5 Questions to Ask Before You Believe a Nutrition Claim Online

TL;DR: How Do You Spot Questionable Nutrition Advice Online?

Before changing what you eat because of something you saw online, ask five things: What exactly is being claimed? Is the person qualified to make that particular claim? Can the numbers be traced to a credible source? What kind of study supports it, in whom and at what dose? And finally, what mechanism would have to explain the result — and who benefits if you believe it?

If the claim remains vague, untraceable or dramatically larger than the evidence behind it after those checks, it probably should not be changing your diet.

Is This Article for You?

Read this if you have changed something about how you eat because of something you saw online; you find yourself unsure which of two contradictory posts to believe; or you would like a quick way to sort useful advice from noise without having to read a research paper.

What this article is not: a criticism of people who share nutrition information online. A great deal of it is careful, well-sourced and generously given, much of it by qualified dietitians and nutritionists doing genuinely valuable public work. The difficulty is that a feed presents the careful and the careless in identical packaging. These questions are for telling them apart.

The most common question I am asked is not about a nutrient. It is some version of: I saw this online — is it true? And underneath it sits a more useful question, which is: How would I work that out myself?

That second question is the one worth answering, because nutrition advice arrives faster than anyone can fact-check it, and it will keep arriving long after you have finished reading this. What follows is not a list of myths to memorise. It is a method — five questions that between them catch most of what goes wrong.

1. What Exactly Is Being Claimed?

Start by pinning the claim down, because a surprising number of them dissolve at this stage.

“Supports gut health” is so broad a claim that it is almost a mood. What does gut health mean in this case — bowel regularity, digestive symptoms, microbial composition, something else? Until an outcome is specified, there is very little to test.

A claim you can actually evaluate has a subject, an effect and ideally a size: this food, doing this thing, by roughly this much, in these people. Compare “turmeric is anti-inflammatory” with something specific enough to check, and you will notice how much of what you scroll past belongs to the first category.

Watch the verbs especially. In India, claims made in food advertising are regulated. Under the FSSAI Food Safety and Standards (Advertising and Claims) Regulations, nutrition and health claims require scientific substantiation, while language claiming that an ordinary food treats, cures or prevents a disease is restricted. Disease-risk-reduction claims are a separate category and have their own requirements.

So when you see dramatic curative language attached to an ordinary food, that is a good reason to examine the claim very carefully.

2. Who Is Saying It — and Are They Qualified?

This question has become somewhat easier to ask in India, although there is an important distinction between ordinary editorial content and influencer advertising.

What ASCI Guidelines Say About Health Influencer Advertising

Under the Advertising Standards Council of India’s influencer advertising guidelines, influencers providing technical advice about, or promoting or commenting on technical aspects of, commercial health and nutrition products or services are expected to have qualifications relevant to the advice being given. Those qualifications must be disclosed prominently and upfront — on the visual or at the beginning of a video, and before the reader gets into a text-based post.

The current guidelines distinguish technical health advice from generic promotion. An influencer does not need to become a nutritionist merely to appear in a general advertisement for a food product; the qualification requirement becomes relevant when the communication moves into technical health or nutrition advice.

So the reader’s move is simple: look for the disclosure when somebody is asking you to trust their technical expertise.

If someone is offering specialised dietary advice and there is no relevant qualification stated anywhere, that absence is information. It does not automatically make the claim false, and a qualification does not automatically make every claim true. But credentials give you something verifiable against which to judge the person’s scope of expertise.

The second half of this question is about fit rather than paper. A qualification in one area does not automatically extend to another — a fitness credential is not a clinical nutrition credential, and neither is a certificate from a weekend course. The question is not “Is this person impressive?” but “Is this specific claim inside what they were trained to assess?”

3. Where Did the Number Come From?

Numbers travel further than the studies they came from, and they mutate on the way.

Any confident-sounding statistic should be traceable to a named source — a survey, a composition table, a published paper or another clearly identified dataset. If the post does not say where the number came from, and you cannot find a credible source for it, treat the precision with caution.

Worked Example: A Statistic You Have Probably Seen

A figure circulates widely in Indian health content along the lines of “seventy per cent of Indians do not get enough fibre.” It sounds authoritative and it may well be directionally plausible, but when you go looking for one national source for that exact figure, it becomes difficult to trace. Different studies measure fibre intake differently, use different populations and arrive at different estimates.

Now compare that with a number that does have a specific document behind it.

