Does quality research make the news?

by | Healthy Living, Tests and Investigations

A recent audit of the types of medical studies that make it into the media has found that lower-quality research is more likely to feature.

News coverage of medical research ranks high for public interest in the potential new treatments and cures it details.

Looking past attention-grabbing headlines promoting ‘breakthroughs’, it’s important for the public to know if the research is actually good quality.

A team of researchers looked at the press given to 75 medical articles that featured in widely circulated newspapers. They compared the studies profiled in the newspapers to a corresponding set of 75 articles published in high impact medical journals over the same period of time.

The types of articles audited were all considered ‘clinical research’ in that it had direct human relevance.  Using a systematic method to assess each of the studies, the researchers were able to make some broad comparisons.

Newspaper coverage of research was only half-as-likely to profile randomised-controlled clinical trials (considered high quality research) compared to the expected frequency of publication in the journals.

Observational studies were by far the likely choice for newspaper stories (at 75 percent of stories) compared to journal publication rates of 47 percent.

The observational studies that received the most media attention were also more likely to have a smaller sample size of people and were more likely to be cross- sectional i.e. at one point in time. Cross-sectional studies are considered very weak types of observational research and can find associations that may just be random connections.

Observational studies can be attention grabbing for the interesting associations they find between health and behaviour. The key is that even the best quality observational studies can only confirm an association, not prove something, for example that “eating two servings of blueberries a day will halve your risk of cancer”.

Observational studies can help lead to designing randomised controlled trials but often the subsequent findings don’t agree with the lower-quality research. This is why you don’t often see headlines stating that “eating two servings of blueberries a day won’t cut your risk of cancer” because research that doesn’t find much doesn’t normally make for interesting headlines.

Implications

The media will choose to report on medical research that likely has the most interest to their audience, but it’s important to consider that the personal implications may not be as strong as promoted.