Breast Cancer Claims

As breast cancer screening evolves in the digital age, so do the issues confronting lawyers. Advances in medicine offer women more effective screenings with greater opportunities for early detection and diagnosis, and as digital screening becomes more widely available, it has been asserted that cancer is being detected earlier and with greater accuracy. While this should reduce the number of missed claims, unfortunately, errors might still occur. As lawyers representing women in cases involving misdiagnoses, we have to keep up with technological developments.

Anatomy of a Breast Cancer Claim

According to the Physicians Insurers Association of America, since 1985, more than $304 million has been paid in breast cancer claims on behalf of physicians nationwide. In fact, a study involving 25 insurance companies found malpractice claims against radiologists comprise 33% of all claims, with “mammogram misread” being the most common allegation.

Most often, plaintiffs' attorneys encounter the following basic fact pattern: A woman is diagnosed with breast cancer. She has received yearly screening mammograms, but nothing was ever mentioned about a problem or a suspicious finding. She questions if there was a misread on one or more of her prior films. The attorney handling the case gets a full history from the client and orders all prior mammogram films, corresponding reports and medical records from all doctors who saw the patient prior to the diagnosis. After the medical information is obtained, the plaintiffs' attorney submits it to a qualified radiologist who reviews it to determine if, indeed, the screening mammogram film was misinterpreted. If so, a qualified oncologist must review the case materials for causation, to see if the delay in the diagnosis made a real difference in case outcome.

Understanding Screening Protocol

In order to understand the current screening process, it's important to know what mediums are used. Screening mammography is done with either a plain film or a digital image. A plain film is interpreted by a radiologist or, sometimes, two radiologists (known as a double read). This plain film can also be turned into a digital film for computer analysis. A digital film can be interpreted by a radiologist using computer-assisted detection, or CAD. It is important to note that all of these methods ' plain film, plain film digitized, double read and CAD ' are within the standard of care because the American College of Radiology accepts them. Even though the radiology field is moving toward CAD, at the present time, one cannot successfully argue that the failure to utilize digital mammography, computer assisted detection or a double read is below the standard of care.

The American College of Radiology Practice Guideline for the Performance of Screening and Diagnostic Mammography 2008 advises that screening mammograms should generally be given at yearly intervals to women over age 40. This guideline also sets forth the qualifications for personnel, exam specifications, required documentation and the requirements for proper communication of results to the patient.

Issues surrounding the standard of care focus on who reviewed the film and whether it was properly interpreted (See 67 FR 5446 ' 900.12(a)(1) setting forth the requirements for a radiologist who reviews mammograms). To meet this standard, any suspicious area that could indicate cancer must be reported to either the physician who ordered the mammography or to the patient herself, depending upon who initiated the screening. Typically, primary care physicians order screening mammograms; however, sometimes women go to a facility or mobile unit for screening without a referral. Suspicious areas that must be reported include spiculated (spiked or pointed) masses, ill-defined masses and microcalcifications.

Appropriate recommendations must also be made in terms of follow-up tests, such as diagnostic mammograms, magnified views of suspicious areas, ultrasounds, MRI studies or biopsies with a surgical consult. Each recommendation is case-dependent, based upon the patient's history, risk factors, presence of high-density breast tissue and location, size and number of the suspicious findings.

Overview of CAD and Its Role in Litigation

The FDA has approved numerous CAD systems for screening mammography since the first one in 1998. As more facilities embrace digital technology, plaintiff and defense practitioners will encounter digital films reviewed with CAD, sophisticated pattern recognition software used by a growing number of radiologists to help them read medical images. CAD uses complex algorithms to identify suspicious features on the images.

The initial medical literature on CAD seemed to support the claim that it improves accuracy in making breast cancer diagnoses. See, e.g., Freer, Timothy W. and Ulissey, Michael J., “Screening Mammography with Computer-Aided Detection: Prospective Study of 12,860 Patients in a Community Breast Center.” 10.1148/radiol.2203001282 Sept. 1, 2001 Radiology, 220, 781-786. However, more recent studies have concluded that CAD's effect on the accuracy of interpretation is unclear. See, e.g., Fenton, Joshua J., “Influence of Computer-Aided Detection on Performance of Screening Mammography,” New Engl J Med, Volume 356:1399-1409, April 5, 2007, Number 14.

With CAD, the radiologist first looks at the mammogram, then uses the computer to check the review. The computer marks calcifications and masses, which the radiologist then re-evaluates before issuing a final report. The images on the computer screen can be adjusted for size, brightness and contrast to see certain areas more clearly.

In a screening mammogram where CAD is used, certain issues might arise in litigation. For the plaintiff, discovery must be obtained to identify the manufacturer, model number and the specifics of the computer software program that was utilized in CAD.

The following questions must be answered: 1) Is CAD data saved? 2) Is the data retrievable? 3) What abnormalities were flagged by the computer, and why? 4) What adjustments were available in the program itself to highlight a particular type of abnormality? and 5) Were the available adjustments utilized and, if not, why not?

