pageok
pageok
pageok
Pre-1966 Deaths from Illegal Abortion:

An April 24 article in the Rocky Mountain News states:

The University of California School of Public Health estimated that before 1966, an estimated 5,000 to 10,000 women died each year in the U.S. from complications of illegal abortions.
Trying to find out more about this study, I found that it was cited in a 1966 book by Lawrence Lader, Abortion. (See note 21 here for a secondary citation.)

Do any readers have additional information about this study, or know of additional research on the levels of pre-1967 maternal deaths from illegal abortion in the U.S.

Please confine your comments to this factual issue, and do not argue the broader pro/con merits of the abortion question.

redlightgrnlight:
I have a citation, but I'm unsure about its accuracy given that I've never examined it. It cites "800 to 10,000 deaths a year." U.S. Senate, Committee on Governmental Operations, Subcommittee on Foreign Aid Expenditures, Hearings on S.1676, Eighty-Ninth Congress, Second Session, 1966, Part 5-B, 1385-89.
5.1.2007 10:26pm
cathyf:
Another interesting statistic is the number of maternal deaths associated with legal abortions in the same time period.
5.1.2007 10:34pm
U.Va. 2L:
Another interesting statistic is the number of maternal deaths associated with legal abortions in the same time period.

After 1900 and before 1967, wasn't it illegal in every single state with only a few life-of-the-mother exceptions? I'm not sure the sample size would be meaningful.
5.1.2007 10:45pm
WonkoTheSane (mail):
Impressive. It only took two posts for someone to disregard the poster's request regarding keeping comments on-topic.
5.1.2007 10:48pm
JonC:
This 5,000 deats/year from illegal abortions figure apparently dates back to 1936, long before the widespread use of modern antibiotics. This FactCheck.org article discusses the pre-1966 5,000 deaths/year claim, and other relevant research, in detail.

Also, on p.73 of The Party of Death, Ramesh Ponnuru writes: "The National Center for Health Statistics attributes 1,313 women's deaths to illegal abortions in the year 1940. That number had declined to 159 by 1966, and to forty-one by 1972." It's not exactly clear what study or reports those figure came from, though.
5.1.2007 11:27pm
Andrew Hyman (mail) (www):
I grabbed the following information off Wikipedia. From 1940 through 1970, annual U.S. deaths of pregnant women during abortion fell from nearly 1,500 to a little over 100. See "Induced termination of pregnancy before and after Roe v. Wade" JAMA, 12/9/92, vol. 208, no. 22, p. 3231-3239. According to the Centers for Disease Control, the number of women who died in 1972 from illegal abortion was thirty-nine (39). See Lilo T. Strauss, et al. Abortion Surveillance --- United States, 2001 (Table 19). In 1960, Dr. Mary Calderone, a former director of Planned Parenthood, said:

Abortion is no longer a dangerous procedure. This applies not just to therapeutic abortions as performed in hospitals but also to so-called illegal abortions as done by physician. In 1957 there were only 260 deaths in the whole country attributed to abortions of any kind....90 percent of all illegal abortions are presently being done by physicians....Whatever trouble arises usually arises from self-induced abortions, which comprise approximately 8 percent, or with the very small percentage that go to some kind of non-medical abortionist....

Mary S. Calderone, M.D.: "Illegal Abortion as a Public Health Problem." American Journal of Public Health, 50:948, 1960.
5.1.2007 11:35pm
Anderson (mail) (www):
In 1957 there were only 260 deaths in the whole country attributed to abortions of any kind....90 percent of all illegal abortions are presently being done by physicians....

Interesting, but how many deaths from illegal abortions were never reported or diagnosed as such?

Remember that the "underclass" would figure disproportionately in that figure, and compare the situation to reports of rape.
5.1.2007 11:43pm
William Oliver (mail) (www):
It's a completely bogus statistic. As a reality check, the total number of abortion-related deaths in the world was about 70,000 the last time I looked. Before 1972, no rigorous surveillance data separated legal from illegal abortion-releated mortality. In 1966, the total combined maternal mortality from illegal, legal and spontaneous abortions was around 191 (I have read it as high as 250, but when I looked at the data myself, I could only find 191).

