The Shifting Landscape of Cardiac CT Angiography Reimbursement

Many imaging providers have been wishing for changes in reimbursement policies for cardiac CT angiography (CCTA) for some time. Changes are about to be made, but those wishes may not be granted. In December 2007, CMS proposed major restrictions on coverage for CCTA that could take effect as early as March 2008. The conclusions that CMS discussed in the proposed coverage determination would, if implemented, have a profoundly negative impact on Medicare beneficiaries by limiting needed access for clinically appropriate indications for using this technology (see sidebar). Whether or not the new CMS National Coverage Determination (NCD) for CCTA is implemented, other reimbursement controversies have the attention of providers. Prominent among these is the issue that providers face of how to report appropriately and accurately on the data obtained during CCTA studies. These collaborative or supplemental readings often constitute a gray area for payors, so imaging providers may need to adopt a team approach. It is important to note that these are not separately billable services. Using a team approach provides an accurate and timely interpretation of the data obtained and provides our patients the highest quality care. There are several business models that are in use. All models need to ensure that all providers are appropriately trained and credentialed to provide the interpretation of these studies. Let’s discuss some of these models. Using a competitive reimbursement model, the referring physician who is ordering the test, selects the study's reader, who bills for the full interpretation; this is especially likely if neither specialty concedes primary reader status to the other. Some feel if the selected reader is a cardiologist, the noncardiac portion of the study may still require interpretation; conversely some wonder if a radiologist can completely interpret the coronary tree. CCTA Coverage Update A surprise proposal issued by CMS on December 13, 2007, would radically change Medicare reimbursement for cardiac CT angiography (CCTA). If adopted, it could also set a precedent that would create major obstacles for the coverage of future imaging technologies and applications. The proposal would use a restrictive National Coverage Determination (NCD) to replace the Local Coverage Determinations that now cover CCTA throughout the United States. Under the NCD, Medicare reimbursement would be disallowed for all but two CCTA indications. For those two, reimbursement would be permitted only for patients enrolled in CMS-approved controlled clinical trials. In addition, the trials would be required to determine whether patient outcomes were improved by CCTA. This is the first time that CMS has linked outcomes and imaging in this way, and the influence of this requirement could be detrimental to imaging as a whole. Because outcomes are highly dependent on diagnosis, treatment, and patients' behavior, their connection to prior imaging procedures is indirect and difficult to assess. Even the best possible imaging could lead to poor outcomes, if followed by inadequate treatment and/or insufficient patient compliance. Unless huge trials can be designed to control for all the variables that come between imaging and outcomes, it may simply be impossible to connect the two, for CCTA or any other kind of imaging. The two groups covered within trials under the NCD proposal are symptomatic patients with chronic stable angina who are at intermediate risk for coronary-artery disease and symptomatic patients with unstable angina who are at low risk for short-term death and at intermediate risk for coronary-artery disease. Considerable public comment (which was accepted by CMS until January 12, 2008) has focused on this choice of indications, which are not thought to have the highest diagnostic yield for CCTA. Instead, its high negative predictive value makes CCTA especially valuable for ruling out cardiac origins of chest pain in patients (particularly those in emergency settings) in who heart problems are unsuspected and unlikely. By preventing unnecessary cardiac catheterization, this application of CCTA saves both money and time. This may be the most common use of CCTA, and is probably among the reasons that it is now being performed at more than 2,000 US facilities. Professional societies have submitted extensive commentaries to CMS, pointing out these problems and adding that much of the recent evidence was not considered by CMS in formulating its proposal. The diagnostic value of CCTA performed using 64-slice scanners is, of course, superior to that of CCTA on earlier equipment, yet CMS used studies performed with older (and even four-slice) scanners as its background material. There is some speculation that the proposal may not be issued as the final rule without major changes, given the overwhelmingly negative responses received during the public comment period. In fact, only the insurance industry appears to favor the measure, perhaps because it sees CCTA as an addition to conventional angiography instead of a replacement for it. CMS expects to issue the final NCD on March 12, 2008. Under a collaborative reimbursement model, a cardiologist reads the cardiac portion of the study and a radiologist reads the soft tissue. They then