Insulin Resistance: the metabolic shift that hides for years before your blood sugar moves

A 47-year-old woman comes in for her annual screen. Her fasting glucose is 5.2. Her HbA1c is 5.4 — comfortably normal. Her cholesterol is in range. Her blood pressure is 122/78. By every standard measure on the report, she’s fine.

But she’s also gained six kilograms over five years that she can’t explain. Her father has type 2 diabetes. She runs three times a week. She doesn’t eat particularly badly. And she feels, in her own words, like her body has stopped responding to the things that used to work.

What none of her standard markers are showing is that her fasting insulin is 16 mU/L — substantially higher than it should be — and her HOMA-IR, the simple calculation that combines insulin and glucose, places her well into insulin-resistant territory. This is a state that often precedes type 2 diabetes by ten or fifteen years, drives stubborn weight gain, and quietly raises cardiovascular risk. It is rarely tested in standard health screens.

At Advantage Medical Group, the VITAL Longevity Screen (from SGD 2,588, available at our Bukit Timah clinic) measures fasting insulin, fasting glucose, HbA1c, HOMA-IR, uric acid, hsCRP, and homocysteine — the panel that catches insulin resistance early, often years before the standard markers move. This article explains what insulin resistance is, why standard screens miss it, and what catching it early actually changes.

How it works, and why standard screens miss it

When you eat carbohydrate, it becomes glucose in your blood. Your pancreas releases insulin — a hormone that tells your cells to take the glucose up. In a healthy system, a small amount of insulin gets the job done.

In insulin resistance, the cells stop responding properly. The pancreas compensates by producing more insulin. With enough extra insulin, blood sugar still clears — so on a standard blood test, fasting glucose looks normal. But insulin itself, which most panels don’t measure, is quietly elevated. This state can run for ten or fifteen years before the pancreas falters and glucose starts to rise. That’s when most people get told they have prediabetes — by which point the underlying process has typically been present for a decade.

The central asymmetry: insulin rises early. Glucose rises late. Standard panels measure the late one.

In Singapore, Screen for Life and most private health screens check fasting glucose and HbA1c. Both are useful once blood sugar has drifted — neither catches the years before. Fasting insulin and HOMA-IR (the simple calculation that combines them) come from a single morning blood draw, but aren’t part of standard panels here. Patients with meaningful insulin resistance get told their screen is “normal” — and the process keeps progressing.

How insulin resistance shows up in real life

This is the part that makes insulin resistance harder to recognise than diabetes: in its early stages, it doesn’t feel like a disease. It feels like life getting a bit harder.

Most patients we see don’t come in saying “I think I have insulin resistance.” They come in saying things like:

• “I haven’t changed what I’m eating, but I’ve gained ten kilos in five years.”
• “I exercise the same amount I did at 35, but my body responds completely differently now.”
• “I get tired in the afternoon in a way I didn’t used to. Sometimes I crash an hour after lunch.”
• “My waist is going up even though my weight isn’t changing much.”
• “My doctor said I’m fine. But I don’t feel fine.”

None of these on its own is diagnostic. Each of them, taken seriously, can also be the early ground signal of something measurable. The clinical question is whether what’s underneath is worth catching now, while it’s still cleanly reversible.

There are physical signs too — sometimes subtle, sometimes obvious. Skin tags, particularly around the neck and underarms. A darkening of the skin in the body’s folds (acanthosis nigricans). A waist circumference that’s drifted up while overall weight has stayed flat — a sign that body composition is shifting toward visceral fat. Each of these is a clue, not a diagnosis. The diagnosis comes from blood.

