Protocol for suicidal users on social networks?

May 19, 2008 10:44 AM

I work as an admin for a social networking site, and I'm trying to figure out protocol for how to handle users who seem suicidal.

Do you work for a social networking site or know anything about protocol for this kind of situation? If a user is posting information and/or pictures on his or her public profile that suggests this person is thinking about killing him/herself, or is currently engaging in self-harm, do we have a legal responsibility to intervene in some way? (I know you are not my lawyer.)

I'm also curious as to what you think the moral responsibility is here as well.

(By the way, I do plan to bring this up to the higher ups but I'd also like to get the hive mind's input.)

Thanks, guys n gals.

posted by infinityjinx to Computers & Internet (3 answers total)

Not sure how helpful it is for you, but here's Livejournal's proposed policy... they seem to only intervene when there is no question that the suicide is definitely intended, planned for a particular time, and written about by someone they can track down in real life.
posted by rivenwanderer at 11:33 AM on May 19, 2008 [1 favorite]

First off - what a crappy situation to be in, for you, your company, and the user.

I don't work for a social networking site, but have many yeras of experience with maintaining sites with large numbers of users (millions+) that can post on message boards, update profiles, post pictures, etc. What I'm about to type is the protocol I would follow if something like this were to come up.

Moral obligation: This all depends on the person/company - for most companies I'd imagine it's more of a legal/PR issue than a moral issue. Me? I'd try to intervene, but only if I was 100% positive that suicide or threat of harm is absolutely imminent. Otherwise, you could get into a sticky legal/bad publicity situation. I would not go on a massive crusade for this person, however. You never know when someone is just pulling a prank online, unfortunately. Anyway, if I were in your shoes, here's what I'd do, and I separated it into 3 categories...

Here's how I'd intervene from a social standpoint:

I'd approach them via the social networking site if there is some type of messaging that's private and I'd follow the steps outlined here: http://www.save.org/index.cfm?fuseaction=home.viewPage&page_ID=705E1907-C4DD-5D32-2C7087CE5924CCA4

Then, if they straight up said that they were thinking of killing themself, or causing themlself harm, I'd continually urge them to call the National Suicide Prevention Lifeline 1.800.273.TALK or at least to visit http://www.suicidepreventionlifeline.org/

Here's how I'd intervene from a technical standpoint:

Is there a requirement for people to put down contact information beyond an email address before they use your site? If not, this makes it hard to contact the local authorities for the user, should they be in imminent danger as you don't know where the user lives. You can always call your local police dept and ask them how you should proceed (turning over log-files with the users IP address, so they can be traced, etc). Before doing any of this though, I urge you to check out your privacy policy to ensure you don't get sued/bad publicity for sharing their information with a 3rd party, no matter how noble your reason is.

Alternatively, I'd run a traceroute on the IP address that shows up when they post to see where it resolves (An IT admin can help with this.)

1.) If the person is posting from their job, you might get lucky and be able to track it to an office where you can alert the IT staff. "A user with the IP Address xxx.xxx.x.x which resolves at your company has been posting at my site and appears to be suicidal or inflicting self-harm. Please investigate if this person needs help." Same goes if they are posting from a school, etc.

2.) If the person is posting from home, you can sometimes figure out at least the city location, such as "Madison, WI" and what provider they use such as Comcast. At that point, I'd check out the website for their service provider to see if they have anything listed for how to report a user. The admins at their service provider can then track the IP address to a physical location. I'd explain the situation and ask for their assistance, too.

After that, it's up to either the service provider, or their workplace to intervene.

Here's how I'd intervene from a legal standpoint, 'cause we live in a lawsuit-happy society:

Before doing any of the above, I'd first run it by my boss, or HR department (as you stated you're going to do). I'd also make sure our terms-of-use portion of the website completely absolves my company of any liability from such a case. Basically something that says you use the site at your own risk, blah blah blah. Good idea to get lawyer advice here. As long as you have such a statement, I do not believe you are under any legal obligation to intervene, even if the threat is 100% imminent. I can't stress enough that I am not a lawyer, so it's really best to involve one and ask them to be sure. This likely wont be the last time your company has to deal with an issue like this if your site is a popular one, or growing in popularity.

Good luck!
posted by xotis at 12:15 PM on May 19, 2008 [1 favorite]

As someone who's got suicide helpline training, I second everything xchmp says. In referral, one of the big things you have to look at in the US is if the hotline calls 911 for the caller or not. The internet tend to collect people who want to be heard and not helped, hotlines are a hit or miss.

I think that a safe move all around would be a private message to a suicide line and a note that the site is concerned about the things the user is posting online, recommending the user talk to someone about their feelings who is trained to listen to people in his/her situation.

I have no idea about the legal or site aspects of it, but I would always lay on the side of caution. It's always better to have an annoyed user because you care, instead of angry parents/friends because you did nothing.
posted by Gular at 3:14 PM on May 19, 2008