Week 18: A1c down, fasting glucose stubborn. Anyone else?

I had the same split around week 16. A1c fell, fasting hung out right where yours is. My CGM showed the 4 a.m. rise too. It finally drifted down after a few more weeks.

Dawn phenomenon or less suppressible hepatic glucose? The flat overnight then 4-5 a.m. rise sounds like the latter to me.

Mine was the opposite. Fasting first, A1c lagged.

What’s your post-meal look like? My fasting was noisy until I stopped late snacks. A1c was still the cleaner signal.

I track fasting, 2h post-meal, and monthly A1c. CGM average helped me see trends, but the fingerstick fasting number bounced 15 points day to day.

That bounce is why I stopped weighing every single fasting reading. The CGM overnight curve is more useful for the 4 a.m. question.

I’m at week 72 of a long run, and my A1c went from 7.4 to 5.8 while fasting went from 130s to low 100s, then plateaued for months. What moved it for me was consistency in meal timing and sleep, not any one lab. I also learned to compare CGM average glucose to lab A1c, because anemia and some other things can skew A1c. If your CGM average and A1c disagree, that’s worth asking your clinician about. Not saying it’s your case, just a puzzle piece that helped me stop obsessing over one fasting number.

Anemia skewing A1c? That’s a good point. My ferritin was low and my A1c looked better than my CGM average suggested.

Did your ferritin change after starting? Mine dropped, but I also changed diet a lot.

CGM lag acn make that 4 a.m. rise look worse.

Mine did the opposite: fasting improved fast, A1c took almost 6 months. The 72-week data makes me think slow responders exist.

The part people forget about slow responders: week 12 can look like nothing happened at all. Then somewhere around week 24 it actually shifts. Fasting is the last thing to move, basically always the last marker to budge.

Mine did the same—A1c dropped but morning number stayed ugly for weeks. I tracked seep and stress instead.

Week 20 here: A1c 6.0, fasting 108–118, weight down 31 lb. My fasting stayed higher on Saturday and Sunday mornings, and I finally connected it to Friday takeout plus two drinks. I stopped logging every single morning and just did Mon/Wed/Fri plus waist, and the weekly average told a cleaner story than the daily number.

My fasting finally budged after I stopped eating after 7 and walked 15 min after dinner—down 8 points, not dramatic.

Honestly, I’d trust A1c and waist over one stubborn fasting number; mine lagged 10 weeks then dropped.

Week 16, down 22 lb, resting HR down 8. I quit checking fasting every morning because it made me obsessive; now I log it three times a week and watch the 7-day average. My fasting still hangs around 108-114, but my CGM average dropped about 18 points. The daily number is the noisiest one for me.

My jeans got loose before my fasting number moved at all. Bodies are weird.

Water is my cheat code: if I'm dehydrated, my morning number looks 5-7 points worse.

Weekends are my tell. If I sleep under 6 hours two nights in a row, Monday fasting is 114-118; after solid sleep it's 104-108. I started logging sleep hours and stress levels, and the pattern was ridiculously clear. A1c at week 19 is 6.0, weight down 24, so I try not to spiral over one bad Monday.

Week 19 here and my fasting number was the last domino to fall — it sat between 108 and 118 from week 10 to about week 22 while my A1c went 7.3 to 6.1 and I dropped 31 lb. Weirdly, it got worse during my two fastest-loss weeks, which scared me enough to start logging it twice a day instead of once. What actually predicted the eventual drop was my 2-hour post-dinner readings, not the morning one: once those stopped regularly hitting 160+, fasting slid into the 90s about three weeks later. So I track fasting, 2hr post-dinner, and the CGM 24-hour average, and the order of movement was average first, post-meal second, fasting dead last by a mile. Your pattern reads normal to me, just slow, and I'd bet the overnight flatline plus a 4am bump is your liver doing its morning thng rather than anything you ate.

Mine sat at 112 for two solid months, then one rndom Tuesday it was 94 and stayed there. Bodies are petty.

