medicare policy.

Some physicians have reported that auditors are questioning the medical necessity of defibrillator implants for patients who experience ventricular tachycardia, ventricular fibrillation, or sudden cardiac arrest while in the above-specified wait periods. In effect, auditors may be applying a hybrid restriction that prevents a patient from qualifying for implant under the secondary prevention guidelines if the qualifying VT, VF, or SCA occurs within the mandated wait period following revascularization or MI.

This hybrid standard is in direct conflict with the standard of practice as I perceive it. Unlike the standards related to inpatient status and the need for a dual chamber device, there are no wild cards in this NCD that permit physicians to override the wait periods if medical necessity is documented or if the patient converts to secondary prevention.

Before depriving patients of the therapy you feel they require, make sure to consider that the current NCD does not technically recognize elective replacement interval (ERI) or end of life (EOL) as appropriate indications for generator change out. If a patient’s heart failure classification falls to Class I or if their ejection fraction rises to above 35%, they have effectively reverse remodeled themselves out of the indications for a replacement defibrillator. If left unchecked, RAC auditors might actually assert that you should turn off the ERI/EOL device and wait to see if the patient decompensates their way back into line with the NCD.

Closing Points

Many of the penalties already assessed by auditors will be overturned during the appeal process. Providers just need to dedicate enough time to piece together all of the documentation required to illustrate compliance with the applicable NCD. In many cases, the critical piece of documentation will be a handwritten note specifying “admit to inpatient,” a two-month-old echocardiogram confirming a 30% ejection fraction, or a year-old EKG tracing illustrating an acute myocardial infarction.

Whether appealing negative audit findings or responding to new audits, don’t assume that the auditor will know how to interpret the documentation you provide. In most cases, the auditors are non-clinicians with only a limited background in coding. In addition to providing the necessary supporting documentation, it helps to provide a copy of the applicable NCD and a cover letter with a ‘paint by numbers’ illustration of your compliance.

The current NCDs force physicians to deprive or delay lifesaving and life-sustaining therapy to millions of Medicare patients. Physicians who attempt to do what is obviously right for the patient at the expense of circumventing an NCD requirement are juicy targets for RAC auditors.

Without prompt correction and periodic maintenance of the NCDs, the RAC racket will prevent physicians from practicing evidence-based medicine in the United States.

References

1. Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877–883.
2. Personal communication, August 19, 2010.
3. Medicare National Coverage Determination (NCD) for Implantable Automatic Defibrillators, Section 20.4.
4. Medicare National Coverage Determination (NCD) for Cardiac Pacemakers, Section 20.8.