Medical researchers have begun to amass data showing that abortion may actually be harmful to women’s health. Studies upon studies are finding there is a link between women who have had abortions and an increases risk of breast cancer. Why aren’t alarm bells ringing? Whose interests are being protected?

The Interim

Because this article is very technical, it will be necessary to explain some terms. Relative Risk (RR) and Odds Ration (OR) also called ‘elevated risk,’ in this context refer to the risk of getting the disease of breast cancer.

The first study to show a significant association between induced abortion and breast cancer was done in 1957 in Japan, (RR=2.6). Another large age-matched Japanese study, 1982, showed the risk to rise with the number of induced abortions (RR=2.5 for one abortion rising to 4.9 for four or more abortions).

In 1981, Pike et al reported that a first trimester abortion before a first full term pregnancy, whether spontaneous or induced, was associated with 2 to 4 fold increase in breast cancer risk (2). In 1986, another study by O.C. Hadjimichael et al, (3) showed that abortion prior to the first live birth was associated with a 3.5 fold increase in the risk of breast cancer. The elevation of risk was independent of some of the major other risks of breast cancer and became more pronounced as the number of years since the abortion increased.

In 1980, Russo et al (4), did experimental work in rats which suggested that full-term pregnancy, with or without lactation, reduced the susceptibility of these animals to benign lesions and carcinomas by means of induction of full differentiation of the mammary gland. They also found that pregnancy interruption prevents sufficient differentiation in the gland to be protective. They reported that 77% of their experimental animals developed carcinomas and that all of them developed benign lesions.

They saw a parallel between their animal model and the human experience since in women, during gestation, there is an increased secretion of estrogen, progesterone and prolactin, which together leads to incomplete development of the mammary gland which may render it susceptible to carcinogenesis.

Vessey et al, 1982 (5) and Rosenberg et al, 1988 (6) found no association between first trimester abortion, spontaneous or induced, and breast cancer. However, Rosenberg et al did not further divide their cases into abortions done very early in the reproductive life from others. This would seriously alter the significance of the results. Kvale et al, 1987 (7) showed as apparent protective affect against breast cancer in groups with an early first birth but not in women with their first birth at age 35 or later. M. Ewertz and S.W. Duffy, 1988 (8) found an increased relative risk of breast cancer in women who were never pregnant and in those who had an early terminated first pregnancy.

The study confirmed that pregnancies must continue to term to offer protection against breast cancer. On the other hand, La Vecchia et al, 1987 (9) found little relation of breast cancer risk with abortions or miscarriages. A study by Britt Marie Lindefors Harris et al (10) 1989 did not indicate an overall increases risk of breast cancer after induced abortion in the first trimester in young women, (younger than 30 years at the time of abortion).

Holly Howe et al (11) 1989, found that matched pair analysis of 1,451 cases showed an increased risk of cancer in patients who had had induced or spontaneous abortion of the first pregnancy. For induced abortion the odds ratio (OR) = 1.9: for spontaneous abortion (OR) = 1.5. Elevated risk (OR) = 4.5 was associated with a history of repeated interrupted pregnancies with no intervening live births. Pair matching was achieved by a computer linkage of the identity of the mother and that of the aborted fetus.

Larissa I. Remennick (12), 1989, reported in a general study of Russian data that the most important statistical determinants of the regional variants of both breast cancer and cervical cancer in the USSR are low parity and high prevalence of induced abortions. Induced abortions seemed to be the only factor common to both sites of cancer. She also said (13) that there are 2 approaches to studies of induced abortion and cancer:

  1. To focus on pregnancy per se. Five or six interrupted pregnancies are deemed equivalent to one full-term pregnancy. Terminated pregnancies are considered protective or neutral.
  2. To focus on termination with concomitant hormonal and immunological stress, incomplete differentiation and growth of breast tissues.

In the opinion of Dr. Remennick, the second approach is more plausible biologically and is backed up by certain clinical and biological data.

The National Cancer Institute (NCI) study 1994, (14) stated that an induced abortion in the last month of the first trimester is associated with nearly a doubling of subsequent breast cancer risk. They found no increased risk of breast cancer was associated with spontaneous abortion. Among women who had been pregnant at least once, the risk of breast cancer in those who had experienced an induced abortion was 50% higher than among other women.

They also said that because all epidemiological studies regarding the hypothesis that an induced abortion can adversely influence a woman’s subsequent risk of breast cancer—the risk of breast cancer should be examined in future studies of women who have legal abortion available to them throughout the majority of their reproductive years, with particular attention to the potential influence of induced abortion in early life.

Matti A. Rookus et al (15) in the Netherlands found the use of oral contraceptives for four or more years was associated with double the risk of breast cancer found with shorter use. Shorter use, involved young women less than 36 years of age at diagnosis. Long term use (twelve or more years) was also associated with double the risk of breast cancer (related to no use), at age 45 to 54. They also raised the possibility that the use of low dose oral contraceptives in general may increase the cancer risk. M. Ewertz. (16), 1992.

