Randomised controlled trialindependent cross-checkfull text checked I-FAST-14 in-depth review
DiRECT after five years
5-year follow-up of the randomised Diabetes Remission Clinical Trial (DiRECT) of continued support for weight loss maintenance in the UK: an extension study.
Overview
Lean ME et al.. Lancet Diabetes Endocrinol 2024 · PMID 38423026 · DOI 10.1016/S2213-8587(23)00385-6
Only 13% of those continuing support were still in remission at 5 years; people with long diabetes or on insulin were not studied.
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?
Outcomes at 5 years with continued low-intensity support
- How was the study done?
5-year extension study of a cluster-randomised trial (DiRECT)
- How many people, animals or samples were studied?
85 extension; 149 controls; DiRECT participants
- What was actually found?
Remission declines over time.
- How large was the observed effect?
13% remission at 5 years
- What do the results mean in everyday life?
Remission needs ongoing weight maintenance.
- What are the study's limitations?
Extension group self-selected; many missing data (especially controls); open-label; only people with diabetes under 6 years and not on insulin.
- What does it say about Elawady's specific claim?
Supports and limits C-DIAB-03 (durability).
- Do the results apply to Germany and Europe?
UK.
- What can explicitly not be concluded from this study?
Cure; remission in long-standing insulin-treated diabetes; effect of any diet other than this low-calorie programme.
Scientific in-depth review
1. Full reference
- Authors
- Lean ME et al.
- Journal
- Lancet Diabetes Endocrinol
- Year
- 2024
- Identifier
- PMID 38423026 DOI 10.1016/S2213-8587(23)00385-6
- Verification
- verified – PubMed efetch 2026-10-25 (this agent)
- Retracted
- no
- Corrections
ErratumIn: Lancet Diabetes Endocrinol. 2024 Jun;12(6):e17. doi: 10.1016/S2213-8587(24)00128-1.
- Extent checked
- full text
- Conflicts of interest and funding
Funded by Diabetes UK (no role in design, analysis or writing). Authors declare grants/fees from Counterweight (intervention provider), Cambridge Weight Plan, Novo Nordisk, Eli Lilly, Novartis and others; one author employed by and holds shares in Counterweight.
2. Research question
Outcomes at 5 years with continued low-intensity support
3. Study design and population
- Study design
- Randomised controlled trial – 5-year extension of DiRECT, an open-label cluster-randomised controlled trial in UK primary care (ISRCTN 03267836): observed outcomes for the original randomised groups (from routine NHS records) plus a non-randomised comparison of self-selected 'extension' participants (continued low-intensity support) vs 'non-extension' intervention participants.
- Population
Adults aged 20-65 with type 2 diabetes diagnosed within 6 years (mean duration 3.0 years), BMI 27-45 kg/m2 (mean ~34-35), NOT receiving insulin, in Scotland and Tyneside.
- Country
UK
- Period
Recruitment 25 July 2014 - 5 August 2016; follow-up to 5 years.
- Sample size
298 randomised (149 intervention, 149 control); 95 entered extension (85 with year-5 weight), 54 non-extension; year-5 data: control weight 82 (55%), HbA1c 93 (62%); intervention weight 116 (78%), HbA1c 118 (79%).
- Inclusion and exclusion criteria
Age 20-65, T2D <6 years, BMI 27-45, not on insulin (details in DiRECT protocol).
4. Methods and endpoints
- Exposure / intervention
Counterweight-Plus: withdrawal of glucose-lowering and antihypertensive drugs, total diet replacement 825-853 kcal/day for 12-20 weeks, stepped food reintroduction (2-8 weeks), structured maintenance; extension: low-intensity support every 3 months for 3 further years.
- Comparator
Original control: best-practice NHS care (encouraged to lose weight after year 2); non-extension intervention participants.
- Endpoints
Bodyweight, HbA1c, remission (HbA1c <48 mmol/mol without glucose-lowering drugs), medications, serious adverse events, MMACE/MMADE, quality of life.
- Statistical methods
Descriptive observed-data analyses without imputation; t-tests, bootstrapping, Fisher's exact tests, negative binomial/Poisson for event rates; no multiplicity adjustment; clustering checked in mixed models.
