Dr. Raskin is a practicing diagnostic radiologist in the subspecialty of neuroradiology in Fort Lauderdale, FL and is the Internal Legal Counsel to the Florida Radiological Society. He is also a member of the Editorial Board of this journal.

Signing a radiology colleague's interpretive report becomes problematic if that colleague is sued for malpractice, as you will almost certainly be sued as well. Substituting your signature for the radiologist who actually did the interpretation is an invitation to the plaintiff's attorney to name another physician in the lawsuit. Some have called this a proxy signature, but a proxy is an authorization by one person to another so that the second person can act for the first. A proxy binds the first person by the act of the second person. A substituted signature can bind you both. Whatever we wish to call it, substituting your signature for a colleague's signature may place you in their shoes in the eyes of the law.

Why sign?

Why do we bother to sign reports in the first place? According to the Health Care Financing Administration (HCFA), the radiologist who performs the service must sign the interpretive report. ¹ In addition, the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) mandates that the interpretive report have the author identified. ² It is not clear that JCAHO requires the author of the report to actually sign the report. However, it is quite clear that HCFA fully intends that the author actually sign the report. The Standard Guide for Electronic Authentication of Health Care Information indicates that the signature is the act of taking responsibility for a document and the purpose of the signature is to determine accountability. ³

How does this work in practice?

Most reports are signed by the interpreting radiologist. However, there is a "gray area" when the author of the interpretive report is not available in a timely matter and the report is signed by a colleague in order to expedite delivery of the report. The timeliness of reporting any radiologic examination varies with the nature and urgency of the clinical problem. When a colleague's signature is substituted for the author of the interpretive report, that colleague may or may not have read the report. The American College of Radiology (ACR) Standard for Communication indicates that the final report should be proofread to minimize typographical errors, deleted words, and confusing or conflicting statements. ⁴ However, most reports are produced using word processing software that checks spelling and even grammar. Any "deleted words and confusing or conflicting statements" may not be recognized by anyone other than the author. Rarely will a colleague review the actual films of a report rendered by someone else, which further compounds the problem. ⁵

When named as a defendant in a lawsuit, the radiologist who substituted his or her signature for the author of the interpretive report will rarely ever claim that they actually looked at the films but will claim that they proofread the report. They will often claim that they signed the report to expedite timely delivery to the referring physician. However, bear in mind that if a significant error has occurred in the written report, expediting this erroneous report only further compounds the initial error.

Radiologists are often reluctant to come in to sign reports when they are off for the weekend. The radiologist on call usually winds up signing all the reports that were generated by those who are off duty. We learned in kindergarten (unfortunately, some never learned) that we clean up our own mess. If you dictated it, you should sign it and not expose a colleague as another "deep pocket" because they signed your report.

Consequences of signing a report

Regardless of any disclaimers you may include on the report, if you sign it, you will most likely be held responsible for the content of the report. Although no Appellate Court decisions have been heard regarding the substituted signature of a radiologist, the New Jersey Appellate Court concluded that by signing a report, a physician attests to the accuracy of the information that the report contains. ⁶ This is an important holding as it does not matter what disclaimers the physician may put before or after the signature. Even a juror knows that you "don't sign something unless you have read it and agreed to it." To afford a physician anything less would let them off the hook for what every juror knows is "common knowledge."

Report of a case

A 55-year-old woman was referred to a surgeon for complaints of a persistent dripping and running nose. The surgeon ordered a cysternogram, and a CT scan of the ethmoid sinus. Based upon the results, the patient was told that the left side of her dripping and running nose was due to cerebral spinal fluid leaking as a result of sinus disease and a defect in her left cribriform plate. The patient was told that the easiest and fastest approach to seal the defect in the cribriform plate was to undergo an operative procedure through her nose, internasal endoscopy, because of decreased anesthesia time.

During the course of the surgical procedure, the surgeon entered the ethmoid sinus and found a mass present. A biopsy was taken and sent for analysis, which was indeterminate as to cell origin due to hemorrhage, although the pathologist thought it might be neural tissue on frozen examination. The surgeon continued to remove the mass from the roof of the ethmoid. The defect in the cribriform plate was visualized, and a fat graft and flap closure was performed. The patient failed to awaken following surgery, and an emergency CT scan showed subarachnoid and interventricular hemorrhage. A ventriculostomy was performed, and the patient remained comatose until her death approximately 1 week later. The pathology report confirmed normal brain tissue.

The cisternogram showed a leak from the left cribriform plate. The CT scan was interpreted as showing "lytic erosive changes" of the left cribriform plate, "likely representing a CHF leak or inflammatory changes." In deposition, the radiologist testified that she considered that this might represent an encephalocele but since an encephalocele was so rare, and she had never seen one before, she did not include it in her report.

As you might imagine, the case was settled by the surgeon and the radiologist without even deposing the expert witness radiologist for the plaintiff. However, the story doesn't end here. The author of the report had been out of the office the day after dictating it, and one of her colleagues readily signed the report. Despite his protestation that he was merely signing the report for "content and expedience," and not attesting to its accuracy, his attorney wisely advised him to settle as well. The "content" defense fell short when the radiologist had to admit, in deposition, that "CHF" stood for congestive heart failure and not cerebral spinal fluid, considerably changing the content of the report. Expedience matters little when the report is patently wrong.

What to do?

* No one signs the report. This is not really possible or feasible since HCFA requires that the radiologist who interprets the study must sign his or her own report.

* Sign your colleague's report. If you do this, do it with the understanding that this will result in legal exposure to you and you may be held liable for the report. Therefore, it would be prudent in those cases in which there is no alternative except to sign a colleague's report for you to review the actual films and correlate them with the report.

* Have the author of the report authenticate the report. This is what we are actually trying to accomplish, but it may become burdensome when the author of the report is not available or may be away. In today's high-technology world, there is little excuse for not being in touch and not being able to communicate. Reports are routinely faxed to referring physicians. The report can be sent to the author of the report by fax or e-mail. These devices are readily available virtually all over the world, in airport lounges, and on cruise ships.

In January 2001, Air Canada began offering e-mail and Internet access on their North American flights. In April 2001, Singapore Airlines began offering e-mail on all of their international flights. Virgin Atlantic and Cathay Pacific soon followed. ⁷ Boeing will be introducing a broadband e-mail and Internet system that can deliver e-mail at speeds about 25 times faster than current modems. ⁸ Soon all passengers will be able to send and receive e-mail no matter where they are flying. So there will be no place you can hide and not get your report.

HCFA allows authentication to include signature, written initials, or computer entry. In this way, the author of the report can authenticate his or her own report. Some radiology departments have started the process of installing software on their home computers that will avoid the necessity of a radiologist signing a colleague's report (L. Berlin, personal communication, May 29, 2001). In the rare situation in which the films must be correlated with a report, a colleague could then review the films. Sending reports electronically to the author of the report will minimize the need for a radiologist to sign a colleague's report.

Conclusion

If the author of the report will not be available within 24 hours after dictating the report, a system should be in place to either fax or electronically transmit the reports to the author. The author can then provide authentication by signature, written initials, or computer entry. By using presently available technology, it should be rare that the author of a report cannot authenticate it within 24 hours. In these rare situations, the radiologist who substitutes his or her signature for the author should review the films and make corrections to the report, if necessary, with the full understanding that he or she will most likely be held responsible for the contents of the report. AR