Confusing Correlation with Causation

We’re all familiar with the breathless tone media often take with news about health and medicine. Bad enough that click-bait headlines fail to describe the science, if any, behind the latest breakthrough, but it’s even worse when the stories that follow can’t get it right.

A recent example was described on HealthNewsReview.org (HNR), a tenacious watchdog of accurate health and medicine media coverage. The story concerned a study that ran on the BBC website under the headline “Being overweight ‘reduces dementia risk,'” and on the Washington Post site as “Being fat in middle age reduces risk of developing dementia, researchers say.”

Uh, no, they don’t. And, no, heft does not ward off dementia. At least not according to this research, which was published in The Lancet Diabetes & Endocrinology.

As HNR explained, such a conclusion is a common mistake when journalists confuse “association” with “causation.”

The research was an observational study, not a clinical trial. The latter sets the “gold standard” of science by assessing the safety and effectiveness of something. Observational studies can be useful in leading researchers toward more specific studies and trials by determining associations between specific factors and outcomes.

Clinical studies require researchers to intervene, and observational studies require them to observe.

As HNR explained, “It is precisely these kinds of stories about observational studies – one day suggesting that overweight raises risk … another day suggesting that overweight lowers risk – that contribute to the decline in public confidence in journalism and in science.”

Observational studies can be interesting, and useful, but they are scientifically insufficient to support the sweeping claims made by the BBC and The Post. For a more accurate interpretation of The Lancet study, HNR suggested, see the story by HealthDay: “Could Obesity Help Protect Against Dementia? Large study suggests that possibility, but experts call for further research into surprising finding.”