The Indian Food Composition Tables 2017, published by ICMR-NIN, list whole-wheat atta at 11.36g of dietary fibre per 100g and refined wheat flour, or maida, at 2.76g per 100g. The product documentation supplied for Aashirvaad Chakki 100% Khapli Atta reports 11.4g dietary fibre per 100g for that particular product.

Those figures tell us something specific and limited: the measured fibre values are traceable, and the tested Khapli product is close to the whole-wheat atta value in the national composition table.

They do not, by themselves, prove that Khapli produces a particular clinical outcome, is superior to every other wheat or will affect every person in the same way.

That distinction is the lesson.

The first number sounds precise but is difficult to pin to one source. The second set can be traced to named documents. If a number is going to influence what you believe — or what you buy — being able to follow it back to where it came from matters.

4. If There Is a Study Behind It, What Kind of Study — and in Whom?

“Studies show” is doing enormous work in a great deal of nutrition content. Three sub-questions dismantle most of it.

• In whom? A great many exciting findings come from cells in a dish or from animals. Those studies matter — they are how hypotheses get generated and mechanisms explored — but they are not automatically evidence of what happens when a person eats the food.
• How many, and for how long? A finding in twelve people over two weeks may be useful preliminary evidence. It is not the same thing as a large, long-term trial.
• At what dose? If an effect required an isolated extract at a concentration you could never reasonably reach by eating the food itself, then a headline written about the whole food may be giving you the wrong impression.

Association and Causation

An observational study can show that two things occur together — for example, that people who eat more of a particular food tend to have a particular outcome. It cannot, on its own, establish that the food caused that outcome, because groups who eat differently often differ in many other ways too.

Randomised controlled trials are designed to address some of those problems by assigning participants to groups by chance, making causal inference more defensible. Even then, the size, duration, population, intervention and outcome still matter.

This is why nutrition headlines so often overstate their sources: association is easier to find, and once compressed into a headline it can begin to read like causation. Nutrition researchers themselves have argued for clearer communication of that uncertainty, particularly when research findings move from papers into press releases.

What Does Peer Review Mean?

Peer review is the process by which a study is assessed by researchers with relevant expertise before a journal publishes it. It is a quality-control mechanism, not a guarantee. Peer-reviewed work can be flawed, contradicted or subsequently revised, and publication does not mean a finding is settled.

But a claim based on a peer-reviewed research paper generally gives you much more to inspect than one resting on an unsourced blog, a press release or another social-media post.

A practical shortcut is to look for reviews and systematic reviews, which assess multiple studies together and can often tell you more about where the evidence as a whole stands than one dramatic experiment.

Applying the Question: What Does the Khapli Research Actually Show?

Khapli, or emmer wheat, provides a useful example of why the details of a study matter.

One older human intervention compared diets incorporating dicoccum whole-wheat flour with common bread-wheat flour in 22 people with diabetes over six weeks. Thirteen participants received dicoccum flour and nine common bread wheat. The study reported favourable changes in several lipid measures in the dicoccum group.

That is legitimate human evidence worth knowing about. It is also a small, older study in a specific clinical population, not a licence to say that Khapli has been conclusively proven to improve metabolic health in everyone.

Notice how much the sentence changes when you ask in whom, how many and for how long.

“Studies prove Khapli is better for metabolic health” and “a six-week trial involving 22 people with diabetes reported favourable changes in some lipid measures” are not the same claim.

The latter may sound less exciting. It is also the one the study actually gives us.

5. What Would Have to Be True — and Who Gains If You Believe It?

The last question has two halves, and the second is more uncomfortable than the first.

Ask what mechanism the claim requires. If a single ingredient is said to fix several unrelated problems at once, the claim needs biological mechanisms capable of explaining those effects, and physiology rarely works as neatly as an advertising sentence.

Claims that a food resets, rebalances, cleanses or purifies something are especially worth slowing down over. Ask what, specifically, is being reset, what measurable process is changing and how anyone knows it changed.

Then ask who benefits.

Someone with something to sell is not thereby lying. Brands, clinicians, researchers, publishers and influencers can all have interests; an interest is not the same thing as dishonesty.

But commercial interest should increase the demand for transparency, not reduce it. If a claim happens to require you to buy the exact product being shown, ask whether the evidence behind that claim is accessible, relevant to the finished product and proportionate to the promise being made.

Including this article.

