When Healthy Becomes a Moral Category

There is a small linguistic trick hiding inside a lot of modern health advice. Food stops being something we eat and starts becoming evidence about who we are.

A meal is no longer merely nutritious, convenient, comforting, celebratory, cheap, expensive, familiar or delicious. It is clean. Or dirty. Good. Or bad. A choice becomes a rule, the rule becomes a standard, and before long the standard begins behaving like a verdict.

That is the live wire running through new research from Flinders University in Australia, where researchers analyzed 300 popular TikTok and Instagram videos found under #healthylifestyle, #nutrition and #cleaneating. Sixty-seven percent emphasized clean or healthy eating, almost one-third promoted food restrictions or rules, and one in five explicitly framed foods as “good” or “bad.” Nearly two-thirds featured body objectification, and many linked health to achieving a lean or toned appearance.

The study’s title captures the implied promise the researchers kept encountering: “Eat Like Me = Look Like Me.” Lead author Katie Porteous summarized the message even more plainly: “if you eat like me, you can look like me.”

That promise is useful to social media because it compresses a complicated subject into a visible transaction. Here is my body. Here is my bowl. Replicate the bowl and perhaps you can replicate the body. Genetics, medical history, income, culture, stress, sleep, exercise, medication, disability, access to food and the hundred other variables that make human health stubbornly individual disappear outside the frame.

Clean Is Not a Neutral Word

The problem is not that people want to eat well. The useful distinction is between care and purity.

Care is usually flexible. It makes room for context. A person can prefer vegetables, protein, whole grains or less-processed foods without needing those choices to become a complete philosophy of worth. Care can survive a birthday cake, a road-trip dinner, a cultural tradition, a cheap meal at the end of an expensive week or simply the decision to eat something because it tastes good.

Purity is different. Purity needs a border. Once a category is declared clean, something else has to sit on the other side of the line. The vocabulary carries its own moral architecture: clean implies contaminated; good implies bad; a cheat meal implies that eating can become an act of dishonesty; guilt suggests an offense requiring emotional payment.

None of those words automatically produces disordered eating. Language is not diagnosis. But repeated often enough, the language can make rigid thinking feel ordinary because it disguises judgment as hygiene and self-control as virtue.

Where Health Advice Starts Looking Like Orthorexia

The Flinders researchers were specifically looking for themes associated with orthorexia nervosa: an excessive preoccupation with eating foods perceived as healthy or pure, rigid dietary rules and distress when those rules are violated. Their research record also notes that more severe patterns can include eliminating entire food groups and avoiding social situations where “impure” foods may be present.

Orthorexia is a useful concept here, but it requires precision. It is discussed widely in eating-disorder research and clinical settings, yet it is not currently a formal diagnosis in the DSM-5-TR or ICD-11. Researchers continue to debate its exact boundaries and whether it should be classified as a distinct eating disorder or understood in relation to other conditions.

That uncertainty is not a reason to dismiss the behavior. It is a reason not to flatten it. The point is not to diagnose someone because they meal-prep on Sunday or read an ingredient label. The point is to notice the threshold where a health preference becomes inflexible, where deviation produces disproportionate guilt or anxiety, or where eating rules begin shrinking a person’s social and emotional life.

Behavioral Crackers territory begins exactly there: not with the label, but with the distinction.

The Credentials Gap Is Part of the Story

The study also found that 78 percent of creators in the sample lacked recognized qualifications in health, nutrition or fitness. Across the sample, 68 percent of the health information was classified as misleading or deceptive.

The difference between qualified and unqualified creators was large. Qualified creators provided information the researchers coded as accurate in 45.5 percent of their posts, compared with just 4.7 percent among unqualified creators. Meanwhile, 75.2 percent of content from unqualified influencers was classified as misleading or deceptive, compared with 40.9 percent from qualified creators.

That does not mean every credentialed professional is correct or every uncredentialed person is useless. Social media would be much easier if expertise worked like a blue checkmark for truth. It does not. But the gap matters because short-form wellness content often borrows the visual authority of expertise without carrying its obligations: uncertainty, individual variation, contraindications, evidence quality, tradeoffs and the deeply unviral sentence, “It depends.”

Three-quarters of the posts also promoted supplements or superfoods. On a platform built to reward immediacy, a product, protocol or perfect plate is easier to package than a discussion of probability.

The Body Becomes the Citation

There is another reason “eat like me” works so well online: the creator’s body can be presented as the evidence.

The format quietly moves the burden of proof. Instead of asking whether a recommendation is supported by good nutrition science, the viewer is invited to look at the person delivering it. A lean body becomes a testimonial. Visible muscle becomes a footnote. Appearance stands in for a medical chart we cannot see.

The researchers found body objectification in nearly two-thirds of the posts, with Instagram placing more emphasis on objectified bodies and muscular ideals while TikTok more often framed its content through general wellness. Either route can arrive at the same destination: health becomes visually legible, and the person who looks the part gets to speak with the authority of the result.

But a body cannot tell you, by sight, whether someone is adequately nourished, metabolically healthy, chronically ill, anxious around food, genetically predisposed to a certain shape or simply posing in good light. The visual certainty is part of the trick.

What This Study Does — and Does Not — Show

The research should not be stretched past its design. It analyzed content; it did not follow viewers and establish that exposure to these posts causes orthorexia or another eating disorder.

The sample was also a snapshot. Researchers collected the first 50 videos returned for each of three hashtags on each platform between May 25 and May 27, 2025, producing 150 TikToks and 150 Instagram Reels. They used hashtag searches and steps intended to reduce algorithmic personalization, but 300 videos cannot represent every nutrition creator, every user’s personalized feed or every form of health information on either platform.

That limitation is important, but it does not make the findings trivial. A content analysis is well suited to a different question: what kinds of messages are readily available, repeated and normalized inside a popular information environment? On that question, the pattern is hard to miss.

The Character Test Hiding in the Refrigerator

The most interesting thing about the study is not a new warning that TikTok contains bad nutrition advice. We have accumulated enough versions of that sentence to fill the wellness aisle.

It is the way the language can quietly transform eating from a practical human behavior into a character test.

Once food becomes moralized, the person eating it becomes moralized too. A clean meal can make someone feel disciplined. A “cheat” can make someone feel weak. “Good all week” can turn Saturday dinner into a confession. The food has not merely entered the body; it has entered the identity system.

This is where healthy living becomes especially compatible with social media. Platforms are exceptionally good at turning behaviors into identities because identities are sticky. “I eat vegetables” is a sentence. “I am a clean eater” is a community, an aesthetic, a shopping list, a content category and potentially a set of rules that can keep generating more content tomorrow.

The distinction worth defending is almost boring in its moderation: healthy eating can be valuable without being pure. Nutrition can matter without becoming morality. People can make intentional choices without needing every plate to prove something about their discipline, virtue or worth.

Professor Ivanka Prichard, a co-author of the study, put the practical standard simply: “Healthy eating should be flexible, enjoyable and sustainable.”

That may be less seductive than a perfect bowl under perfect lighting. It is also much harder to turn into a purity test.

SOURCE NOTES

• Flinders University News — “Eat like me to look like me” (Sept. 8, 2026)
• Flinders University ResearchNow — study record and open-access publication details
• Journal of Technology in Behavioral Science — DOI 10.1007/s41347-026-00685-5
• National Eating Disorders Association — Orthorexia overview and DSM-5-TR status