Discrepancies in sample size calculations and data analyses reported in randomised trials: comparison of publications with protocols

Mayo Clinic, Rochester, USA.
BMJ (online) (Impact Factor: 16.38). 02/2008; 337:a2299. DOI: 10.1136/bmj.a2299
Source: PubMed

ABSTRACT To evaluate how often sample size calculations and methods of statistical analysis are pre-specified or changed in randomised trials.
Retrospective cohort study. Data source Protocols and journal publications of published randomised parallel group trials initially approved in 1994-5 by the scientific-ethics committees for Copenhagen and Frederiksberg, Denmark (n=70).
Proportion of protocols and publications that did not provide key information about sample size calculations and statistical methods; proportion of trials with discrepancies between information presented in the protocol and the publication.
Only 11/62 trials described existing sample size calculations fully and consistently in both the protocol and the publication. The method of handling protocol deviations was described in 37 protocols and 43 publications. The method of handling missing data was described in 16 protocols and 49 publications. 39/49 protocols and 42/43 publications reported the statistical test used to analyse primary outcome measures. Unacknowledged discrepancies between protocols and publications were found for sample size calculations (18/34 trials), methods of handling protocol deviations (19/43) and missing data (39/49), primary outcome analyses (25/42), subgroup analyses (25/25), and adjusted analyses (23/28). Interim analyses were described in 13 protocols but mentioned in only five corresponding publications.
When reported in publications, sample size calculations and statistical methods were often explicitly discrepant with the protocol or not pre-specified. Such amendments were rarely acknowledged in the trial publication. The reliability of trial reports cannot be assessed without having access to the full protocols.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Previous studies investigating the association between adenomatous polyposis coli (APC) gene promoter methylation and colorectal cancer (CRC) have yielded conflicting results. The aim of this study was to comprehensively evaluate the potential application of the detection of APC promoter methylation to the prevention and treatment of CRC. PubMed, Embase, and MEDLINE (results updated to October 2014) were searched for relevant studies. The effect size was defined as the weighted odds ratio (OR), which was calculated using either the fixed-effects or random-effects model. Prespecified subgroup and sensitivity analyses were conducted to evaluate potential heterogeneity among the included studies. Nineteen studies comprising 2,426 participants were selected for our meta-analysis. The pooled results of nine studies comprising a total of 740 subjects indicated that APC promoter methylation was significantly associated with CRC risk (pooled OR 5.53; 95% confidence interval [CI] 3.50-8.76; P<0.01). Eleven studies with a total of 1,219 patients evaluated the association between APC promoter methylation and the presence of CRC metastasis, and the pooled OR was 0.80 (95% CI 0.44-1.46; P=0.47). A meta-analysis conducted with four studies with a total of 467 patients found no significant correlation between APC promoter methylation and the presence of colorectal adenoma (pooled OR 1.85; 95% CI 0.67-5.10; P=0.23). No significant correlation between APC promoter methylation and patients' Dukes' stage, TNM stage, differentiation grade, age, or sex was identified. In conclusion, APC promoter methylation was found to be significantly associated with a higher risk of developing CRC. The findings indicate that APC promoter methylation may be a potential biomarker for the carcinogenesis of CRC.
    OncoTargets and Therapy 01/2015; 8:211-22. DOI:10.2147/OTT.S75827 · 1.34 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Outcome reporting bias has received widespread recognition and been considered to pose two threats to the validity of clinical decision making because they overestimate the effect of treatments or distort the results of trials. However, the problem of outcome-reporting bias has not been systematically studied among randomized clinical trials of acupuncture. Our objectives were to evaluate the consistency between the registered records and subsequent publications with respect to outcomes and other data as well as to determine whether outcome-reporting bias favors significant primary outcomes.MethodsA systematic search of 15 registries was conducted from their inception to January 2014 to identify randomized clinical trials on acupuncture for which the status was listed as `completed.¿ The subsequent publications were retrieved by searching PubMed and three Chinese databases. Basic characteristics and the registration information were extracted from the registered records and publications. We performed comparisons regarding primary outcomes and other data between the registered records and subsequent publications to assess the consistency and selective outcome reporting.ResultsEighty-eight trials on acupuncture with 96 published reports were identified. Only 19.3% (17/88) were registered before the start of the trial, suggesting prospective registration. The trial registration number was unavailable in 36 published reports (37.5%). A comparison of registered and published primary outcomes could be conducted in 71 publications (74.0%), and the inconsistency of the primary outcomes was identified in 45.1% (32 of 71); 71.4% (15 of 21) had a discrepancy that favored statistically significant primary outcomes, while 28.6% (6 of 21) favored nonsignificant primary outcomes. Furthermore, the other inconsistencies between the registry records and subsequent publications involved the inclusion criteria (54.7%), exclusion criteria (47.9%) and controls (22.9%).Conclusions We find that prospective registration for randomized clinical trials on acupuncture is insufficient, selective outcome reporting is prevalent, and the change of primary outcomes is intended to favor statistical significance. These discrepancies in outcome reporting may lead to biased and misleading results of randomized clinical trials on acupuncture. To ensure publication of reliable and unbiased results, further promotion and implementation of trial registration are still needed.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Qualitative evidence suggests patient-reported outcome (PRO) information is frequently absent from clinical trial protocols, potentially leading to inconsistent PRO data collection and risking bias. Direct evidence regarding PRO trial protocol content is lacking. The aim of this study was to systematically evaluate the PRO-specific content of UK National Institute for Health Research (NIHR) Health Technology Assessment (HTA) programme trial protocols.
    PLoS ONE 10/2014; 9(10):e110229. DOI:10.1371/journal.pone.0110229 · 3.53 Impact Factor

Full-text (2 Sources)

Available from
May 28, 2014