Why I Started a Metabolic Health Lab at 59

The data-driven story of how a technology risk manager debugged his own metabolism — and why I'm building Metabola to help others do the same.

November 2024. Annual health screening. The numbers came back and they were unambiguous.

HbA1c: 6.4%. Pre-diabetic, bordering on diabetic. Blood pressure: 155/100. Stage 2 hypertension. Fasting glucose: 7+ mmol/L. Consistently elevated. Body weight: 74 kg at 170 cm. BMI overweight. Body fat estimated above 35%. O2 ring data: An oxygen desaturation index of 38.5 events per hour. I'd bought the ring on a whim — I like health gadgets — and it revealed what I hadn't suspected: severe obstructive sleep apnea.

Full-blown metabolic syndrome. At 58 years old. After three decades in enterprise technology and risk management.

My doctor was straightforward: "We need to start metformin for the glucose. An antihypertensive for the blood pressure. Possibly a statin."

I asked him for time. "Give me a chance to try lifestyle intervention first."

He agreed. Reluctantly.

The Comfortable Lie I Told Myself

Here's the part that's hard to admit. I wasn't inactive. I wasn't ignorant. I had been doing what I thought were all the right things — for years.

Intermittent fasting? I'd been doing 16:8 for years. Longevity supplements? NMN, resveratrol, the full stack. I was a diligent consumer of health podcasts and metabolic research. I owned a continuous glucose monitor. I tracked things.

And yet. HbA1c 6.4%. Blood pressure 155/100. An O2 ring showing 38 desaturation events per hour every night.

The CGM data told the truth I was avoiding. Here's what a typical day looked like, eating normal Singaporean food:

• Breakfast (I skip)
• Lunch (fried noodles): peaked at 11.4 mmol/L
• Dinner (chicken rice): peaked at 9.8 mmol/L
• Time spent above 7.8 mmol/L: 8+ hours daily
• Overnight fasting glucose: drifting to 6.5–7.0 mmol/L by morning

I was eating 250+ grams of carbohydrates daily. Walking maybe 4,000–5,000 steps "when I felt like it." Exercising sporadically at best.

The intermittent fasting? It was just skipping breakfast. I was still flooding my system with glucose during every eating window. The supplements? Expensive insurance against a house that was already burning. The CGM? I was watching my own metabolic disaster unfold on a screen — and not doing anything about it.

I was doing the easy things. Skipping a meal is easy. Swallowing pills is easy. Reading about metabolic health is intellectually satisfying. Wearing a sensor feels proactive.

What I was avoiding was the hard stuff. Actually restricting carbohydrates at every single meal. Exercising every single day without negotiation. Losing the weight. Being consistent when it wasn't convenient.

I was optimizing the 10% while ignoring the 90%.

From November 2024 through June 2025 — seven months — nothing changed. Weight: still 74 kg. Fasting glucose: still 7+. CGM: still showing the same damaging patterns. My tonsillectomy for the sleep apnea was scheduled for July.

Seven wasted months.

The Catalyst

July 2025. Tonsillectomy.

The surgery was to address the severe OSA — enlarged tonsils blocking my airway during sleep, 38 oxygen desaturation events per hour, every single night, for years. The downstream damage from that alone was staggering: chronic systemic inflammation, impaired glucose regulation, elevated liver enzymes, cardiovascular stress.

Post-tonsillectomy recovery is brutal. Two weeks of liquid diet, severe throat pain, minimal food intake. I lost 6 kg in three weeks. Dropped from 74 kg to 68 kg.

And then a moment of clarity hit me, lying in that recovery bed, 6 kilograms lighter.

I did the math. My target weight was around 65–66 kg. A 10% reduction from my starting point. I was already most of the way there — accidentally, through surgical recovery. My appetite had been reset. My airway was clear for the first time in decades. I could breathe. I could sleep.

I had two choices. Let the weight come back, resume old patterns, and waste the surgery. Or use this as a launchpad to reprogram everything.

There was no middle ground. I chose to reprogram.

The Simultaneous Multi-Variable Intervention

I didn't ease back into my old diet. I didn't "gradually reduce carbs." I didn't "start exercising a little more." I changed everything. Simultaneously. In August 2025.

Diet: High-carb to high-fat, low-carb. Under 50g of carbohydrates per day. No transition period. Immediate and complete. The post-surgical appetite suppression made the first two weeks surprisingly manageable — a window of reduced hunger that I exploited ruthlessly.

Exercise: Sporadic to non-negotiable. 8,000+ steps every single day. Structured strength training three times per week. Zone 2 cardio two to three times per week. Started as soon as I was recovered enough to move. No exceptions, no "I'll do it tomorrow."

Sleep: From severe apnea to resolved. The tonsillectomy removed the mechanical obstruction. The weight loss resolved the remaining positional component. For the first time in likely a decade, I was getting restorative sleep. This alone changes everything — cortisol regulation, growth hormone secretion, glucose metabolism, appetite signaling.

