Spunk Science: Bad Science Journalism - Exhibit B.

So the notion that human semen may act as an antidepressant is rearing its semi-flaccid head again.

Broadsheet from Salon.com (May, 2007) asks are you addicted to semen?

Someday they'll have a patch for that. Dr. Gordon G. Gallup theorizes (via Feministing) that women have a "chemical dependency" on semen. He's based this conclusion on a survey that found that women who regularly had sex without condoms became increasingly depressed the longer they went without sex (read: semen). Women who regularly used condoms didn't have this experience. So, it could be that if you consistently have sex with a condom, you'll not only ward off STDs but depression via "semen withdrawal." (I'm curious whether the couples who went without condoms used birth control and whether a raised risk of pregnancy -- real or perceived -- during regular unprotected sex could be a factor. Might there be a craving for frequent reassurance through frequent sex -- paradoxically, since it could increase your chances of pregnancy -- that your partner is planning on stickin' around in case a baby comes along?)

The Feministing link references a Psychology Today article with the precious catchy title Crying Over Spilled Semen.

So much fisking, so little time.

To start, this research is five years old. Gallup et al., as of my last check of MedLine at noon today, have not published a follow up study. So why the heck is Psychology Today recycling this "news?" And why are Feministing and Salon so woefully behind the times to consider a 2002 article to be such hot stuff?

I found the original paper by Gallup et al. (2002) "Does Semen Have Antidepressant Properties?" Archives of Sexual Behavior 31: 289-293. What I read wasn't terribly compelling: a mild correlation with preliminary results. The hypothesis is that vaginal absorption of testosterone, estrogen, and prostaglandins (predominantly E1, E2, and F2alpha) among other hormones in seminal fluid may affect mood. Gallup el. reference a paper by P.G. Ney (1986 - yes, the Jurassic) "The intravaginal absorption of male generated hormones and their possible effects in female behaviour" in Medical Hypothesis 20:221-231 as the basis for their hypothesis.

Here's an excerpt from the Gallup team's results:

Significant correlations were found between BDI
scores and the length of time (in days) since engaging in
sexual intercourse. For females who did not use condoms,
length of time since their last sexual encounter was correlated
with depressive symptoms (r D :229; p same was true for females who reported using condoms
some of the time (r D :318; p who used condoms most or all of the time these correlations
were near zero and not significant. Thus, for sexu ally active females who did not typically use condoms,
depressive symptoms increased as the amount of time
since their last sexual encounter increased.

OK, for groups with a substantial number of human subjects, maybe there is real "wowza!" in those r values, but in my more defined in vitro world, it takes r values of 0.7 or higher to arouse us. Also, the standard deviations in the groups are massive, and although there's "significance," it's not all that strong.

That said, Gallup et al. offer this caution:

It is important to acknowledge that these data are preliminary and correlational in nature, and as such are only suggestive. More definitive evidence for the antidepressant effects of semen would require more direct manipulation of the presence of semen in the reproductive tract and ideally the measurement of seminal components in the recipient's blood.

Of course, the Psych Today article totally misses stuff like this.

Gallup et al. note that a large number of questions loom, and discuss alternate explanations for their results. There's more in the paper, but suffice it to say that I was not impressed.

Far more egregious is Ney's paper from which the Gallup team drew their hypothesis.. It consists of one - one - case study. Here's the abstract:

In this paper I hypothesize that a woman through her vagina absorbs sufficient quantities of the prostaglandins from her partner's semen, to affect her mood. The vagina appears to have an active transport mechanism which readily absorbs the hormones found in seminal plasma. The seminal secretions include estrogens, F.S.H., LH, testosterone, and at least 13 prostaglandins. Prostaglandins have a modulating effect on neurotransmitters. Evening Primrose Oil (E.P.O., Efamol) may improve depressed mood by facilitating the production of prostaglandins. I have described a depressed, child abusing mother who was possibly getting insufficient exogenous prostaglandins post-partum to maintain her pre-partum mood state because she was avoiding intercourse with her husband. When her mood improved on E.P.O., she began enjoying her husband and her child. If these observations and deductions are correct, regular amounts of seminal plasma may be important in maintaining a woman's affective health and E.P.O. may be useful in treating depressed abusing mothers.

I'm not even going to touch the potential woo factor of evening primrose oil.

Effective concentrations in various body fluids are not addressed in any critical manner in Ney's paper. It's purely conjectural. The potential impact of the pharmacokinetics of absorbed testosterone, estrogens and prostaglandins is not considered at all. Keep in mind that sex hormone concentrations tend to be tightly regulated by proteins that bind to them specifically, even if they escape first pass metabolism by intravaginal absorption. As for prostaglandins, both synthetic and natural, 10-43% of a given dose is absorbed intravaginally (see Benzinger & Edleson, 1983, Drug Metabolism Reviews, 14:137-168). It beggars the mind to imagine how the low concentrations of the sex hormones and prostaglandins in 3.5 milliliters (on average) of seminal fluid could have a clinically significant effect.

But the following two paragraphs from Ney's article really made my jaw drop. Aside from a strange, and to me, trivialized, approach to the very serious condition of post-partum depression (1), I'm thinking this dude has an agenda (check out the text that I have bolded):

When a woman is sufficiently depressed and irritable to abuse her
child and then given large amounts of gamma-linolenic acid, there may
be an improvement in the prostaglandin level and improvement in mood.
Consequently, the mother would be more interested in intercourse with
her mate and there would be a return to the usual route of obtaining
male made prostaglandins. In the presented case there appeared to be
a close temporal relationship between the improvement in the mother's
mood, treatment with E.P.O., increased interest in sex, and later the
return of normal sexual relations. She was advised that, instead of
immediately springing from the bed following intercourse, she should
relax. She should ensure her bladder is empty before getting into bed
and try hard to lie still after intercourse.

While the general ethos in many Western societies is critical of men,
particularly because of their sexuality, this hypothesis, lf
corroborated, would place a somewhat greater value on men and their
sexual activity. Although the role of exogenously given
prostaglandins in improving libido and mood has not been established,
there appears to be some link. The change in this woman's wellbeing
may be due to the absorbed estrogen or testosterone hormones, but in
the amounts available from the sperm, it would seem that the
prostaglandins would have a greater impact. If the deductions of this
paper are correct, the oil of the evening primrose may be an
important adjunctive treatment for postpartum depression and infant
abuse or neglect.

So there's the paper trail from a weird little 1986 paper in Medical Hypothesis to a highly preliminary publication in the Archives of Sexual Behavior to a fluff piece in Psychology Today. As for the commentary in Salon and Feministing, believe me, I am on your side, but please take these studies with a bigger dose of skepticism, or at least get a bona fide scientist to do a little background check for you.

(1) To be fair, the first selective serotonin reuptake inhibitor was not on the market until 1988, and the antidepressants available in 1986 were ickier in terms of side effects, so maybe psychiatrists were desperate at the time to seek treatment for their patients. This doesn't address the issue of psych meds that are compatible with breastfeeding. But still, the "general ethos is critical of men?" Yeah, it must be really validating for you guys to know you're good for something: walking, churning vials of antidepressants.