The only food thing that budged my mornings was the clock, not the carbs. Lst bite by 6:30pm and a genuinely protein-heavy dinner, and my fasting averaged 99 over six weeks versus 114 when I was eating the same calories but finishing at 8:30. I didn't believe it, so I tested it with the same pasta bowl at 6:15 one night and 8:30 another — 14 point difference the next morning, twice. Late protein snack before bed did nothing for me either way, which surprised me since everyone swears by it. It's not a miracle, a 9pm pizza still shows up as a 120 the next day, but shifting dinner earlier was the one lever that actually moved a number that had been welded in place since about week 12.

Fasting was the last thing to move for me too, and the 4-5am bump is textbook liver-waking-up stuff. My clinician said the same thing about labs mattering more and I argued with her for a month before admitting she was right. Now I check fasting maybe twice a week and lean on my 14-day CGM average and waist measurement instead. Way less data, way less crazy.

I cook every meal for my mom now that she's living with us, so I get the constant bites — tasting while I season, finishing what she leaves on her plate. None of that ever made it into my log. Two weeks of writing down sleep hours plus four or five bites a day I'd have sworn I never took, and the picture changed. Rough nights, fasting up, identical dinner. Were you tracking sleep or stress at all? And congrats on the A1c — 28 lb is huge.

have you looked at sleep midpoint instead of dinner timing? For me, fasting glucose correlated way more with sleep midpoint variance and overnight HRV than with carbs. I logged both next to the CGM, and the 4–5 a.m. rise shrank on weeks my midpoint was stable within ~30 min. Alcohol was noise compared to sleep debt. Does your overnight line do a small 2–3 a.m. dip before the rise, or just a clean ramp? The dip pattern was the tell for me.

Veteran here. Mine tracked sleep way more than dinner timing. I added columns for bedtime, wake time, and overnight wakeups, and my fasting sat 108–114 until I had two weeks of consistent 7+ hours. A1c had already improved; mornings just lagged. A thread last year about “sleep debt and dawn numbers” finally made it click for me. Are you logging sleep consistency, or just food and CGM? That’s where I’d poke next.

Road warrior here too. The weird one for me was sleep continuity, not just bedtime. I started tagging hotel nights vs home nights in my CGM app and noting wakeups (HVAC, neighbor doors, time zones). My fasting stayed 108ish after decent dinners but dropped into the 90s on trips where I actually got 7 unbroken hours in the same zone two nights running. Airport days with 5-hour sleeps? Forget it. So now I track “hours slept before 3am” as a dumb metric. Anyone else see fasting follow sleep debt more than supper? (Also: hotel minibar nuts at midnight are a trap, ask me how I know.)

Congrats on the A1c and 28 lb—that’s a lot of quiet consistency. I’m 20 weeks in and my fasting also lagged. The new clue for me was logging sleep duration and wake time right next to my CGM. On nights under 6 hours, or when I woke at 3–4 a.m. and checked, my “fasting” looked 8–10 points higher. When I slept 7+ and stayed horizontal until my normal wake time, it was lower. Have you looked at sleep/wake timing yet? Curious if your 4–5 a.m. rise shifts on rough-sleep nights.

I’m about a year into the empty-nest reset too, and the surprise for me was how much my morning number tracked sleep, not supper. Same dinner, but after under 6 hours of sleep or a glass of wine, my CGM morning average ran 8–10 points higher. I started jotting bedtime, wake-ups, and stress 1–10 next to the graph. Over a month, consistent sleep and a boring wind-down routine moved my average more than any dinner tweak. Have you compared your stubborn mornings to sleep and alcohol at all?