See:
The 1966 Vital Statistics report


The CDC didn't start doing surveillance until 1972, and there were no standard reporting criteria in 1966 — the CDC didn't develop criteria until 1978. According to the CDC:


Since CDC's surveillance of abortion mortality began in 1972, the annual number of deaths associated with legal induced abortion has decreased by 54% (as of 1991) (31). In 1972, 63 women died as a result of induced abortion. Of those deaths, 24 were associated with legal abortion and 39 with illegal abortion. In 1991, 12 women died as a result of induced abortion: 11 of these deaths were associated with legal abortion and one with illegal abortion. The case-fatality rate decreased approximately 80% between 1972 (4.1 deaths per 100,000 legal induced abortions) and 1991 (0.8 deaths per 100,000 legal induced abortions). These rates, consistent with previously published data for the 1970s and mid-1980s (32-34) indicated that the risk for death from legal induced abortion continues to be extremely low.



Similarly in another review:


The CDC abortion mortality surveillance system documented the profound public health effects of the availability of legal abortion. The shift from illegal to legal abortions dramatically reduced the number of women dying from abortion.[18] After 1970, the overall abortion mortality rapidly declined and generally continued declining through 1976 (Figure 3). Nearly all of this decline occurred in the illegal abortion category; the number of deaths decreased from 39 in 1972 to 2 in 1976 and remained at this level thereafter. The reduction in illegal abortion deaths had a distinct temporal association with the increasing availability of legal abortion.[19] In contrast, the opposite trend occurred in Romania when abortion was made inaccessible. [20]

(
Willard Cates Jr et al. Abortion surveillance at CDC
Creating public health light out of political heat
American Journal of Preventive Medicine
Volume 19, Issue 1, Supplement 1, July 2000, Pages 12-17)

The decline in deaths in the 70s and 80s was dramatic, but in the US it meant a decrease of a couple of hundred deaths a year. Nothing to sneeze at, but not the genocide pro-abortion folk like to pretend. The primary cause of the drop is the introduction of broad spectrum antibiotics (most deaths due to illegal abortions involve sepsis, not exsanguination) as much as anything else.
5.1.2007 11:49pm
JB:
I agree with Anderson. There was probably woeful underreporting, sufficient to make any statistic meaningless.
5.1.2007 11:54pm
CaptDMO:
Estimated, unreported, assumed, attributed, underclass, in Mexico, in Canada, in Europe, reportedly, deduced, educed,-

Do any readers have additional information about this study, or know of additional research on the levels of pre-1967 maternal deaths from illegal abortion in the U.S.

In My Humble Opinion-
Andrew Hyman, "I grabbed the following information off Wikipedia." an appropriate disclaimer!

Carry on!
5.1.2007 11:55pm
William Oliver (mail) (www):
"Interesting, but how many deaths from illegal abortions were never reported or diagnosed as such?

Remember that the "underclass" would figure disproportionately in that figure, and compare the situation to reports of rape."

The analogy is not a good one. A pregnancy-related death is not easily misdiagnosed. It's not something you "hide" the way you would hide rape.
5.1.2007 11:58pm
William Oliver (mail) (www):
"I agree with Anderson. There was probably woeful underreporting, sufficient to make any statistic meaningless."

No. The pre-1972 statistics are for all abortion-related deaths, including spontaneous abortion. Illegal abortions might have been misreported as spontaneous abortions, but they would not have been reported as, say, gunshot wounds. If an illegal abortion was reported as a spontaneous abortion instead, then it would have still been included in that statistic.

Wheat do you believe the death would be reported as? Car crash?
5.2.2007 12:02am
TWL (mail):
Why would there be significant underreporting?

Were the dead women fearfull of prosecution?

Can a doctor tell when the cause of death was "botched abortion"? Assuming these women underwent any kind of post-mortem examination - hospital, morgue, whatever - wouldn't that cause of death be reported?

