Home » Social Media » ‘Twitter Rage’ Study Correlates Social Stress With Fatal Heart Disease

‘Twitter Rage’ Study Correlates Social Stress With Fatal Heart Disease

23 January 2015

From the Huffington Post UK:

Twitter rage could be bad for your health.

Researchers say there is a correlation between people who are often angry on the social media site and incidence of heart disease in the United States.

The study analysed language from public Tweets sent between 2009 and 2010, and compared it to existing public data on where heart disease is most prevalent across the country.

It found that communities who frequently used ‘negative’ language had a greater rate of death from heart disease – indicating wider problems with stress in those areas, which could contribute to cardiological problems.

The opposite correlation was found in communities which used positive words like ‘wonderful’ more frequently.

. . . .

The Twitter rage study looked at Tweets sent from 1,300 US counties, and compared an automatic ‘sentiment’ analysis with heart disease data from the US Centers for Disease Control and Prevention.

. . . .

“These people are the canaries of the psychological profile of their communities,” Johannes Eichstaedt, the study’s lead order, told WaPo.

Link to the rest at the Huffington Post UK and thanks to Jan for the tip.

Social Media

15 Comments to “‘Twitter Rage’ Study Correlates Social Stress With Fatal Heart Disease”

  1. Excellent news.

  2. I think rage is bad for your heart, period.

    • Exactly. I have a feeling that the scant possible data related to twitter is less relevant than how those same people experience stress and anger in the rest of their life.

      Sounds like hack science or hack science reporting.

  3. So the trolls will all die early, hm? I think that’s wishful thinking.

    However, I urge everyone to follow my current internet mantra. Every time I want to get into a hot discussion, I remember this:


    • You know the old gag about the guys in prison, who have heard all of each other’s jokes so often, they tell them by number?

      We must be getting pretty close, when I start to recognize XKCD comics by URL. 386 was a very good number.

      ObOnTopic: Unfortunately, it’s more likely that the people who are being trolled will die early. A lot of the trolls seem to have no feelings at all.

    • Oh, the trolls aren’t stressing. They’re provoking it in others. Let’s frame it like this instead:

      “By increasing stress, anger, and heart strain, trolls are a danger to public health. Their homes should be quarantined and then sterilized by fire.”

  4. I’ve had to block people on facebook and drop twitter followers because their opinions were hazardous to my health.

    Especially enraging are the tweets. It’s easy to slam someone in 144 characters. Challenging debate? Impossible. Best to avoid it.

  5. Ah, correlation vs causation. From a page called “Spurious Correlations”:

    * Number people who drowned by falling into a swimming-pool
    correlates with number of films Nicolas Cage appeared in

    * Divorce rate in Maine correlates with per capita consumption of margarine (US)

    * Age of Miss America correlates with Murders by steam, hot vapors and hot objects


  6. Or, ‘Stressed people with other problems tweet too much!’ …


    (Mark Twain?): “Figures don’t lie, but liars figure.”

  7. @ Sarah, I had the same thought. There’s a HUGE problem with causality here.

    It could just as easily be:

    Angry people who die of heart disease, vent their anger on Twitter if they had an account.

    I’m guessing the problem has to do with anger, stress and heart disease, not Twitter. Sheesh.

    • I’m replying to myself because I want to address it from a different angle.

      This study could, for example, be proving that if you express rage on Twitter, you will develop heart disease and die.

      Or maybe it’s saying that if you do NOT express rage on Twitter, you will NOT develop heart disease, nor will you be angry.

      Or maybe it is saying if you have heart disease, you will express yourself negatively on Twitter. And die.

      I just love causality.

      • Or maybe it is saying nothing at all.

        Remember the Swedish study about cellphone use causing particular cancers? Turns out they ran the numbers on disease after disease, looking for something that showed a correlation. This method is guaranteed to produce false positives, especially if you use the standard .95 confidence interval.

        Since we’re quoting XKCD in this thread:


        Unfortunately for the international health panic industry, the alleged correlation disappeared after the study was published. Cellphone usage skyrocketed, but the particular cancer to which they said it was linked did not become any more frequent.

        Most correlations are just coincidences, and will disappear if you study a larger sample of the population. At this stage, trying to explain the correlation in terms of causation either way is a matter of telling Just So Stories. This is how superstitions get started.

  8. Correlation is not causality. It could be that people in stressful circumstances are more likely to exhibit both rage and heart disease.

    Fundamental attribution bias in action.

Sorry, the comment form is closed at this time.