Mammography Screening Interval for Women Older Than 70 Years of Age

Abstract & Commentary

Synopsis: In the experience from a single institution, screening mammography was found to result in the diagnosis of breast cancer in a substantial number of older patients. In many of these, the diagnosis was made despite nonpalpable lesions. When examined by age group it was discovered for those younger than 50 years, mammography was falsely negative 26% of the time, compared to 16% for those between 50 and 70 years and 13% for those older than 70 years of age. For those older than 70, despite an increased number of tumors detected, tumor size was also larger and interval between screening examinations was longer. Annual screening mammography in those older than 70 years of age appears to be justified.

Source: Sener SF, Am Surg 1999;65:731-736.

Screening for breast cancer in older women remains a challenge. Nonetheless, the risk of developing breast cancer increases with each decade and the average age at diagnosis is now 70 years. It is likely that regular mammography would be effective in this age group but the issue has not been carefully studied (the famous large studies on mammography did not include patients older than age 70) and confounding coexisting medical and social issues often preclude regular evaluation. In this report from the Department of Surgery at Evanston Hospital, Northwestern University, a single institution’s experience with breast cancer detected by mammography is reviewed.

A goal of the review was to compare size of tumors detected only by mammography with those detectable by physical exam and to determine if there were any differences with regard to patient age. There were 609 patients with primary breast cancer evaluated during a two-year period (1994-96) within the Evanston Hospital system. Of these, 496 patients (81%) had their screening mammography at that institution and the remainder had mammography elsewhere. The median age of the breast cancer patients was 57 years.

Of the total, 252 (41%) cases were detected by mammography alone. Physical examination alone (with negative mammography) led to the diagnosis in 112 (18%) of the cases. Younger women (< 50 years old) were less likely to have mammographically detected, nonpalpable breast cancer than those in the older age groups. Thus, younger women were more likely to have tumors detected by physical examination alone.

Carcinoma in situ was discovered in 92 (15%), tumors less than 1 cm in 156 cases (25%), tumors between 1 and 5 cm in 328 (54%) cases, and tumors more than 5 cm in 33 (5%) cases. Of the 41% that had cancers detected by mammography alone (i.e., without palpable lesion), as age increased the proportion of those with in situ lesions decreased and the proportion of those with tumors larger than 1 cm increased. In fact, most carcinoma in situ lesions were detected by mammography alone, and this was more common in younger women.

Sener and associates also calculated mammography yield at their institution. During the two-year study period, 49,585 women had mammograms and 496 individual breast cancer patients were diagnosed (1% of those screened). In 212 of these (43%), the lesion was detected by mammogram alone. Thus, there were approximately 10 breast cancer cases (including those with false negative mammograms) per 1000 mammograms performed and 4.27 patients detected by mammography alone per 1000 mammograms. For women older than 70 years, there were 6.05 cases detected by mammography alone per 1000 mammograms. For all cases, including those detected by physical exam and mammography, there were 8.15 cases per 1000 mammograms and for those 70 years and older, there were 11.38 cases per 1000 mammograms.


There appear to be three functional age groups for breast cancer screening by mammography. For those younger than 50 years, screening has been recommended but controversy still abounds.1-3 In this age group, the incidence of new cancers is small and, for technical reasons, sensitivity and specificity are less.4 Therefore, a large fraction of the mammographically detected abnormalities will lead to biopsies that do not detect cancer in this group and women who have annual mammography between the ages of 40 and 50 years have a significant likelihood of coming to a negative biopsy during that period. In the current series, of the 184 breast cancer cases detected, 26% were not visualized by mammography. For the older age groups, the false negative rates are approximately half that seen in the younger age group. Of the 17,905 mammograms performed (in 2 years) in women younger than 50 years, there were 2.51 nonpalpable cancers per 1000 mammograms. In contrast, among the 8258 mammograms performed in women 70 years and older, there were 6.05 nonpalpable breast cancers per 1000 mammograms.

For the second age group, those women older than 50 but younger than 70 years of age, there is consensus that annual mammography is indicated. The rising incidence of breast cancer in this age group and the increased sensitivity of the procedure after menopause have resulted in an appreciation of this procedure as effective screening with life-saving potential.

However, for those women older than 70 years of age, the data remain incomplete. Some feel the procedure is indicated, perhaps on a less frequent interval (e.g., every other year), presumably based upon the more indolent growth of breast tumors in the elderly. Nonetheless, this report raises concern with less frequent screening of women older than age 70. The duration between repeat mammography was longer for those older than 70 years compared to younger groups and the size of the tumor detected was larger. As expected, the number of tumors was also greater. Despite the absence of data from prospective, randomized controlled trials, it would appear from the available data that annual mammographic screening should continue past the age of 70 years. Mammograms are both more sensitive and more specific in this group and breast cancer incidence is highest in this group. If mammography works in women older than age 50, it should be even more effective in women older than age 70. Thus, despite the absence of "official" recommendations by the American Cancer Society or the National Cancer Institute, the available data suggest it would be wise to continue annual mammography in women over age 70 years. There is no precedent for any screening test to become less effective despite increased sensitivity, specificity, and incidence in the target population.

The question of the optimal interval for screening mammography in older women needs to be determined. However, short of a definitive study, data from reports such as this would support annual screening for women in this age group. Nevertheless, physicians and other providers are left with the challenge of overcoming the multiple inherent obstacles existent in geriatric populations and the institution of an accepted screening program.5

Is there an age at which annual mammographic breast screening should cease? Unfortunately, there are no data. However, as the life expectancy of the population increases, we shall be forced to reexamine our notions about how close to the end of her life an 80-year-old woman actually is. Demographic data have documented that the death rate of older people actually declines! Living through the period when the most common maladies take their toll seems to identify a group more hardy than the average person. In 1997, the life expectancy of an average 80-year-old woman was almost nine years. Thus, in the absence of an active acute illness with short-term mortality as the expected outcome, a case can be made for continuing mammographic screening indefinitely. The issue really should be examined in a large-scale clinical trial.


1. American Medical Association, Council on Scientific Affairs. JAMA 1989;261:2535-2542.

2. American Cancer Society Mammography Guidelines. CA Cancer J Clin 1983;33:255-259.

3. US Preventive Services Task Force. Guide to Clinical Preventive Services, 2nd ed. Alexandria, VA: International Medical Publishing; 1996.

4. NIH Consensus Statement, J Natl Cancer Inst 1997; 89:1015-1120.

5. Blustein J, et al. J Am Geriatr Soc 1998;46: 941-946.

With regard to breast cancer screening in women older than the age of 70 years, which of the following statements is true?

a. Screening is less accurate when compared to younger women and false-negative studies occur more frequently.

b. Screening is technically comparable to other age groups and the false-negative rate is less than that observed for studies in women younger than the age of 50 years.

c. Screening intervals should be longer in older vs. younger individuals due to the slow growth rate of tumors in older women.

d. Screening results in the appearance of smaller tumors in older women when compared to younger age groups.

e. Screening should not be undertaken in women older than age 70 years.