Am I completely off base? Is cause of death frequently self-reported . . . ?
5.2.2007 12:06am
Dr. T (mail) (www):
Even with underreporting, the 5-10 thousand estimate seems far too high. If we assume an illegal abortion rate of one million per year (which probably is much higher than the actual rate), the high end figure means that 1% of illegal abortions were fatal. The vast majority of illegal abortions were performed by physicians, so a 1% fatality rate is far too high.
5.2.2007 12:06am
James Fulford (mail):
Cecil Adams at The Straight Dope agrees with Ponnuru, more or less.

That's because they're looking at the same study:

Trends in national abortion mortality, United States, 1940-1974: implications for prevention of future abortion deaths.

* Cates W Jr,
* Rochat RW,
* Smith JC,
* Tyler CW Jr.

PMID: 1001804

As for why doctors might deliberately under report on death certificates, (assuming they weren't abortionists themselves) it might be to spare the feelings of the survivors. Abortion was crime, and shameful as such. So was the state of being pregnant with an illegitimate child.

However, I don't know that such under-reporting did take place, it's just a guess.
5.2.2007 12:25am
Serge13102 (mail) (www):
Here is an article by the medical director of Planned Parenthood from 1960. If you look at p949 and read fact#3, it is clear that she did not consider deaths from illegal abortion to very numerous. In fact, she claims that "Abortion is no longer a dangerous procedure. This applies not just to therapeutic abortions as performed in hospitals but also to so-called illegal abortions as done by physicians. In 1957 there were only 260 deaths in the whole country attributed to abortions of any kind.

and also

"So remember fact number three; abortion, whether therapeutic or illegal, is in the main no longer dangerous, because it is being done well by physicians."

It seems that if 5000 to 10,000 women were dying of complications from illegal abortions, the medical director of PP would certainly have known of it.

Serge
LTI Blog
5.2.2007 12:49am
JB:
"Internal bleeding," "fever," "infection," something else. The deaths wouldn't necessarily occur right then, the abortion-botcher, doctor or not, wouldn't admit to having performed one (and if it was the mother herself, well, she's dead)...remember, these are deaths of the mother, not the baby, and deaths "from complications of illegal abortions," not "soon after losing babies" and occurring for possibly unrelated reasons.

Also, if the available statistics cover spontaneous abortion, then they'll be of little value for estimating illegal abortions, except as an upper bound.
5.2.2007 1:55am
Harry Eagar (mail):
I was a newspaper reporter then (and still am). 5K deaths would be around 2/county/yr.

If there had been anywhere near that many, we'd have noticed.

We didn't.
5.2.2007 1:56am
Viscus (mail) (www):
Harry Eager,

Your 2/county/yr statistic is meaningless, since the number of deaths would not be evenly distributed among counties. Los Angeles county has a much higher population (and thus presumably much higher number of deaths from all causes) than Podunk County, Nowhere, USA.

That you in particular don't notice a problem living in "Podunk County" doesn't mean much.
5.2.2007 2:45am
pmen (mail):
Folks,

I don't know what the figure was, but some background...

The cut off date is important.

As early as the 1940s, abortion was provided in many places in what were essentially the equivalent of modern out-patient clinics. The procedure itself was pretty much as simple and safe then (from a mortality of the mother standpoint - I'm not trying to troll), as it is today.

This depended on authorities looking the other way, as they did in many places with both abortion and prostitution.

With widespread strict enforcement, the predominant method of illegal abortion in the late 1960s became induction of a miscarriage, after which the woman was sent to the emergency room immediately. Doctors performed D&C procedures in the hospital setting, and there were hospital review boards set up to look into whether a particular doctor was performing too many procedures in contexts where what the medical necessity was questionable.

So there is both more potential for underreporting by the late 1960s than the scenario that most people imagine (botched D&C), because what you have is a woman presenting at the emergency room with a miscarriage, and also doctors with an incentive to underreport.

Back to the question at hand, though. I do not have a cite, unfortunately. If no one can help David, then he can email me and I could ask around. However, my understanding is that it is widely accepted that mortality from illegal abortion prior to widespread enforcement of abortion bans was low relative to what one might expect. Numbers for loss of fertility would be interesting to look at but probably impossible to obtain.

