I believe myself fortunate to have a soapbox-this column in DI-from which to opine my views on the legal and regulatory issues encountered in my practice representing radiologists. Few issues have been as nettlesome, or as hot politically, as the one involving shared arrangements for radiologists and cardiologists to interpret cardiac CT angiography studies. It is my strongly held view that there are significant regulatory issues with many of these shared reading or overread arrangements between cardiologists and radiologists. If such arrangements are not structured correctly, I am concerned radiologists could face legal exposure.
Of equal, if not greater, concern is the precedent potentially set by shared reading arrangements. Could not orthopods seek hospital privileges to interpret MSK magnetic resonance studies in hospitals, focusing on the skeletal images and backed up by an overread of the soft tissue by radiologists? Or could the CCTA shared model apply to virtual colonography with gastrointestinal physicians who review only the colon and bill for the studies, while relying on the overreads of radiologists to document not only the polyps but any other noncolonic abnormalities on the axial source images as well? And on and on-you just name the specialist requesting an overread by a radiologist.
It is my hope that this column will encourage those who have embarked on these arrangements to reevaluate them. Here are several things to consider.
New CPT codes for CCTA. Last year, the American College of Radiology, the American College of Cardiology, and the Blue Cross and Blue Shield Association collaborated to create new CPT codes for various types of cardiac CT and CTA studies. The CPT Editorial Panel approved eight new CPT Category III codes (0144T-0151T) that became effective in January.
Each of these codes and the proper method for using them are described in an article published in the joint AMA/ACR publication Clinical Examples in Radiology. The article repeatedly noted that the new CPT codes are designed to describe not just the interpretation of the cardiac and coronary portions of the CT studies, but the interpretation of all axial source images contained in the CT study. The joint AMA/ACR article observed that "[i]t is important to note that the physician interpreting the study is responsible for the interpretation of all information on the axial source images of the precontrast, arterial phase sequence, and venous phase sequence, as well as the 2D and 3D reformatted images resulting from the study, including cine review." In other words, the physician interpreting (and billing for the interpretation under one of the new CPT codes) must read and report the entire axial source image, not just the cardiac portion.
Similarly, the ACR-proposed credentialing guidelines for these studies provide that "[i]n addition to an examination of the cardiac structures of interest, the interpreting physician is responsible for examining all the visualized noncardiac structures and must report any clinically relevant abnormalities of these adjacent structures."
Medicare billing requirements. As a general rule, Medicare requires that the name of the physician who actually performed a professional service correspond to the physician's name appearing on the Medicare claim form CMS-1500. As designed, that form permits only one physician's name and physician identification number. Every CMS-1500 claim form submitted to the Medicare program must include a certification signed by the physician stating that the services listed on the claim were both medically necessary and personally furnished by the physician.
In addition, in order to bill for the professional component of any service, the physician's name appearing on the CMS-1500 must correspond to the name of the physician indicated on the written, signed interpretation report included in the patient's file. The signed report demonstrates that the services were, in fact, performed. The billing physician should sign the report that fully describes the CCTA findings.
When a cardiologist signs and submits the CMS-1500 for a CCTA interpretation that overlooked noncardiac structures, one must ask if the certification is accurate and complete if it states that all services billed were personally furnished by that physician. I believe it is not.
Thus, should a cardiologist read only the cardiac portion of the CT studies, he/she will not be performing all of the professional services provided to the patient and submitted to Medicare for reimbursement. Instead, some portion of the professional service is actually performed by a second physician (e.g., by a radiologist who reads all other axial source images such as the lung, mediastinum, and spine). The concern is that the signed certification could be characterized as false because the physician claiming the services did not personally perform all of those services that will be reimbursed by Medicare.
Note, however, this is not to say that genuine overreads are not possible. Under appropriate circumstances, an overread by a radiologist may be permissible, but only if the cardiologists fully interpret and report all of the images for the CCTA study. If the overread follows a fully performed interpretation, it is my view that the radiologist can perform an overread report.
Fraud and abuse exposure: false claims. The Federal Civil False Claims Act has become one of the government's primary-and most lucrative-enforcement tools in combating fraud related to government funds, particularly in connection with Medicare and Medicaid reimbursement. False claims allegations are also the principal source of qui tam relator law suits filed against healthcare providers. The act calls for civil penalties in the amount of $5500 to $11,000 per false claim, as well as damages totaling three times the amount of damage sustained by the government as a result of the false claims.
My concern is that a so-called overread arrangement that is not a genuine overread could facilitate the submission of potentially false claims to the Medicare program by the cardiologist. Radiologists would share in the liability for their role in facilitating what could be construed as a false claim. Each time a cardiologist submits a CMS-1500 for payment of a CCTA that is not fully interpreted, he or she certifies to the Medicare program that the services identified on the form were personally performed by that cardiologist. With the assistance of the overread by a radiologist, these arrangements could well lead to false claims allegations either by the government or by a qui tam relator.
Fraud and abuse exposure: Stark. There are also Stark issues affecting such shared reading arrangements. In order for cardiologists who have an ownership interest in the group practice to refer the professional component of the CTA study to the cardiology group when interpreted by an independent contractor, the referral must meet the "physician services" exception to the Stark antireferral prohibition. Because a radiologist's so-called overread service as a "physician in the [cardiology] group practice" is as an independent contractor but is not provided pursuant to a valid reassignment, these arrangements do not appear to fit the required exception to the Stark rules. For the Stark requirements to have been met, the cardiology group's CMS-1500 form would have to include the personal identification number of the interpreting radiologist who validly reassigned to the cardiology group. Those required elements of the Stark exception are not present in shared reading arrangements.
Other risk areas. This column has not even touched upon the potential professional liability risks associated with these arrangements. I have confined my comments only to those issues that should make these arrangements absolutely unacceptable. The ACR legal department has prepared a white paper that identifies other risk areas radiologists should evaluate when considering such arrangements.
In short, I caution radiologists against entering into overread arrangements unless they are genuine overread arrangements with a written agreement that includes the following safeguards:
- The interpreting physician is solely responsible for generating the official, complete, signed interpretation report for the patient's file with the interpreting physician having read all of the axial source images; and
- Any overread report should indicate that it is a secondary read of a previously interpreted, and complete, study.
Until new codes are promulgated to separate cardiac from noncardiac services as separate and distinct studies, or until Medicare accepts separate billing for the cardiologist's and the radiologist's shares of these studies using the -52 modifier, my recommendation is that radiologists avoid these ad hoc procedure-sharing/fee-sharing arrangements.
Mr. Greeson is a partner in the healthcare group of Reed Smith LLP in Falls Church, VA. He can be reached at 703/641-4242 or tgreeson@reedsmith.com.
