Question explored with the scientific record
Describe the methology in the paper "Acetic acid chromoendoscopy for detection of neoplastic Barrett’s oesophagus: stepped wedge cluster randomised controlled trial (ACID-trial)"
The ACID-trial's methods are not in this retrieval. What I can describe is the stepped-wedge design it uses, and what that design can and cannot prove.
The question asks for the methodology of one specific paper. None of the retrieved records is that paper. What the records do cover is the stepped-wedge cluster randomized trial as a design, and that is what I can honestly describe [1][2][4].
In a stepped-wedge trial, whole sites or groups (clusters) all start in the control condition, then switch to the intervention in a randomized sequence over several time periods, so every cluster eventually gets treated [2][3][6]. The ACID-trial name implies this shape: endoscopy units switching to acetic acid chromoendoscopy in random order. The design is chosen when you cannot randomize individuals, only whole units [8].
The catch is statistical. Time itself can change outcomes, and adjusting for that time trend reduces power [1]. The widely used power formula overestimates real power, so these trials are often underpowered [4]. Small numbers of clusters make the analysis fragile, and the choice of correction method changes whether a result looks significant [5][7].
| Design feature | What it means for the reader |
|---|---|
| Clusters, not patients, randomized | Few independent units, so chance can drive results [1] |
| Time trend adjustment | Can cut statistical power [1] |
| Power formula | Commonly overestimates true power [4] |
| Small cluster counts | Results sensitive to analysis choice [5][7] |
The outcome matters more than the design. If ACID-trial measured detection of dysplasia or cancer at biopsy, that is a real clinical endpoint, not a lab surrogate. I cannot confirm from these records what its primary outcome was.
My call: I can describe the stepped-wedge method and its weaknesses, but I cannot describe the ACID-trial's actual protocol, sample size, or endpoints from this retrieval. Confidence: low, because the paper itself is absent.
Sources used 8
-
Stepped wedge cluster randomised trials: a review of the statistical methodology used and available
This systematic review analyzes the statistical methods and sample size considerations used in stepped wedge cluster randomised trials, revealing widespread use of GLMMs and GEEs, frequent underpowering due to small numbers of clusters, and a need for clearer distinctions and im…
DOI: 10.1186/s12874-016-0176-5 -
Implementing and evaluating visual key information pages compared to text-based key information: protocol for a stepped-wedge cluster randomized trial
Evaluating a health-literacy–driven visual key information page for informed consent across three US centers using a stepped-wedge cluster randomized design to assess participant understanding, satisfaction, decisional confidence, and coordinators' readiness to adopt the approac…
DOI: 10.1186/s12910-025-01341-y -
Effect of immediate kangaroo mother care (iKMC) on neonatal mortality and culture-positive sepsis in low-birth-weight neonates in district hospitals in Chhattisgarh, India (PRISM study): protocol for a stepped-wedge cluster randomized trial
This protocol outlines a stepped-wedge cluster randomized trial (PRISM) to assess whether immediate kangaroo mother care initiated within 12 hours of birth reduces all-cause neonatal mortality and culture-positive sepsis among 1000–1799 g birth-weight neonates admitted to distri…
DOI: 10.1186/s13063-025-09083-3 -
Sample size determinations for stepped-wedge clinical trials from a three-level data hierarchy perspective
This methodological study derives and simulates power functions for stepped wedge cluster randomized trials with three-level data, showing that second-level correlations must be accounted for and that the widely used HH power formula overestimates empirical power.
DOI: 10.1177/0962280216632564 -
Maintaining the validity of inference in small‐sample stepped wedge cluster randomized trials with binary outcomes when using generalized estimating equations
This study evaluates how to preserve valid inference in small-sample stepped-wedge cluster randomized trials with binary outcomes by comparing several small-sample bias corrections and degrees-of-freedom adjustments within generalized estimating equations, through extensive simu…
DOI: 10.1002/sim.8575 -
Supporting Policy In health with Research: an Intervention Trial (SPIRIT)—protocol for a stepped wedge trial
Protocol for SPIRIT: a 30-month stepped-wedge cluster randomized trial evaluating a multifaceted program to increase health policy agencies' use of research evidence, with objective (SAGE) and self-reported (SEER) outcomes, plus process evaluation and cost analysis.
DOI: 10.1136/bmjopen-2014-005293 -
Comparison of small-sample standard-error corrections for generalised estimating equations in stepped wedge cluster randomised trials with a binary outcome: A simulation study
Extensive simulations compare small-sample standard-error corrections for generalized estimating equations in stepped-wedge cluster randomized trials with a binary outcome, finding KC and FG with independent working correlation provide near-nominal coverage across most scenarios…
DOI: 10.1177/0962280220958735 -
Evaluation of injury prevention interventions using the stepped wedge cluster randomised trial design: key considerations
Seven methodological challenges associated with stepped-wedge cluster randomized trials in injury prevention are highlighted, with mitigation strategies illustrated by the prehospital Canadian C-Spine trial.
DOI: 10.1136/ip-2025-045788