The Puzzling Link Between HIV Diagnoses And Craigslist

[Steven_Draker]

Now, as casual hookup apps begin to replace online ads, experts worry that HIV rates might continue to rise.

 

 

A new study has linked a rise in HIV diagnoses with the emergence of Craigslist personal ads throughout the US. The research, conducted between 1999 and 2008, found that HIV cases rose in a city when Craigslist was introduced.

 

“Online platforms offer access to a larger social group than is generally available through offline contacts, making the internet an emerging venue for seeking casual sex partners. The ease of seeking sex partners through classified ad sites may promote risky behaviors that increase transmission of STDs,” study authors Jason Chan and Anindya Ghose write.

 

The idea behind the HIV study arose when Chan and Ghose were researching Craigslist as a means to sell used goods. They noticed a large number of ads for one-night-stands, both in major cities — not so surprising, they thought —and smaller cities, where casual hookups aren’t quite as common. Chan and Ghose eventually focused their research on 33 states, and found that when Craigslist came to town, HIV rates rose.

 

Related: Should We All Be In ‘Monogamish’ Relationships?

 

"Craigslist’s entry into a market is related to a 16 percent increase in HIV cases," Chan tells Yahoo Health. "This maps out to an average of 6,200 cases of HIV infection in the US each year, which amounts to $62 million in annual treatment costs. We further found that the increase in HIV trends is linked with the personal ad section, not the escort section."

 

Now, as casual hookup apps like Grindr and Tinder begin to replace online ads, experts worry that HIV rates might continue to rise. “With apps like Tinder and Grindr being used more and more, the likely consequence is that we will see a even greater spread of the disease,” Chan tells Yahoo Health. “However, this is only a speculation because human behavior is dynamic and can be subjected to changes over time. With the release of this study result and the increase in awareness raised about the risks involved in hooking up online strangers, individuals who use these apps might take more precautions or eliminate their current risky behaviors.”

 

source: https://www.yahoo.com/health/the-puzzling-link-between-hiv-diagnoses-and-109987894272.html

[marylander1940]

http://media.tumblr.com/61d3a668b21eb2e37132b04e92c22a17/tumblr_inline_mii6q7Kz4s1qz4rgp.gif

[+ quoththeraven]

I downloaded the paper this is based on and started reading it shortly after this was posted, but only made it halfway through and am not sure when I'll get back to it. While I therefore can't make a completely definitive statement, particularly about the statistical analysis, I can say that what I read was unimpressive. The authors artificially manipulated the contours of the graphs in Figure 1: The relationship between HIV trends and Craigslist entry and activity level, p. 4. They don't make logical connections where they should, so their argument based on what other studies say about the link between casual sex and HIV risk is not rigorous. (The study is a PDF, so I can't copy and paste any of it here for critique.) And the authors aren't public healthcare or HIV experts, and as such I'm skeptical of their ability to negotiate the nuances. This looks like a paper that arose out of a desire to use Craigslist as the basis for a study without regard to how meaningful the study is.

 

Without having read the statistical analysis, I can't say that they've confused correlation for causation, but it's definitely a consideration. But what bothers me most is this: More Craigslist ads may mean more casual sex, but what does that really mean in the absence of clearly documented links between increased casual sex and, say, fewer safe sex precautions? Cases of HIV would increase even if the percentage of acts during which precautions were taken remained the same because more sex acts were engaged in. Whether more casual sex acts correlate with increased risk-taking is something this study can't and doesn't purport to measure. So mostly I feel like this is a tempest in a tea cup -- something that might be cited to confirm prior studies about casual sex and risky behavior (in other words, the real-world effects of those studies' conclusions). But if that were its aim, one would expect the studies' summary to say so. It doesn't.

 

For an example of what a well-written and rigorously argued research paper in the social sciences looks like, see any of the David Lisak papers cited or linked to here.

[SteveEscort]

I would be very surprised if it leads to more instances of hiv transmission because my experience of Craigslist is 93 timewasters per 100 users

[+ quoththeraven]

I would be very surprised if it leads to more instances of hiv transmission because my experience of Craigslist is 93 timewasters per 100 users

 

Steve - They studied non-commercial uses of Craigslist. They concluded that commercial uses were correlated with a reduced incidence of HIV, which makes intuitive sense, but since I haven't read the meat of the study, I don't know how they proved it.

[SteveEscort]

Steve - They studied non-commercial uses of Craigslist. They concluded that commercial uses were correlated with a reduced incidence of HIV, which makes intuitive sense, but since I haven't read the meat of the study, I don't know how they proved it.