The patients most likely to have early insulin resistance hiding behind a “normal” screen are:

• Adults with stubborn central weight gain that doesn’t respond as cleanly to diet and exercise as it used to
• Anyone with a parent, sibling, or grandparent with type 2 diabetes
• Women in perimenopause or post-menopause, when oestrogen-related metabolic protection drops off
• Patients with sarcopenia, where less muscle means less storage capacity for blood sugar
• Patients with fatty liver detected on an ultrasound (NAFLD), elevated triglycerides, low HDL, or rising blood pressure with no other explanation
• Patients on medications that promote insulin resistance (long-term steroids, certain antipsychotics, certain HIV medications)

The marker that matters early: HOMA-IR

HOMA-IR stands for Homeostatic Model Assessment of Insulin Resistance. The name is more intimidating than the test. In practice, it’s a simple calculation from two numbers you can get from one morning blood draw:

• Fasting insulin (taken after an overnight fast)
• Fasting glucose (the same draw)

These two numbers go into a formula that produces a single index. The higher the number, the more insulin resistance is present.

Roughly speaking:

• Below 1.0 is excellent insulin sensitivity
• 1.0 to 1.9 is normal
• 2.0 to 2.5 is the early borderline range
• 2.5 and above suggests meaningful insulin resistance
• Above 5.0 is significant insulin resistance

These thresholds vary slightly between labs and populations, and a HOMA-IR result is best read alongside the rest of the picture — body composition, lipid profile, hsCRP, family history, what someone’s actually eating and how they’re moving. But the calculation itself is unambiguous, cheap, and one of the earliest signals you can get.

The reason HOMA-IR is so useful is precisely because it captures the period that fasting glucose alone misses. A patient with fasting glucose of 5.0 and fasting insulin of 18 looks “normal” by every standard guideline, but their HOMA-IR will be around 4.0 — clearly elevated. The clinical picture is meaningfully different from a patient with the same glucose and a fasting insulin of 5.

This is the test that changes the conversation from “your screen is normal, see you next year” to “your insulin is doing more work than it should, and here’s what we can do about it.”

Why catching it early matters

Insulin resistance is, in itself, mostly invisible. The reason it warrants attention is what it drives over time, and how much of that downstream load is reversible if it’s caught while the system can still recalibrate.

Type 2 diabetes. This is the most direct trajectory. Insulin resistance precedes type 2 diabetes by ten to fifteen years on average. Catching it during that window — when blood glucose is still normal but insulin is rising — gives the largest possible head start. The further along the trajectory, the harder it becomes to reverse, and at some point the pancreas’s capacity to compensate begins to fail.

Stubborn weight gain — particularly central. High circulating insulin makes it harder for the body to release stored fat for energy. This is why patients with insulin resistance often describe the experience of gaining weight despite eating and exercising in ways that should produce a flat or downward trajectory. The fat that accumulates is also disproportionately visceral — around the organs — which carries its own metabolic risk.

Fatty liver (NAFLD). Insulin resistance is the main driver of non-alcoholic fatty liver disease, now one of the most common chronic liver conditions globally. Many patients are surprised to learn they have fatty liver — usually picked up on an abdominal ultrasound — and even more surprised when it traces back to insulin metabolism rather than alcohol or diet alone.

Cardiovascular risk that BMI doesn’t capture. Insulin resistance shifts the lipid profile in characteristic ways: triglycerides rise, HDL falls, and the LDL particles themselves become smaller and denser — a more atherogenic pattern. Combined with rising blood pressure and chronic low-grade inflammation, the cardiovascular risk runs ahead of what a standard cholesterol panel suggests. This is why the Lp(a) and ApoB markers we discuss elsewhere become particularly relevant in patients who also have elevated HOMA-IR.

Two patient groups in particular sit at the intersection of these forces. Women in perimenopause and post-menopause: oestrogen has a protective metabolic effect, and as it declines, insulin sensitivity tends to drop with it — which is why the same lifestyle that worked at 38 stops working at 48. And patients with sarcopenia: skeletal muscle is the body’s largest sink for blood glucose after a meal, so less muscle means less storage capacity, and insulin resistance and muscle loss feed each other.

The pattern across all of these is the same: insulin resistance is doing real work in the body, even when nothing on the standard report suggests it. Catching it early is the difference between a recalibration and a chronic disease.

