Why Is My Sleep Still Bad After Losing Weight?

My blood tests were coming back better every month. My GP was happy. And I was still waking up at three in the morning wondering what was wrong with me.

Nothing looked broken from the outside. The numbers were moving in the right direction. But I’d lie there at three, four in the morning — completely awake, no reason for it — and think: if I’ve fixed all these other things, why is this still happening?

I’m Neil. NHS blood data going back to 2012, off all medication since February 2024. And it took me six months of improving metabolic markers before sleep finally followed. In the next ten minutes I’m going to show you why sleep is always last in the repair sequence — and why that’s not failure, it’s physiology. If your blood sugar, inflammation, or weight are heading in the right direction and sleep is still broken — this is for you. That’s exactly how it works.

We’ll cover three things. First — why metabolic damage disrupts sleep architecture at a deeper level than most people realise, and why fixing the blood tests doesn’t automatically fix the sleep. | Second — what my own six-month sleep lag looked like, month by month, including the one morning in May 2024 when I woke up naturally at 5:47am and understood that something had finally reset. | And third — the one thing I was doing wrong in months two and three that was actively delaying the process, and what I changed.

Here’s what most people believe, and I believed it too for a long time.

Sort your diet. Lose a bit of weight. And sleep will sort itself out. Simple cause and effect — improve the inputs, sleep follows. It’s in every lifestyle article, every wellbeing guide. “Eat better, sleep better.” Clean living leads to clean sleep.

And on the surface? Makes complete sense. There’s nothing wrong with that logic.

The problem is the timeline it implies. Fix the food, fix the sleep — and the assumption underneath it is that they happen together. Simultaneously. That you clean up what you’re eating and within a few weeks the sleep catches up.

That’s the bit nobody mentions. The lag.

Kayla Barnes-Lentz — longevity specialist, over a decade in this space — put it plainly: poor sleep increases ghrelin, suppresses immune function, and directly undermines whatever nutrition work you’re doing the following day. What she’s describing is a two-way street. Sleep disrupts metabolic health. But metabolic damage also disrupts sleep. And when you’ve spent years building up that damage, unwinding it takes longer than a fortnight of good meals.

The conventional framing misses that. Completely.

My sleep started going wrong somewhere around 2021. I think. Covid period, working from home, routines completely collapsed.

Before that, I’d been a heavy sleeper. Proper deep sleep, nothing woke me. Then — gradually — that changed. I went from sleeping deeply to being restless. Not every night. If I was genuinely exhausted, I’d go down and stay down. But most nights, I’d wake somewhere between two and four in the morning. Completely awake. No obvious reason. Just — awake.

And I’d lie there thinking about nothing in particular, which is somehow worse than lying there worrying about something.

This carried on for years. And during those same years, here’s what my blood tests were showing.

July 2020 — HbA1c 43.2 mmol/mol. Pre-diabetic range. Cholesterol 7.4. October 2022 — HbA1c climbed to 47. December 2023 — cholesterol peaked at 8.0. Highest it had ever been.

I was on six medications. Statins for the cholesterol. Allopurinol for gout. Asthma inhalers. Antihistamines. I was doing what I was told. And the markers were still going in the wrong direction. The sleep was broken, the bloods were getting worse, and I couldn’t work out which was causing which.

Turns out — both. They were feeding each other.

Boxing Day 2023. I made a decision. Changed everything — what I was eating, when I was eating, how much I was moving.

The specifics: I moved to one meal a day. Cut out ultra-processed food almost entirely. Started walking every morning, most days. And — this is relevant to the sleep — I stopped eating late.

Not deliberately at first. Just naturally. When you compress your eating window, you stop sitting on the sofa at midnight with a biscuit and a full-caffeine coffee. That behaviour just — went. And I hadn’t quite registered it as a sleep intervention at the time. I thought I was changing my diet. I didn’t realise I was also changing my circadian signals.

