Two examples of misreporting

The media provides an important forum for different stakeholders to educate, inform, pursue, and influence media audiences. Questioning media credibility is not a new phenomenon, especially when new technology has opened up access to information and knowledge. It’s becoming more and more difficult to discover the truth in the news in all sorts of areas, including food and nutrition.

Let’s look at examples of how the results of scientific articles have been presented differently by journalists:

Example 1: Milk

.

Independent website. 29 October 2014.

This headline is taken from the Independent website.

There were similar headlines such as ‘Dangers of more than 3 glasses of milk a day: High intake may not protect against broken bones and could actually increase chance of death’ in the Mail Online and ‘Three glasses of milk a day can lead to early death, warn scientists’ in The Daily Telegraph (please note this article is behind a paywall).

These headlines are a result of the study Milk intake and risk of mortality and fractures in women and men: cohort studies. which was carried out in 2014.

Reading the original scientific article, we see that the researchers indeed concluded that ‘high milk intake was associated with higher mortality in one cohort of women and in another cohort of men, and with higher fracture incidence in women’. However, the researchers also emphasised that ‘given the observational study design, with the inherent possibility of residual confounding and reverse causation, a cautious interpretation of the results is recommended’. An observational study design can link milk consumption with particular health outcomes, but it can’t prove direct cause and effect because researchers can’t be sure that it’s just milk consumption that’s having the effect on mortality and fracture incidence – there may be other lifestyle factors affecting the results. These are termed ‘confounding factors’ and we’ll look at another way such links can be misinterpreted in Week 2.

Example 2: Bread

The Sun website. 6 June 2017.

The ‘Bread affects clinical parameters and induces gut microbiome-associated personal glycaemic responses’ study is a randomised trial which compared the effect of consumption of different types of bread (white versus brown sourdough) for one week in 20 healthy people. The results showed no overall differences in glycaemic control when people consumed white bread compared with wholemeal sourdough bread, however, the researchers found there were very diverse responses to consuming either bread which may have been caused by their individual gut bacteria.

When we read the original publication, it is straightforward to find out the limitations of the study, such as a small sample size of only 20 people, and the trial only lasted one week. Also, the authors declared a conflict of interest as paid scientific consultants for a company that promotes personalised microbiome and nutrition. However, the current results were reported by newspapers in headlines such as ‘Sliced white bread is “just as healthy as brown’’, shock findings reveal’ in The Sun newspaper, which isn’t really what the research found, and the newspaper did not mention the authors’ conflict of interest.

The problem with such reporting is that the stories are very quickly discredited when other studies reach different conclusions and readers don’t know which report to believe.

The longer term implication of such reporting is a crisis in credibility and widespread distrust of the media and science.

You’ll be exploring a scientific paper in detail towards the end of this Week. If you would like to read the full studies on which the headlines in this Step are based, you can find the links in the ‘See also’ section below.