I do on-call locksmith work — a 2 a.m. lockout call and my whole sleep block gets scrapped. Noticed my CGM starts climbing before I'm even awake. What moved the needle wasn't food, it was tracking when I slept. Rough sleep or a shifted block: fasting ran 10-15 points higher. Hold a consistent sleep window two weeks and it came down. Now sleep hours and wake time go in the log right next to the fasting number. See if your worse days track

Week 22 here. A1c 7.0→6.1, fasting still 100–112. What helped me was tracking the overnight delta: bedtime CGM vs. 3 a.m. vs. wake. My 4–5 a.m. rise correlated more with late dinners and evening stress than total carbs. Moving dinner 90 min earlier plus a 10-min post-dinner walk shrank the rise, but fasting stayed stubborn. Have you tried isolating last-meal timing vs. composition instead of chasing the morning number?

Fellow dorm-dweller here. My A1c also dropped while fasting camped out at 103–112. What gave me a clue was tracking my last meal’s protein-to-carb vibe and the gap before bed—not just the overnight CGM line. Cheap-protein rotation: cottage cheese cup, tuna packet, boiled eggs swiped from the dining hall, lentil pouch. When I had that instead of 10 p.m. ramen, my 4–5 a.m. rise looked less like a ski slope. Fasting still isn’t pretty, but it shifted a few points. Did you log what you ate 10–12 hours before those morning numbers, or only the

Two kids here, so my fasting number ended up being less about overnight and more about the 8:30pm snack-drawer raid after cleanup. I started jotting only two things: time of last bite and whether it was leftover carbs vs cheese/nuts. Same bedtime, but late crackers = 6am number 8-10 points higher for me. The 4-5am CGM rise stayed either way. Not saying it’s your cause, just the one variable that actually moved my morning number. Have you logged late snacks, or only the overnight window?

Fasting glucose is often the last metric to budge—A1c reflects 24/7 glycemia, so it can improve while dawn phenomenon masks the morning number. I’m week 30: A1c 7.3→6.1, fasting 104–116. What finally correlated for me was sleep fragmentation and late meals, not weight. I logged last bite time, alcohol, sleep efficiency, and CGM. My 4–5 a.m. rise was ~15–20 mg/dL worse after <6h sleep or a late dinner. Have you compared fasting to sleep efficiency and last meal time?

Hi—brand new here, only week one, so I’m mostly reading and taking notes. This thread is really reassuring. One thing I’ve started logging (maybe overkill?) is dinner timing: my fasting number is consistently 8–10 points

Cafe owner here — same stubborn fasting, A1c dropped but mornings hang around 104–112. What finally gave me a clue was tracking my “last bite” time, not just dinner. I graze while closing: broken cookie, sauce taste, latte I didn’t log. If I nibble after 8:30, my 4–5 a.m. rise looks noticeably higher. Not saying that’s your story, but has anyone else separated a true dawn rise from late-kitchen-temptation rise? I’m trying to be honest about the tasting spoons now.

Road warrior here—airport pretzels are a food group. My fasting line only made sense when I split logs into hotel days vs. home days. Weirdly, my worst 4–5 a.m. rises were travel days with a late lounge “dinner” and hotel thermostat set to arctic. When I swapped in a grocery-store rotisserie chicken + bagged salad, even at 9 p.m. local, the dawn bump got less dramatic. Not magic, just a pattern. Are you separating travel vs. home days in your CGM?

Empty-nester here, about 5 months in. My A1c dropped while fasting hung around 108–114. What finally gave me a clue was logging my last bite time, not just overnight. Dinner at 6 with a 15-minute walk after moved my morning number a little; dinner at 8 didn’t. My CGM still shows that 4–5 a.m. rise, so I mostly watch the weekly average now. Curious—does yours drift down by mid-morning, or does it stay up until you eat?

i’m in maintenance now, but I had that same split: A1c kept improving while fasting camped in the low 100s. What helped me stop obsessing was tracking my 24-hour average and time-in-range separately from that single morning number. For me, a later dinner or a rough night’s sleep reliably added 8–12 points, even when my CGM looked calm otherwise. Have you compared fasting mornings after an earlier, protein/fiber-heavy dinner versus a late one? That pattern was way more useful to me than any one fasting reading.