Discovery should be done before the deposition of the interpreting radiologist. This will ensure that the plaintiffs' attorney has all the information that was available to the defendant radiologist at the time of the alleged misread. It will also prepare the plaintiffs' attorney to intelligently ask the following questions of the deponents: What abnormalities, if any, did the radiologist see or report before using the CAD system? If CAD highlighted any suspicious areas, did the radiologist consider, accept, reject or report each abnormality? Why, or why not?

If CAD showed an area of concern that was rejected by the radiologist, and ultimately proved cancerous, it could serve as strong evidence at trial of a misinterpretation of the screening film. The plaintiffs' attorney should be prepared to prove the scientific reliability of the computer software algorithm through competent evidence. A request for admission may suffice, but expert testimony may also be required.

Even if CAD flagged the area as a suspicious lesion, and the radiologist rejected the computer's “diagnosis,” the defense can still argue that the radiologist's rejection was a judgment call, but not a deviation from the standard of care.

On the other hand, if the defense can show that CAD did not flag the alleged suspicious area, this can be powerful evidence that the mammogram was not misinterpreted. The dilemma for plaintiff's counsel is that CAD evidence cannot be obtained pre-suit. A writ could be filed to obtain this discovery pre-complaint, but in my experience this approach is cumbersome and inefficient. If plaintiff's counsel learns in discovery that CAD did not flag the lesion in question, it may be advisable to discontinue the case.

Conclusion

As sophisticated computer algorithms are utilized in breast cancer digital screenings, lawyers must be educated on the technology itself to determine what a doctor really did and did not do during the interpretation of the mammogram. The attorney must be able to effectively communicate the facts learned in discovery to a jury. To make an effective, compelling case, it's important to re-create the radiologist's analysis of the film in real time, highlight the suspicious area flagged by the computer, and show that it was dismissed by the medical professional on trail. And, keep in mind that technology isn't limited to the medical field; your firm's information technology professionals may be able to help you present the necessary evidence to build your case.

Howard J. Levin is a shareholder at Anapol Schwartz Weiss Cohan Feldman & Smalley in Philadelphia and concentrates his practice in medical malpractice and products liability. He is also a volunteer member of the 2009-10 Philadelphia affiliate of the Susan G. Komen for the Cure Grant Review Panel. This article first appeared in the New Jersey Law Journal, an Incisive Media sister publication of this newsletter.

As breast cancer screening evolves in the digital age, so do the issues confronting lawyers. Advances in medicine offer women more effective screenings with greater opportunities for early detection and diagnosis, and as digital screening becomes more widely available, it has been asserted that cancer is being detected earlier and with greater accuracy. While this should reduce the number of missed claims, unfortunately, errors might still occur. As lawyers representing women in cases involving misdiagnoses, we have to keep up with technological developments.

Anatomy of a Breast Cancer Claim

According to the Physicians Insurers Association of America, since 1985, more than $304 million has been paid in breast cancer claims on behalf of physicians nationwide. In fact, a study involving 25 insurance companies found malpractice claims against radiologists comprise 33% of all claims, with “mammogram misread” being the most common allegation.

Most often, plaintiffs' attorneys encounter the following basic fact pattern: A woman is diagnosed with breast cancer. She has received yearly screening mammograms, but nothing was ever mentioned about a problem or a suspicious finding. She questions if there was a misread on one or more of her prior films. The attorney handling the case gets a full history from the client and orders all prior mammogram films, corresponding reports and medical records from all doctors who saw the patient prior to the diagnosis. After the medical information is obtained, the plaintiffs' attorney submits it to a qualified radiologist who reviews it to determine if, indeed, the screening mammogram film was misinterpreted. If so, a qualified oncologist must review the case materials for causation, to see if the delay in the diagnosis made a real difference in case outcome.

Understanding Screening Protocol

In order to understand the current screening process, it's important to know what mediums are used. Screening mammography is done with either a plain film or a digital image. A plain film is interpreted by a radiologist or, sometimes, two radiologists (known as a double read). This plain film can also be turned into a digital film for computer analysis. A digital film can be interpreted by a radiologist using computer-assisted detection, or CAD. It is important to note that all of these methods ' plain film, plain film digitized, double read and CAD ' are within the standard of care because the American College of Radiology accepts them. Even though the radiology field is moving toward CAD, at the present time, one cannot successfully argue that the failure to utilize digital mammography, computer assisted detection or a double read is below the standard of care.

The American College of Radiology Practice Guideline for the Performance of Screening and Diagnostic Mammography 2008 advises that screening mammograms should generally be given at yearly intervals to women over age 40. This guideline also sets forth the qualifications for personnel, exam specifications, required documentation and the requirements for proper communication of results to the patient.

