How Do I Know If My GLP-1 or Metformin Is Working Without Waiting for A1C

In clinical experience, a CGM can show early glucose changes within one to three weeks, long before an A1C lab moves. This timing is illustrative, not a sourced threshold. You are looking for three changes: a lower fasting baseline, smaller and often later post-meal rises, and less time above your usual upper band. A1C still matters for the three-month average, but it cannot show which part of the day changed or which meal improved.

How do I know if my GLP-1 or metformin is working without waiting for A1C?

Compare three stable days before you started with three stable days after ten to fourteen days. Do not compare one isolated reading. The early signs to look for are:

• Fasting and overnight glucose: the median between 2am and 6am drifts down by a few mg/dL, and the early morning rise is smaller. This is an illustrative pattern, not an established range.
• Post-meal glucose: the rise from the pre-meal value to the peak is smaller for the same meal. In clinical experience, the reduction is often around 10 to 30 mg/dL, and with a GLP-1, the peak may arrive 15 to 45 minutes later. These numbers are illustrative, not established ranges.
• Time above your usual upper band: less time above 140 mg/dL. In one multicentre study of adults without diabetes, about 96% of the day was spent between 70 and 140 mg/dL (Shah et al. 2019).

For metformin the overnight and fasting change is usually the first thing visible. For a GLP-1, the meal curve changes first, often before appetite or weight changes become large.

What the two drugs change in the glucose curve

Metformin reduces how much glucose the liver releases and makes muscle and fat slightly more responsive to insulin. The liver is the main source of glucose between meals and overnight, so metformin lowers the floor, especially from late night into morning. It does not blunt a meal as much as it lowers the baseline.

A GLP-1 receptor agonist increases glucose-dependent insulin, reduces glucagon, and slows stomach emptying. The slower emptying changes meal timing, so food leaves the stomach more slowly, glucose enters the blood later, and the rise is smaller. That changes the shape of the meal curve, not just the height.

A curve that stays above 140 mg/dL for less time is more informative than a single lower post-meal number. A lower baseline and a smaller meal rise both contribute to that.

What to watch in the first two weeks

Use the same meals as your comparison. If you change your breakfast, you cannot separate the drug from the food. The exact size of an early drop is not well standardised across people, so treat these as patterns, not pass marks.

• Metformin: in clinical experience, some people see a downward shift in overnight baseline within 3 to 7 days. This timing is illustrative, not an established range. The change may be a few mg/dL on the median and still meaningful across a full day.
• GLP-1: meal effects can appear within the first days, but the full effect builds over weeks as the dose is adjusted. A peak that is lower but later is still an effect.
• Ignore the first day of a sensor. It is the least reliable.
• Compression lows, where the reading falls fast overnight and recovers quickly, are sensor artifacts, not a medication effect.
• Average glucose drops of 5 to 15 mg/dL over a week can matter in clinical experience, but this is illustrative and individual variation is wide. Do not expect a flat line.

The mistake people make with a lower one-hour reading

A lower one-hour reading is not the same as less glucose. Look at the full two to three hours after a meal.

A GLP-1 that slows stomach emptying can make the same food produce a lower reading at 60 minutes but a longer tail at 90 to 150 minutes. The total glucose exposure may be lower, but not as much as the one-hour point suggests. If you only check one hour after eating, you will overread the effect. Check the area under the curve, or at least the full two to three hours.

What A1C can and cannot tell you later

A1C reflects glycated haemoglobin, a weighted average of the last two to three months. After a medication change, the lab value moves slowly because older red blood cells from before the change are still in circulation. Some change can appear at four weeks, but the three-month mark is the fair comparison. CGM gives the process: what happened, when, and after which meal. It does not replace A1C, and it is not a diagnostic device. The sensor's glucose management indicator can estimate A1C from 14 days or more, but it is not a lab result and can differ from a measured A1C.

A medication can be doing something even while A1C is still unremarkable. That is not failure. Measure the process with the sensor, and confirm the average with the lab.

If you want to see these patterns in your own numbers, drop a LibreView, Dexcom Clarity, Stelo, Lingo or SiBionics export at the free CGM report. It shows your AGP, time in range and biggest rises, in the browser, no account. If you do not have a sensor, HealthOS Sensor is a month of continuous glucose with the app for $70, no membership. HealthOS Sensor is a wellness product, not a medical device, and does not diagnose or treat.