I stopped all medication on the 31st of October 2023 — I told my GP afterwards, with the trajectory to show her. And from January 2024, I was tracking blood results regularly.

Sleep, though? I was monitoring that myself. My GP and I never discussed it once. Not one conversation. Which is either a gap in the system or a sign that my numbers were the thing they were focused on. Possibly both.

What I was doing wrong in months two and three: eating close to bedtime — and caffeine, full strength, late into the evening. I genuinely didn’t clock that having a strong coffee at ten or eleven at night and then wondering why I couldn’t sleep was part of the same problem. Moved to decaf after midday. Stopped the late snacking. That shift mattered.

The blood markers moved first. Significantly.

January 2025 — HbA1c 40. Back in normal range for the first time in years. CRP — inflammation marker — dropped from 6.2 in January to 1.0 by April. Liver enzymes, which had been elevated on the statins, normalised completely. ALT went from 69 down to 10.

But sleep? Sleep was lagging behind.

Months one and two post-reset — still broken. Still waking at three or four. Still lying there. I was eating better, moving more, losing weight — and the three-o’clock wake-ups hadn’t stopped.

Month three, four — I started noticing I was falling asleep at a more reasonable time. Not dramatically earlier. Just — earlier. And the waking was becoming less frequent. Less predictable.

Month five — still some disruption, but I was waking and getting back to sleep rather than lying there for an hour.

And then — May 2024. Month seven, roughly. I woke up at 5:47am. No alarm. Room still dark. And I just — lay there feeling completely clear. No grogginess. No wondering why I was awake. Just awake, properly, at the right time.

I genuinely thought something was wrong. That’s how unfamiliar it felt.

It happened the next morning. And the morning after that.

By the end of 2025, I tracked 57 nights rated nine or ten out of ten for sleep quality across the full year. No sleeping pills. No melatonin supplements. No alcohol as a sleep aid. Just metabolic health that had — finally — caught up.

My GP has started looking forward to my blood tests almost as much as I do. Which probably says something about the state of the NHS — but also about what’s possible when you actually change the inputs. data going back to 2012, all of it documented here. Subscribe if you want to follow it.

THE LESSON

What the data suggests

What my six months of broken sleep followed by genuine reset suggests — and it’s one person’s data — is that sleep is downstream of metabolic health, not parallel to it. You don’t fix them together. You fix the inputs first, and sleep follows on its own timeline. In my case, that was six months. Not six weeks.

When the conventional advice still applies

This doesn’t mean everyone with poor sleep should start fasting and wait. If your sleep is disrupted by sleep apnea, by chronic anxiety, by medication side effects, by shift work — those are different problems with different solutions. And if you’re on medication of any kind, the conversation about changing your eating patterns starts with your GP. Not with YouTube.

The prevention-stage angle (Metabolic Cascade: YES)

Here’s the thing worth sitting with if you’re not yet in the clinical conversation. The mechanism — the way blood sugar instability and low-grade inflammation interfere with sleep architecture — tends to be running in the background years before anyone puts a number on it. The 3am wake-up often comes before the HbA1c creeps up. Before the cholesterol letter. It’s the body signalling early that something’s off. If that sounds familiar, it might be worth paying attention to — not after the diagnosis, but now.

I’ll say it plainly. Everything I changed was done with my GP aware of the trajectory. Blood tests, regular monitoring, no changes to medication without review. If you’re watching your numbers and you’re curious about what lifestyle changes might do — the place to start that conversation is with your GP. That’s not a disclaimer. That’s what I actually did.

If this landed for you, my 7-Day Fasting Blueprint is free in the description. It covers the eating window changes that, for me, turned out to be doing double duty — metabolic repair and sleep, both at once. I didn’t plan it that way. The data just showed up like that.

Link’s below.

Educational content only — not medical advice. Never stop or change medication without your GP’s supervision. I stopped my own medications in February 2024 on my own initiative and told my GP afterwards. That is not the approach I am recommending. Speak to your GP before changing your fasting pattern, diet, or medication.