So, why isn’t Foursquare actively being used to improve health?  It has enough data about me to provide personal recommendations to improve the way I live.  Many people have recognized the potential impact Foursquare could have, but there are very few (if any) health-related applications developed around this service.

Tony and I, in consultation with our resident behavior designer (Steph), worked to address this issue during the first-ever Foursquare Global Hackathon. In less than 40 person-hours (and, combined, 10 cans of Mountain Dew Throwback), we created Fourhealth.

The goal is simple: we want to help users make healthier decisions about where they eat out.  We do this in a couple of ways.  First, when a user signs up for Fourhealth, we ask them for height and weight information (and continue to collect weight via text message on a weekly basis).  Then, about 90 minutes after each checkin to a restaurant, we text the user to find out how they’re feeling after eating there.  This information is aggregated to the food venue, giving us a data source of objective (i.e., BMI) and subjective (i.e., feelings) health elements.

Immediately after each checkin (thanks to Foursquare’s new Push API), we look for previous feelings the user has had about the venue, and try to find a potentially healthier restaurant near where they are.  If we find either, we immediately text them to let them know, giving them the knowledge to make a potentially healthier decision:

Of course, you can’t always just leave a restaurant because you got a text message telling you to go somewhere else.  Maybe you’re meeting a group there, or the suggested venue is closed, or you just really want a triple whopper with cheese value meal.  But, there are times when you’re just hungry and go for the first place you see (only to find the organic cafe just around the corner).  Or are sorta thinking about your health, but just need a gentle nudge.  Or are in a completely new city and don’t know where to go.  We can help with all those situations.

Future Development

What I’ve described so far is our first iteration, and was completed in less than two days.  It’s live, and you’re welcome to sign up and try it out

However, there are many more things that need to be developed/thought out/changed, including:

• Improved analytics.  Imagine a Klout-esque score, only for how healthy a restaurant is.
• Ability to search.  We want to help people find healthier restaurants when they’re in unfamiliar cities.
• Personalized visualizations and data.  Some people really like exploring how their data affects their health.
• Withings integration (so you don’t even need to enter your weight).
• Weekly email summary.
• Venue-specific healthy food recommendations.

The Minnesota Department of Health (MDH)¹, from analysis of data collected by their partner services program, says:

Commonly reported risk factors were meeting partners on the internet, anonymous sex, and no condom use

Now, I don’t want to be the debbie downer, but if I’m not using condoms and am meeting partners on the internet for anonymous sex, then being told to use condoms and being handed a brochure is not going change my behavior.

I’ve been working in sexual health education/testing for nearly a decade, and I have yet to meet a young man that didn’t know he should probably be wearing a condom.  Education is not the problem².  The problem is that having sex without a condom feels really good.  Much better than with a condom.  So much better that, in that moment, the risk of getting syphilis or HIV feels like a pretty equal trade for the feeling of sex without a condom³.

A brochure can not compete against our biologic drive to exchange body fluids.  Ever.

Craigslist Explored

What can compete, however, is sex.  But before I explain, let’s look at a few personal ads from Craigslist (sorry, this is sexually explicit):

22 one fifty 6′ 7cut bottom/vers
gl.ddf, into jo, oral, making out, want to bottom safely
looking for ddf, good shape, top under 30 unless really hot
host/travel won’t respond to replies w/ no pic

Looking for a btm who’s willing to let me go wild….flip on your stomach, put your ass up, I slide in… I am 5’10″, 145lbs, blond, 7″ cut, nonsmoker, DDF, HIV-, popper friendly. Would prefer u be nonsmoker and definitely DDF/neg. Hit me up! Please send age pics.

looking to have some fun now, 29yo hiv+ looking for some fun, vers, oral, body contact, open to whatever, bb or safe cool, lets get together

A few definitions/codewords:

• “bb” or “wild” = bareback, anal sex w/o condoms
• safe = generally anal sex w/ condoms
• ddf = drug and disease free
• neg = HIV-negative
• popper = amyl nitrate
• gl = good looking
• jo = jacking off
• top = insertive partner; bottom or btm = receptive partner; vers = either

So, each of these personal ads share two characteristics: 1) they explicitly state what they want (e.g., “let me go wild”, “bottom safely”, “bb or safe cool”), and 2) they state their HIV or STD status (e.g., “DDF/neg”, “hiv+”).  Of course, not every Craigslist post does this, but many do — if you don’t believe this is an accurate representation, then start browsing (and comment on this post with your findings).