issue a blended report. This may represent a reimbursement quandary in order to ensure that both readers are compensated fairly for interpreting the same study, understanding that only one may bill. Medicare's basic position is that patient care is paramount and a team approach may accomplish this. It is then incumbent on the providers to ensure appropriate coding, interpretation, and reporting. It is equally important to check with your local carrier, however, before assuming that this will be the case everywhere. In general, supplemental readings will not be stand-alone services for which separate billing can be made. CCTA is currently covered under Category 3 “T” codes, but their application may vary based on the final memo from CMS and whether it “allows” the local LCDs to stand, or if there will be a Coverage with Evidence Determination. Since private payors often follow Medicare’s lead, their coverage may change as well. Among commercial payors, Wellmark, Cigna, Anthem, and UniCare, at this juncture, consider CCTA investigational. Blue Cross/Blue Shield covers CCTA in seven states and has pilot projects in three more. Humana, Aetna, and United cover CCTA under certain conditions, with appropriate indications and diagnosis, if it is performed on an advanced scanner (with at least 40, 64, and 16 slices, respectively). It is also critical to review each carrier’s medical policy on this. In 2008, many hospital-based imaging services are now bundled for Medicare purposes. This, in conjunction with the effect of the Deficit Reduction Act (DRA) on imaging-center reimbursement, poses many concerns in the imaging community as it relates to their ability to offer access to their communities. CMS has stated, however, that most services that are bundled in the hospital setting will not be subject to DRA comparison if they are provided in imaging centers billing under the Medicare Physician Fee Schedule (MPFS). In short, a service that has no “value” in the ambulatory payment classification (APC) hospital portion of coding, will have no DRA impact (reminder: under DRA, providers are paid the lesser of the MPFS or the Hospital Outpatient Prospective Payment System (HOPPS) payments). CCTA Experience The CT scanner used for CTA by Illinois Heart and Vascular, Hinsdale Illinois, has been operational since April 2005 and is owned by the practice. In 28 months of experience, 3,008 scans have been performed; of these, 2,438 were CCTAs, 570 were calcium scores, and 570 were peripheral vascular studies (150 carotid, 205 abdominal, and 215 lower extremity). Of total patients undergoing CTA, 41% were female. The CTA payor mix differed by type of study, with the Medicare population being 75% of peripheral studies and 42% of cardiac studies. Of the remaining peripheral studies, the payor mix was: PPOs 19%, commercial insurers 5%, and other payors 1%. Of the cardiac studies not in the Medicare population, PPOs accounted for 41%, commercial insurers for 4%, and the remaining Medicaid or private paying patients. The primary indications for cardiac CTA, on retrospective review, showed that approximately 60% had a history of chest pain and 30% had indeterminate stress tests. The risk factors exhibited by CCTA patients, in order of prevalence, included dyslipidemia, hypertension, a family history of coronary-artery disease, diabetes, and smoking. CCTA was ordered due to abnormal stress-test results in 49% of cases, chest pain in 32%, cardiomyopathy in 2%, and other reasons in 17%. It is now being used for pulmonary vein isolation prior to bi-ventricular pacemaker insertion or ablation procedures. The CT scanner of Cardiovascular Associates, located in Elk Grove Village, Illinois, became operational in August 2005, and 2,445 scans had been completed within two years. The scanner is owned in a 50–50% joint venture with Alexian Brothers Medical Center, with Cardiovascular Associates assuming 75% of the scanning time. CCTAs accounted for 1,561 studies, calcium scores for 226, and peripheral vascular examinations for 884. Their payor mix was similar to the prior case study. Medicare - 68% of peripheral studies and 47% of cardiac studies. Of the remaining peripheral studies, PPOs accounted for 28%, commercial insurers 3%, and other payors 1%. The remaining cardiac studies, were PPOs accounting for 44%, commercial insurers for 5%, and other payors or self pay 4%. Results of the CCTAs performed on both scanners indicate that images obtained during an examination ordered for an appropriate cardiovascular indication and diagnosis can provide a clear-cut diagnosis in many cases, or an alternative explanation for patients' symptoms. On analysis, perhaps its most amazing capacity is to rule out cardiac disease with an often-quoted 97% negative specificity rate. It is critical, however, to also note that noncardiac findings were seen in 26% of patients, and significant incidental findings such as malignancies were discovered in 6% of patients in one site. The most common noncardiac findings, as one may expect, were lung nodules, but it is important to note that primary malignancies were also identified. Clearly, these findings need to be weighed against the benefit of early detection in the context of a large prospective registry, but it is obvious that full use should be made of the noncardiac data obtained during CCTA.