One in Every 3 Articles Shared on Social Media Contains Misinformation

According to a new analysis, one in three articles published on the most popular news and social media pages contains misinformation, including articles with the potential to cause serious harm to patients.

Of the 200 most popular news articles on prostate, lung, breast and colorectal cancer, with 50 articles for each, approximately one third (32.5%) contained misinformation. Furthermore, 76.9% of these misinformation-containing articles included harmful information.

Researchers involved in the analysis stated: "The internet is a leading source of health misinformation. This is particularly true of social media, where misinformation spreads faster and more widely than verified information." This was their explanation.

Skyler Johnson of the Huntsman Cancer Institute at the University of Utah in the United States, where the study was conducted, says: “We need to address these problems directly" he explains.

"As a medical community, we cannot ignore the problem of cancer misinformation on social media or ask our patients to ignore it. We should empathize with our patients and help them when they encounter this information. My goal is to help answer their questions and provide accurate information that will give cancer patients the best chance of the best outcome."

Social Media: Popularity and the Search for Miracles Instead of Accuracy

Published on July 22 in the Journal of the National Cancer Institute , the online study screened articles posted on Facebook, Reddit, Twitter and Pinterest between 2018-2019. Popularity was assessed through reader responses, such as upvotes, comments, reactions and shares.

As an example of misinformation, Johnson pointed to an article headlined "44-Year-Old Mother Claims CBD (Kannabinoidol, Cannabis) Oil Cured Her Breast Cancer in 5 Months". The author labeled the February 2018 truetheory.com article an "opinion". At the end, it also linked to another article by the same author published in the Daily Mail newspaper.

Ideas and claims in such articles can be very influential. After 20 years as an oncologist, it is unfortunately no longer surprising when a patient rejects treatment recommendations and chooses "alternative" treatment instead. Another "trap" for patients and relatives is the search for miracles. As is evident, the proportion seeking to exploit this search is far from small.

• Related topic: Intelligence and Money May Not Be Enough to Find the Right Path in Cancer Treatment!

Brian Southwell of Duke University, who studies patients and misinformation, says misinformation is sometimes not sensational, yet remains influential through cleverly written wording and presentation. “It is not the lie that is magically appealing; the way misinformation is usually framed can make it appealing.” Southwell recommends that doctors be proactive in taking a stand against medical misinformation.

In a recent article in the American Journal of Public Health, Southwell says: "Rather than waiting for patients to voice concerns without prompting, healthcare professionals should talk with patients about possible misinformation" . In short, he recommends asking patients what they know about their cancer treatments. "Patients generally do not know that the misinformation they encounter is misinformation. Approaching patients with compassion and empathy is a good first step."

Study Details

In the study published by Johnson and colleagues, two National Comprehensive Cancer Network panel members were selected as content experts for each of the four cancers. Their task was to review each article's primary medical claims. The experts then completed a series of ratings to determine the misinformation and potential harm in each article.

Of the 200 articles, 41.5% came from nontraditional news (digital only), 37.5% from traditional news (online versions of print and/or broadcast media), 17% from medical journals, 3% from crowdfunding sites and 1% from personal blogs.

The review by these experts concluded that approximately one third of articles contained misinformation. Misinformation was defined as misleading information (a headline unsupported by the text or statistics/data, also not supporting the conclusion – 28.8%), mischaracterized strength of evidence (evidence presented as strong or, conversely, weak – 27.7%), and unproven treatments (not studied or insufficient evidence – 26.7%).

Unfortunately, articles containing misinformation had more engagement than factual articles. The median number of engagements, such as Twitter likes, was 2300 versus 1600, respectively. Overall, 30.5% of the 200 articles contained harmful information.

The study defined this harmful information as follows:

• Harm due to lost time (may cause delay in a treatable condition or failure to seek medical help – 31.0%)
• Economic harm (treatment costs or travel costs to obtain treatment – 27.7%)
• Harm due to an intervention (potentially toxic effects of the recommended test/treatment – 17.0%)
• Harmful interactions (known/unknown medical interactions with modern treatments that could lead to survival – 16.2%)

One limitation is that the study included only the most popular cancer articles published in English. Nevertheless, it clearly shows that not everything we access online is accurate; as much as one third may be wrong, and sensational news warrants particular caution. It is especially important to remember that people differ and treatments can vary from person to person. Individual assessments from your doctor and reliable sources are more appropriate than general information.