Posted by Eugene Volokh:
Half the Story:
[1]Slate's new "Human Nature" column (on "Science, culture, and
politics") reports:
SEX
Item: Morning-after pills don't lull women into risking unprotected
sex.
Source: Center for Reproductive Health Research and Policy
Outlet: Washington Post, Jan. 5
Gist: Last year the Food and Drug Administration refused to allow
nonprescription sales of Plan B, a morning-after pill, on the
grounds that its easy availability might lull young women into
having more unprotected sex. A study by CRHRP finds no such effect.
Liberal spin: This cuts the scientific legs out from under what was
really a political decision. Now give us the pill.
Now of course this is a short item, which necessarily omits some of
the story. Here's an excerpt from the longer story, in the
[2]Washington Post:
Providing women with easy access to the emergency contraceptive
Plan B [the "morning-after pill"] did not lead them to engage in
more risky sexual behavior, a study of more than 2,000 California
women has concluded.
The study did find that women given a supply to keep at home were
more than 1 1/2 times as likely to use the drug after unprotected
sex as those who had to pick it up at a clinic or pharmacy. The
findings led the study authors to conclude that easy access to Plan
B, also called the morning-after pill, could reduce the number of
unwanted pregnancies while posing no apparent risk to women.
The study contradicts a key claim made by opponents of easier
access to Plan B at a time when the Food and Drug Administration is
preparing to decide on a second application to allow
nonprescription sales of the drug. . . .
"Our findings were that women don't change their sexual behavior
when the drug is easily available, but rather that they're more
likely to use it if access is easier," said lead author Tina R.
Raine of the Center for Reproductive Health Research and Policy at
the University of California at San Francisco.
The study looked at the experiences of 2,117 San Francisco-area
women ages 15 to 24 who were randomly put into one of three groups
-- one was given the drug to take home, the second could pick it up
without a prescription at a clinic, and the third could get it
without a prescription at a pharmacy.
The study found that about the same percentage of each group had
unprotected sex over a six-month period, that incidence of sexually
transmitted disease was equal, and that about the same percentage
became pregnant.
Nonetheless, about 37 percent of women in the group given Plan B to
take home used it at least once during the six months, compared
with 21 percent in the clinic-access group and 24 percent in the
pharmacy-access group. . . .
The new study, being published today in the Journal of the American
Medical Association, supports the position taken by much of the FDA
review staff and 23 of 27 members of the FDA advisory panel that
the drug could be safely and properly used without a prescription.
. . .
So what's missing from the Slate item, and buried in the Post story?
While the study didn't find an increase in unprotected sexual
behavior, it also didn't find a decrease in pregnancy. As the [3]study
abstract says, "compared with controls, women in the pharmacy access
and advance provision groups did not experience a significant
reduction in pregnancy rate (pharmacy access group: adjusted odds
ratio [OR], 0.98; 95% confidence interval [CI], 0.58-1.64; P = .93;
advance provision group: OR, 1.10; 95% CI, 0.66-1.84, P = .71)." The
study's text, which my library got for me, says the same.
The Post story, as I mentioned, does note this in the 8th paragraph --
but the 2nd paragraph says "The findings led the study authors to
conclude that easy access to Plan B, also called the morning-after
pill, could reduce the number of unwanted pregnancies while posing no
apparent risk to women." Say, wouldn't it have made sense to mention,
in the same paragraph, that the study failed to find a reduction in
the number of unwanted pregnancies?
Also, note that the research doesn't precisely lead to the
"conclu[sion]" that "Providing women with easy access to [the
morning-after pill] did not lead them to engage in more risky sexual
behavior." Rather, it fails to support the contrary conclusion: It
finds no statistically significant evidence that providing women with
easy access does lead them to engage in more risky sexual behavior. As
they say, absence of evidence (here, lack of a statistically
significant correlation between access and risky behavior) isn't
evidence of absence (a statistically significant showing of no
correlation). If enough studies fail to find a correlation, we may
become fairly confident that the correlation, if there is one, is very
weak. But when one study fails to find a statistically significant
link, that's hardly dispositive.
But that's a fairly small quibble compared to the broader point --
sure, the pill that aims to reduce unwanted pregnancies doesn't seem
to have some kinds of harmful side effects, but it doesn't seem to
have much of an effect in reducing unwanted pregnancies, either. Seems
to me that both points should be noted prominently.
I should stress, by the way, that I have nothing against morning-after
pills, and I think that women should have access to them. Even if
these pills don't do much good overall, they may help some women in
some circumstances, and I think women should have this choice. (I
don't take the view that a day-old unimplanted zygote has rights; and,
no, I'm not going to take the time to debate this point, which is in
any event tangential to the heart of this post.) I thus support the
bottom-line policy conclusion that the Post article and the Slate item
point to. I'm just not wild about how the study is being described.
As always, please correct me if I'm misreading either the study or the
stories.
References
1. http://slate.com/id/2111786/
2. http://www.washingtonpost.com/wp-dyn/articles/A48377-2005Jan4.html
3. http://jama.ama-assn.org/cgi/content/short/293/1/54
_______________________________________________
Volokh mailing list
[email protected]
http://highsorcery.com/cgi-bin/mailman/listinfo/volokh
