The Observatory — January 22, 2008 09:00 AM

The MRSA-Gay Connection

Press slaps its own wrist for mischaracterizing research

By Curtis Brainard

• Single Page
• Print
• Email
• Comments
• Digg Facebook Reddit StumbleUpon Delicious

• Single Page
• Print
• Email
• Digg Facebook Reddit StumbleUpon Delicious

Subscribe to the Columbia Journalism Review at our special Web rates.

Comments Post a Comment

How ridiculous is this? CJR now has a blog devoted to science journalism whose mission statement is spotlight science news when “it falls victim to spin, engages in alarmism, perpetrates false balance, misrepresents the science in peer-reviewed literature, or displays questionable priorities in news judgment.”

So to illustrate this new dedication to Science Journalism, Mr Brainard highlights the recent spat about activist group who got all riled up about the report, lodged a complaint with the UC San Francisco whos PC drones fell in line, and this is a positive step for science journalism? While the report did not specifically say that “gay sex is a risk factor of USA 300”, the conclusion is certainly not a stretch by any means. Or to quote from the report:

In Boston, multidrug-resistant USA300 was recovered exclusively from men who have sex with men.
Multidrug-resistant USA300 MRSA infection is especially common among men who have sex with men. It might be sexually transmitted in this population.
Twenty-nine of the 30 patients with multidrug-resistant USA300 infection had a history of having male–male sex, consistent with the high incidence of multidrug-resistant USA300 observed in San Francisco ZIP codes with high percentages of male same-sex couples

Scientific studies are by nature very conservative with their conclusions, and unless the researchers fall victim to the ACT-UP lobby, a future study will most likely conclude in no uncertain terms that gays are more susceptible to USA 300 because of a general pattern of more risky sexual behavior, which would appear to be effective vector for transmission of USA300. Sticking your fingers in your ears and setting your blinders to full PC mode does not change this fact, a fact you try you hardest to gloss over in this post.

Posted by TDC on Tue 22 Jan 2008 at 11:19 AM

I am appalled by what can only be described as partisan cherry-picking of findings by Mr. Brainard to obscure the truth. He should be ashamed; worse, he should be censured.

In objecting to a headline that reads “Drug-resistant staph passed in gay sex - U.S. study,” Brainard says "That is not what the study found," and proceeds to quote a portion of the study's "Discussion" section acknowledging limitations in the data. "To peg 'gay sex' as the culprit in a headline is completely misleading and journalistically irresponsible," concludes Brainard.

Brainard was quoting a section of limitations acknowledging that the study cannot pinpoint which of several possible specific behaviors are at fault. That gay sex generally was implicated was clear and unmistakable, however. From the same section of the study Brainard quotes, the researchers tell us this:
Data from this study suggest that multidrug-resistant USA300 has spread rapidly among men who have sex with men in San Fransisco and Boston, and that having male-male sex seems to be a risk factor for multidrug-resistant USA300 infection independent of HIV infection.

The researchers continue in the next paragraph:

Our findings that 27% (32 of 118) of men who have sex with men from the SFGH HIV clinic and 39% (47 of 121) of men who have sex with men from Fenway Community Health had infections involving buttocks, genitals, or perineum are consistent with sexual transmission of USA300 in this population.

In other words, the study found precisely what Mr. Brainard says the study did not find. The headline about which Brainard complains is, in fact, accurate.

It gets worse. Again, from the discussion section of the findings (same section Brainard quotes):

It is not clear whether the behavior potentiating these infections among men who have sex with men is anal sex ..., skin-abrading sexual practices, or increased frequency of intimate skin-to-skin contact; prevention messages may therefore need to suggest caution in each of these practices.

In other words, the limitation on the study's specificity, which Brainard represents as absolving gay sex as the culprit, the researchers themselves claim is a reason to include all possibly implicated types of gay sex in their warnings!

The caveat Brainard quotes occurs at the end of the "Discussion" section of the report, after the sections I quote; it seems highly unlikely that he found his quote without reading the rest of the section. The most likely cause of his error is that he is lying outright -- that he wants to obscure the clear findings of the report. The only other plausible explanation is that he wants so badly for the report to say other than what it says that he's simply incapable of reading accurately. In either case, Mr. Brainard should be severely chastised.

This is scandalous. If the Columbia Journalism Review endorses clear, deliberate misstatements of fact as egregious as Mr. Brainard's in this article, then it cannot be trusted at all.

Posted by philwynk on Mon 28 Jan 2008 at 06:10 PM

I believe I was a bit too rough on Mr. Brainard. He could have committed the error he committed without lying.

Rereading the various quotations from the study, it appears possible to me that he simply misread the sentence he quoted, and took “cannot comment on… specific male–male sexual practices” to mean “cannot comment on homosexuality as a cause.” Brainard could have reached his conclusion without lying.

He's wrong, though; the sentence he quoted does not absolve gay sex; the authors are very clear about implicating male--male sexual contact, and simply can't specify which specific types of contact are at fault.

Posted by philwynk on Tue 29 Jan 2008 at 08:35 AM

Fortunately, four recent medical studies have demonstrated that an inexpensive and completely natural substance called colloidal silver literally decimates MRSA infections. You can read a very interesting report on it here…

Posted by Spencer Jones on Wed 6 Aug 2008 at 02:25 AM