 

 

You can't "officially" run a commercial craigslist advert anymore in any country where it's legal or otherwise. You have to selective on how you word it. They've banned escort adverts for a number of years

 

The non commercial adverts are hard work too. I know from the number of clients who have mentioned it over the years. It's all pic collectors and people who never get back to you.

[+ quoththeraven]

Steve - Trying again. They were primarily interested in hookups that didn't involve money.

[SteveEscort]

Steve - Trying again. They were primarily interested in hookups that didn't involve money.

 

Trying again ..... Me too. I'll be a bit clearer this time in case you missed it the first time

 

I didn't state my experience involved money if you look. I said "because my experience" and then added in another post that they don't allow commercial adverts from escorts

 

Escorts have a life outside of hustling you know

 

You really "shouldn't presume" escorts live a 24/7 escort existence. I never once implied my experience was commercial but because I am an escort you immediately thought that to be the case as both of your replies prove. Now I wonder if you apply that logic to married men who book gay escorts, the logic that they must be gay like your logic that I was hustling on the site.

 

Anyone can place a craigslist advert but because I'm an escort, you came to the immediate conclusion that I used the site to get clients. Very presumptious of you and very wrong.

 

Try again

[+ quoththeraven]

I wasn't thinking of it that way because I was thinking you were responding in your capacity as an escort and not a civilian, so to speak. Thanks for clarifying.

[marylander1940]

Trying again ..... Me too. I'll be a bit clearer this time in case you missed it the first time

 

I didn't state my experience involved money if you look. I said "because my experience" and then added in another post that they don't allow commercial adverts from escorts

 

Escorts have a life outside of hustling you know

 

You really "shouldn't presume" escorts live a 24/7 escort existence. I never once implied my experience was commercial but because I am an escort you immediately thought that to be the case as both of your replies prove. Now I wonder if you apply that logic to married men who book gay escorts, the logic that they must be gay like your logic that I was hustling on the site.

 

Anyone can place a craigslist advert but because I'm an escort, you came to the immediate conclusion that I used the site to get clients. Very presumptious of you and very wrong.

 

Try again

 

Stay away from backpage and craigslist, good escorts don't try cheap sites.

[+ FreshFluff]

I downloaded the paper this is based on and started reading it shortly after this was posted, but only made it halfway through and am not sure when I'll get back to it. While I therefore can't make a completely definitive statement, particularly about the statistical analysis, I can say that what I read was unimpressive. The authors artificially manipulated the contours of the graphs in Figure 1: The relationship between HIV trends and Craigslist entry and activity level, p. 4. They don't make logical connections where they should, so their argument based on what other studies say about the link between casual sex and HIV risk is not rigorous. (The study is a PDF, so I can't copy and paste any of it here for critique.) And the authors aren't public healthcare or HIV experts, and as such I'm skeptical of their ability to negotiate the nuances. This looks like a paper that arose out of a desire to use Craigslist as the basis for a study without regard to how meaningful the study is.

 

Without having read the statistical analysis, I can't say that they've confused correlation for causation, but it's definitely a consideration. But what bothers me most is this: More Craigslist ads may mean more casual sex, but what does that really mean in the absence of clearly documented links between increased casual sex and, say, fewer safe sex precautions? Cases of HIV would increase even if the percentage of acts during which precautions were taken remained the same because more sex acts were engaged in. Whether more casual sex acts correlate with increased risk-taking is something this study can't and doesn't purport to measure. So mostly I feel like this is a tempest in a tea cup -- something that might be cited to confirm prior studies about casual sex and risky behavior (in other words, the real-world effects of those studies' conclusions). But if that were its aim, one would expect the studies' summary to say so. It doesn't.

 

For an example of what a well-written and rigorously argued research paper in the social sciences looks like, see any of the David Lisak papers cited or linked to here.

 

 

These days, economists are the ones writing many of the papers requiring complex statistical analysis (though the analysis here is pretty simple).

 

Based on a quick skim, I see some issues with the paper. But I don't think they need to prove that increased casual sex correlates with increased risk taking. Statistically, if people take the same precautions but increase the number of partners they have sex with, the incidence of HIV should increase.

 

Let's make a simplifying few assumption to make this easier to discuss: First, let's use pregnancy as an example instead of HIV since the probabilities are larger. Let's suppose condomss are always 98% effective against pregnancy. (Of course, in real life, their level of effectiveness would vary with the woman's fertility level and where she is in her cycle, but the basic argument remains the same.) Let's take the example of Jill, who always uses condoms, though it makes no difference here, has a different partner every time she has sex.

