Clinical trials, epidemiology, and public confidence

Area of Science:

• Biostatistics
• Research Integrity

Background:

• Growing concerns regarding exaggerated claims in clinical trials and epidemiological studies.
• Published research frequently exhibits poor methodological quality, particularly in statistical analysis.

Purpose of the Study:

• To address the issue of questionable research findings stemming from investigator bias in data analysis.
• To explore the impact of conflicts of interest on reported research outcomes.

Main Methods:

• Review of statistical literature on investigator outcome preferences and data analysis.
• Analysis of published papers to identify associations between financial conflicts of interest and reported results.

Main Results:

• Investigator outcome preferences, including financial interests and commitment to hypotheses, contribute to questionable research findings.
• Evidence suggests a link between financial conflicts of interest and the nature of reported research results.

Conclusions:

• Enhancing statistical review competence in clinical journals is necessary.
• Graduate education must prepare statisticians for surrounding conflicts of interest.
• Individual responsibility in employing unbiased analytic procedures is paramount for research credibility.