When a diet stops working, the usual explanation is that the diet was not followed properly. Indian obesity guidance does not treat it that way past a certain point. The Endocrine Society of India directs that the key cause of weight gain should be identified in each individual seeking evaluation for obesity, screened across three domains: slow metabolism, increased caloric intake, and limited physical activity. Two of those three are things a food diary cannot measure. This piece covers what happens at the handoff, which categories of blood-panel findings change what a clinician does, and which ones a good clinician will decline to order.

• Self-reported food intake runs roughly 20% below measured expenditure on average, and the gap widens to 34% in people already living with obesity, the very group most likely to be told to try harder.
• A blood panel measures biological processes that sit entirely outside a dietary assessment: insulin resistance, thyroid function, and drug-induced weight gain.
• Indian obesity guidance specifies a response threshold: if 5 to 10% body weight loss is not achieved within the first year of lifestyle intervention, a wider clinical workup is the next step.
• Some findings on a blood panel change what gets prescribed. Others correlate with weight gain but do not redirect treatment on their own.
• Indian guidance names several tests as not recommended in obesity (free T3, thyroid ultrasound, routine cortisol, and HOMA-IR in PCOS), and they are ordered anyway.

The belief has a real basis, which is why it survives. Self-reported food intake is systematically wrong, and it is wrong in one direction. Measured against doubly labelled water, the gold-standard method for true energy expenditure, self-reported energy intake runs measured expenditure below by around 21% in men and 19% in women on average, rising to as much as 34% in people already living with obesity. Lean subjects underreport by roughly 2%.

Read that gradient carefully, because it is the point. The error is not random. It scales with body size, with dietary restraint, and with socioeconomic status, which means the people most likely to be told to try harder are the ones whose self-report is least reliable. Individual-level error is wider still: one comparison against doubly labelled water found individual differences running from minus 76% to plus 25%, and its authors concluded that self-reported energy intake is not recommended for use in obesity research at all.

So the honest version of the conventional wisdom is not "you are eating more than you think, therefore eat less." It is that nobody in the conversation, including the person taking the diet history, knows what you are eating to within useful accuracy. That is an argument for a different instrument, not for more of the same one.

It is worth adding that no Indian doubly-labelled-water validation of dietary recall or food-frequency questionnaires was found in the research behind this article. Indian questionnaires have been validated against Indian 24-hour recalls, which validates the tools against each other rather than against measured truth, so the size of the gap in Indian diets and portion sizes is genuinely unquantified.

Insulin resistance is the clearest example of a finding that sits entirely outside a dietary assessment, and it is worth understanding because most of the rest of this piece rests on it.

Insulin's job is to move glucose out of the bloodstream and into muscle, liver and fat cells. Insulin resistance means those tissues have become less responsive to that signal. The pancreas detects that glucose is not clearing fast enough and compensates by producing more insulin. For a long stretch, this works. Blood glucose stays inside the normal range because the body is producing more and more insulin to hold it there.

Nothing about that process is visible in what you eat, how disciplined you are, or what the scale says on a given morning. It is a change in how the body handles food, not in how much food arrives.

That distinction is what separates the two instruments. A nutrition assessment measures inputs, which are self-reported and imprecise. A blood panel measures what the body is currently doing with those inputs, which is not self-reported. Neither is a substitute for the other. But when the input side has been worked on repeatedly with no result, continuing to measure only the input side is a deliberate choice, not a default.

Indian guidance does not leave the escalation to indefinite individual judgement. It specifies a trial period and a response threshold.

The RSSDI-ESI consensus sets the target for lifestyle intervention at 5 to 10% body weight loss within the first year. Failure to approach that threshold, or continued weight gain despite the intervention, is the point at which the recommendation shifts from repeating dietary counselling to widening the clinical workup. The Endocrine Society of India directs that endocrine causes be tested for when there is clinical suspicion or when obesity is resistant to intervention.

The 2025 Delphi-based Asian Indian obesity consensus changes what counts as far enough along to warrant a workup. Obesity is now staged from a BMI of 23 kg/m² for Asian Indians, with Stage 2 defined by the presence of a comorbidity or an organ-function symptom. Stage 2 is associated with recommendations for a fuller biochemical assessment. The two-stage Indian definition is new enough that its long-term utility as a management trigger has not been established in outcome data, which is worth knowing before treating a stage label as settled.

