PE R S PE C T IV E

Screening Mammography and the R Word

ing a false choice. The choice is not between health care ration ing and some undefined alterna tive, since there is no alternative. Rather, the choice concerns what principles we will use to ration health care. In the United States, we have traditionally rationed health care in the same way we ration expensive cars: those who can afford to pay for them are those who can have them. The al ternative currently being consid ered in health care reform would involve a shift to other principles, such as those rooted in consid erations of fairness, efficiency, and efficacy. Unfortunately, many support ers of the new mammography guidelines have been reluctant to call a spade a spade. Efforts to disguise the guidelines under the

cloak of false reassurances that mammographic screening for women in their 40s does not work only fuels suspicions that these experts are being evasive, or even misleading. In a Novem ber 18 interview on the NBC Nightly News, Susan Love candidly ac knowledged, This is rationing but its rationing of the best kind. If the debate about health care reform is going to progress with clarity, transparency, and honesty, we must lose our fear of the R word and discuss how, not whether, we should ration health care.
Financial and other disclosures provided by the authors are available with the full text of this article at NEJM.org. From the Departments of Anesthesia and of Global Health and Social Medicine, Harvard Medical School; and the Division of

Critical Care Medicine, Childrens Hospital Boston both in Boston. This article (10.1056/NEJMp0911447) was published on November 25, 2009, at NEJM.org. 1. Preventive Services Task Force. Screening for breast cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2009;151:716-26. 2. Grady D. Second Opinion quandary with mammograms: get a screening, or just skip it? New York Times. November 3, 2009. 3. Nelson HD, Tyne K, Naik A, Bougatsos C, Chan BK, Humphrey L. Screening for breast cancer: an update for the U.S. Preventive Services Task Force. Ann Intern Med 2009;151:727-37. 4. Ahern CH, Shen Y. Cost-effectiveness analysis of mammography and clinical breast examination strategies: a comparison with current guidelines. Cancer Epidemiol Biomarkers Prev 2009;18:718-25. 5. Crewdson J. Rethinking the mammogram guidelines. The Atlantic. November 19, 2009. (Accessed November 23, 2009, at http:// www.theatlantic.com/doc/200911u/ mammograms.)
Copyright 2009 Massachusetts Medical Society.

Cervical-Cancer Screening New Guidelines and the Balance between Benefits and Harms
George F. Sawaya, M.D. ne of the greatest challeng es in developing cancerscreening guidelines is devising strategies that maximize screen ing benefits and minimize screening harms. The benefits of cancer screening decreased cancer-related morbidity and mortality are well known and widely promoted; the harms of cancer screening receive less at tention and can take many forms: direct complications from screening and confirmatory tests; the expense, anxiety, and life disruptions incurred with new diagnoses with unclear clinical significance; and the prolonged surveillance endured by patients with positive tests

but no evidence of disease. Whereas benefits can be maxi mized by initiating screening early and repeating testing fre quently over a persons entire life span, harms can be minimized by adopting a more focused scope for screening. The tension be tween these two screening goals is apparent and often leads to controversies regarding the age at which to begin cancer screening, the age at which to end screen ing, and the appropriate screen ing interval. New guidelines for cervicalcancer screening issued on No vember 20 by the American College of Obstetricians and Gy necologists (ACOG)1 address all

these controversies (see table). Three major changes have been made that are designed to mini mize screening harms while pre serving high benefit. The approach is in sync with contemporary evi dence-based guideline develop ment, which gives equal atten tion to screening benefits and harms and makes transparent judgments about the balance be tween the two.2 The first major change, and perhaps the most striking, involves the age at which to begin cervical cytologic screening (with the Pap anicolaou, or Pap, smear). The new guidelines state that screening should begin at 21 years of age and that screening should be
2503

n engl j med 361;26 nejm.org december 24, 2009

The New England Journal of Medicine Downloaded from nejm.org on August 28, 2012. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

PERS PE C T IV E

Cervical-Cancer Screening New Guidelines

Cervical Cytologic Screening Guidelines from the American College of Obstetricians and Gynecologists, 2009.
Age Under 21 yr 21 to 29 yr 30 to 65 or 70 yr Between 65 and 70 yr Recommendation for Cytologic Screening Avoid screening Screen every 2 yr May screen every 3 yr* May discontinue screening

* This recommendation applies only to women with three consecutive negative cytologic tests; exceptions include women with human immunodeficiency virus infection, compromised immunity, a history of cervical intraepithelial neoplasia grade 2 or 3, or exposure to diethylstilbestrol in utero. This recommendation applies only to women with three or more consecutive negative cytologic tests and no abnormal tests in the preceding 10 years; exceptions include women with multiple sexual partners.

avoided before that age. The evi dence supporting this recommen dation is compelling. Although cervical cancer is rare before the age of 21, cytologic abnormali ties are common and can lead to labeling, anxiety, extended sur veillance, and invasive procedures, such as colposcopy. If colposcopy is performed, the ACOG guide lines devoted to the management of histologic abnormalities rec ommend restraint in the treat ment of most biopsy-confirmed precancerous lesions identified in young women.3 For example, the most common type of cervical le sion, cervical intraepithelial neo plasia grade 1 (CIN 1), is consid ered a manifestation of acute human papillomavirus (HPV) in fection, and treatment is discour aged. For young women with CIN 2, surveillance rather than treatment can be offered, since spontaneous regression of this le sion is common. The most prox imal cervical-cancer precursor, CIN 3, is rare and may persist for a decade before becoming in vasive. If CIN 3 develops before a woman is 21 years old, screen ing after that age affords multi ple opportunities for such lesions to be detected and treated. Pre
2504

