Cohort studyindependent cross-checkabstract checked★ key study I-INS-129 in-depth review
Ketoacidosis during very-low-carbohydrate diets
Illnesses associated with ketosis including diabetic ketoacidosis during very low carbohydrate and ketogenic diets.
Overview
Roberts CGP et al.. Diabetes Obes Metab 2025 · PMID 39996364 · DOI 10.1111/dom.16252
DKA was rare (about 1 per 1000 patient-years in type 2 diabetes) in a supervised programme, but almost three times higher with SGLT2 inhibitors.
Does the study support the specific claim? Partly · Methodological quality: low
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Measure ketosis-related illness on ketogenic diets.
- How was the study done?
Retrospective programme data.
- How many people, animals or samples were studied?
72,751 patient-years.
- What was actually found?
DKA 1.01/1000 PY in T2D; 2.90 with SGLT2 inhibitors.
- How large was the observed effect?
Rates as reported.
- What do the results mean in everyday life?
Diet changes in diabetes need medical supervision and ketone checks.
- What are the study's limitations?
Programme data; no T1D.
- What does it say about Elawady's specific claim?
Supports supervision and monitoring, not stopping medicines alone.
- Do the results apply to Germany and Europe?
Applicable.
- What can explicitly not be concluded from this study?
Cannot address type 1 diabetes.
Scientific in-depth review
1. Full reference
- Authors
- Roberts CGP et al.
- Journal
- Diabetes Obes Metab
- Year
- 2025
- Identifier
- PMID 39996364 DOI 10.1111/dom.16252
- Verification
- verified – PubMed efetch 2026-10-26; Crossref 2026-10-26
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
Not stated in abstract (authors linked to the programme per affiliation; not verified).
2. Research question
How common are illnesses with ketosis and DKA during very-low-carbohydrate ketogenic diets?
3. Study design and population
- Study design
- Cohort study – Retrospective real-world cohort of a telemedicine ketogenic-diet clinic
- Population
Patients with and without T2D in a VLCKD programme
- Country
USA
- Period
Not stated in abstract
- Sample size
72,751 patient-years
- Inclusion and exclusion criteria
Programme participants
4. Methods and endpoints
- Exposure / intervention
VLCKD (with or without SGLT2 inhibitors)
- Comparator
Subgroups
- Endpoints
BHB >=3 mmol/L illness (IAK); DKA
- Statistical methods
Incidence rates
5. Actual results
- Results
IAK 1.18 per 1000 person-years overall; without diabetes 0.04 per 1000 PY and no DKA; with T2D IAK 1.69 and DKA 1.01 per 1000 PY; with T2D on SGLT2 inhibitors DKA 2.90 per 1000 PY.
- Absolute risks
not reported
- Authors' conclusion
VLCKDs are generally safe with low rates of IAK and DKA; higher DKA with SGLT2 inhibitors may be manageable with home BHB monitoring.
6. Strengths
- Large person-time
7. Weaknesses
- Programme-run data; possible underascertainment
- No control group
- Excludes T1D
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
Selection and ascertainment bias possible.
9. Assessment of the statistical interpretation
Incidence rates only.
10. Relevance to the specific claim
Diet-based programmes in T2D can be run with low DKA rates when supervised with ketone monitoring and medication adjustment; DKA risk rises with SGLT2 inhibitors.
11. Independent contrary findings
12. Transferability to Germany and Europe
Directly applicable; same biology and clinical standards in Europe.
13. Overall methodological judgement
Supports supervised, monitored dietary therapy, not unsupervised medication stopping.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | low |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | unknown |
| Consistency with the overall evidence | consistent |
| Regional transferability | moderate |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
15. Linked fact checks
- People with diabetes can stop insulin and diabetes medication by following the Tayebat system ('zero medicine'), as shown by followers who stopped insulin and are 'perfectly fine'. – Studies with mixed or unclear results
- Laboratory values such as blood glucose and acetone are only signs of the body's changing strategy and not causes; there is no need to measure glucose or blood pressure. – Studies contradicting or limiting the claim