# Oral Adjuvant Curcumin Therapy for Attaining Clinical Remission in Ulcerative Colitis: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.

- **DOI:** [10.3390/nu10111737](https://doi.org/10.3390/nu10111737)
- **Rating:** Sound (3.4 out of 5)
- **Verdict:** A methodologically careful curcumin meta-analysis whose conservative rare-event correction usefully tempers enthusiasm, constrained by an evidence base of only three small, heterogeneous trials.
- **Framework version:** v5.0
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-06-15
- **Evaluation page:** https://www.nabu.science/paper/10.3390%2Fnu10111737

## Strengths

- **Conservative rare-event method drives conclusion** Supplementing the Mantel-Haenszel model with a beta-binomial random effects model for zero-event cells correctly dissolves an apparent per-protocol significance, and the conclusion follows the more conservative result rather than the superficially positive one.
  - Dimension: methodological rigour
  - Location: Methods §2.6; Results §3.3
- **Transparent, well-justified trial exclusion** The exclusion of the Banerjee trial is documented with specific registry-versus-results discrepancies in intervention duration, dose, and outcome measures, demonstrating reproducible quality control.
  - Dimension: reporting
  - Location: Results §3.2
- **Honest engagement with competing evidence** The discussion directly compares its findings against a concurrent meta-analysis and identifies specific methodological differences, including that study's pooling of enema trials and failure to address rare-event false significance.
  - Dimension: positioning
  - Location: Discussion §4

## Limitations

- **Thin and heterogeneous evidence base** The meta-analysis rests on only three small RCTs with high heterogeneity (I2=73% for ITT) and heterogeneous dosing from 450 mg to 3 g daily, precluding subgroup analyses and definitive conclusions.
  - Dimension: contribution
  - Location: Results §3.3; Discussion §4
- **Incomplete reporting against 2019 standards** Risk-of-bias terminology mixes RoB 2.0 with older high/unclear/low categories, no outcome-level GRADE certainty summary is presented, and the post-protocol search expansion is described narratively without a reproducible updated strategy.
  - Dimension: methodological rigour
  - Location: Methods §2.5; Discussion §4
- **No sensitivity analysis for key exclusion** Because excluding the Banerjee trial materially shapes the pooled estimates, the absence of a sensitivity analysis including and excluding it under explicit rules weakens the robustness of the synthesis.
  - Dimension: methodological rigour
  - Location: Results §3.2

## Reliability checks

Caveats (3 of 4 checks).

- **Internal coherence:** Noted. The synthesis is largely internally consistent across abstract, methods, and results, with consistent OR/CI values and transparent documentation of the Banerjee exclusion. Two reporting watch-outs appear: the risk-of-bias method names RoB 2.0 but applies older high/unclear/low risk terminology, and the post-protocol search expansion is described narratively without a reproducible updated search strategy.
- **Research conduct:** Noted. Protocol registration (PROSPERO CRD42018098996) and duplicate screening/extraction support basic review conduct. No conflict-of-interest or funding statement is visible in the supplied text, and the disclosed protocol deviation is not fully operationalised in an updated methods record.
- **Reference integrity:** Pending

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
