Current issues of ACP Journal Club are published in Annals of Internal Medicine


Induced abortion did not increase the risk for breast cancer

ACP J Club. 1997 Jul-Aug;127:18. doi:10.7326/ACPJC-1997-127-1-018

Source Citation

Melbye M, Wohlfahrt J, Olsen JH, et al. Induced abortion and the risk of breast cancer. N Engl J Med. 1997 Jan 9;336:81-5.



To determine whether an association exists between induced abortion and the risk for breast cancer.


Cohort study of women followed from birth through 1992.


Population-based study in Denmark.


All Danish women born from 1 April 1935 through 31 March 1978. Overall, 1 529 512 women were included in the cohort.

Assessment of risk factors

Through linkage with the National Registry of Induced Abortions, information on the number and dates of induced abortions was combined with information on the gestational age of each aborted fetus. The induced abortions included in this analysis were those that occurred between 1973 and 1992. Data on the obstetric history (age, parity, age at delivery of first child) were taken from the cohort database.

Main outcome measures

New cases of breast cancer identified through linkage with the Danish Cancer Registry. Follow-up for breast cancer began on 1 April 1968 or on the woman's 12th birthday, whichever came later.

Main results

Of the more than 1.5 million women in the cohort, 280 965 had a total of 370 715 induced abortions and 10 246 developed breast cancer. After adjustment for age, parity, age at delivery of first child, and calendar period, the overall risk for breast cancer in women with a history of induced abortion did not differ from that of women without such a history (relative risk [RR] 1.00, 95% CI 0.94 to 1.06). No increases in the risk for breast cancer were found in subgroups of women defined according to age at induced abortion, parity, number of abortions, time since abortion, or age at diagnosis of breast cancer. The RR for breast cancer did increase with increasing gestational age of the fetus. With each 1-week increase in the gestational age, a 3% increase occurred in the risk for breast cancer. The RR increased from 0.81 (CI 0.58 to 1.13) among women whose most recent induced abortion was at < 7 weeks of gestation to 1.38 (CI 1.00 to 1.90) among women whose most recent abortion was at > 12 weeks of gestation (reference category, induced abortion at 9 to 10 weeks of gestation).


History of induced abortion did not increase the risk for breast cancer.

Sources of funding: Danish Cancer Society; Danish National Research Foundation; U.S. Department of Defense.

For article reprint: Dr. M. Melbye, Department of Epidemiology Research, Danish Epidemiology Science Center, Statens Serum Institut, 5 Artillerivej, DK-2300 Copenhagen S, Denmark. FAX 45-32-68-31-65.


Spontaneous abortions have not been associated with an increased risk for breast cancer. Whether induced abortions are associated is controversial. The results of case-control studies and reviews of the evidence have yielded variable results, with some showing an increased risk (1, 2). Most of the increased risk can be accounted for by different types of biases, including differential recall between case and control participants (3, 4). Cohort studies have found no increased risk for breast cancer.

The cohort study by Melbye and colleagues adds more information to this controversial area. Mandatory reporting of induced abortions and linkage to a breast cancer registry strengthens the validity of the conclusions. Further, the number of abortions was not associated with an increased risk for breast cancer, which might be anticipated if a dose-response relation existed.

Despite the strong study design, some methodologic limitations exist. Reporting of induced abortions has been mandatory in Denmark since 1939. By limiting the inclusion of induced abortions from 1973 to 1992 (when computerized data were available) and assuming that most induced abortions are done in women < 30 years old (data not given), the follow-up of this relatively young cohort of women with induced abortions is short. Breast cancer is predominantly a disease of older women. It is possible that the risk (if any) for breast cancer with induced abortions may manifest itself years or decades later. The authors of this study did not take into account other important risk factors, such as family history.

Although this study represents a step forward, the question remains open. One can, however, say with increasing confidence that induced abortion is not associated with an increased risk for breast cancer, at least in the short term. Further studies are needed to address the issue of long-term risk.

Edmond E. Chouinard, MD
Hamilton Regional Cancer CentreHamilton, Ontario, Canada

Edmond E. Chouinard, MD
Hamilton Regional Cancer Centre
Hamilton, Ontario, Canada


1. Michels KB, Willett WC. Epidemiology. 1996;7:521-8.

2. Brind J, Chinchilli VM, Severs WB, Summy-Long J. J Epidemiol Community Health. 1996;50:481-96.

3. Weed DL, Kramer BS. J Natl Cancer Inst. 1996;23:1698-700.

4. Rookus MA, van Leeuwen FE. J Natl Cancer Inst. 1996;23:1759-64.