5. Actual results
- Results
Extension group year 5: mean weight -6.1 kg (SD 5.2); 11/85 (13%) in remission (59/95, 62%, at year 1); off glucose-lowering drugs 40%; HbA1c 5.5 mmol/mol above baseline. Of those in remission at year 2, 11/43 (26%) remained in remission at year 5. Extension vs non-extension: visits in remission 34% vs 12%; HbA1c <48 36% vs 17%; off medication 62% vs 30%. Original intervention vs control: remission at year 5 12/118 (10%) vs 5/93 (5%); visits in remission 27% vs 4%; weight >5% below baseline 61% vs 29%; HbA1c at year 5 not different (66.7 vs 64.4 mmol/mol). Year-5 weight change -5.6 vs -4.6 kg (difference -1.1, 95% CI -2.8 to 0.6). With >10 kg loss, remission 78% at year 2 but 14% at year 5. Serious adverse events 4.8 vs 10.2 per 100 person-years (IRR 0.5, 0.3-0.8); 8 new cancers in control vs 1 in intervention group; no significant difference in MMACE/MMADE.
- Absolute risks
Remission at 5 years: 13% (extension), 10% (all original intervention with data; 7% of all 149), 5% (control). Serious adverse events 4.8 vs 10.2 per 100 person-years.
- Authors' conclusion
The extended intervention was associated with greater aggregated and absolute weight loss and suggested improved health status over 5 years; remission declined substantially with weight regain; better long-term weight maintenance strategies are needed.
6. Strengths
- Randomised origin with long (5-year) follow-up in routine primary care
- Objective outcomes (weight, HbA1c, prescriptions) from NHS records
- Transparent reporting of observed data and limitations
7. Weaknesses
- Extension group self-selected (engaged participants) - non-randomised comparison
- Large missing data, especially controls (45% without year-5 weight)
- Controls encouraged to lose weight after year 2 - not usual care
- Open-label; trial powered only for 1-year remission; many unadjusted comparisons
- Population restricted to early T2D (<6 years) not on insulin
- Intervention provider (Counterweight) linked to authors
8. Risk of bias
- Orienting tool
- RoB 2 (randomised comparison) and ROBINS-I (extension vs non-extension)
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- high
- Rationale
Single-reviewer structured judgement from the full text; formal assessment needs two independent reviewers. Changed from 'some_concerns' to 'high' because of substantial missing outcome data and the self-selected extension comparison documented in the full text. The randomised 2-year DiRECT results remain the stronger evidence.
| Domain | Judgement | Rationale |
|---|---|---|
| D1 randomisation process (original groups) | low | Cluster randomisation by practice in original DiRECT; baseline similar except age. |
| D2 deviations from intended interventions | some concerns | Open-label; controls received weight-loss encouragement after year 2; extension offered only to engaged participants. |
| D3 missing outcome data | high | Year-5 weight missing for 45% of controls and 22% of intervention; remission computed only for those with data. |
| D4 measurement of the outcome | low | Objective HbA1c, prescriptions and weights from records (some self-reported during COVID-19). |
| D5 selection of the reported result | some concerns | Many descriptive comparisons without multiplicity correction. |
| ROBINS-I: confounding/selection (extension comparison) | high | Extension vs non-extension compares self-selected responders with withdrawers (baseline HbA1c 57.8 vs 65.0; year-1 weight loss 12.1 vs 5.3 kg). |
Full-text review (25 October 2026): Population (T2D <6 years, not on insulin), intervention details (825-853 kcal formula diet), data availability, randomised 5-year comparisons (remission 10% vs 5%, HbA1c no difference), >10 kg subgroup, SAE and cancer counts added; funding (Diabetes UK) and author COI (Counterweight etc.) documented; bias overall some_concerns -> high (missing data, self-selected extension); directness -> indirect for C-DIAB-03.
9. Assessment of the statistical interpretation
Remission after intensive diet-induced weight loss is achievable but wanes with weight regain (62% at 1 year to 13% at 5 years in the most engaged group; 5-year difference vs control small, 10% vs 5%). Extension benefits are confounded by self-selection.
10. Relevance to the specific claim
Limits C-DIAB-03: remission (not cure) occurred only in early T2D not on insulin, via strict calorie restriction, and mostly relapsed by 5 years.
11. Independent contrary findings
12. Transferability to Germany and Europe
UK.
13. Overall methodological judgement
Shows that type 2 diabetes remission in early, non-insulin-treated disease is driven by substantial, sustained weight loss via a calorie-restricted formula diet, and that it is often not durable. Does not support 'cure' claims, especially not for long-standing insulin-treated diabetes or for a diet unrelated to calorie restriction.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | high |
| Precision of results | moderate |
| Directness | indirect |
| Reproducibility | not replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | moderate |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
None beyond those stated.
15. Linked fact checks
- Diabetes can be cured by following the Tayebat system without medication, illustrated by a patient who stopped insulin after 30 years of insulin use. – Studies with mixed or unclear results