All of this begs the question a bit, though, doesn't it?
5.2.2007 8:31am
dearieme:
In summary then: to first order, people have only the vaguest idea of the numbers.
5.2.2007 8:46am
AppSocRes (mail):
I got so annoyed at the anecdotal data on this, that I did a back-of-the-envelope analysis of the time-series of female, age-specific, cause-specific, mortality from 1950 to whenever I did the analysis in the late 1980s. There's a slow but steady decline in mortality from all causes that might be associated in any way with pregnancy, birth, or abortion. I did not observe any significant deviation from the trends in the five years or so after Roe v. Wade. Just eyeballing the data makes it obvious to anyone without an agenda that annual abortion deaths in the US pre Roe v. Wade couldn't have exceeded a couple of hundred and may well have been less. I'd even be willing to enteretain the possibility that the high current incidence of abortion has increased the toal number of abortion-related deaths. But I couldn't support that assertion witn the data that are available.
5.2.2007 9:50am
William Oliver (mail) (www):
A last couple of points: Mortality from all causes was high in 1940, before the antibiotic age. As noted by Cates, et al.:

Both non-abortion-related maternal and abortion-related mortality have declined in the U.S. Between 1940-1950 and after 1965, deaths from abortion declined more rapidly than deaths from other pregnancy and birth-related causes. 1970-1976 average annual rate, 25%). Between 1951-1965, maternal mortality related to pregnancy or childbirth declined more rapidly than abortion-related mortality (average annual rate, 13% until 1965, and 5% thereafter). It is suggested that 5 reasons exist which explain the decline in abortion deaths since 1965: 1) increased use of effective contraceptives, 2) availability of legal abortions, 3) possibility of increased safety of illegal abortion, 4) selected underreporting, and 5) changes in coding practices.


Cates,W Jr, et al. Legalized abortion: effect on national trends of maternal and abortion-related mortality (1940 through 1976).Am J Obstet Gynecol. 1978 Sep 15;132(2):211-4.


More important, however, I may have tracked down the original source of the "5000 to 10000." I can't vouch for the data, but at least it's a claim. Apparently, it's an extrapolation from a single NYC clinic:


The 5,000 - 10,000 claim is one of the standard abortion promotion tricks: misleading citing. Often you'll see abortion advocates citing some recent (or relatively recent) "research." But when you check their source, you'll find that the source cites an even older source. And when you check that source, you'll find that it cites yet another, older source. You'll go round and round. (I've often joked that tracking down pro-choice original source material gives me motion sickness.) Eventually, you'll find the original source. If you're lucky.

[editorializing snipped]

In the case of the 5,000 - 10,000 claims, the original source was a book -- Abortion, Spontaneous and Induced -- published in 1936 by Dr. Frederick Taussig, a leading proponent of legalization of abortion. Taussig calculated an urban abortion rate based on records of a New York City birth control clinic, and a rural abortion rate based on some numbers given to him by some doctors in Iowa. He took a guess at a mortality rate, multiplied by his strangely generated estimate of how many criminal abortions were taking place, and presto! A myth is born!

Even if Taussig's calculations, by some mathematical miracle, had been correct, they still would have been out of date by the end of WWII. Antibiotics and blood transfusions changed the face of medicine. And you will notice that abortion proponents are all too aware of how dated Taussig's numbers are -- why else would they play Musical Cites instead of simply citing Taussig in the first place? But not only are the Taussig numbers dated, they were never accurate to begin with. At a conference in 1942, Taussig himself appologized for using "the wildest estimates" to generate a bogus number.

Although it took Taussig six years to reject his own faulty calculations, at least he did admit that he'd been wrong. Other abortion enthusiasts lacked Taussig's compunctions. Bernard Nathanson, co-founder of NARAL, admitted that he and his associates knew that the claims of 5,000 to 10,000 criminal abortion deaths were false. They bandied them about anyway, Nathanson confessed, because they were useful. This, too, is old news -- Nathanson came clean over twenty years ago.