 

If Jill has sex once per year using a condom, her chances of getting pregnant would be 2%. If Jill has sex ten times per year using a condom each time, her chances of getting pregnant are 1-(.98)^10=18.3% . Using similar math, if she has sex twenty times per year under the same assumption, she has a 33.2% chance of getting pregnant.

 

So again, even if the precautions she takes remain the same, her chances of getting pregnant rise with the number of times she has sex.

 

It's late, so if the math is incorrect or if I've misunderstood your argument, let me know.

 

I only skimmed the paper, but I have a different question. CL wasn't introduced randomly into different cities. Demand for the service, including the "sex partners wanted" area of the site, likely had an effect on the timing of introduction. So it may be that CL was introduced more quickly into "sluttier" cities.

 

Also, re your comparison to the paper you linked in the "Why we pay" thread: The authors of this paper had to make do with data that's far less rich than the survey data that Lisak and Miller had access to.

[SteveEscort]

I wasn't thinking of it that way because I was thinking you were responding in your capacity as an escort and not a civilian, so to speak. Thanks for clarifying.

 

That's fair enough but escorts do have life experiences outside of escorting and it's unfair to presume that our input to any subject is purely limited to sex work experiences.

 

My experience of craigslist is that it is mainly timewasters and fantasists that frequent the site, echoed by clients I have seen who have replied to sex contact adverts on there and found the site to a load of nonsense.

 

I can see how you presumed I was talking about paid adverts but I never implied I was. I've no idea how they identified commercial sex adverts in there because they've not been allowed for about 4 years. Yes they exist everyday but not as outright commercial adverts.

[+ quoththeraven]

These days, economists are the ones writing many of the papers requiring complex statistical analysis (though the analysis here is pretty simple).

 

Based on a quick skim, I see some issues with the paper. But I don't think they need to prove that increased casual sex correlates with increased risk taking. Statistically, if people take the same precautions but increase the number of partners they have sex with, the incidence of HIV should increase.

 

Let's make a simplifying few assumption to make this easier to discuss: First, let's use pregnancy as an example instead of HIV since the probabilities are larger. Let's suppose condomss are always 98% effective against pregnancy. (Of course, in real life, their level of effectiveness would vary with the woman's fertility level and where she is in her cycle, but the basic argument remains the same.) Let's take the example of Jill, who always uses condoms, though it makes no difference here, has a different partner every time she has sex.

 

If Jill has sex once per year using a condom, her chances of getting pregnant would be 2%. If Jill has sex ten times per year using a condom each time, her chances of getting pregnant are 1-(.98)^10=18.3% . Using similar math, if she has sex twenty times per year under the same assumption, she has a 33.2% chance of getting pregnant.

 

So again, even if the precautions she takes remain the same, her chances of getting pregnant rise with the number of times she has sex.

 

It's late, so if the math is incorrect or if I've misunderstood your argument, let me know.

 

I only skimmed the paper, but I have a different question. CL wasn't introduced randomly into different cities. Demand for the service, including the "sex partners wanted" area of the site, likely had an effect on the timing of introduction. So it may be that CL was introduced more quickly into "sluttier" cities.

 

Also, re your comparison to the paper you linked in the "Why we pay" thread: The authors of this paper had to make do with data that's far less rich than the survey data that Lisak and Miller had access to.

 

Thanks for taking a look at the paper.

 

This is positioned as a paper with public health implications. I don't think economists or B-school professors are well-positioned to address them.

 

I recognized that the incidence of HIV should increase as the number of casual sex acts (or partners, as you put it) increases. That's what "Cases of HIV would increase even if the percentage of acts during which precautions were taken remained the same because more sex acts were engaged in" means. I'm saying the study, with its sensationalist title "Internet's Dirty Secret: Assessing the Impact of Online Intermediaries on HIV Transmission" and a summary and style skewed toward alarmism for a scholarly paper, oversells itself. The real reason for alarm -- at least to my mind -- is if more casual sex equals riskier behavior. What I've read so far implies as much, but doesn't make the case for it cleanly, and I don't see how it could. That's why I wrote "This looks like a paper that arose out of a desire to use Craigslist as the basis for a study without regard to how meaningful the study is" and called it a tempest in a tea cup.

My comment about meaningfulness (or the lack thereof) goes to your note that Lisak was working with richer data. The weakness of the data is much of the problem I have here (and the reason why I said it looked like they decided to do this because it allowed them to use Craigslist data; this topic isn't where they started out). So it seems we're more in agreement than it would appear.