The Effectiveness of Screening Mammography

To the editor:

The Patient-Oriented Evidence that Matters (POEM) review entitled “Screening Mammography May Not Be Effective At Any Age”¹ is not helpful to clinicians and family medicine educators. Although I agree with the assessment of Drs Wilkerson and Schooff that the reviewed study casts doubt on the effectiveness of screening mammography in mortality prevention, neither the study nor the reviewers’ assessment appears to address questions about cost-benefit analysis relative to breast cancer–related morbidity secondary to delayed diagnosis. The review states: “Two studies evaluated morbidity, finding surgery and radiotherapy to be performed more frequently in the screened patients.” What were the long-term outcomes in the unscreened patients? Were they more likely to require aggressive chemotherapy or surgery secondary to delayed diagnosis? These questions were not clearly answered in the original article.²

As we work to advise our patients accurately about the risks and benefits of various screening procedures, we must define benefit more broadly than mortality prevention. Financial and emotional costs of extensive therapy and quality of life during the life span of our patients and their families must also be considered. Therefore, the most accurate conclusion from this important meta-analysis is not that “studies that started with truly equal groups showed no benefit to screening.” Rather, when counseling patients about mammography, we must state that although mammography may not extend their life span it may be beneficial in terms of quality of life and other morbidity related to treatment of breast cancer diagnosed at different stages of the disease.

POEMs are useful to clinicians and educators when they analyze research accurately, point out its strengths and weaknesses, and draw appropriate conclusions. Physicians in clinical practice depend on reviews like POEMs to change their practice patterns. The implicit conclusion in the recommendations for clinical practice section of this particular POEM—that this analysis should dramatically alter clinical practice regarding screening mammography—is not justified by the text of the review. Also, the title suggests a much stronger conclusion than is warranted by the review. This certainly renders the POEM less effective and understates the complexity of the question. In many ways my concerns echo those of Dr Boyd in his letter to the editor in the April issue.³ POEMs will take their rightful place in the medical literature and in shaping practice patterns only when they strive to adhere to even more rigorous criteria than the articles reviewed.

Elizabeth Steiner, MD
Oregon Health Sciences University
Portland

REFERENCES

• BF, Schooff M. Screening mammography may not be effective at any age. J Fam Pract 2000; 49:302, 371.
• P, Olsen O. Is screening for breast cancer with mammography justifiable? Lancet 2000; 355:129-34.
• MA. ERT and breast cancer. J Fam Pract 2000; 49:376.

The preceding letter was referred to Dr Schooff who responded as follows:

I appreciate the comments from Dr Steiner. Her letter illustrates several concerns about the current state of clinical evidence regarding many screening practices, including mammography.

Dr Steiner is correct in stating that the article we reviewed¹ does not provide information regarding cost or quality-of-life benefits (or harms) of screening mammography. The 2 methodologically sound studies did find, however, that screened patients underwent more surgery (including more biopsies with benign results) and radiotherapy than patients in the unscreened groups, all with no impact on mortality. It would be very interesting to see the results of a study comparing medical costs and quality of life in women with and without breast cancer who were randomly assigned to a mammography screening program or no such program. Unfortunately, to the best of my knowledge no such study exists.

Dr Steiner also questions the conclusions of our review. She mentions the need to use the information from this study in counseling women about mammography. That was also what we hoped to say in our final sentence—that instead of rigidly suggesting mammography to all women we should inform them of what we currently know and do not know about the utility of screening mammography and allow them to make informed decisions on the basis of that information.

Along this line, a recent article by Ewart² discusses the ethics of screening programs based on the current quality of evidence. He argues that in screening programs (where it is suspected that the patient does not have the disease) as opposed to investigations (where the patient is felt to have a significant likelihood of having the disease), there is a requirement for better evidence that the benefits of the program (the screening test and all possible follow-up tests or treatments) significantly outweigh the costs and potential harm to both the individual patient and to society. He compares this with the evidence necessary in a criminal law case (beyond any reasonable doubt) versus that in a civil case (on the balance of the probabilities). The article we reviewed clearly shows that we cannot say that mammography screening is effective beyond any reasonable doubt.