In public health terms, they’re practicing harm reduction by attempting to serosort.  But there are two major questions that are raised:

1. Are you actively ensuring your partner is “DDF”?  Or are you relying on your post to sort out all those who aren’t “DDF”?
2. How do you know your partner knows their HIV/STD status?  When were they last tested?(I’ve met more than one person who assumes they’re HIV-negative, yet has never been tested).

Data-Driven Sex

Qpid.me provides HIV/STD test verification for online dating⁴.  They work with your clinic to get your HIV, syphilis, gonorrhea, and chlamydia test results, and then allow you to share them with whomever you like.  Right now, it’s done via text message, although there are plans to expand soon. So, for example, you can get my test results by texting “Verify John Demo” to 774363.  ”John” is my username, and “Demo” is my access code — this allows me to ensure I’m only sharing my results with the people I want to.

This could be a huge game-changer for sexual health.  Now there’s a tool available to help those no-condom-wearing, internet-using, anonymous-sex-having heathens men serosort using actual, verified data.  And, of course, all of the normal just-want-to-be-close-to-someone guys and gals could benefit from Qpid.Me, too.

Even more important, there’s a sexual incentive to get tested more frequently.  If the status quo is to use self-reported HIV/STD status as a criteria for selecting sexual partners (e.g., “I will only have sex with you if you say you don’t have HIV”), what happens when you can verify the status along with approximately how long ago they were tested?  A potential partner would be more attractive if they’ve been tested more recently, meaning: getting tested = getting laid.  And who doesn’t want to get laid?  (it sure beats having your doctor hand you a brochure about syphilis).

For people living with HIV, you can verify your viral load and also share that with potential partners.  With an undetectable viral load, one becomes less infectious (it is still possible, but less likely).  So, this allows people living with HIV to have informed conversations with their partners about the risk of HIV transmission.  Done correctly, this could actually alleviate some of the stigma of HIV-infection, get more people laid, and decrease HIV incidence.  Win-win-win.

Use fire to fight fire.  [insert your own joke about red-heads or gonorrhea].  That is, use sex to fight sexually transmitted diseases by redefining how people have sex.  Instead of having a traditional information campaign or continuing to hand out condoms, actually arm people with a tool to be able to make informed decisions.  Even better, give them a tool to allow them to do something they’re already doing, a little bit better.  By encouraging people to make informed decisions with actual data, you’ll see them get laid more often while getting fewer STDs and getting tested more frequently⁵.  And, most importantly, it requires essentially no change in current behavior.

If you haven’t figured it out by now, Qpid.Me is the health tech start-up I took a job with.  We’re currently in private beta, while we work on building out a bunch of the back-end.  If you’re interested in signing up, head to the website, or comment here/email me — I might be able to hook you up

Seriously, who footnotes a blog post?
1: I pick on MDH because I’m from Minnesota.  They’re actually quite good at what they do, given the constraints of politics.  The issues I raise are pretty ubiquitous among all major public health departments.
2: While I argue education is not the problem, I do believe comprehensive sexuality education in schools would help lower HIV/STI rates.  That’s more of a shift in attitudes than providing students with facts.
3: Of course, I wouldn’t know from first hand experience…
4: Now, if only there were a Qpid.Me for verifying penis size.
5: There are a couple of research studies in the works regarding this, but right now this statement is based on qualitative information from users and theory.

I couldn’t find much about the DigMe program in particular, but here’s the video from the U: