Science1 publisher2 min readPublished
Fresh Instagram account's #womenshealth search surfaces sales pitches in 39% of posts
Researchers searching #womenshealth from a new Instagram account found 39% of 472 posts selling a product or service. Only 7% cited scientific evidence, and because the account was new, the figures describe the first view a stranger gets from a search.
The Scientist · Science desk
What happened
- Researchers opened a new Instagram account late last year, searched #womenshealth and analysed 472 of the first 500 posts after removing duplicates and broken links.
- Hormonal and reproductive health was the most common topic, appearing in almost half of posts, and menopause and perimenopause alone made up 12%.
- Some 39% of the sampled posts were selling a product or service directly.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Because the account was brand new, the 39% describes what a newcomer searching the hashtag is shown; the share in a long-time user's personalised feed was not measured and could be higher or lower.
- exposure Among individual posters, a woman starting from this hashtag meets about 1.6 alternative health practitioners for every doctor, and the researchers cite evidence that influencers can change health behaviour.
- decision Writers of health guidance aimed at this audience can stand out by naming their evidence, since roughly one post in 14 among these top results did so.
- precedent The researchers want platforms to flag advertising and marketing posts. That is a disclosure remedy, and it fits what this study can support: it measured commercial intent and left accuracy untested.
Taking the first 500 posts a search returns measures ranking. Here the ranking had no personal history to draw on, because the account was new [1]. That design has a clear purpose: it shows what Instagram put in front of a stranger typing the hashtag. A woman looking up perimenopause for the first time starts from much the same place.
I think the 39% selling share is a fair description of that first view as it stood late last year [1][4]. The sample is also a thin slice of the pool. The hashtag held more than 8.8 million posts when the researchers searched it, so the 472 they analysed were about one in every 19,000 [3][1]. The count has since reached 9.6 million [3].
Most of the selling came from accounts that were not companies [2]. Company accounts made up 16% of all posts, and individuals 63% [6][8]. Even if every company post were a sales pitch, at least 23 of the 39 percentage points of selling posts came from other accounts, which is at least 59% of the pitches [2]. A broader test, whether the account holder had a clear financial interest in the subject, was met by 51%. That share is counted over account holders, not posts [5].
Put the claims next to the citations and the sample has about ten posts making an explicit health claim for every one that referred to scientific evidence [7][3]. The thing this doesn't tell you is whether any of those claims were wrong. "Our study did not assess whether specific health claims were accurate," the researchers wrote in their account of the work on phys.org [11]. A claim with no citation can still be correct. The study counted commercial intent and citation, both of which can be measured without a clinician grading each post, and it stops short of a misinformation rate. The researchers drew their conclusion from the mix: "But the combination of frequent health claims, financial interests and very few references to scientific evidence is concerning," they wrote [12].
The researchers say they cannot be certain the pattern holds on other platforms, though they have seen similar trends in other recent studies of Instagram and TikTok [13]. They also cite large-scale reviews finding that misleading health information is common across social media platforms [15].
What to watch
- A repeat of the sample from accounts with real viewing histories, to test whether personalisation raises or lowers the 39% selling share.
- A follow-up that grades the accuracy of the health claims, the step this study left out.
- Whether Instagram adds advertising or marketing labels to health posts, as the researchers propose.