Slashdot Mirror

← Back to Stories (view on slashdot.org)

Mobile Phone Data Can Track the Spread of Infectious Diseases

Posted by timothy on from the hashtag-#germanmeasels dept.

jan_jes writes: Researchers have used anonymous mobile phone records for more than 15 million people to track the spread of rubella disease in Kenya and were able to quantitatively show that mobile phone data can predict seasonal disease patterns. The researchers compared the cellphone analysis with a highly detailed dataset on rubella incidence in Kenya. They matched; the cellphone movement patterns lined up with the rubella incidence figures. In both of their analyses, rubella spiked three times a year. This showed the researchers that cellphone movement can be a predictor of infectious-disease spread.

21 comments

  1. Pre-diction by Anonymous Coward · · Score: 1

    No matter what requirements we try using to distinguish successful theories vs wild speculations, apparently people will come up with some sleazy way around them to sound important. If they pre-dicted something, that means that they published a paper earlier claiming a certain relationship between cell phone use and infectious disease. Did this happen?

    1. Re:Pre-diction by Anonymous Coward · · Score: 0

      What are you talking about? Do you just not know what "predict" means?

    2. Re:Pre-diction by Anonymous Coward · · Score: 1

      Epidemiological mechanisms driving this within year triple peak in incidence are the occurrence of three yearly troughs in transmission

      This data is quite interesting. I had never heard of three peaks in incidence per year. Apparently that is what this data shows, although the above statement is a tautology. They estimated transmission rates from the incidence data, so it makes no sense to say that seasonal changes in transmission rates cause the incidence data... I do wonder if this is due to people being more likely to bring their child to the doctor at times they are traveling more (perhaps the roads are safer some times of year?).

      Nevertheless, reporting rates are estimated to be very low in our analysis of this data, consistent with previous analyses (15) and the known mildness of this infection in children. This precludes an analysis of metapopulation dynamics (15), but for investigation of seasonal transmission, this will only bias results if there is additionally seasonal variation in reporting.
      [...]
      reporting rate (7e-5 to 9e-4).

      Yes, there is good reason to think that reporting rates will be linked to travel. If it is difficult or a hassel to travel to the doctor, the rubella won't get reported. I'm not saying that IS the explanation, but it seems plausible to me. The reporting rates are estimated to be so low (1 in 700,000 cases...) that minor changes here will have a huge effect.

    3. Re:Pre-diction by Anonymous Coward · · Score: 0

      I am sure it has taken on some new confusing statistical meaning like significance and confidence that just happens to be much weaker than the usual one. Whenever one of these new definitions is adopted, it *just happens* to make people think the work is more important and informative than it really is.

    4. Re:Pre-diction by Anonymous Coward · · Score: 1

      Thinking about this more, I can replace every reference to "seasonal changes in rubella transmission rates" to "seasonal changes in rubella reporting rates". Even in their SIR model, hold transmission rates constant and vary the reporting rates and you can get the exact same output. The exact same paper could be published with totally different conclusion just by switching two terms around a few times. It is not possible to distinguish between the two explanations from this data.

    5. Re:Pre-diction by Ol+Olsoc · · Score: 1

      What are you talking about? Do you just not know what "predict" means?

      my prediction is he doesn't.

      --
      The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
    6. Re:Pre-diction by Anonymous Coward · · Score: 0

      I predicted you would say that.

  2. Mobile Phones Can Spread Infectious Diseases. by zenlessyank · · Score: 1

    FTFY. At least that is what I thought it read when I first glanced at the headline.

    1. Re:Mobile Phones Can Spread Infectious Diseases. by ItsJustAPseudonym · · Score: 1

      Infectious diseases on public phones wiped out the Golgafrincians in Hitchiker's Guide to the Galaxy.

      We also know that diseases mutate and evolve.

      Apparently, public phone diseases have evolved from public phones to cell phones, and also from a fictional story to the real world.

  3. Study is not that valuable... by Anonymous Coward · · Score: 0

    All they found is:

    1. Kids get more infections when they come back from vactions - almost every parent knows it

    2. People make phone calls on vacation - this is also pretty deep

  4. Mobile phones spread like a disease themselves by Anonymous Coward · · Score: 0

    And look at all the sufferers with their necks craned down.

  5. Causality by Anonymous Coward · · Score: 0

    Actually, infectious disease causes cell data.

    It's interesting enough, but it seems to me like cell data is simply the lowest hanging fruit on the tree of detective work, despite being more invasive than not.

  6. Any excuse by PopeRatzo · · Score: 0

    It's for the kids! They're tracking my mobile phone data so my kids don't get Ebola. Thank you, NSA. What would we do without you?

    --
    You are welcome on my lawn.
    1. Re:Any excuse by fustakrakich · · Score: 1

      I prefer fashionable earrings

      --
      “He’s not deformed, he’s just drunk!”
    2. Re:Any excuse by Chikungunya · · Score: 1

      It would be very interesting to see the Ethical committee discussion when they approved the study, I bet they are convinced that "anonymizing" the calls means impossible to identify by any means.

  7. Re:Diseases are for (mad) cows. by fustakrakich · · Score: 1

    The cell phones are our ear tags. Saves having to implant them. Not that it matters much anymore since the batteries aren't removable now.

    --
    “He’s not deformed, he’s just drunk!”
  8. Outlaw phones! by Anonymous Coward · · Score: 0

    Then diseases will no longer spread

  9. That's what I heard by AndyKron · · Score: 1

    Never take your cell phone when buying weed. That's what I heard.

  10. Not really sure what it shows by 140Mandak262Jamuna · · Score: 1

    Cell phone movement patterns coincided with annual peaks of German measles (actually renamed Liberty measles during World War I anti-German frenzy in USA) in Kenya. Not sure what else would correlate with annual events. Rains, migratory animals, pollen, food prices, seasonal foods all have some annual rhythm. So there is going to be high correlation between all annual events. So I really don't get what this cell phone movement correlation withe G measles really means.

    --
    sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
    1. Re:Not really sure what it shows by Anonymous Coward · · Score: 0

      Yes, everything is correlated with everything else. Merely observing a correlation is not interesting. Most infectious diseases are seasonal, and this remains unexplained since so many things are correlated. It has driven many a researcher mad, to the point where you can find papers where the author states that his colleagues warned him to stay away from this problem.