Reply to JJ Meerpohl et al.

Dear Editor:

The statements “[…] that any potential bias, inconsistency, indirectness, imprecision or reliance on study type other than randomized trials can result in downgrading of the quality of evidence means that formally identifying effects which are regarded as important and based on high quality evidence using the GRADE system may be unattainable in the context of nutritional determinants of chronic disease […]” and that “This needs to be taken into account when developing nutritional recommendations” (1), which were published in a highly cited meta-analysis in the British Medical Journal in 2013, highlight the need for developing and evaluating new tools to assess the quality of meta-evidence in the field of nutrition research. In conceiving our article, “NutriGrade: A Scoring System to Assess and Judge the Meta-Evidence of Randomized Controlled Trials and Cohort Studies in Nutrition Research” (2), we reasoned that the development of a scoring or grading system to assess the quality of evidence in nutrition research represents a scientific contribution that is part of a continuous process to improve global efforts in an important field of public health. With regard to developing criteria for a grading system of meta-evidence in the field of nutrition research, our concept differs from the well-established and widely used Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach in the following aspects:

• Better balanced evaluation of prospective observational and intervention study designs

• Assessment of nutrition-specific aspects, such as dietary assessment methods and their validation, calibration of FFQs, or the assessment of diet-associated biomarkers

• Consideration of the conflict of interest and funding bias as a separate item

• Introduction of a scoring system

In our article (2), we outlined the rationale for each of our decisions, including those criteria for which GRADE has a different view. The different views may be explained by the distinct perspectives of the groups. Our group is mainly composed of scientists with expertise in the field of nutrition, whereas GRADE is historically composed of mostly clinical research scientists. Other scientists from related disciplines have already found that processing evidence in the clinical research compared with the public health research areas follows slightly different approaches. For example, the US government established task forces for both evidence in clinical settings as well as evidence in public health (3). We feel that there is still a need for a scientific debate as to whether both areas could be combined in a unified system that is not dominated by one perspective.

In their letter, Meerpohl et al. “encourage[d] the authors of this article and interested readers to further explore how GRADE works and to join in advancing the methods in a unified approach.” We acknowledge this kind offer from the GRADE group to open up the process of developing timely solutions, particularly in areas in which the groups have different views and also in light of the recent exponential increase in the number of meta-analyses published in the field of nutrition research. In this way, we also appreciate the actual invitation of a scientific debate and potential collaboration with GRADE as proposed by Meerpohl et al., which also includes a wider discussion of our own positions. Overall, NutriGrade should not be considered as a competitor to GRADE but rather as an approach suggested by nutrition specialists that adapts the GRADE philosophy of assessing the trustworthiness of evidence for publications to an area beyond clinical research and with specific needs. In conclusion, we regard the letter by Meerpohl et al. as the starting point for consensual discussions on different views in the same area of interest, with a clear perspective of further refining the grading approach for meta-analyses in the field of nutrition research.