You are reading a nutrition article on a food brand’s health-content platform, and this piece has used one of that brand’s products as a worked example. Apply the fifth question here exactly as you would anywhere else: can the numbers be traced, are the sources identified, have product-specific measurements been kept separate from clinical outcomes, and have we told you where the evidence stops?

That is the standard these five questions are asking of everyone. It would be a strange article about scepticism that exempted itself.

The Ninety-Second Version

You will not run five formal questions against every post you see. In practice, the shortcut is this: Does the claim say something specific? Is a source named? Does the person’s expertise fit the claim? Does the study actually resemble the person, food and dose being discussed? And would you still find the claim convincing if nothing were being sold?

That takes about as long as watching the reel did.

And if a claim survives all of that and you still are not sure — which happens, because plenty of genuine questions in nutrition are unsettled — the honest answer may simply be that the evidence is mixed.

That answer performs badly online. It is often the correct one.

Frequently Asked Questions

How do I check whether a nutrition influencer in India is qualified?

If the post is influencer advertising that gives technical health or nutrition advice, look for an upfront qualifications disclosure. ASCI’s guidelines require relevant qualifications for influencers giving technical advice about commercial health and nutrition products or services and specify that those credentials should be prominently disclosed.

A qualification still needs to match the subject. Someone being highly qualified in one field does not automatically make them an expert in another.

Does “studies show” mean a nutrition claim is proven?

No. Ask what the study actually was: whether it involved cells, animals or humans; how many participants were included; how long it lasted; what was measured; and what dose or form of the food or nutrient was used.

One study can be useful evidence without settling an entire question.

What is the difference between association and causation in nutrition research?

An observational study may show that two things occur together, but that does not establish that one caused the other. People who eat differently can differ in income, activity, smoking, healthcare access, other foods and many additional factors.

Randomised trials can make causal interpretation more defensible, although their own design and limitations still matter.

Is peer-reviewed nutrition research always trustworthy?

No. Peer review is a useful quality filter, not a guarantee that a study is correct or that its findings will never be challenged. The strongest picture usually comes from looking at a body of evidence rather than treating one publication as the final word.

Can a food legally be advertised as curing a medical condition in India?

Ordinary food advertising cannot simply claim that a food treats, cures or prevents a disease. FSSAI’s Advertising and Claims Regulations require scientific substantiation for nutrition and health claims and distinguish permitted disease-risk-reduction claims from impermissible treatment or prevention language.

If someone is selling something, does that mean their nutrition claim is wrong?

No. Commercial involvement does not automatically invalidate information. It means transparency becomes particularly important: check the evidence, whether the source is independent or brand-generated, whether the study examined the actual product being promoted and whether the strength of the claim matches the evidence.

A laboratory measurement can establish a product’s measured fibre or moisture content, for example; it does not automatically establish an unrelated clinical benefit.

What if two qualified nutrition experts say opposite things?

Sometimes one person is working from stronger evidence than the other. But sometimes disagreement exists because the evidence itself is uncertain, incomplete or applicable only to particular groups.

Look for good reviews, systematic reviews or established professional guidance rather than deciding solely on the basis of whichever individual sounds more certain.

How can I tell whether a nutrition statistic is trustworthy?

Look for the original source. A percentage should ideally lead you to a survey, official dataset, research paper or composition database explaining who was measured, how and when.

If the statistic travels from article to article but nobody can show where it originally came from, its repetition does not make it evidence.

The Short Version

1. Pin the claim down. If nobody can tell you what measurable outcome it refers to, don’t treat it as established.
2. Check the expertise. The qualification should fit the advice being given.
3. Chase the number. A precise statistic should lead somewhere you can inspect.
4. Interrogate the study. In whom, how many, how long, and at what dose?
5. Ask what would have to be true — and who gains if you believe it.
6. Leave room for uncertainty. “The evidence is mixed” is an unpopular answer and often the right one.

Sources

• Advertising Standards Council of India: Guidelines for Influencer Advertising in Digital Media, Addendum II
• Food Safety and Standards Authority of India: Food Safety and Standards (Advertising and Claims) Regulations, 2018
• Indian Food Composition Tables 2017, ICMR-NIN
• Yenagi NB, Hanchinal RR, Patil CS, Koppikar V, Halagi M: Glycemic and Lipidemic Response to Dicoccum Wheat (Triticum dicoccum) in the Diet of Diabetic Patients, 2001
• “Interpretation of observational studies: the good, the bad and the sensational,” Proceedings of the Nutrition Society, 2022