Edited to add: Births worldwide per year is 137 million infants. Abortifacient deaths from IUDs are 1 to 2 losses per woman per year [Mechanisms of action of intrauterine devices; Am J Obstet Gynecol 2002;187:1699-70]. Worldwide, 163 million women use IUDs, resulting in as many as 163 to 326 million abortifacient deaths per year. (The entire U.S. population is only 326 million persons.) And then we have to add deaths from oral, injected, or implanted contraceptives at about 50.4 million, and induced abortions at 55.9 million (106.3 million more). The grand total is 269.3 to 432.3 million deaths from abortifacients and abortion.
The birth control pill is abortifacient contraception. Most of the time, the pill works by preventing ovulation, thereby preventing pregnancy. However, sometimes the contraceptive action of the pill fails, and ovulation occurs. This is termed “break-through ovulation”. The approximate rate of break-through ovulation is 20% per month. [1, 2] When break-through ovulation occurs, the woman ovulates, and conception becomes possible.
That figure of 20% per cycle for the break-through ovulation rate is an average. Some oral contraceptives have a higher rate. And these rates are based on “perfect use” — the case where the woman takes the pill daily, at the same time of day, never misses a dose, and does not do any of the things that can make the pill less effective (taking antacids, taking antibiotics, having a stressful disruptive daily schedule). When there are dosing errors or other disruptive factors, the break-through ovulation jumps higher.
“Dosing errors have the potential to reduce the efficacy of a hormonal contraceptive agent. Missing one or more pills per cycle has been estimated to result in a 2.6-fold greater risk of unintended pregnancy when compared with women who use OCs correctly.” 
But even at 20%, oral contraceptives are very ineffective as a method of contraception. Condoms have a failure rate of 18% per year, not per month. The effectiveness of chemical contraceptives comes mainly from its ability to abort conceived prenatals at a very early stage of development, prior to implantation.
What is the chance of conception when any form of contraception fails? It is the rate of conception without any contraceptive method times the failure rate of the method. The chance of pregnancy per month with regular sex for couples under 32 year of age is about 20%. [Source]
A woman under 32 years of age, using the BCP, has a 4% chance per month of conceiving a child: The 20% rate of break-through ovulation times the 20% chance of conception equals 4% (0.2 x 0.2 = 0.04). That works out to a 48% chance per year of conception. However, the BCP only has a 9% published failure rate, because when the contraceptive action fails, the pill works as an abortifacient. [CDC.gov] Therefore, about 39% of the time per year [48% minus 9%], the BCP mainly works by killing a conceived prenatal. For every 100 women using the BCP, under 32 years, there will be 39 prenatal deaths per year due to the abortifacient action of the pill.
For women from 32 to 49, using the BCP, the chance of conception, when break-through ovulation occurs, is 10 to 15%. At an average of 12.5%, with a 20% per month rate of break-through ovulation, there is a 2.5% chance of conception per month, which is 30% per year. Should we subtract the 9% of live births that occur despite the use of BCP? No, the decreased fertility also decreases the 9%, by about the same as the chance of conception, that is, to 5.6%. So when we subtract the 5.6% from 30%, the rate of abortifacient action per year, for older women, is 24.4%. For every 100 women using the BCP, 32 years or older, there will be about 24 prenatal deaths per year due to the abortifacient action of the pill.
In the U.S. there are about 54 million women aged 20 to 32 and about 71 million women aged 33 to 49. [Census.gov, 2014 numbers]. If we weight the above break-through ovulation numbers by age, 43.2% of women of fertile age are in the first category (39% rate for abortifacient action), and the rest, 56.8%, are in the second category (24.4%). The weighted average is then 30.27%.
About 10 million women use the BCP [Guttmacher.org]. This results in about 3,027,000 deaths from abortifacients per year. Actually, that number is very conservative, since the break-through rates discussed earlier assume “perfect use” of the contraceptive. Break-through rates can go much higher, when dosing errors are made by the woman taking the pill. For that reason, the number of deaths from abortifacients in the U.S. each year certainly exceeds 3 million.