In spite of the tone of the article , it looks like they've tracked it down. This isn't surprising to me. This kind of thing is pretty common in advocacy statistics. I remember tracking down a claim made by the Clinton Administration that something like 40% of female ER admissions were due to domestic violence. When that number was tracked down, it was found to have come from an article in the fashion magazine Mirabella; the author had just made it up. Similarly, even further back, I remember the missing children hysteria of the 1970s. I was a fellow in forensic pathology back then, and couldn't help but wonder why, if so many kids were being killed and abducted, we saw so few. The answer turned out to be bogus statistics again -- this time based on the way they organized the data. In the missing child case, every time a child ran away and every time it was reported, it counted as a different kid. Thus, a chronic runaway who ran away from the home 5 times a year and who had mulitple family members report it might end up counting as 15 or 20 different "missing children."
5.2.2007 10:36am
pmen (mail):
David,

There is a table in a (pro-choice) book, based on a 1965 journal article, showing rates of "abortion death" in the early 1960s of nearly fifty deaths per year in the New York City area, a little more than 3 per one thousand live births.

The author bases this statistic on an article regarding "therapeutic abortion" in NYC, but does not discuss the journal article itself in much detail. See Reagan When Abortion Was a Crime 211 (citing Edwin M. Gold et al. "Therapeutic Abortions in New York City, A Twenty-Year Review" AJPH 55 (1965) 965).

Because this data is probably based on hospital mortality statistics -- I have not looked at the underlying journal article itself -- which would have included abortions performed for reasons other than the induced miscarriage scenario described above, the extent to which these numbers reflect mortality from "illegal abortions" is somewhat questionable. Also, I am not sure how valid it would be to extrapolate these NYC numbers to the national live birth rate, which was a little more than 4,000 thousands in 1960.

But, "back of the envelope," this appear to support you 5-10k figure.

What the statistics (apparently) show is that, for the period from the 1950s to the early 1960s, when mortality should have been declining, it was in fact increasing -- along with, significantly, hospital admissions. At least in some urban areas. The argument is that this was due to a new set of more desperate abortions techniques arising from stricter enforcement of abortion restrictions.

The statistics also apparently show, for whatever you want to make of it, that mortality was decreasing again substantially during the years leading up to Roe, and did not decrease as dramatically as some believe during the years immediately following Roe.


[DK: Thanks for the lead. I found the 1965 article on the American Journal of Public Health website. To the AJPH's great credit, their old articles are on-line, and free. I read the article, and it deals exclusively with therapeutic abortion data from New York City. Therapeutic abortions (for serious reasons of maternal health) were legal, and all the data in the article come from NYC hospitals. The authors make a few passing references to the problem of criminal abortions, but they do not supply data on the subject.]
5.2.2007 11:37am
pmen (mail):
David,

Again, though, not as easy a distinction as you may be implying. Some significant proportion of those "therapeutic abortions" apparently were criminal abortions that ended up in the emergency room, because that was SOP at the time -- start a miscarriage and put the woman in a cab to the hospital. Although I am sure this varied from place to place, these women were apparently treated by D&C under rubric of "therapeutic abortion," at least in the big cities. There were then committees at the hospitals that attempted to ensure that doctors did not push this envelope too far, and I have heard anecdotally that you were in some places essentially allowed to do x number of procedures per month before they started to investigate whether what you were doing was medically justified.
5.2.2007 1:18pm
amy (mail):
There's no way to know the real numbers. Pro-legal abortion people will want higher numbers; anti-legal abortion people will want lowers numbers. You could do interviews with doctors from emergency wards in the relevant period; but again there would probably be bias. I don't think doctors today would be as involved as doctors in the past — I predict a black market RU-486 if abortion is made illegal.
5.2.2007 1:39pm
Aultimer:

Dr. T wrote:

The vast majority of illegal abortions were performed by physicians, so a 1% fatality rate is far too high.


