Treatment allocation in ophthalmological randomised-control trials (TAO-RCT): A cross-sectional meta-research study
摘要
Ophthalmological randomised-control trials (RCTs) are complicated by inter-eye correlation, contralateral crossover effects, and heterogeneity in treatment allocation patterns. This study examined the prevalence of uncontrolled treatment allocation and its impact on outcomes in ophthalmological RCTs.
MethodsAll ophthalmological RCTs published in 2022 were analysed in a cross-sectional study that adhered to a preregistered protocol (CRD42023474661). Eligible trials were assessed for treatment allocation patterns and clarity of reporting using a simple nomenclature system. Associations between uncontrolled allocation and study characteristics, reporting clarity, funding, and trial outcomes were examined.
ResultsFrom 359 RCTs, 42 distinct allocation patterns were identified, with 306 trials (85.2%) using controlled treatment allocation. Uncontrolled treatment allocation was associated with unclear reporting (χ2 = 44.7, p < 0.001) and two-eye allocation patterns (Fisher’s exact test, p < 0.001). The distribution of p-values was similar between controlled and uncontrolled trials (t = −0.603, p = 0.547), suggesting no increased likelihood of statistical significance. Uncontrolled allocation was more frequent in non-English-speaking countries (χ2 = 4.681, p = 0.030) and studies of surgical interventions (χ2 = 4.287, p = 0.038).
ConclusionUncontrolled treatment allocation is prevalent in ophthalmological RCTs and is associated with unclear reporting, two-eye study designs, non-English-speaking settings, and surgical interventions. While no signs of deliberate misuse are evident, these patterns may compromise the validity of trial analysis. Transparent reporting and careful consideration of treatment allocation should be prioritised in trial design, protocol registries, and appraisal frameworks. A standardised nomenclature system could improve clarity and reproducibility.