cancerous lesions are often treat ed with excisional procedures; ob servational studies have shown consistent associations between these treatments and adverse pregnancy outcomes, including preterm delivery and low-birthweight infants.4 Since most women under the age of 21 have not yet begun or completed child bearing, these adverse effects were weighted heavily in balanc ing benefits with harms, thereby prompting this unprecedented recommendation. The second major change in volves screening intervals. Where as annual screening has been standard practice for many de cades, the new guidelines state that women with an average level of risk do not require such fre quent testing. Specifically, the guidelines recommend testing ev ery 2 years for women 21 to 29 years of age. Thereafter, screening may be performed every 3 years among women who have had three consecutive negative cyto logic tests. This recommendation is based on evidence showing that among women in this age group, as the number of previous normal tests increases, the likelihood of underlying cervical neoplasia de

creases substantially; continued frequent screening of these lowrisk women is associated with many positive tests and needless interventions but has little effect on the overall incidence of cervi cal cancer.5 The third major change involves the age at which screening should end. Previously, the ACOG had de termined that the evidence was in conclusive and so was not useful in establishing an age at which testing should stop. The new guide lines state that it is now reason able to discontinue screening in women between 65 and 70 years of age who have had three or more consecutive normal tests and no abnormal results within the pre ceding 10 years. This conclusion reiterates that of an ACOG Com mittee Opinion issued in May 2009 and is in agreement with current guidelines published by the Amer ican Cancer Society and the U.S. Preventive Services Task Force. The factors supporting this recom mendation are familiar: screening benefits are small in women at low risk, and harms are incurred when tests are positive. Moreover, coexisting medical conditions that become more common with ad vancing age may increase the risks associated with surgical treat ments. Important areas of uncertain ty remain. The effects that HPV vaccination will have on screen ing initiation and screening in tervals are unknown; immunized women should therefore continue to be screened in the same way that nonimmunized women are screened. It is also unclear what appropriate care consists of for women over 65 who are deemed to be at low risk for cervical neo plasia because of previous normal cytologic test results but who have multiple sexual partners (and pre

n engl j med 361;26 nejm.org december 24, 2009

The New England Journal of Medicine Downloaded from nejm.org on August 28, 2012. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

PE R S PE C T IV E

Cervical-Cancer Screening New Guidelines

sumably new sexual exposures); given the current uncertainties, the new recommendations call for continued routine screening in such women. It is now also sug gested that women with a history of CIN 2 or 3 who have under gone hysterectomy with removal of the cervix continue to be screened a modification of the recommendation put forth in the previous guidelines. A lack of high-quality evidence to support the previous recommendation to discontinue screening after three normal cytologic tests is cited as the basis for this new recommen dation. Modeling studies and in dividual risk calculators may be useful in determining whether the presence of certain risk fac tors increases the risk of cervical neoplasia to a level that justifies continued screening. Although annual cytologic testing is no longer recommended for most women, the guidelines state that physicians should inform women that an annual gynecologic ex amination may still be appro priate. The precise content of this examination is undefined, and more evidence regarding the benefits and harms of this wide spread practice is needed. How should clinicians discuss these changes with patients?

Guidelines promoting a recom mendation to do less are often viewed with suspicion; individu al women may feel as if they are being asked to accept greater personal risk as part of an over all effort to contain costs and con serve resources. Clinicians should inform women that the changes in the guidelines have not been prompted by financial consider ations but by careful consideration of the estimated balance between benefits and harms. Women should be made aware that health recommendations are updated pe riodically as newer, more robust evidence becomes available. Changes are expected to occur over time. Women should also be aware of components of peri odic health examinations that have been identified as providing greater benefits than harms, such as those actively promoted by the U.S. Preventive Services Task Force. Clinicians can have the greatest effect on minimiz ing the harms of screening by taking seriously the recommenda tions to do less screening among women in low-risk groups, espe cially young women. Finally, in terms of screening, clinicians should seek out and offer screen ing to unscreened and poorly screened women, a group that

accounts for at least half of the estimated 11,000 cases of cervi cal cancer that occur annually in the United States.
Financial and other disclosures provided by the author are available with the full text of this article at NEJM.org. From the Department of Obstetrics, Gynecology, and Reproductive Sciences, the Department of Epidemiology and Biostatistics, and the Helen Diller Family Comprehensive Cancer Center all at the University of California, San Francisco. This article (10.1056/NEJMp0911380) was published on November 25, 2009, at NEJM .org. 1. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin number 109, December 2009: cervical cytology screening. Obstet Gynecol 2009;114:140920. 2. Sawaya GF, Guirguis-Blake J, LeFevre M, Harris R, Petitti D. Update on the methods of the U.S. Preventive Services Task Force: estimating certainty and magnitude of net benefit. Ann Intern Med 2007;147:871-5. 3. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin number 99, December 2008: management of abnormal cervical cytology and histology. Obstet Gynecol 2008;112:1419-44. 4. Kyrgiou M, Koliopoulos G, Martin-Hirsch P, Arbyn M, Prendiville W, Paraskevaidis E. Obstetric outcomes after conservative treatment for intraepithelial or early invasive cervical lesions: systematic review and metaanalysis. Lancet 2006;367:489-98. 5. Sawaya GF, McConnell KJ, Kulasingam SL, et al. Risk of cervical cancer associated with extending the interval between cervicalcancer screenings. N Engl J Med 2003;349: 1501-9.
Copyright 2009 Massachusetts Medical Society.

n engl j med 361;26 nejm.org december 24, 2009

2505

The New England Journal of Medicine Downloaded from nejm.org on August 28, 2012. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.