The National Cancer Institute article provoked an uproar not only in the media but in the professional medical press. As I have pointed out, the scientific literature is not unanimous, and we must be cautious not to foment public fears on inadequate grounds. Breast cancer is potentially lethal, and has a high incidence, striking 12% of American women (over 1.5 million a year). Even if the modest NCI study is confirmed, that would result in 40,000 to 50,000 additional cases of breast cancer a year in the U.S. alone. Canadian figures are usually 10% of U.S. figures.

This posited risk factor is almost always the result of personal choice, and is avoidable in a way environmental risk may not be. (17)

The NCI study (by Daling) was attacked by an editorial of the NCI Journal in the same issue which had published the study written by Lynn Rosenberg who had done a study of her own (Ref.6). Her own work was seriously flawed because the breast cancer patients in her study were, on average, 12 years older than the cancer-free control patients. Since the risk of cancer increases with age, this was an egregious methodological error. (18)

The American Journal of Epidemiology, which had published Rosenberg’s article in 1988 (ref. 6) declined to publish the age-matched study of New York state women by Dr. Holly Howe et al of New York State’s own dept. of health (ref. 11). The Howe study was finally published in the English International Journal of Epidemiology, 1989.

The idea was put about in a Swedish study in the American Journal of Epidemiology in 1991 that “response bias” explained the tendency towards increased cancer risk associated with induced abortion. (Lindeford-Harris et al Response Bias in a Case-control Study: American Journal of Epidemiology. 299. 1430, 1989).

The theory is that cancer-free women are more likely to deny abortions they had while women with breast cancer are more likely to report their abortion history accurately. The only evidence to support this hypothesis was that cancer patients reported abortions of which the computer had no record. The Swedish study, 1991, had based its assertion that patients tend to “over report abortions,” that is, imagine abortions they never had. This evidence is absurd. The NCI study destroys this bias theory. (19)

False information was provided by Dr. Devra Lee Davis, Senior Advisor to the U.S. Assistant Secretary of Health. She placed side by side the long standing legality of abortion in Japan, with the fact that Japan has “the world’s lowest breast cancer rate.” This despite the fact that all four Japanese epidemiological studies show a higher incidence of breast cancer among the small minority of Japanese women who actually have had any abortions. (20)

Some discussion of the methodology of these abortion research projects is necessary at this point. There are two types of studies, the first, the retrospective case—control type. Hundreds or thousands of cases are compared, which are similar except for the fact that one group has breast cancer and the other, or control group, has not. A relative risk (rr) is calculated. RR = 2 means that having an abortion doubles the risk of getting the disease, as previously explained. RR less than 1, indicates a protective effect. Published RR values are usually corrected or adjusted for other factors affecting risk that controls are not matched for. In the second type of study there is good case control matching and sufficient numbers to make the study statistically significant. Poor matching reduces the ability of the study to show up small RR. The study authors may conclude that their postulated risk factor is not significant. Whereas the risk factor in a study with closer matching and sufficient numbers may make the same factor statistically significant. No RR above 1 can be regarded as insignificant in a disease like breast cancer because its incidence is so great.

Dr. Joel Brind, a professor of biology and endocrinology who works from New York University and has been studying this phenomenon since 1992, has pointed out that there is an ample opportunity for researchers with a pro-abortion bias to design studies of deliberately low statistical power. (21) “…This trick is usually performed through inappropriate age matching or adjusting…The incidence of breast cancer increases by 30% per year in women between 30 and 35 years old…a randomly selected group of patients in that age range might have a median age of about 34, whereas a randomly selected group of controls might have a median age of about 33, thus, matching and adjusting for age, in 5 year intervals 9 instead of matching within 1 month), an increase in RR (1.3) shows up as no risk increase (RR = 1.0)” (22)

“Dr. Vessey’s study, 1982 (ref, 5) had almost no patients in their study of the same age or with the same abortion history. Similarly Drs. Ewertz and Duffy in their study of 1988 (ref. 8) found an RR of 1.2 and 3.8 for parous and mulliparous women, and Rosenberg et al, 1988, found an RR of 1.3 for nulliparous women, for women under 40, RR = 1.4 for any abortion history (1.7 for those with more than one abortion)—this despite an egregious age mismatch that permitted the data to avoid statistical significance and the authors to ‘suggest that the risk of breast cancer is not materially affected by abortion.’

The previously mentioned study by Dr. Lindefors-Harris, 1991 (ref. 10) tried to dismiss an entire body of case-control evidence with the notion that cancer patients remember and report more abortions that controls and are likely to imagine abortions that never happened! In their 1989 study, they actually misinterpreted the protective effect of having children as a protective effect of having an abortion.”

In 1993, the African-American Journal of the National Medical Association was the first to breach the wall of silence of the American Medical Association. Amelia Laing et al noted in their age matched Howard University study of over a thousand black women, “Breast cancer is the leading cause of cancer mortality in black women. Black women are more heavily represented among abortion clients. Their RR = 2.7 for the cancer of the breast and rose to 4.7 in the 50 and over group. The popular press did not publish this.” (ref. 17)

The American Medical Association attacks the theory that abortion increases the risk of breast cancer. The Journal of the American Medical Association (JAMA), Jan., 1996 published a study by Polly A. Newcomb et al (23) which showed a weak positive association between abortion and the risk of breast cancer.

