Glucose spikes in healthy people: what the research actually says

A glucose "spike" is just the rise in blood (or interstitial) glucose that follows a meal. In healthy non-diabetic adults, spikes are normal and expected. The interesting questions are about size, duration, frequency, and whether they predict anything about long-term health.

What a normal response looks like#

Several large CGM studies in non-diabetic adults (Hall et al. 2018; Zeevi et al. 2015; PREDICT 1 / Berry et al. 2020) report that healthy people commonly:

• Reach peak glucose 30–60 minutes after a mixed meal
• Peak somewhere in the 7–9 mmol/L range for typical Western meals
• Return to baseline (4–6 mmol/L) within 2–3 hours

The same meal in the same person produces a markedly different curve on different days depending on sleep, prior exercise, time of day, and whether it was paired with fat, fibre, or protein.

Inter-individual variability is huge#

The biggest finding from large CGM cohorts is how much the same meal varies between people. Zeevi et al. fed identical meals to 800 non-diabetic adults and found that postprandial response varies widely between people. PREDICT 1 (the ZOE study) replicated this in over 1,000 UK and US participants.

In other words: there is no universal "good" or "bad" food for glucose response; there's your response to a specific food in a specific context.

What spikes correlate with#

In non-diabetic populations:

• Larger and more variable postprandial responses correlate with higher fasting insulin and triglycerides in cross-sectional studies.
• Whether reducing spikes in a non-diabetic person produces any long-term cardiometabolic benefit is not yet established. The intervention studies are short and rely on surrogate markers.
• HbA1c (a 3-month average) remains the most-validated long-term measure for glucose-related disease risk in non-diabetics, not CGM-derived spike size.

What we'd actually take from this#

A CGM in a healthy adult is useful as a personalisation tool, not a verdict. If you find a meal you eat regularly produces a much larger response than your other meals, the literature suggests options that often blunt that response on retest: a 10-minute post-meal walk, eating protein and vegetables before the carbohydrate portion, or simply not eating that food on its own.

If the same CGM shows fasting glucose persistently above 7 mmol/L or post-meal readings above 11 mmol/L, the appropriate response isn't a tweak: it's an HbA1c test with your GP.

What the evidence doesn't support#

The claim that any post-meal glucose rise above 1.7 mmol/L is "bad" has weak grounding in the published literature. It's a marketable threshold, not a research-validated one. The PREDICT 1 paper specifically noted that single thresholds are misleading given inter-individual variability.