The number of surgical or medical abortions in 2017 in the U.S. was about 919,800. This number is derived from the number of women in the U.S. ages 15 to 44 (63 million) times the abortion rate of 1.46 for 2017. [Source] So the rate of death by abortifacients is about 3 times higher than by surgical or medical abortion.
But when we add the deaths from abortifacients and surgical/medical abortions together, the estimate is 3,946,800. How does this compare to the birth rate? “There were 3,853,472 births in the U.S. in 2017.” [Source] So the number of unborn children killed by abortifacients plus abortion is approximated the same, or possibly a little higher than the number of live births.
It’s a balance. We are killing half of our children in the womb. (Thanos would be proud.)
The above numbers are approximate. However, the break-through ovulation rates used above are very conservative, as are the estimates for the number of unborn children killed by abortifacients. What is apparent from those numbers is that abortifacient contraception represents a danger to prenatal lives that is greater than the danger from abortion. Whatever the number of lives lost to abortion is, a multiple of that number additionally die from abortifacients.
When Catholic priests, theologians, apologists and online commentators approve of the use of abortifacient contraception while sexually active, regardless of the reasoning, many prenatals lose their lives as a result. The change of break-through ovulation is not small, and the change that the BCP will work as an abortifacient, rather than as a contraceptive is substantial.
Note: In her latest book on contraception (Self-Gift: Humanae Vitae and the Thought of John Paul II), professor of Catholic ethics Janet E. Smith couldn’t be bothered to even mention in passing that oral contraceptives are abortifacient. She terms them “infertility-causing hormones”. She proposes that they are less sinful than condom use. She justifies their use for a medical purpose, without considering the moral weight of the deaths of prenatals. She bemoans the effect that chemical contraceptives are having on the environment, on fish and on amphibians. She also claims that “hormonal contraceptives use may alter women’s ability to attract a mate” [Kindle Location 8335]. But no mention of the millions of innocents who are killed by the use of abortifacient contraception. Not a word.
Abortion is genocide. It is a modern-day holocaust, which has so far killed over a billion, perhaps as many as 2 billion unborn persons — not counting those killed by abortifacients, which could be a higher number. In the U.S., about three times as many prenatals die from abortifacients than from abortion. (Worldwide, the numbers seem to be only a little higher for the abortifacients.) Catholic pro-lifers who ignore or even support the use of oral contraceptives are a part of the problem. They argue for an end to abortion, but not an end to abortifacients. That’s like arguing for saving only one out of every four endangered innocent lives. Ignoring the prenatals who die from abortifacient contraception is unethical.
Can abortifacient contraception be used in cases of rape? Mere contraception can be used in cases of rape, because its use is “indirect”, not direct, and therefore is not intrinsically evil. But what about abortifacient contraception? Chemical contraceptives may be used in cases of rape, if the physician or woman can be morally certain that the intervention will work as a contraceptive, and not as an abortifacient. A physician, who is aware of where the woman may be in her cycle, might be able to ascertain that the chemical contraceptive will not work as an abortifacient for the one cycle at issue. And that is all that is needed when treating most rape victims. The chance of conception from the rape occurs within 5 or 6 days of intercourse.
However, and this next point is very important, the use of Emergency Contraception, i.e. single or double dose chemical contraceptives (e.g. LNG or UPA), are almost always abortifacient when used as emergency contraception. So this type of use would be gravely immoral. The aforemented use would be ordinary oral contraceptives (or the patch) used in the usual manner, which then has a very low risk of working as an abortifacient.
[updated 3 August 2018]
 Gardner, Ph.D. and Miller, M.D., Journal of Women’s Health, Vol. 14, n. 1, 2005 PDF]
 Pierson, et al., “Ortho Evra/Evra versus oral contraceptives”, Fertility and Sterility, July, 2003, 80:1 [PDF]
Please take a look at this list of my books and booklets, and see if any topic interests you.