Tracking: From passive observation to active feedback loop. Same CGM, same blood tests, same devices — but now they were part of an intervention loop, not a spectator sport. Measure, analyze, adjust, re-measure. Monthly lipid panels. Regular body composition scans. Every data point feeding back into protocol refinement.

Why all at once? Because compound effects only work when the variables move together. HFLC drives weight loss. Weight loss resolves residual sleep apnea. Better sleep improves insulin sensitivity. Better insulin sensitivity amplifies the glucose response to HFLC. Exercise builds muscle, which acts as a glucose sink. The whole system reinforces itself — but only if every piece is in motion.

The Data Speaks

Four months in. November 2025. Return visit to the same doctor, same lab.

Glucose metabolism — reversed:

• HbA1c: 6.4% → 5.5% (37 mmol/mol). Normal. Solidly out of the danger zone.
• Fasting glucose: 7+ mmol/L → 5.0–5.5 mmol/L
• CGM time in range (3.9–7.8 mmol/L): from ~60% → 98–99%
• Average glucose coefficient of variation: 15–18% (exceptionally stable)

Blood pressure — normalized:

• 155/100 → 120/80. No medication.

Lipids — transformed:

• Triglycerides: dropped to 0.75 mmol/L (66 mg/dL)
• Triglyceride-to-HDL ratio: 1:1. This ratio is one of the strongest markers of insulin sensitivity. Mine went from metabolically broken to metabolically elite.

Inflammation — healing:

• hsCRP trending down from 5.7 → 2.1 mg/L (consistent with years of hypoxic injury from untreated OSA)
• Liver enzymes (GGT) improving on a healing trajectory

Body composition:

• Weight: 74 kg → 66 kg (11% reduction)
• Visceral fat: 0.75 kg — 33% below peer average
• DEXA body fat: 36% (still high — the spare tire is the last to go, but metabolically, the visceral compartment is clean)

Cardiovascular:

• Coronary artery calcium score: 7 (minimal). After decades of untreated metabolic syndrome and severe OSA, the structural damage to my coronary arteries was almost negligible. I got lucky. Not everyone does.
• Lipoprotein(a): <3.1 nmol/L (very low genetic risk)

Sleep:

• OSA: resolved. Pre-surgery ODI 38.5 → post-intervention: resolved through surgery plus weight loss.

No metformin. No antihypertensive. No statin. Just food, movement, sleep, and consistency.

What I Learned

Lesson 1: The easy interventions are often the least effective. Intermittent fasting without dietary quality is just skipping breakfast. Supplements without lifestyle change is expensive theater. Tracking without action is voyeurism.

Lesson 2: Compound interventions compound. No single variable explained my reversal. Diet alone wouldn't have resolved the sleep apnea. Surgery alone wouldn't have fixed the insulin resistance. Exercise alone wouldn't have normalized the blood pressure. It was the simultaneous intervention across all variables that created the cascade.

Lesson 3: Your metabolism is talking. The question is whether you're listening. A continuous glucose monitor doesn't lie. Monthly lipid panels don't lie. But the data only matters if it closes the loop — if you actually change behavior in response to what the numbers are telling you. For seven months, I watched my glucose spike to 10+ mmol/L after every meal and did nothing. The data was there. The action wasn't.

Lesson 4: Windows of opportunity are real, and they close. The post-tonsillectomy period gave me a physiological and psychological launchpad — reduced appetite, momentum from weight loss, the investment of surgery creating a sunk-cost motivation to follow through. If I'd waited another month, two months, three months — the window would have closed. Motivation decays. The body re-equilibrates. You have to move when the door opens.

Lesson 5: N=1 experimentation is powerful, but only with rigor. I'm a sample size of one. My results don't generalize automatically. But the process generalizes: measure baseline, intervene systematically, track outcomes, adjust. This is how any engineer would approach a system that's malfunctioning. Your body is a system.

Why Metabola Exists

I spent 30 years managing technology risk for a major bank. Identifying systemic failures. Building monitoring systems. Designing intervention protocols. Running post-mortems when things go wrong.

Metabolic health is the same discipline applied to a different system. The inputs are diet, movement, sleep, and stress instead of code, infrastructure, and processes. The monitoring is CGM, bloodwork, and wearables instead of dashboards and alerts. The interventions are protocol-driven and data-validated, not guesswork.

Metabola is where I document this work. It's a lab — part personal experiment log, part educational resource, part community for anyone who wants to approach their metabolic health with the same rigor they'd bring to any serious engineering challenge.

If you're a busy professional in your 40s, 50s, or 60s, eating the same food everyone around you eats, getting the same annual checkup everyone gets, and watching the same numbers creep in the wrong direction — this is for you.

I was you. Twelve months ago. The numbers are reversible. The process is learnable. The data is available to anyone willing to look at it honestly.

Welcome to the lab.

Koon Seng is a technology risk leader with 30 years of enterprise engineering experience, a Precision Nutrition Level 1 certified coach, and the author of "The 10,000 Year Detour: How We Lost Our Metabolic Rhythm — and How to Find Our Way Back." He lives in Singapore, where he continues to debug his metabolism one data point at a time.