Why your "healthy" diet might not be working for you

You did everything the magazine told you. Switched to brown rice. Swapped ghee for olive oil. Added a morning smoothie. Dropped the second chai. Started counting macros on a phone app. Three months in, the weight has not moved. Your fasting glucose is the same. Your energy is, if anything, slightly lower. You feel quietly betrayed.

Here is the part the magazine left out. The "healthy" diet you read about, the one with the testimonials and the before-and-after photos, was tested on a population that does not eat your food, does not metabolise it the way you do, and in many cases does not share your genetic risk profile.

What personalised nutrition actually means

Personalised nutrition is a bigger field than the supplement industry suggests. At its serious end, it is a research area that studies how individual genetic variants interact with specific nutrients to produce different metabolic outcomes. At its sillier end, it is a quiz that recommends a different shade of green smoothie.

A small number of gene variants do most of the heavy lifting in the serious version. TCF7L2 and FTO sit close to type 2 diabetes risk. PPARG and APOE influence how your body handles dietary fat. MTHFR shapes folate metabolism. CYP1A2 decides whether the espresso you just ordered will help you focus or wreck your sleep. None of these variants are deterministic. Each of them tilts a probability. Stacked together, they explain a meaningful slice of why two friends on the same meal plan get very different results.

What the research shows about Indian biology

The Indian metabolic phenotype is its own thing. Lower BMI thresholds for cardiometabolic risk. Higher abdominal fat at given body weights. A genetic tilt towards insulin resistance that shows up in cohorts well before symptoms do. The PURE study, the ICMR-INDIAB series, and the more recent South Asian genome work have all pointed at the same conclusion. A typical urban Indian adult is metabolically older than their chronological age suggests, and the dietary advice imported from European trials is calibrated for someone who is not.

The practical implication is unflattering and useful in equal measure. If your "healthy" diet centres on grains, processed dairy, and snacking, your insulin curve probably does not look like the curve in the textbook. The fix is not a fad. It is a smaller plate of refined carbohydrate, more protein than you are probably eating, fibre with every meal, and a willingness to test instead of assume.

Three changes you can run this week

The first one is the cheapest. Start every meal with vegetables. Order matters. Beginning with fibre changes the glycaemic shape of what comes next. It is the dumbest, most reliable intervention in the modern nutrition literature, and Indian thalis make it easy to do without thinking.

The second is protein. Add a fist of it to every meal. Dal counts. Eggs count. Paneer counts in modest amounts. Most Indian plates underdeliver on protein. The deficit shows up as poor satiety, then as overeating, then as a metabolic story that does not fit your gym schedule.

The third is the second chai. Replace it with water. Caffeine and sugar in the late afternoon can be the difference between a flat insulin curve and a spiky one, and the difference between a deep sleep and a restless one. Watch how you wake up over the next two weeks.

When you are ready to test

Three numbers are worth more than any wellness quiz. Fasting insulin, not just glucose. HbA1c, not just lipid. And, if you can find a lab that runs it, a DNA wellness panel that covers the metabolic variants for a South Asian reference. The genes do not change. The report you get at thirty is the report you have at fifty. It is one of the few pieces of biology you only need to buy once.

The promise of personalised nutrition is not that the right diet will be exciting. It is that the right diet will be obvious, once you stop guessing. That is the version worth pursuing.