How We Misinterpret Medical Statistics

If I told you that reported chlamydia infections in Australia had risen eightfold over a ten year period and that the majority of infections were in people under 29, you might conclude that those pesky youngsters are routinely indulging in unsafe sex. You'd be wrong — a reminder that you need to consider the full context of statistics before drawing conclusions.

Image courtesy of Shutterstock

So this is the data we have. Reported cases of chlamydia in 2011 in Australia were around 80,000, and 66,000 of those were in the 16-29 age group. In 1999, the total number for everybody was just above 10,000.

What is driving those numbers up, however, is not necessarily increased infection, but increased detection. "While chlamydia numbers are increasing, these increases are largely because we're doing more testing, " associate professsor Jane Hocking, an epidemiologist at the Centre for Women’s Health, Gender and Society at the University of Melbourne, explained in a recent media briefing. "The more you test, the more infections you're going to find. And the number of tests have also been steadily increasing each year." Young people are more likely to be sexually active and seeking tests. Unless everyone gets tested, we can't conclude that they're more likely to transmit infections than other age groups.

Statistics are useful tools for understanding the world, but they always need to be examined carefully so you understand what's being said.

That doesn't mean there isn't a problem to be addressed. Hocking's own research suggests that around 5.2 per cent of men and 4.3 per cent of women in Australia have chlamydia. If applied across the relevant population, that suggests 210,000 people should have been diagnosed — a big gap from the 80,000 identified.

As 85 per cent of people with a chlamydia infection show no symptoms, many of them won't get tested. To actually reduce levels of infection, even more testing is needed, so people who do have the infection can be treated and avoid passing it on to others. "We need to get testing rates up to about 40 per cent before we can expect to see a decline," Hocking said.


    The proportions still hold true my friend. If 66% of chlamydia is in a 10 year segment of the population I would be worried. Also chlamydia can leave you infertile so I wouldn't play down these stats considering how bad the outcome could be

      You're still misunderstanding. The proportions are meaningless.

      It's not that 66% of infections is in a 10 year segment, its 66% of the REPORTED infections. And reported infections are going to be higher in the age bracket the tests more.

      There's really nothing here to suggest that younger people don't have a LOWER infection rate. They may just be more diligent in getting tested.

        There is no screening process for chlamydia. People who get tested are the ones who have urinary symptoms. Other groups have less symptoms and get tested less because, guess what, they have lower rates of chlamydia. This argument only applies to screening a population. However this is a diagnostic test. 2 very different things

          Not necessarily the case, j - STI testing, including testing for chlamydia, is part of routine sexual healthcare. More young people tend to seek that healthcare than older people who have settled down, because of the perception that it isn't necessary when in a monogamous relationship. Some of the people who get tested are those who have symptoms and are looking for the cause, but plenty of others are tested because it is the responsible thing to do in terms of your own health and the health of your partners, even if you believe you haven't been exposed to a risk.

            That's a good point. When the wife and I first got shacked up, we both got tested to make sure neither of us were bringing anything undesirable in.

    Also, speaking of misinterpreting stats, you give 2 reference points a decade apart. Notifications are mandatory to the health board so stats are available for every year. You purposely excluded all other data sets which show a clear trend of increase despite no change in diagnostic method in the past 4 odd years (earlier depending on the centre). Don't rubbish medical stats mate, especially ones that are legit

    You're still missing the point of the article j

    Sigh. I guess you're all experts. I guess lifehacker linked the study and I guess you all checked it out. Aww no? You can say I'm missing the point all you want. Just disregard the fact you cant see the yearly trends. I'm not a maths student, I've just worked in sexual health at one of these clinics that you know actually does the testing. But far be it from me to tell you that young folks are spreading stis because you don't want to hear it. FYI I'm 24 and I care because this has jumped like 3 fold in the past few years. But yeah, tell me again im missing the point

      j, if Angus had given us the whole story then he wouldn't seem so uber intelligent now, would he? He is, after all, a *journalist*, you know - they don't just give the title out in weeties boxes!

      The sad fact is, unless you're sourcing your information directly from a peer-reviewed journal article, or have the raw statistical data, you can't trust what you read. Lifehacker/today tonight/ACA can put any spin on it they want, and the more sensationalist it is the more hits they'll get, the more sponsorship from Microsoft or Vodafone. I do so love the irony of being misled in an article warning us about misinterpreting data!

      And while there was nothing in the article to refute the assumption that Angus placed into our mouth regarding the kiddies and unsafe sex, he was still happy to make an a$$ of u and me...

Join the discussion!

Trending Stories Right Now