Related Experiment Videos
Confronting publication bias: a cohort design for meta-analysis
Statistics in Medicine
|January 1, 1987
Summary
Publication bias in clinical trials can skew results. This study proposes an international clinical trial registry to enable unbiased pooled analysis, improving therapeutic effectiveness evaluations, as demonstrated in ovarian cancer chemotherapy research.
Area of Science:
- Clinical research methodology
- Biostatistics
- Medical informatics
Background:
- Published clinical trial literature may exhibit publication bias, favoring positive results, potentially leading to inaccurate conclusions about therapeutic effectiveness.
- Publication bias is amplified in pooled analyses, where minor treatment differences can achieve statistical significance.
- Existing literature reviews may not adequately address the impact of publication bias on therapeutic evaluations.
Purpose of the Study:
- To develop and present a model for pooling clinical trial results that mitigates publication bias.
- To propose the establishment of an international registry for all clinical trials with clearly defined objectives and endpoints.
- To enable researchers to select cohorts of clinical trials independently of their results for unbiased analysis.
Main Methods:
- Development of a novel model for pooling clinical trial data, designed to be free from publication bias.
- Utilizing the International Cancer Research Data Bank (ICRDB) registry as a case study for cancer clinical trials.
- Applying the model to evaluate the effect of chemotherapy on survival in advanced ovarian cancer using registered trials.
Main Results:
- A pooled analysis of registered trials for ovarian cancer chemotherapy yielded different conclusions compared to traditional reviews of published trials.
- The proposed registry-based approach allows for the selection of trials independent of their outcomes, thus avoiding bias.
- Demonstrated significant differences in therapeutic effectiveness conclusions when publication bias is addressed.
Conclusions:
- Establishing an international clinical trial registry is crucial for unbiased therapeutic evaluations.
- The proposed model effectively addresses publication bias in pooled analyses of clinical trial data.
- Implementing such a registry can significantly improve the reliability of conclusions regarding treatment efficacy.