Box Turtle Bulletin

Box Turtle BulletinNews, analysis and fact-checking of anti-gay rhetoric
“Now you must raise your children up in a world where that union of man and box turtle is on the same legal footing as man and wife…”
This article can be found at:
Latest Posts

The alarmist and misleading new headlines about HIV transmission

Timothy Kincaid

March 12th, 2010

The Centers for Disease Control has issued a “new look at disproportionate impact of HIV and syphilis among U.S. gay and bisexual men”. And this “new look” is quite alarming:

The data, presented at CDC’s 2010 National STD Prevention Conference, finds that the rate of new HIV diagnoses among men who have sex with men (MSM) is more than 44 times that of other men and more than 40 times that of women.

That certainly sounds frightening. And it lends itself easily to anti-gay activist who argue that gay men and women do not deserve equal rights under the law because gay men are crawling with disease. (Iowa Independent)

Homosexual activity is “more dangerous for individuals who engage in it than is smoking,” and because of this, state lawmakers need to pass a constitutional amendment overturning last year’s Iowa Supreme Court decision legalizing same-sex marriage, according to Iowa Family Policy Center President Chuck Hurley.

Naturally, anti-gay activists like Hurley fail to distinguish between “individuals who engage in homosexual activity” and individuals who engage in unsafe sex. And, having only the vaguest of notions about those same-sex couples who marry, he leaps at a conclusion that is out of his reach.’

Unsafe sex significantly increases the odds of seroconverting. And considering that the CDC lumps all men who have sex with men (MSM) together regardless of their relationship status or sexual practices, single gay men who whoop it up sans condom with a stranger tonight have a FAR MORE than 44 times likelihood of becoming HIV positive than does a heterosexual dude that does the same. Which is one reason we should encourage relationships and responsible sex.

But setting aside the lunacy of opposing committed relationships so as to stop promiscuous behavior, let’s look at the numbers themselves. And let’s consider the responsibility of the CDC and those who use their statistics to report them in a way that is meaningful and useful rather than pointlessly frightening.

Yes, “44 times that of other men” is scary. But what does it mean?

For that we have to look at the following paragraphs.

The range was 522-989 cases of new HIV diagnoses per 100,000 MSM vs. 12 per 100,000 other men and 13 per 100,000 women.

The rate of primary and secondary syphilis among MSM is more than 46 times that of other men and more than 71 times that of women, the analysis says. The range was 91-173 cases per 100,000 MSM vs. 2 per 100,000 other men and 1 per 100,000 women.

Unpacking the medical language, this is what we find:

There were about 33,750 new cases of HIV infection in 2007. Assuming that gay/bi men make up about 4% of the population, this number is about three quarters of one percent (0.75%) of all gay/bi men.

For syphilis, the number is about 5,900 cases and an infection rate of about one tenth of one percent (0.13%).

Previous analysis suggests that about 12% of all gay/bi men are currently living with HIV. As syphilis is treatable, the total number of gay men living with syphilis is less easy to determine.

Further, if we wish to be credible, we have to recognize that the gay community is not homogeneous. HIV and syphilis infections impact certain demographics far more than others. Factors such as race, location, education, and self-worth can have almost as much a factor on one’s likelihood to contract HIV as sexual orientation.

Too often we can read alarming headlines and think, “why fight it, it’s inevitable.” This isn’t true. Most gay men don’t have HIV and never will. And while these frightening statistics can remind us to be careful, they should be taken in perspective.

When we hear “HIV diagnoses… 44 times that of other men” or “2000 times higher than repeat blood donors”, we have to remind ourselves to step back and take a better look at the numbers. We need to recall that the purpose of information of this sort is to fight the virus, not the people who are infected. And we need to understand that we each are individual and not a statistic.



Mark F.
March 12th, 2010 | LINK

Why don’t the anti-gays love lesbians? Virtually no HIV or STD transmission at all. Plus they make hot movies straight men love! ;-)

March 13th, 2010 | LINK

When I was in college some 30 years ago, I took a course in statistical analysis. On the first day of class, the professor told us he was going to teach us how to “lie with statistics” and he did just that.

Today, when I see something based on such analysis, I think about the logic behind it. Was the sample random? Was it equal (in the number of people, etc.,) for each group being analyized and does the outcome benefit any certain group?

Just as the examples given in this story, most of the time I find that the number of items in each group do not represent a cross section of the population.

So who is lying with these statistics?

Timothy (TRiG)
March 13th, 2010 | LINK

There’s an excellent analysis on “lying with statistics” in Ben Goldacre’s Bad Science, a book I can’t recommend enough.


Jim Pickett
March 13th, 2010 | LINK

Thank you for a smart article.

The disparities gay/bi and other men who have sex with men are appalling.

In Chicago, it is estimated one in three gay black men have HIV, one if four gay latino men, and one in 5 gay white men.

On LifeLube yesterday, we posted an item putting some context to these numbers, including things like historical underfunding of prevention activities for gay men, despite the fact we make up the biggest chunk of the epidemic. And some other facts that help us understand these numbers – which by the way, are not lying.

March 13th, 2010 | LINK

My lecturer once said: There are three kinds of lies exist in this world: 1. Bad lies, when you lie for doing harm to others; 2. White lies, when you lie for the good of others; 3. Statistics.

Every student who gets statistics in college always gets this message: When being used as a justification or even an information, statistics are always a lie.

Statistics are purely numbers. Nothing else.

March 13th, 2010 | LINK

MSM != gay

While I understand the need for the term, I wish they would include stats for those who actually identify as gay to show the disparity between the two..