What a proper assessment includes

A clinical insulin resistance assessment isn’t about a single number. It’s about building a picture, because the same HOMA-IR can mean different things in different bodies. The full panel that informs the picture:

The first three rows are the diagnostic core for insulin resistance specifically. The remaining rows fill out the metabolic and cardiovascular context — because once HOMA-IR is elevated, the next clinical question is always “what else has shifted with it?”

The VITAL Longevity Screen at AMG includes all of these markers, alongside the cardiovascular, hormonal, and physical-function panels that together form a complete metabolic picture. A comprehensive assessment takes a single morning at our Bukit Timah clinic, with results reviewed in a 60-minute physician consultation.

For patients who specifically want metabolic screening without the broader longevity panel, our Executive Health Screening options can include the core insulin resistance markers as part of a more focused package.

What you do with the result

Most patients we screen for insulin resistance fall into one of three patterns. The pattern matters because the response is meaningfully different.

Pattern one: clean. A 42-year-old man, normal weight, exercises regularly, no family history of diabetes. His HOMA-IR comes back at 0.9. Fasting insulin is 5.5. His glucose, lipids, and hsCRP are all clean. There is nothing to fix, and that’s worth knowing.

For him, the screen is a confirmation — and a baseline. We agree to repeat the panel every two or three years to track the trajectory.

Pattern two: drifting. A 47-year-old woman — the patient at the top of this article. Fasting glucose 5.2, HbA1c 5.4 (both “normal”). But fasting insulin is 16, HOMA-IR is 3.7, hsCRP is 2.8, triglycerides are climbing, and her DEXA shows visceral fat at the high end of normal. Nothing on a standard screen would have flagged this. Everything together tells a clear story.

This is the largest opportunity. The interventions are concrete: whole-food protein and fibre at meals, less refined carbohydrate, resistance training to build the muscle that absorbs glucose post-meal, and addressing sleep where it’s contributing (poor sleep alone worsens insulin resistance). Pharmacotherapy may be part of the conversation depending on the full picture. Six months of focused intervention can move HOMA-IR substantially at this stage.

Pattern three: established. A 58-year-old with HOMA-IR of 6.5, fasting glucose at 6.4 (just below the diabetes threshold), HbA1c at 6.0, fatty liver on ultrasound, and a waist circumference well above the metabolic cut-off. He has been told for years that his “sugars are fine.”

The interventions are largely the same, but the slope is different. Lifestyle work still helps. Pharmacotherapy is more often part of the conversation — the evidence for medications that improve insulin sensitivity at this stage is strong, with the right combination depending on the full picture. The clinical priority has shifted from recalibrating the system to slowing the trajectory toward diabetes.

In every one of these patterns, the heavy lifting is done by what someone eats, how they move, and how well they sleep. Medications can help when indicated, but they work better against changed metabolism than as a substitute for it. The earlier the intervention, the larger the response.

Why we measure this at AMG

Most patients who eventually develop type 2 diabetes — or fatty liver, or insulin-driven cardiovascular disease — had a standard medical record that read as unremarkable for years before things actually changed. Their fasting glucose was fine, until one day it wasn’t. Their HbA1c was fine, until one day it wasn’t. The insulin doing the work behind the scenes — the metric that could have flagged the trajectory five or ten years earlier — was simply never measured.

That’s the gap longevity screening exists to close.

The clinical logic is the same one we apply to Lp(a), sarcopenia, and the rest of what the VITAL Screen measures: by the time the standard tools detect the problem, the most useful interventions are harder to apply. A first fasting insulin and HOMA-IR at 40 gives you a baseline. The repeats at 43, 46, and 49 turn that baseline into a trajectory. The trajectory is what tells you whether the system is staying steady or drifting.

For most people, the result will be reassuring — insulin metabolism is doing what it should, with specific targets for keeping it that way. For others, it will catch something earlier than it would have been caught otherwise, while the window for clean reversal is still open. Both outcomes are useful. Both are part of why we do this.

The cost of measuring is low. The cost of not knowing, or only knowing too late, is the part that most cannot afford.