Cohort studyindependent cross-checkabstract checked★ key study I-FRU-109 in-depth review
Fruit, fruit glycaemic load and diabetes in Korean adults
Fruit intake, fruit glycemic load, and type 2 diabetes risk: findings from a community based Korean cohort study.
Overview
Ha AW. Nutr Res Pract 2026 · PMID 41694466 · DOI 10.4162/nrp.2026.20.1.101
In Korean women, very high fruit intake and higher watermelon and melon intake were weakly linked to diabetes; in men there was no link.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Study fruit-GL and diabetes.
- How was the study done?
19-year cohort.
- How many people, animals or samples were studied?
761 diabetes cases.
- What was actually found?
Positive association in women only.
- How large was the observed effect?
Watermelon HR 1.02 per SD.
- What do the results mean in everyday life?
A weak signal that needs confirmation; not a reason to ban melons.
- What are the study's limitations?
Multiple testing; one population.
- What does it say about Elawady's specific claim?
Weak support for C-FRU-01.
- Do the results apply to Germany and Europe?
Low.
- What can explicitly not be concluded from this study?
It cannot show that melons cause diabetes.
Scientific in-depth review
1. Full reference
- Authors
- Ha AW
- Journal
- Nutr Res Pract
- Year
- 2026
- Identifier
- PMID 41694466 DOI 10.4162/nrp.2026.20.1.101 · PMC12893857
- Verification
- verified – PubMed efetch 2026-10-26 + Crossref works API
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
No conflicts declared.
2. Research question
Are fruit intake and fruit glycaemic load associated with incident T2D in Korean adults?
3. Study design and population
- Study design
- Cohort study – Prospective community-based cohort (KoGES Ansan and Anseong), sex-stratified
- Population
Korean adults in the Ansan and Anseong cohorts
- Country
South Korea
- Period
2001-2020 (19-year follow-up)
- Sample size
761 incident T2D cases over 73,545 person-years
- Inclusion and exclusion criteria
Baseline participants with FFQ and follow-up.
4. Methods and endpoints
- Exposure / intervention
Fruit intake and fruit-GL (12 fruits, FFQ)
- Comparator
Lowest quartile
- Endpoints
Incident T2D
- Statistical methods
Cox models (3 adjustment levels); per-SD analyses for 12 individual fruits without mutual adjustment
5. Actual results
- Results
Women: highest vs lowest fruit-intake quartile HR 1.45 (1.04-2.03); fruit-GL HR 1.36 (0.98-1.91). Men: no association after full adjustment. Individual fruits (women): watermelon HR 1.02 (1.00-1.04) per SD, P=0.024; melon/chamoe HR 1.09 (1.00-1.19), P=0.041; other fruits not significant.
- Absolute risks
Incidence 10.4 per 1,000 person-years overall.
- Authors' conclusion
In Korean women, excessive fruit consumption (even low fruit-GL) increased diabetes risk; quantity, quality and macronutrient balance may matter.
6. Strengths
- Long follow-up
- Includes watermelon and melon specifically
7. Weaknesses
- 24 individual-fruit tests without multiplicity correction
- Women only
- Single baseline FFQ
- Contrasts with most cohorts
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Multiplicity; residual confounding (higher-intake women older, lower SES).
9. Assessment of the statistical interpretation
Watermelon and melon findings do not survive Bonferroni correction (threshold 0.0021); with 24 tests about 1.2 false positives are expected (stats/C-FRU-01_ha2026_multiplicity.py).
10. Relevance to the specific claim
The only new evidence found that specifically links watermelon and melon to higher T2D risk (C-FRU-01); weak, subgroup-based and contrary to most data. The authors do not attribute it to the GL of melons.
11. Independent contrary findings
I-FRU-106, I-FRU-03, I-FRU-01.
12. Transferability to Germany and Europe
Low (Korean dietary context with rice-heavy diets).
13. Overall methodological judgement
Hypothesis-generating support for C-FRU-01; insufficient to justify avoidance.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | low |
| Directness | direct |
| Reproducibility | not replicated |
| Consistency with the overall evidence | inconsistent |
| Regional transferability | low |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Replication in other cohorts.