Timothy Kincaid
March 13th, 2010 | LINK


Interestingly enough, when the CDC released their huge study in 2005 (based on 2002 numbers) they found that the percentage of men who identify as gay/bi was very similar to the percentage of men who have sex with men.

When I looked at the numbers to try and get a good sense of the gay population, I came up with similar conclusion.

I don’t think that 4% is the definitive answer and that we can just accept it and use that figure from now on. But I do think it is the best current estimate for the base minimum of people who are gay/bi.

March 13th, 2010 | LINK

Well I mean more in their behavior. I would imagine someone who identifies as gay has better sex practices than someone who merely fits the category of MSM, given how many different people the latter includes (to my understanding it includes those who have had it non-consensually or as a means of income).

March 13th, 2010 | LINK

“The range was 522-989 cases of new HIV diagnoses per 100,000 MSM vs. 12 per 100,000 other men and 13 per 100,000 women.”

So the rate for MSM was much higher than for other people, but still very low: about 0.5% to 1%. The “still very low” aspect is part of the balance that’s being left out of the reporting by mainstream media (also MSM for short), and possibly not made clear enough in the CDC report.

Because they represent new diagnoses, these rates are a very current indicator. If we looked at the total number of HIV cases instead, we would be including a large number of people who were infected years ago, and the results wouldn’t necessarily be indicative of current risk.

Nonetheless, the rates are a few poor basis for comparing MSM with other groups – which is exactly what the conservatives and seemingly the CDC are doing. Group comparisons require, firstly, a good indicator of HIV cases as a proportion of the different populations. These rates don’t provide that, because only a few HIV cases are included.

Incidentally, it’s anal intercourse that actually is a high-risk behavior, whether between men or between a man and a woman. The CDC doesn’t seem to be acknowledging this. Maybe they still have Bush appointees at the helm.

On the other question – I’m not sanguine to the CDC’s use of MSM to represent gay and bisexual men. This approach, even if technically sound, could add fuel to antigay and antibisexual rhetoric. Plus, two percentages can be about equal without actually representing the same group.

March 13th, 2010 | LINK

Another perspective on HIV as a reason to discriminate:

The stats on the above-linked graphic (which is hardly funny, but I didn’t create the site) come from the CDC:
For 2008

Deaths from
AIDS: 8,423
Lung Cancer: 96,835
Heart Disease: 316,968
Car Accidents: 17,463
Suicide: 26,132

Using Hurley’s reasoning, smokers should be barred from marriage, and so should obese people, people who drive or ride in cars – their activities are more dangerous for those who engage in it than unprotected sex.

And since societal condemnation is a leading factor in the depression that leads to suicide, condemning homosexuality is more dangerous to society as well.

March 14th, 2010 | LINK

And yes, Gay men are much more likely to have sexual health check-ups than heterosexual men. And many women still have to insist on an HIV test against the ‘conventional wisdom’ of standard practitioners. The gay community have been forced to come out of the closet post HIV which is a good thing, however there will be no reliable stats until the general community, including doctors, are educated about transmission risks & stop assuming the nature of sexual practise based on ‘out’ presentation.

March 14th, 2010 | LINK

Timothy: The fact that two groups are roughly the same size doesn’t by itself say anything about the degree of overlap between them. For example, about 13% of the US population is black. About 13% of the US population is poor. It’s often stated that poverty is a predominantly black problem in the US. But it turns out that roughly 25% of black people are poor and, because the black and poor proportions are about the same, that means that roughly 25% of poor people are black.

This means that most poor people are unlikely to be black and most black people are unlikely to be poor. It also means that black people are at a much greater risk of poverty than white people. The innumerate, however, can’t tell the difference.

This can lead to stereotyping. While it’s true that stereotypes seldom have zero basis in reality (e.g. the Jewish blood libel or the gay pedophilia libel) it’s also true that most stereotypes don’t follow the “51% rule.” Most stereotyped traits, while they’re more common inside rather than outside the stereotyped group, are still uncommon inside it (Gordon Allport, in his classic The Nature of Prejudice, called this a “categorical differential.”

Also, as others have been mentioned, the risk is inhomogeneous: not all gay men, or MSM, are at equal risk (and trying to be “inclusive” by saying “LGBT” is even less accurate). It’s quite possible for a group as a whole to have a high risk of something at the same time that the majority of the group has a low risk for that same thing. That’s a form of Simpson’s Paradox in which aggregates behave statistically differently than individuals (consider, for example, a city that has 10 times the murder rate of the nation as a whole, but where almost all murder victims are gang members).

March 14th, 2010 | LINK

The half-to-one percent number doesn’t strike me as low at all.

If 0.75% are seroconverting every year and 12-13% are living with HIV, that puts the number at 20% in a decade if things stay the same.

That’s an awful trend in my book.

March 16th, 2010 | LINK

Thank you ebohlman. Great summary.

Mr. Kincaid mentioned:

There were about 33,750 new cases of HIV infection in 2007.

And then today an AP writer Edith M. Lederer said: Michel Sidibe, the head of UNAIDS,: “… said it is “shocking” that more than 50 percent of new HIV infections last year [in the United States] occurred among homosexuals.

Maybe we need the alarmist headlines.

Leave A Comment

All comments reflect the opinions of commenters only. They are not necessarily those of anyone associated with Box Turtle Bulletin. Comments are subject to our Comments Policy.

(Required, never shared)

PLEASE NOTE: All comments are subject to our Comments Policy.