Allulose and blood sugar
What 7 graded rows on allulose agree on for blood sugar, and where they differ. Counted side by side, never averaged.
Consensus finder Rows from the Allulose page, reviewed Oct 21, 2026 How it decides
What the rows agree on
Each block and row is one graded row from the Allulose page, with its PubMed record.
Your question
Allulose and blood sugar
7 rows · 1,298 peopleMostly an improvement: 6 of 7 rows, and most of the people. 1 found no difference.
- BGrade B× 1
- CGrade C× 1
- DGrade D× 4
- ∅No effect× 1
- Rows
- 7
- Trials
- 25
- People
- 1,298
- Pooled reviews may count some twice
Where they differ
From the rowsThe best-graded row, grade B, is a meta-analysis of 1,033 people (2026) and found an improvement. 1 row is graded No effect, our grade for adequately powered trials that found no effect.
Only the 6 rows that found an improvement include pooled reviews of several trials (2 of 6); the row that found no difference is a single study.
The rows
6 improved · 1 no difference- BGrade BAfter-meal glucose and insulin (pooled trials) · Improved
Meta-analysis, 12 studies · n 1,033 · 2026 · Adults in 20 controlled trials (12 of allulose, 8 of tagatose); 1,033 participants across both sugars, the allulose-only total is not given in the abstract
As tested in the pooled trials (not stated in the abstract) · Single doses and longer trials
Allulose lowered the after-meal glucose rise (incremental AUC SMD −0.66; 95% CI −0.92 to −0.39) and insulin (SMD −1.27; −2.14 to −0.40; I² 96%), both moderate certainty.
- CGrade CAfter-meal blood glucose in healthy people · Improved
Meta-analysis, 8 studies · n 145 · 2023 · Healthy people in 8 small randomised crossover experiments (8 to 30 per group; 145 per group in total)
5 g, or up to 10 g, with a meal · Single doses
Incremental 2-hour glucose AUC was lower with 5 g (SMD −0.28; 95% CI −0.51 to −0.05) and with up to 10 g (SMD −0.26; −0.49 to −0.03), both rated moderate certainty.
- DGrade DGlucose and insulin after a sugar drink · Improved
Crossover RCT · n 30 · 2021 · 30 people without diabetes
2.5, 5, 7.5 or 10 g, single doses (vs placebo) · Single test days, 7 to 14 days apart
Glucose at 30 minutes was lower with 7.5 g (mean difference 11; 95% CI 3 to 19) and 10 g (12; 4 to 20); overall glucose and insulin excursions fell significantly only with 10 g.
- DGrade DAfter-meal glucose in borderline diabetes, and 12-week safety · Improved
Crossover RCT · n 26 · 2010 · 26 adults including people with borderline diabetes (crossover), plus 17 healthy adults in a 12-week safety arm
5 g with a meal; 5 g three times a day in the safety arm · Single meals; 12 weeks for safety
Glucose was lower 30 and 60 minutes after the meal (p<0.01, p<0.05) with a smaller AUC, mainly in borderline diabetes; 12 weeks of 15 g/day caused no abnormal effects or clinical problems.
- DGrade DPeak after-meal glucose on a diabetic diet in type 2 diabetes · Improved
Crossover RCT · n 24 · 2023 · 24 hospital inpatients with type 2 diabetes in Japan (full text)
8.5 g of allulose in the diet · 2 days on each diet, at least 1-day washout (full text)
Peak after-meal glucose, measured by continuous glucose monitor, was lower on the allulose diet than on the standard energy-controlled diabetic diet; C-peptide fell, which the authors read as a lower insulin requirement.
- DGrade DGlucose after a glucose drink in type 2 diabetes · Improved
Crossover RCT · n 24 · 2018 · 24 adults with type 2 diabetes in Toronto
5 g or 10 g, single doses · Single test days, more than 1 week apart
10 g lowered the glucose incremental AUC by 8% vs 0 g (717.4 vs 777.5 mmol×min/L, P = 0.015) with a linear dose response; fructose had no effect.
- ∅No effectGlucose control in type 2 diabetes · No difference
Crossover RCT · n 16 · 2023 · 16 patients with type 2 diabetes (Thai trial)
7 g twice daily (14 g/day) · 12 weeks per arm with a 2-week washout
No significant effect on glucose tolerance, incretin levels or body composition; during allulose, HDL cholesterol fell from 51 to 41 mg/dL and MCP-1 rose from 259 to 297 pg/mL.