They conceded that an increased risk of cancer was “somewhat greater among women with a history of induced terminations.” However, they said “this association may be due to reporting bias and was not significantly different from the slight risk for spontaneous terminations.”

The dubious Swedish study quoted above was used by Newcomb to buttress her own notion that persons who think they don’t have cancer will not confess to having had an abortion because abortion is, in her words, “an extremely sensitive procedure.”

The Swedish data provided no evidence that the information supplied was biased. Perhaps the researchers projected their own bias in the matter on to the control group. Despite the fact that they acknowledged that abortions nowadays are at less than 8 weeks gestation, most often, and that the evidence in the literature suggests that abortion in the first trimester has the highest correlation to subsequent breast cancer. Newcomb et al (24) chose to define abortion as any pregnancy lasting less than 6 months. The fact that the protective effect of pregnancy might have cancelled out the carcinogenic potentiality of an early abortion seems not to have crossed their minds, or perhaps was deliberately concealed.

In the same issue of JAMA the editorial itself shows bias. It approves of studies which indicate a low risk of cancer with abortion. The “modest” 10% to 50 % correlation of the NCI study was labeled statistically insignificant. For the JAMA, 40-50,000 extra cases of breast cancer do not merit their concern.

Dr. Brind has told us that the most avoidable risk factor of breast cancer is induced abortion. There is an increasingly obvious effort to prevent the public from getting to know this. The motivation is clear. The notion of abortion as safe for women is of paramount importance to the pro-abortion claque.

The American Medical Association vociferously proclaims that the risk of a mother’s dying in childbirth is 12 times greater than the risk of dying from an abortion. The fact is that the risk of dying in childbirth is less than 5 per 100,000. If the over-all increase in breast cancer risk were even the “modest 10-50% increase (to quote JAMA), it would raise the life time risk of cancer from 12% to 18% an increased incidence of 6,000 per 100,000 women. Even with a breast cancer cure rate of 75%, the increase in the death rate from induced abortion calculates out to 1,500 per 100,000, making abortion 300 times more likely to result in the death of a woman than childbirth.” (25) Mathematically correct statistics have more statistical power, but perhaps less political power than those which are politically correct!

The evidence suggests that abortion, the use of oral contraceptives, and even delaying pregnancy, all increase the risk of cancer in the breast. This raises serious questions about a lifestyle many young women today adopt, in a society which has come to accept as normal pre-marital sexual activity and the use of the contraceptive pill. Women who contemplate careers involving long, arduous years of training, which will demand all of their energies, and who intend to delay pregnancy because of this, would do well to ponder these facts.

Why is abortion being so assiduously protected?

Part of the reason is that political, medical, legal and media establishments are committed to the idea that the most fundamental and urgent problem we face is world over population. They have proved willing to abandon principle in the interest of population control. As the story unfolds, we can expect further manipulative policies to emanate from our bureaucracies, our government, non-governmental organizations of all kinds and the United Nations Organization.

John Shea, M.D., is the former Chief of the Department of Radiology at Centenary Hospital in Scarborough, Ont.


  1. Joel Brind. Life Advocate, 1993.
  2. M.C. Pike et al. British Journal of Cancer 43.72,1981.
  3. O.C. Hadjimichael et al, 1986, Br. J. Cancer.
  4. J.R. Russo and I.H. Russo, The American Journal of Pathology, 100. p. 497. 1980
  5. M.P. Vessey et al, Br. J. Cancer, 1982, 45. 327
  6. L. Rosenberg et al, American Journal of Epidemiology, 1988, 127. 981
  7. Gunmar Kvale et al, Am. J. of Epidemiol. 1987, 126, 831 and 834 No. 5
  8. M. Ewertz and S.W. Duffy, Br. Med. J. 1988, 58, 99
  9. C. La Vecchia et al, Int. Journal of Epidemiol. 1987, 16.3
  10. Lindefors Harris et al, Br. Med. J. 1989, 299. 1430
  11. Holly L. Howe et al, Int. J. of Epidemiol. 1989, 18, No. 2.
  12. Larissa Remennick, Reproductive Patterns and Cancer Incidence in Women: a Population based correlation Study in the USSR. Int. J. of Epidemiol., 1989
  13. Larissa Remennick, Induced Abortion: Review of Epidemiological Evidence. J. of Epidemiol. Community Health, 1990
  14. Janet R. Daling et al, J. of national Cancer Institute, 1994 86, No. 21
  15. Matti A. Rookus et al, The Lancet, 1994, Vol 344 Sept. 24
  16. M. Ewertz, European J. Cancer, 1992, 28A :1176-81
  17. Joel Brind, National Review, 1995, Dec. 38-41
  18. Ibid
  19. Ibid
  20. Bid
  21. Joel Brind, Life Advocate, Feb. 1993 Ref.
  22. Ibid
  23. Polly A. Newcomb et al, J.A.M.A. Jan 24/31, 1996 Vol. 275, No. 4 Ref.
  24. Ibid

Joel Brind, Life Advocate, Feb. 1993