Public trust in nutrition science has waned over time in several countries (1, 2). Professional bodies persevere to reverse the trend or earn trust by improving rigor in research conduct and dissemination (3–5). While several biases in research are cataloged and remedial measures suggested, “the investigator bias” arising from investigators themselves, poses a major challenge. Investigator bias rarely receives mention in academic literature. There is currently no universally accepted definition or assessment for investigator bias. In 2003, Ernst and Canter in their short letter on investigator bias and false positive findings defined it as a “bias resulting from a conflict of interest arising from passionate beliefs held by the investigators” (6). We define “investigator bias” as acts of omission or commission that arise from ignorance, hubris or excessive attachment to beliefs, based on existing discourse (7, 8). In our viewpoint, we do not focus on research misconduct such as data falsification or unethical publication practices; as their perils are well-recognized and clear-cut mechanisms exist to prevent, and redress these concerns. Without a clear framework to either identify or assess investigator bias, it is unsurprising that there is a paucity of empirical evidence on the extent of its prevalence, or its impact or more importantly to address it.

The lack of formal evaluation of investigator bias is concerning given that investigator bias encroaches into one or more stages in the research process including study conception, design, conduct, analysis, interpretation and dissemination, challenging research credibility (6). It influences what research questions are posed, how hypotheses are tested and interpreted (6, 7). Such positional biases in reviewers and funders can have consequences for the field, by determining what is rewarded and perpetuated. To tackle bias, it would be imperative to understand what drives it. Opportunities and career advancement in academia and research are increasingly competitive and are linked to the ability to attract funding and publish in indexed journals with high impact. Funders and institutions are predominantly defined by “normal” science characterized by its stability, driven by set precedents, and with a tendency to lean toward the wisdom of accepted figures in the field (9). Therefore, questioning current practices and paradigms may be problematic and unproductive for researchers. Additionally, promotional language in grant applications is known to increase their chance of being funded and the citation impact of resulting publications from these grants (10).

Therefore, it is possible that grant applications or publications apart from reflecting the investigators' beliefs and hypotheses, may also be catering to what funders, journals or reviewers are known to value. It is important to note that it is impossible to personally know all authors to sufficiently discriminate between playing to the gallery and reflecting inherent biases in grants and publications. Thus, we are bound to interpret what is written and reported as a proxy for an author's personal biases. When biased grant applications or publications are successful, the success in itself reinforces current behavior in investigators, creating a feedback loop. Thus, current systems and practices lead to the perpetuation of the practice of investigator bias and its consequences.

Without a global understanding of investigator bias, it is impossible to confidently describe or quantify its presence and impact in nutrition. However, the discussion on dissonance between the cultural basis of food and the Eurocentric values of nutrition science (the formal study of food) over a decade ago (11), makes one ponder if investigator bias can affect nutrition in more profound ways than it does to other areas of science. As if in agreement, in 2018, Ionnadis and Trepanowski published a highly provocative opinion piece. The opening lines of the paper read: “nutrition research is among the most contentious fields of science” and the authors argued that non-financial conflicts of interest in nutrition were beyond “allegiance bias and preference for favorite theories” that exist in any field of study (12). The authors proposed that nutrition scientists faced additional challenges that arose from making personal dietary choices that were shaped by familial, cultural, or religious norms. The controversial perspective evoked passionate response from nutrition scientists. A counterpoint totally dismissed these notions (13). The Editorial board of the European Journal of Nutrition responded with an editorial that highlighted the “need for self-reflection” in scientists funded by the industry and those “seeking to prove their favored hypothesis” (14). Thus there exists some consensus that investigator bias in nutrition can misinform dietary guidelines, create disagreements among experts in the public domain, and further erode trust in nutrition (15) (Figure 1). Thus, it is both urgent and important to tackle this bias from the public health perspective. A potential first step to negate investigator bias and to effect change is to create awareness among all stakeholders. As investigator bias rarely features in current scholarly discussions in nutrition, we aim to create awareness, by providing a few field-specific examples of investigator bias and identify or propose some measures to address it (Table 1). The aim of this perspective piece is not to be an exhaustive reference of examples of investigator bias and recommended remedial measures. Through this piece, we aim to begin a dialogue with stakeholders of nutrition research (which includes funders, ethics committees, academics, scientists, trainees, practitioners, and consumers), to create awareness, and galvanize efforts to appropriately tackle the bias.

Since we are but human, we are all touched by ignorance, hubris and beliefs, and therefore are not immune to investigator bias. Established expertise by default does not invite or encourage criticism. Therefore, the current focus of the research ecosystem on financial interests to address conflict of interest, with no clear mandate on managing non-financial interests (36), is inadequate to deal with the bias that can arise from investigators themselves and support good science. Moreover, reporting non-financial conflicts predominantly occurs after study completion, and during publication. To prevent the colossal waste of resources, attempts to identify investigator bias are needed earlier and throughout the research process.

Despite acknowledgment of investigator bias in nutrition, the existing nihilism in certain sections is defeatist and has prevented progress in addressing the bias. As a way forward, we propose the following suggestions. First, in the absence of punitive measures for the consequences of errors in nutrition research and practice, the first line of defense against investigator bias is self and peer regulation. Thus, the earlier we choose to acknowledge and sensitize ourselves to this bias that arises from within, the better prepared we are to prevent ourselves and our peers from falling prey to it. As a starting point for these discussions and to create awareness and encourage introspection and reflection within the nutrition research community, we have collated a few existing safeguards, suggested the repurposing of a few others or propose a few other aligned with steps in the research process in Table 1. Reflective practices, we believe, can result in a more thorough evaluation of the bias, and further deliberation on whether formal processes to recommend best practices to deal with the bias (such as developing assessment tools or checklists) are necessary.

Second, systematic and formalized collective efforts to define and catalog investigator bias and identify where the bias arises from are important, warranting dedicated task forces established by professional societies for the purpose. These efforts should preferably translate, as described earlier, to developing assessment tools to identify the bias, and therefore help quantify its presence and further enable studying trends over time. The task force could also deliberate and recommend appropriate ways to report positional biases (such as those arising from dietary preferences) in authors, reviewers, and editors. While mandating authors to declare conflicts of interest has received major attention in recent decades, it is also important to acknowledge and manage reviewer and editorial bias as their actions are definitive and determine how science is shaped.

Thirdly, a conscious approach to building tolerance to alternative views and improving diversity within research teams, Institutional Review Boards (IRB), editorial boards and reviewer panels should be encouraged at personal and institutional levels. Several of the issues discussed above, would have been prevented if red teaming/adversarial collaborative approaches were practiced (37, 38).

Finally, and importantly, being prescriptive or improving declaration of conflicts of interest has never been the panacea to eradicating biases. Therefore, emphasizing reflection on its occurrence and emphasizing this in training of younger generation of scientists may be the best bet yet.