Measuring Design, Reporting and Funding Bias in Nonclinical Research
Location(s): United States
Minimizing bias in the design, conduct, and reporting of research is essential to research integrity, the topic of RFA-ES-11-004. When biased research outcomes are incorporated into decisions about use of chemicals or drugs, incorrect conclusions can have important implications for health. Evidence shows a strong association between industry funding, investigator financial conflicts of interest (COIs), and biased outcomes in clinical research, but there is little evidence regarding the influence of these COIs on the integrity of nonclinical research, including laboratory and animal studies. Discovering how financial COIs are related to bias in design and reporting of nonclinical research should lead to more methodologically sound studies and results that provide better protection of humans from exposure to drugs tested in clinical trials or chemicals. The long-term goal of the proposed project is to develop an evidence base to inform the development of COI policies for nonclinical research. The objective is to examine the association of research funding sources and financial conflicts of interest of investigators with the design and reporting of nonclinical studies. Aim 1: Determine whether industry research sponsorship and / or principal investigator financial ties are associated with methodological biases or publication of biased outcomes in animal studies. We will conduct a systematic review of methodological quality assessment tools for animal studies. The methodological criteria identified will be applied to 4 cohorts of published animal research to examine the association of research funding source, investigator financial ties, and methodological quality with the direction and effect size of the published research outcomes. Our working hypothesis is that industry funded investigators and those with personal financial ties to their research sponsors are more likely to publish research with outcomes that favor their sponsors than researchers without these ties, even when controlling for other study design features. Aim 2: Assess the extent of reporting biases (failure to publish entire studies or selective outcome reporting) in a sample of animal studies and determine the association of reporting biases with research sponsorship, investigator financial ties, and direction and effect size of results. We will identify a cohort of investigators conducting animal research and obtain information on their funding sources (private and federal) and personal financial ties from their publicly available financial disclosure forms. We will measure the association of type of funding source and financial ties (which may or may not be disclosed in publications) with results reported in their publications. We will also interview the cohort of investigators to identify unpublished data, if any, and reasons for failure to publish. Our working hypothesis that investigators with industry funding for their research or personal financial ties to their sponsors are more likely to selectively report research outcomes that favor their industry sponsor than researchers without these ties. Our expected outcome is to inform policies for reducing bias in nonclinical research. Controlling bias in the design, conduct, and reporting of research is essential to research integrity, the topic of this RFP. The development of appropriate policies to reduce bias and the influence of conflicts of interest in animal studies could result in the protection of humans from toxic compounds tested in animals. Nonclinical studies are also a critical component of translational research, so policies to reduce bias and influence of conflicts of interest should lead to the initiation of appropriate clinical trials that are an efficient use of resources and minimize risk to humans.