Any suggestion of proof for that claim?
5.2.2007 2:37pm
Special Guest:
In Caitlin Flanagan's book review this month in the Atlantic, she speculates that her grandmother's cause of death (listed as "gangrene" on the official papers) was due to an illegal abortion. She also recounts how family lore passed down that the death was due to botulism poisoning, rather than the official cause or what she suspects was the true cause.
5.2.2007 3:32pm
William Oliver (mail) (www):
'"Internal bleeding," "fever," "infection," something else. The deaths wouldn't necessarily occur right then, the abortion-botcher, doctor or not, wouldn't admit to having performed one (and if it was the mother herself, well, she's dead)...remember, these are deaths of the mother, not the baby, and deaths "from complications of illegal abortions," not "soon after losing babies" and occurring for possibly unrelated reasons.'

That's exactly the point. The abortion doc wouldn't be filling it out -- it would either be the doc who attended the death or the pathologist who did the post. Thus the person who did the abortion would not be the one filling out the death certificate (which is where most of the vital statistics comes from). Certainly there has been an evolution of nosologic validity of the DC over the years, but this would not likely be misdiagnosed. Even if if was misdiagnosed as a spontaneous abortion instead of an illegal abortion, it would have been included in the aggregate statistic of 190 or so in 1966.

And, of course, no professional puts "botched abortion" as an immediate cause of death. Instead, the immediate cause of death is always "sepsis," "exsanguination" or whatever. The *proximate* cause, however, would be marked (i.e. the "Due to" lines). Note, by the way, that lawyers and medical examiners often use "proximate" differently -- in many legal definitions, lawyers use "proximate" for what we use "immdiate."

I'm not saying that people won't lie on death certificates -- I've had all sorts of issues with local physicians who do that and call a suicide an accident, for instance. The biggest rash of lies on DCs I've experienced was docs who refused to put AIDS/HIV on DCs in the 1980s. In fact, I remember one young resident who wrote an article in one of the popular science magazines (Discover, I think) about how proud she was of doing exactly that and thus saving the family the trauma of having the decedent's disease on legal paper.

But even given that, it's a stretch to believe that with a reported fatality rate of 190 deaths in 1966, that there were an additional 9,810 misclassifications. There's no evidence for that at all.
5.2.2007 3:34pm
cathyf:
I found this interesting:
SOP at the time -- start a miscarriage and put the woman in a cab to the hospital
because it suggests a chain of causality which might actually make illegal abortion, as practiced pre-Roe-v-Wade, safer than legal abortion as practiced in recent times. In modern times, the majority of legal abortions happen in storefront clinics. Abortion clinics are notoriously difficult to regulate, because any and all regulatory activity is (correctly) viewed with great suspicion by abortion-rights activists and vigorously resisted. If "back-alley" abortions really only started in back alleys but were basically done in hospital emergency rooms and operating rooms, then I would certainly consider it plausible that they could be safer than strip-mall-clinic abortions.

I'm quite surprised that absolute numbers were so low. Certainly the abortion of a tubal pregnancy required abdominal surgery with all of its attendent risks, and a ruptured tubal is a nasty medical emergency. I wonder if those are not being counted as miscarriages at all? Or maybe there are just far fewer of them than I thought?
5.2.2007 4:17pm
pmen (mail):
Cathyf,

Not sure what you mean by "absolute numbers," but note that I am using the term miscarriage loosely.

There are people who have argued that ad hoc "regulation" of abortion by hospital staff was better in some ways than what you have now, but the procedures used outside of the hospital to _induce_ the abortions were not good at all, in many instances) for fertility retention (foreign objects or chemicals, not just dilation in a sterile environment). Not to mention that this was pretty unpleasant.

Some doctors may have encouraged this process by sterilizing patients.
5.2.2007 4:34pm
cathyf:
By "absolute numbers" I mean that I was quite surprised that the number of maternal deaths associated with abortions (both legal and illegal) and miscarriages was so low. If the maternal deaths associated with legal abortion and spontaneous miscarriage had been way higher than the deaths associated with illegal abortion I would be immediately suspicious that the deaths associated with illegal abortion were misclassified or otherwise undercounted. Wonko accused me of going off-topic, but I think that comparing the maternal death rates from all of the different miscarriage, abortion, tubal pregnancy scenarios can give you a much better picture of whether a particular illegal abortion statistic is believable or not.
5.2.2007 4:57pm