Crohn’s Disease Reels on Instagram Often Miss the Mark on Accuracy

As a medical student witnessing the popularity of Instagram and TikTok surge, Sankirth Madabhushi increasingly noticed patients turning to these platforms for health information. Coming across medical videos on social media, he had a realization.

“Unfortunately, not all of them were evidence-based,” said Madabhushi, today an internal medicine resident doctor at the University of Massachusetts Chan Medical School. While his medical background helped him identify unreliable content, he recognized that not everyone could do this, prompting him to assess the quality of health-related content on Instagram. He decided to focus on Crohn’s disease, a type of chronic inflammatory bowel disease, because of his research interest in gastroenterology.

Madabhushi and his colleagues evaluated Instagram reels on Crohn’s disease and found these to be of moderate quality and that they often spread clinically significant misinformation.¹ Their findings, published in PLoS One, highlight the need to critically examine health content on social media as these platforms gain popularity for medical and science communication.

“It's been quite an area of concern generally about disinformation and misinformation about health online,” said Megan Lim, a researcher who studies how digital media impacts public health at Burnet Institute and was not associated with the study. “It's a useful reminder to people how inaccurate this information can be.”

Top Crohn’s Disease Reels Are of Moderate Quality

For their study, Madabhushi and his colleagues made a new Instagram account so that their own social media algorithms would not bias their search. They searched for the hashtag “#crohns”, filtering their search to limit it to the ‘Top Reels’ category on Instagram reels. This left them with 78 videos in English.

Of these, non-medical professionals had created more than 80 percent of the videos, while medical professionals—including physicians, physician assistants, nurse practitioners, physical therapists, PhD holders, pharmacists, registered nurses, and others who regularly interact with patients—created only a fraction of Crohn’s disease content. By further stratifying these videos based on their intent, the researchers found that medical professionals created mostly educational content while other creators focused on lifestyle and patient experience with Crohn’s.

Madabhushi and his team used the Journal of the American Medical Association (JAMA) criteria scores to assess the quality and validity of the videos.² This scoring system awards one point each for content that lists authors with their credentials, sources, disclosures of conflicts of interest or advertising, and clear upload dates.

Overall content created by medical professionals had higher JAMA scores than that created by non-medical professionals, suggesting better validity of information. However, focusing on educational videos revealed a deeper trend: Medical advice content by healthcare professionals and other creators had similar moderate scores, indicating that the information was of ordinary quality.

Madabhushi and his team also scored the videos on a previously established scale of -1, 0, or 1 depending on whether the content promoted potentially harmful advice, neither harmful nor beneficial nonfactual information, or accurate and helpful facts, respectively.³

Only about one-third of all videos had a positive harm/benefit score, indicating that a minority of the videos had offered helpful advice that outweighed potential risks. Forty two percent of the videos with negative harm/benefit scores, or those likely containing misinformation, were created by medical professionals. They also observed that the popularity of content did not correlate with accuracy.

While Madabhushi expected videos to frequently contain misinformation, he said that finding people with a medical background often creating inaccurate content was surprising. However, he emphasized that this result represented only a small subset of self-identified medical professionals, highlighting that this does not mean one should not trust doctors.

A Call for Better Standards and Stronger Public Health Communication Online

Madabhushi also noted that the JAMA scoring system dates back to the 1990s. “Given that it was a while ago, it may not be as applicable to today’s internet age,” said Madabhushi. “[But] it is probably the most reputable option, [indicating that] some updated frameworks are due,” he added, noting that such scoring systems should be compatible with diverse content types such as patient experience videos that are not meant for educational purposes.

Although some medical professionals’ videos being inaccurate was slightly unexpected, “I wasn't surprised by the amount of misleading information,” said Lim. However, she noted that the degree of harm in such content was not immediately obvious to her. “It's unclear if it was just not best practice versus completely incorrect or dangerous,” she explained.

She believes these results underscore the importance of holding social media platforms and their owners accountable. “But we have less control over what they do, [than over] what we do,” she said. “The best practice that we can do as public health professionals is to try to put out…accurate information that’s well designed, well resourced, [and] that speaks directly to the people that we want to reach using the kinds of techniques that successful advertisers and influencers use.”

Madabhushi agreed, “There's a lot of healthcare professionals on social media who do a great job of dispelling myths,” he said. “This is a pretty futuristic thing, but perhaps there could be some scoring algorithms that could provide a score or information to help vet content before it gets spread to the general public on social media.”