Bare Minimum Standards for Science Reporting: A Brainstorming Session | Vol. 3 / No 28.4

Photo: Ecig Click, CC BY 2.0

In response to the wonderful segment done by John Oliver on Last Week Tonight about the atrocious state of science reporting in the news, Phil Plait recently posted his own thoughts. The part that really got me was about how the back and forth reporting of preliminary data as often-contradictory fact doesn’t hurt news media so much as it hurts the reputation of science writ large. So to that end, I’d like to start compiling a list of small things which, if consistently done, will help to minimize the damage. Because most of the flip-flopping in the news is about medical studies, that’s where I’m starting today, but I’m sure this will change over time with contributions from friends and readers alike.

Here’s some pretty simple ideas to begin with:

  • If we’re talking about a study, state what kind. Is it an exploratory study? Is it an observational study? Is it experimental? Is it a meta-analysis of hundreds of studies? Is it a new set of observational statistics out of a government agency? You don’t have to be technical, but you do need to be a little descriptive.
  • State whether were talking about real life or laboratory conditions. Was the study was done in a petri dish or in an animal? Remember that bleach and bullets both kill cancer in a dish, so if something cures cancer in a bundle of cells, eating it won’t help.
  • If a study was performed on animals (including humans), always say what kind of animal. There’s a massive difference between a study performed on humans and one on rats, for instance, and the number of treatments that , honestly, not that high.
  • If something’s in human trials, differentiate between types. A phase one clinical trial doesn’t tell you whether a new treatment is effective, just whether or not it’s safe enough to test in humans. A phase four clinical trial is given after the drug’s gone to market to examine possible long-term side-effects.
  • Express a little skepticism: explain at least one possible weakness of the study. Does the study point out correlation but not causation? Does the study use very small sample sizes? Are the study’s findings still preliminary?

These aren’t always applicable, but we’ll try the best we can. Here’s a story from this week on children being poisoned by ingesting e-cigarette liquid. It’s based on an AP release, which itself is based on an article in the journal Pediatrics.

It’s a pretty unambiguous story: from 2012 to 2015 the number of children poisoned each month from drinking e-cigarette fluid has gone up significantly, probably in line with the spread of e-cigarettes. Though the study doesn’t actually talk about the rate of adoption of e-cigarettes, the fact that the number of “exposures” to the combined tobacco/e-cigarette category seems to remain fairly stable suggests that exposures to e-cigarettes are likely coming at the expense of exposure to tobacco. It’s also lumping together all kinds of exposure, meaning everything from touching to inhaling to consuming, which is a pretty big weakness of the study if you ask me.

Here’s how bad it could get on a morning show:

E-cigarettes are in the news again this week with reports that the number of children being poisoned by the nicotine-laced e-cigarette liquid has skyrocketed 1500% since 2012. According the Dr. Gary Smith, the lead author on the new study, “this is an epidemic, by any description.” Symptoms of exposure to the liquid include jitteryness, and faster heartbeat, and vomiting. New measures are being introduced to force companies to make their packaging display more warnings and be more child-resistant: by the end of the study more than two hundred calls to poison control were reported every month after exposure to the liquid, which often comes in sweet flavours and bright packaging. If you or someone you know uses e-cigarettes, remind them to keep the products securely out of the reach of children, and if you suspect a child has been exposed to this dangerous substance, contact poison control immediately.

At this point you could insert some disapproving banter about people these days and toxins, and you’d be all set.

Meanwhile, here’s how I’d report it, just for a little comparison:

A new study in the journal Pediatrics is serving as a reminder to e-cigarette users to keep the devices — and the liquid in them — safely out of the hands of children. The data for the study come from the National Poison Data System, which recorded phone calls to poison control from 2012-2015. As the number of smokers making the switch from tobacco products to e-cigarettes has risen, the number of children getting into — and sometimes ingesting — the e-cigarette liquid has risen as well. This can be more dangerous than ingesting tobacco because a given amount of the nicotine-laced e-cigarette liquid is easier to ingest than its equivalent in tobacco. The study claims that exposure rates have risen roughly 1500%, though in this case exposure could mean ingesting, inhaling, or touching, so long as it resulted in a call to poison control. While the number of children exposed to tobacco products greatly outpaces the number exposed to e-cigarettes and the numbers involved are still small, the rise in frequency — added to the relative danger from accidental ingestion — is such that the study’s authors are recommending measures be taken, such as warning labels and child-proof containers. In the meantime, smokers of tobacco products and e-cigarettes alike should keep in mind the potential danger to children, and be vigilant in efforts to keep it out of their hands.

I doubt it’s a perfect representation of the study, which once again you can find here, but I think just a little bit of specificity and skepticism could help in the future. Let me know by commenting here, on our Facebook page, or via Twitter if you have other suggestions.


Thanks for reading! I only get paid in my own (and your) enthusiasm, so please like This Week In Tomorrow on Facebook, follow me on Twitter @TWITomorrow, and tell your friends about the site!

If you like our posts and want to support our site, please share it with others, on Facebook, Twitter, Reddit — anywhere you think people might want to read what we’ve written. Thanks so much for reading, and have a great week.


Richard Ford Burley is a human, writer, and doctoral candidate at Boston College, as well as an editor at Ledger, the first academic journal devoted to Bitcoin and other cryptocurrencies. In his spare time he writes about science, skepticism, feminism, and futurism here at This Week In To

One thought on “Bare Minimum Standards for Science Reporting: A Brainstorming Session | Vol. 3 / No 28.4

  1. Including important data for human test subjects, like biological sex and race. When medical studies are done primarily on one group, it may not reveal the effects on other kinds accurately.

Comments are closed.