Anything beyond lifestyle measures is a registered medical practitioner's decision after an assessment, not a step a reader takes independently.

A panel produces a page of numbers, and only some of them redirect treatment. The rest are findings a clinician reads in context.

Prediabetes or type 2 diabetes: Has important implications for metabolic health and may influence nutrition and broader weight-management planning.

Strength: Strong. Directly guideline-driving.

Overt hypothyroidism: Identifies a thyroid condition that may contribute to changes in metabolism and weight.

Strength: Strong. Indian guidance mandates thyroid testing in resistant obesity.

Drug-induced weight gain: Strongest lever after diabetes. Found in a medication history, not on a blood panel at all.

Strength: Strong, but detected by history.

PCOS or hyperandrogenaemia: Can provide relevant context for weight, insulin sensitivity and reproductive health.

Strength: Moderate. Testing restricted to patients with clinical features, not universal.

Dyslipidaemia: Provides additional information about cardiovascular and metabolic risk.

Strength: Strong for risk stratification. Weaker as an insulin-resistance surrogate, and South Asian cut-offs are contested.

Subclinical hypothyroidism: Little, for weight. Levothyroxine in this group reduces BMI by around 0.38 kg/m² and waist circumference by about 1.1 cm.

Strength: Weak to moderate. Population studies find no BMI difference at all.

Insulin resistance on HOMA-IR: Useful in narrow situations rather than as a general explanation for weight gain.

Strength: Moderate for specific uses, weak as a general explanation.

Iron deficiency, vitamin D or B12 deficiency: Corrects fatigue and musculoskeletal symptoms. Not a weight-loss lever in any Indian or international guideline.

Strength: Weak as a weight determinant.
The findings in the top rows can have broader implications for someone's weight and metabolic-health assessment. The bottom four correlate with weight gain without redirecting treatment on their own. That is not an argument for ignoring them. It is an argument for having someone qualified read the whole page rather than reacting to the one number that looks abnormal.
Note what is missing from the top of that list: the single strongest lever after diabetes is not a blood result at all. It is the medication history, because several widely prescribed drug classes cause weight gain, and no amount of dietary work corrects for a drug effect.

A blood panel is not a virtue, and this is where a piece like this can easily become an advertisement for testing. Indian guidance names several tests as not recommended in obesity, and they are ordered anyway:

• Routine free T3 testing in patients with an elevated TSH
• Routine thyroid ultrasound, irrespective of thyroid function
• Routine cortisol testing without clinical suspicion of Cushing's syndrome (the dexamethasone suppression test is reserved for patients with the clinical features)
• Genetic testing for obesity in adults, outside research settings, unless syndromic features are present
• HOMA-IR as a routine tool in PCOS management (it is unreliable once diabetes is established)

Over-testing has costs beyond money. A borderline TSH or a marginal androgen result can trigger a specialist referral, a repeat testing cycle and a good deal of anxiety, none of which was warranted by the finding. Access is not the barrier either: bundled basic panels covering blood count, liver, kidney, thyroid and glucose are listed in the low hundreds to around ₹1,400 at major Indian online diagnostic platforms, most of them bookable without a prescription. Cheap and unsupervised is precisely the combination that produces a page of numbers nobody is interpreting.

The useful question is not "should I get tested" but "has the lifestyle-first attempt been given a fair run, and is there now a reason to look at a different layer." That is a conversation, and the right place to have it is with a registered medical practitioner who can assess what is actually indicated.

The right test, ordered for the right reason, read by the right person. A clinical consultation with Voy puts your history, medication list, and markers in front of a registered clinician who can decide which tests are warranted, rather than a test menu.

This article is for general information and education only and is not medical advice, diagnosis, or treatment. GLP-1 and other medications referenced are prescription-only and are appropriate only for certain people under the supervision of a qualified clinician. Do not start, stop, or change any medication based on this article. Please consult a registered medical practitioner about your individual circumstances. Information reflects what was available at the time of review and may change.