Issues surrounding the standard of care focus on who reviewed the film and whether it was properly interpreted (See 67 FR 5446 ' 900.12(a)(1) setting forth the requirements for a radiologist who reviews mammograms). To meet this standard, any suspicious area that could indicate cancer must be reported to either the physician who ordered the mammography or to the patient herself, depending upon who initiated the screening. Typically, primary care physicians order screening mammograms; however, sometimes women go to a facility or mobile unit for screening without a referral. Suspicious areas that must be reported include spiculated (spiked or pointed) masses, ill-defined masses and microcalcifications.

Appropriate recommendations must also be made in terms of follow-up tests, such as diagnostic mammograms, magnified views of suspicious areas, ultrasounds, MRI studies or biopsies with a surgical consult. Each recommendation is case-dependent, based upon the patient's history, risk factors, presence of high-density breast tissue and location, size and number of the suspicious findings.

Overview of CAD and Its Role in Litigation

The FDA has approved numerous CAD systems for screening mammography since the first one in 1998. As more facilities embrace digital technology, plaintiff and defense practitioners will encounter digital films reviewed with CAD, sophisticated pattern recognition software used by a growing number of radiologists to help them read medical images. CAD uses complex algorithms to identify suspicious features on the images.

The initial medical literature on CAD seemed to support the claim that it improves accuracy in making breast cancer diagnoses. See, e.g., Freer, Timothy W. and Ulissey, Michael J., “Screening Mammography with Computer-Aided Detection: Prospective Study of 12,860 Patients in a Community Breast Center.” 10.1148/radiol.2203001282 Sept. 1, 2001 Radiology, 220, 781-786. However, more recent studies have concluded that CAD's effect on the accuracy of interpretation is unclear. See, e.g., Fenton, Joshua J., “Influence of Computer-Aided Detection on Performance of Screening Mammography,” New Engl J Med, Volume 356:1399-1409, April 5, 2007, Number 14.

With CAD, the radiologist first looks at the mammogram, then uses the computer to check the review. The computer marks calcifications and masses, which the radiologist then re-evaluates before issuing a final report. The images on the computer screen can be adjusted for size, brightness and contrast to see certain areas more clearly.

In a screening mammogram where CAD is used, certain issues might arise in litigation. For the plaintiff, discovery must be obtained to identify the manufacturer, model number and the specifics of the computer software program that was utilized in CAD.

The following questions must be answered: 1) Is CAD data saved? 2) Is the data retrievable? 3) What abnormalities were flagged by the computer, and why? 4) What adjustments were available in the program itself to highlight a particular type of abnormality? and 5) Were the available adjustments utilized and, if not, why not?

Discovery should be done before the deposition of the interpreting radiologist. This will ensure that the plaintiffs' attorney has all the information that was available to the defendant radiologist at the time of the alleged misread. It will also prepare the plaintiffs' attorney to intelligently ask the following questions of the deponents: What abnormalities, if any, did the radiologist see or report before using the CAD system? If CAD highlighted any suspicious areas, did the radiologist consider, accept, reject or report each abnormality? Why, or why not?

If CAD showed an area of concern that was rejected by the radiologist, and ultimately proved cancerous, it could serve as strong evidence at trial of a misinterpretation of the screening film. The plaintiffs' attorney should be prepared to prove the scientific reliability of the computer software algorithm through competent evidence. A request for admission may suffice, but expert testimony may also be required.

Even if CAD flagged the area as a suspicious lesion, and the radiologist rejected the computer's “diagnosis,” the defense can still argue that the radiologist's rejection was a judgment call, but not a deviation from the standard of care.

On the other hand, if the defense can show that CAD did not flag the alleged suspicious area, this can be powerful evidence that the mammogram was not misinterpreted. The dilemma for plaintiff's counsel is that CAD evidence cannot be obtained pre-suit. A writ could be filed to obtain this discovery pre-complaint, but in my experience this approach is cumbersome and inefficient. If plaintiff's counsel learns in discovery that CAD did not flag the lesion in question, it may be advisable to discontinue the case.

Conclusion

As sophisticated computer algorithms are utilized in breast cancer digital screenings, lawyers must be educated on the technology itself to determine what a doctor really did and did not do during the interpretation of the mammogram. The attorney must be able to effectively communicate the facts learned in discovery to a jury. To make an effective, compelling case, it's important to re-create the radiologist's analysis of the film in real time, highlight the suspicious area flagged by the computer, and show that it was dismissed by the medical professional on trail. And, keep in mind that technology isn't limited to the medical field; your firm's information technology professionals may be able to help you present the necessary evidence to build your case.

Howard J. Levin is a shareholder at Anapol Schwartz Weiss Cohan Feldman & Smalley in Philadelphia and concentrates his practice in medical malpractice and products liability. He is also a volunteer member of the 2009-10 Philadelphia affiliate of the Susan G. Komen for the Cure Grant Review Panel. This article first appeared in the New Jersey Law Journal, an Incisive Media sister publication of this newsletter.